MIA PaCa-2 and PANC-1 human PDAC cell lines were purchased by the American Type Culture Collection (ATCC) and maintained at 37°C in a 5% CO₂ humidified atmosphere, using Dulbecco’s Minimum Essential Medium (DMEM) (ECM0728L; Euroclone) supplemented with 10% fetal bovine serum (FBS) (ECS0180L; Euroclone) and 1% penicillin/streptomycin (ECB3001D; Euroclone) as culture medium. Typically, cells were equally seeded and kept under standard growing conditions for 24 h. The following day, AdipoR and Gem were supplemented to fresh media, either individually or in combination, and PDAC cells were incubated for times and concentrations provided in each experimental condition. Ultimately, adherent cells were trypsinized and collected with potential floating ones, before being centrifuged for 5 min at 1,500 RPM. Since AdipoRon and gemcitabine were dissolved in DMSO and H₂O, respectively, an equal solvent rate (% v/v) was used as a negative control.

Chemicals: AdipoRon (#SML0998; Sigma-Aldrich), gemcitabine (#G6423, Sigma-Aldrich), trypan blue (#T8154; Sigma-Aldrich), propidium iodide (#P4864; Sigma-Aldrich), crystal violet (#C0775; Sigma-Aldrich), PD98059 (#P215; Sigma-Aldrich), dimethyl sulfoxide (DMSO) (A3672; AppliChem), and ethanol absolute anhydrous (308603; Carlo Erba). Antibodies: α-Tubulin (#3873; Cell Signaling Technology), cyclin E1 (#4129; Cell Signaling Technology), p44/42 MAPK (#9102; Cell Signaling Technology), phospho-p44/42 MAPK (#9101; Cell Signaling Technology), cyclin A1 (sc-751; Santa Cruz Biotechnology), vinculin (sc-73614; Santa Cruz Biotechnology), and p27^KIP1 (ab3203; Abcam).

A total number of 8 × 10⁴ MIA PaCa-2 and 1 × 10⁵ PANC-1 cells were moved in 6-well plates and kept in a standard growing state for 24 h. AdipoR and Gem, either alone or in combination, were subsequently added to new media and allowed to act in PDAC cells. For each experiment, times and concentrations are indicated in the Results section and Figure legends. Usually, pelleted cells were resuspended in 1.5 ml DMEM and diluted 1:1 with trypan blue, which, crossing damaged membrane, discriminates living from dead cells. Specifically, 10 μl of both media containing cells and blue dye (0.4%, v/v) were mixed, and the relative cell content was established using a Bürker chamber, where the number of unstained (living) and stained (dead) cells was recorded. Each point has been counted at least twice in each experimental procedure.

Cytometric analysis was performed to define the respective cell phase distribution in reaction to different stimuli. A procedure similar to that described in point 2.3 was applied to seed, treat, and collect PDAC cells. Subsequently, pelleted samples were resuspended first in 300 μl PBS (ECB4004lL; Euroclone) and then in 700 μl ice-cold absolute ethanol. Fixed cells were stored at −20°C until analysis. Before investigation, the samples were spun down for 5 min at 1,500 RPM and incubated with PI staining solution containing 15 μg/ml PI and 20 μg RNase A (R5503; Sigma-Aldrich) in PBS for 10 min at room temperature in the dark side. For each experimental condition, at least 20,000 events were acquired and analyzed by FACS-Celesta (BD Biosciences).

PDAC cells were seeded in 6-well plates at a density of 1.5 × 10³ per well (MIA PaCa-2 and MIA PaCa-2 RES) or 2 × 10³ (PANC-1) and exposed to different times and concentrations of AdipoR, Gem, and combination (see Results for more details). At the established endpoint, media was discarded, and newly formed colonies were stained with crystal violet solution (1% aqueous solution) for 10 min. The staining solution was later removed, and wells were washed several times in distillate water. Colonies have been allowed to air dry naturally and acquired by photographic equipment. Quantification analysis has been performed by determining the optical density (OD) of dissolved colony-bound crystal violet staining in 10% acetic acid at 590 nm by an Infinite 200 PRO Microplate Reader (Tecan Life Sciences).

Depending on the target protein, an amount of 10–30 μg of total extracts was loaded and separated by sodium dodecyl sulfate–polyacrylamide gel electrophoresis (SDS-PAGE) for each sample. Subsequently, sample proteins were transferred onto nitrocellulose membranes (GEH10600008; Amersham) by the Mini Trans-Blot system (Bio-Rad Laboratories). After washing in tris-buffered saline (TBS) supplemented with 0.05% Tween 20 (TC287; HIMEDIA), films were blocked 1 h in 5% no-fat dry milk (A0530; AppliChem) aimed at covering potentially free spots into the nitrocellulose membrane. Incubation overnight at 4°C has been chosen for primary antibody binding. In the following days, horseradish peroxidase (HRP)-conjugated secondary antibodies, reacting against the related primary species, were applied to the membrane for 1 h at room temperature. Each incubation step was preceded and followed by three 5-min rinses in T-TBS. Finally, protein-related light signals were acquired by ChemiDoc™ (Bio-Rad Laboratories) using the enhanced luminol-based chemiluminescent substrate (E-IR-R301; Elabscience) as a detection system for HRP.

A number of 4.8 × 10⁵ (MIA PaCa-2) or 6 × 10⁵ (PANC-1) cells were plated in 100 mm plates and left free to attach for 24 h. In the next day, media was replaced with a fresh one containing AdipoR, Gem, and the combination in doses and timelines reported in the Results section, and Figure legends. At every experimental point, cells were collected and spun down at 1,500 RPM for 5 min. Pellets were later resuspended in 3–5 volumes of RIPA buffer (R0278; Sigma-Aldrich) supplemented with protease and phosphatase inhibitors cocktail (#5872; Cell Signaling Technology). After 30 min, samples were further centrifuged at 14,000 RPM for 15 min at 4°C, and the supernatant was recovered and assessed for the relative protein content by Bradford Assay (39222; SERVA). Protein samples were first mixed 1:1 with Laemmli 2× (S3401; Sigma-Aldrich) and later boiled at 95°C for 6 min.

MIA PaCa-2 cells were chronically exposed to increasing Gem concentration over a period of 4 months. Specifically, starting from 1 nM, cells were cultured in media containing Gem until they grew steadily. A higher cumulative Gem dosage was subsequently applied, and the resistant procedure was repeated as long as a final concentration of 200 nM was reached. At each step, cells were amplified, harvested, and cryopreserved in liquid nitrogen or an ultralow-temperature freezer. The obtained MIA PaCa-2 Gem-resistant cells were finally cultured in drug-free medium for up to 2 weeks before performing the reported experiments.

Results are indicated as average value ± SD of biological independent replicates. Significance has been defined using either Student’s t-test, to compare the mean of two samples, or analysis of variance (ANOVA) followed by Turkey’s test, to discriminate differences between more than two experimental groups. In both cases, values of less than 0.05 were recognized as significant. Densitometric analyses have been carried out by ImageJ (NIH, Bethesda).

Recently, two different studies have reported the AdipoR ability in suppressing tumor growth in PDAC (Messaggio et al., 2017; Akimoto et al., 2018). In order to extend and corroborate these findings, herein we first established the AdipoR impact in two distinct human PDAC cell lines, namely, MIA PaCa-2 and PANC-1.

In agreement with the previously published results, AdipoR exposure induced a remarkable cell growth decrease in PDAC cells, almost in a dose-dependent manner, without substantial differences between MIA PaCa-2 and PANC-1 cell types (Figure 1A).

Choosing 10 µg/ml as a subsequent effective working dosage, time course experiments up to 72 h showed a near time dependency in MIA PaCa-2, where a cell number decrease of 20, 42, and 57 percent was recorded at 24, 48, and 72 h, respectively (Figure 1B). A different trend was obtained in PANC-1, in which no considerable responsiveness to AdipoR was observed at 24 h (Figure 1B).

Notably, AdipoR-mediated antiproliferative properties were supported by an increase in the G0/G1 phase and a concomitant decrease of both S and G2/M phases in MIA PaCa-2 (Figure 1C). Precisely, the cell amount in G0/G1 moved from 35 to 48% after 24 h of treatment with 10 µg/ml AdipoR, while both S and G2/M phases diminished approximately 6%, concurrently.

Interestingly, although 10 µg/ml AdipoR was not effective in impacting PANC-1 cell growth after 24 h, changes in cell cycle distribution were detected. Similar to MIA PaCa-2, AdipoR provoked a G0/G1 intensification and an S-phase depletion in PANC-1, albeit in this latter cell model the magnitude was less sharp. Conversely, no G2/M involvement seems to occur in AdipoR-treated PANC-1 cells (Figure 1C). Overall, these findings further recognize AdipoR as an antiproliferative compound in PDAC and support its peculiarity in slowing down cell cycle progression.

Before exploring the consequences of the combination treatment AdipoR plus Gem in PDAC models, we preliminarily addressed the Gem-mediated cell growth impact on both employed cells.

Evaluating a wide concentration range, Figure 1D displays a different aptitude in reacting to Gem between MIA PaCa-2 and PANC-1. While MIA PaCa-2 showed great responsiveness to Gem already at very low concentration, the PANC-1 ability in resisting Gem was further confirmed when high dosages were applied. Exposing MIA PaCa-2 to 50 or 100 nM Gem for 48 h, for instance, nearly affected the totality of the cells, differently from PANC-1, in which the inhibition rate was roughly 40 and 60%, respectively. Taken together, these results remark an effective yet different Gem sensitivity between the examined PDAC cells.

With the purpose of addressing potential cooperating effects in PDAC models, we subsequently combined effective concentrations of both AdipoR and Gem in a constant dilution ratio, and the relative outcome in cell growth was later assessed.

Specifically, three different doses of both AdipoR (5, 10, and 20 µg/ml) and Gem (7.5, 15, and 30 nM) were employed in MIA PaCa-2, exhibiting a clear dose dependency (Figure 2A). But even more interestingly, the concomitant use of AdipoR plus Gem further counteracted MIA PaCa-2 cell proliferation, suggesting a positive interplay between these two compounds. Compared with 5 µg/ml AdipoR and 7.5 nM Gem, combination treatment improved single outcomes by nearly 33% and 20%, respectively. This tendency became even more pronounced at the highest tested doses, raising inhibition values of 47 and 34% versus AdipoR and Gem, individually (Figure 2A).

The different Gem responsiveness has required the use of higher concentrations in PANC-1 (25, 50, and 100 nM), while no changes in AdipoR doses were applied. In line with MIA PaCa-2 results, even in PANC-1, all three tested mixtures enhanced the anticancer effects of single treatments (Figure 2B). Minimal fluctuations were observed in response to the increasing combinations in PANC-1.

CompuSyn analysis was subsequently performed with the purpose of defining drug-drug interaction and the relative combination index (CI). Plotting dose-effect curves of both single and combination agents, the Chou-Talalay method discriminates among additive (CI = 1), synergism (CI < 1), and antagonism (CI > 1) effects, using the median-effect equation (Chou, 2010). The MIA PaCa-2 Fa-CI plot revealed a robust synergistic action already at very low concentrations, maintaining a constant trend even when combination affected 90% of cells (Figure 2C). Albeit in all tested conditions CI estimation supported a synergic action, the Fa-CI plot unveiled a different tendency in PANC-1 (Figure 2D).

Lately, we performed time-course experiments, using 10 µg/ml AdipoR plus 15 nM in MIA PaCa-2 or 50 nM Gem in PANC-1. Although co-administration AdipoR plus Gem improved single drug-mediated cell reduction in both PDAC models, different curves were outlined over time. Whilst a time dependency was revealed in reaction to both single and combination treatments in MIA PaCa-2 (Figure 2E), no clear reliance on treatment duration was observed in reaction to Gem in PANC-1. Moreover, comparing combination versus AdipoR, time exposure did not amplify the gap (Figure 2F).

Collectively, these data show that the combination of AdipoR plus Gem impairs MIA PaCa-2 and PANC-1 cell growth more effectively compared with single ones. In addition, as suggested by CompuSyn analysis, a potential synergism might exist between these two compounds.

The clonogenic assay is considered a valuable in vitro assay for monitoring undifferentiated potential and anchorage-independent growth (Rajendran and Jain, 2018). Given that Gem and AdipoR have been proved to act as effective agents in mitigating colony formation, we successively addressed the potential impact of combination AdipoR plus Gem on this PDAC feature (Messaggio et al., 2017; Alhothali et al., 2019; Zhou et al., 2019).

Aiming at defining the consequences of long-term exposure, PDAC cells were seeded at very low density and treated with AdipoR and Gem, both individually and in combination, until newly-formed colonies became viewable. The employment of a small amount of AdipoR and Gem moderately impaired PDAC colony-forming ability, separately (Figure 3). Conversely, a very strong reduction in PDAC clonogenic potential was observed when the same doses of AdipoR and Gem were put together (Figure 3A). Quantification analysis revealed a further enhancement in colonies reduction of nearly 40% compared to Gem alone in MIA PaCa-2, as a result of both number and size decrease (Figure 3B). Consistent results were also obtained in PANC-1 (Figure 3C).

Altogether, this evidence indicates a stronger and deeper outcome in limiting PDAC clonogenic potential made by combinatory treatment AdipoR plus Gem compared to single-agent administration.

To figure out how the combination treatment AdipoR plus Gem affected PDAC cell growth, we successively performed cell cycle analysis intended to determine the cell phase distribution in reaction to our stimuli. Comprehensively, single and combination treatments were performed in both PDAC models for up to 48 h, and the relative DNA content was later detected by flow cytometry using propidium iodide (PI) as basepair intercalating dye.

Depending on the concentration employed, Gem has been reported to induce both S and G2 phase arrest in PDAC models (Miao et al., 2016; Montano et al., 2017; Passacantilli et al., 2018; Kumarasamy et al., 2020). In agreement with these findings, in MIA PaCa-2, we observed a remarkable S-phase accumulation in reaction to 24h Gem administration (Figures 4A,C). In respect of untreated cells, Gem raised S-phase from 40 to 62% at the expense of G0/G1 (−14%) and partly G2/M (−7%). A similar but more pronounced tendency was observed at 48 h as a result of changes in both cell density and nutrients occurring in control cells, rather than a Gem-mediated action (Figure 4A). Quite the contrary, AdipoR intensified the G0/G1 cell amount and decreased both S and G2/M phases at 24 h, while at 48 h, the G0/G1 enrichment was only supported by S-phase reduction.

Looking at the cell phase distribution in reaction to AdipoR plus Gem, different but intermediate features were detected in comparison with single agents. In this respect, after 24 h, combination displayed a G0/G1 amount closer to AdipoR, while conversely, the simultaneous presence of both AdipoR and Gem for additional 24 h exhibited an S-phase accumulation similar to Gem (Figures 4A,C). A quite comparable pattern was also obtained in PANC-1, especially following 24 h of treatment (Figures 4B,C).

In agreement with the recorded cell phase distribution, considerable changes were also detected in cyclin A1 and E1 levels, and cyclin-dependent kinase inhibitor p27^KIP1, in reaction to both single and combined stimuli (Figure 4D; Supplementary Figure S1).

Analysis of subG1 population, which usually includes hypodiploid cells undergoing DNA fragmentation, showed a substantial increase in reaction to both Gem and combination at 48 h compared with untreated cells (Figure 4E). The absence of significant additive cytotoxic effects between Gem and combination was also confirmed in trypan blue exclusion assay, which revealed only minimal changes in death vs living cells in response to these two conditions (Figure 4F). Overall, these findings reveal a different ability in braking cell cycle progression among AdipoR, Gem, and combination.

As the most frequent mutated gene, abnormal KRAS hyperactivation occurs recurrently in PDAC (Buscail et al., 2020). Consequently, dysregulation of the p44/42 MAPK pathway has been recognized in PDAC, assuming a possible correlation between its expression and tumor prognosis (Furukawa, 2015).

Modulation of p44/42 MAPK has also been detected in response to Gem administration in both in vitro and in vivo PDAC models, and in patients (Jin et al., 2017; Ryu et al., 2021). Correspondingly, although the AdipoR-related molecular mechanisms remain largely unknown, its antiproliferative action has been linked to p44/42 MAPK activation in PDAC (Akimoto et al., 2018). Recently, we also observed AdipoR-mediated p44/42 MAPK stimulation in osteosarcoma cell lines (Sapio et al., 2020).

Taking into account the mentioned findings and the relevance of this pleiotropic pathway in regulating the entirety of cell functions (Guo et al., 2020), we first addressed the involvement of p44/42 MAPK in reaction to our stimuli.

With this purpose, MIA PaCa-2 and PANC-1 cells were treated with AdipoR and Gem, alone and in co-administration, for up to 48 h and subsequently analyzed for p44/42 MAPK phosphorylation status.

In the absence of substantial protein amount variations, we recognized a different combination capability in modulating p44/42 MAPK phosphorylation between these two cell lines. Specifically, while in MIA PaCa-2, the concomitant administration of AdipoR with Gem resulted in p44/42 MAPK activation at 48 h (Figure 5A), in PANC-1, instead phospho-p44/42 MAPK upregulation was already apparent at 24 h and maintained up to 48 h (Figure 5B).

To further investigate the p44/42 MAPK involvement in combination-mediated effects, we subsequently tested the impact of MEK1/MEK2 inhibitor PD98059 on AdipoR plus Gem outcomes in MIA PaCa-2 cells. Bearing in mind that long-term exposure to downstream blockade of MAPK deeply impairs PDAC cell growth (Wong et al., 2016), we chose 10 μM for 24 h as effective dosage of PD98059 and time to mitigate p44/42 MAPK signaling and affect MIA PaCa-2 cell growth, marginally (Figure 5C; Supplementary Figure S2A).

Although the combination of AdipoR plus Gem improved cell growth inhibition compared with single ones, PD98059 partially counteracted combination effectiveness, reducing the inhibition rate of approximately 25% relative to p44/42 MAPK-proficient counterpart (Figure 5D). Comparable experiments performed in PANC-1 also revealed a PD98059-mediated capacity in hindering the combination anticancer action (Figure 5E), albeit MEK1/MEK2 inhibitor alone affected cell growth in a more effective manner with respect to MIA PaCa-2 (Supplementary Figures S2B,C).

On the whole, these findings suppose an involvement of p44/42 MAPK pathway in AdipoR plus Gem combination response.

Although Gem displays one of the highest response rates compared to other anticancer agents in PDAC, resistance outbreak occurs already within few weeks of initiating dosing (Amrutkar and Gladhaug, 2017). As a result of Gem-induced refractivity, PDAC generally becomes more aggressive, causing a further reduction in overall survival (Quinonero et al., 2019).

To further speculate the usefulness of AdipoR-based therapy in PDAC, we first developed stable MIA PaCa-2 cell lines resistant to Gem (Gem-Res). Thereafter, MIA PaCa-2 and MIA PaCa-2 Gem-Res cells were cultured in a medium containing 10 µg/ml AdipoR and 15 nM Gem, both individually and in combination for up to 48 h. As previously described in MIA PaCa-2, combination treatment resulted in a further cell growth reduction compared to AdipoR and Gem singularly, both at 24 and 48 h (Figures 6A,B). Remarkably, even though Gem was ineffective in reducing the cell number in MIA PaCa-2 Gem-Res cells, AdipoR induced a 25% growth inhibition, and even more interestingly, co-administration AdipoR plus Gem affected cell proliferation by another 18% with respect to AdipoR alone at 48 h (Figure 6B).

Using a higher dose from the one previously employed in Figure 3, Gem abrogated the colony forming ability in MIA PaCa-2, while conversely, in Gem-Res cells, the same amount marginally affected the growing colony (Figure 6C). Interestingly, either alone or in combination with Gem, AdipoR administration reduced clonogenic potential by 45 and 55%, correspondingly (Figure 6D).

Despite being less pronounced than MIA PaCa-2, flow cytometry analysis showed AdipoR persistence in braking cell cycle progression even in MIA PaCa-2 Gem-Res. Like the sensitive cells, increased G0/G1 phase was observed in the resistant ones supplemented with AdipoR (Figure 6E). But even more interesting, reducing both S and G2/M phases, the concomitant administration of AdipoR and Gem enhanced the G0/G1 accumulation compared with AdipoR alone (Figure 6E). Remarkably, no substantial changes were detected between Gem-treated and untreated cells, confirming the loss of chemotherapy responsiveness by this cell line.

Taken together, these data indicate that the combination AdipoR plus Gem is effective in preventing growth and colony formation even in Gem-resistant MIA PaCa-2 cells.