The aim of this study was to examine the behaviour of Apitolisib-resistant H1975 lung adenocarcinoma cells from the perspective of metabolic reprogramming and proliferation aggressiveness. The notion was to shed light on how clinicians might proceed when malignant tissues have stopped responding to Apitolisib: continue or stop Apitolisib, or consider a drug combination. Therefore, understanding cancer biology and expanding the treatment team to include experts from different disciplines could improve patient outcomes. In our study, to model tyrosine kinase inhibitor resistance arising in the clinic, we used H1975 lung adenocarcinoma cells resistant to Apitolisib, and compared them to their parent sensitive cells.

Apitolisib reverted the excessive proliferation of resistant H1975R− cells

The confluency test and proliferation rate measurements showed that the H1975R− had an aggressive proliferation rate even higher than that of drug-naïve parent cells. However, the proliferation rate was suppressed by incubation with this drug (Fig. 1). Clinically, this means that stopping Apitolisib allows the tumour to become very aggressive, making tumour relapse highly probable. To elucidate the reasons for the relatively slow rate of proliferation observed in H1975R+ cells compared with H1975R− cells, we conducted cell cycle profiling. Interestingly, the cell cycle patterns in the treated and untreated cells were identical. However, the G0/G1 ratio of the newly dividing resistant cells was reduced by a factor of five compared to that of their parent cells. Moreover, the S and G2/M phases both doubled, indicating that both cell populations were actively dividing. The only discernible difference between the two cell populations was the energy output, which was scarce in the treated cells. To examine such hypothesis, seahorse energy phenotyping and central carbon metabolism assessment were performed (Figs. 3 and 4).

The inverted Warburg effect reflects the increased aggressiveness of H1975R− cells

An imbalance of the tyrosine kinase cascade can result in malignant transformation. This imbalance could be caused by gain-of-function (driver) mutations or increased expression of RTK or NRTK, combined with loss of apoptosis function, and in response to adaptive mitochondrial function. Consequently, in response to the increased demand for glucose for nucleic acid synthesis, somatic cells may activate alternative metabolic pathways to generate energy. This could involve an increased reliance on glycolysis, which accounts for up to 60% of cancer cell energy sources [14]. In cancer cells, the rate of glycolysis is 100× faster than that in normal somatic cells [15, 16], raising the amount and rate of glucose entry into the cells [7,8,9, 17,18,19]. The PI3K signalling pathway plays a role in regulating the cellular metabolism balance, between the aerobic and anaerobic pathways, and the uptake of glucose and other nutrients, as well as mitochondrial function [18, 19]. Inhibition of PI3K signalling by Apitolisib disrupts the balance between the dominant OXPHOS pathway and glycolysis, resulting in suppression of both pathways [20]. A large depression in energy production lowers the proliferation rate (Figs. 1 and 3). As a consequence of PI3K pathway inhibition, cancer cells begin to tolerate and adapt to several mechanisms to cope with this shift in energy. The ability of cancer cells to tolerate a one-log-fold increase in the IC50 of an inhibitor is called resistance. Apitolisib is a PI3K/m-TOR dual inhibitor was introduced in clinical trials after tumour cells become tolerant to PI3K single inhibitors. Resistance to this dual inhibitor has been well-documented in clinical trials to be due to either intrinsic genetic makeup or acquired resistance via exposure and adaptation [21].

H1975 cells harbour activated EGFR and a mutant PI3KCA enzyme downstream (a mutation in the catalytic subunit of the kinase enzyme) in combination with a dysfunctional TP53 gene. Their adaptive response to repetitive exposure to Apitolisib might involve activation of molecular crosstalk with EGFR, such as the EGFR/MAPK axis and other growth factor receptor signalling, which can bypass the PI3K/m-TOR tyrosine kinases, activating the downstream transcription factors [21]. Therefore, these cells can tolerate higher doses of Apitolisib over time and can sustain minimal energy requirements. Our findings show that when Apitolisib-resistant cells are incubated with the drug, the inhibition of PI3K may disrupt metabolic regulation, leading to a shift towards non-glucose substrate aerobic metabolism, which is confirmed by their increased consumption of fatty acids, accumulation of pyruvate and elevated levels of ketone bodies (ketosis), combined with their decreased OCR and ECAR (Figs. 3 and 4) [22]. This could be attributed to Apitolisib, as a PI3K inhibitor, leading to additive suppression of both mitochondrial OXPHOS and glycolysis, as well as to the reduction of glucose uptake by GLUT transporters, which would limit the availability of glucose for metabolic processes and nucleic acid synthesis [10, 23].

In Apitolisib-free media, resistant cells (H1975R) shifted to an aerobic metabolism phenotype (a reversed Warburg effect) driven by energy starvation caused by Apitolisib. These cells became more energetic than their parent cells, as indicated by their increased OCR, increased pyruvate consumption, and lowered ketone body levels (Figs. 3 and 4). These highly energetic cells point to the restoration of PI3K/mTOR- facilitated glucose influx and its utilization via mitochondria, with glucose scarcity caused by Apitolisib acting as a driving force (Fig. 3), as well as extra energy sourced through non-carbohydrate substrates (reversed Warburg effect). This explains why Apitolisib-resistant H1975 cells in drug-free medium exhibit high proliferation rates.

Elevation of ketone bodies explains the slow proliferation rate of treated H1975R+ cells

Accumulation of ketone bodies and increased consumption of fatty acids can slow cancer cell proliferation and invasion, which is the case in H1975R+ cells [24]. The opposite pattern was observed in non-treated cells, in which the diminished levels of ketone bodies reflected the adequacy of their glucose supply for a high proliferation rate (Fig. 1). Ketone bodies have been shown to inhibit the proliferation of cancer cells and induce apoptosis, leading to reduced tumour growth. They can also diminish glycolytic flux and glutamine uptake in tumour cells, resulting in decreased ATP content and survival [25]. Increased ketone bodies in resistant cancer cells disrupt their survival and proliferation, inhibit glycolysis, and block the main pathway of energy production [24, 26]. Ketone supplementation or a ketogenic diet has shown positive therapeutic advantages in various malignancies, potentially targeting tumour metabolism as a new area for drug discovery [27]. The molecular pathways affected by ketone body treatment in cancer cells include the JAK-STAT and mTOR pathways, which lead to decreased cell migration and inhibition of cellular growth and proliferation. This might explain the activated JAK/STAT axis in our treated Apitolisib-resistant H1975 cells published earlier by the thoracic oncology research group at TTMI, Trinity College, Dublin [28]. The acquired reliance on the JAK-STAT signalling pathway might make them less vulnerable to the inhibitory effect of Apitolisib. In the absence of the drug, these cells can still exploit alternative PI3K-independent growth promoting pathways to achieve a higher proliferation rate.

Apitolisib-resistant H1975R− cells tolerate higher doses of vorinostat than H1975R+ 

HDACs play a vital role in the transcription of genes involved in cell cycle progression and survival. Furthermore, HDACs 1–7 are essential for the transcription complexes required for the cell cycle and survival [29]. HDAC7 has been found to be a significant contributor to cancer cell proliferation because it affects cell cycle progression by regulating c-Myc expression [30]. Thoracic oncology research group at TMMI-TCD reported that Apitolisib activates c-Myc expression in Apitolisib-treated resistant H1975 cells [31], which was also the subject of the current study. It has been demonstrated that HDAC inhibitors impede proliferation and induce apoptosis in tumour cells, indicating that they may affect the post-translational modifications of molecules involved in cell proliferation and survival [31, 32]. Western blot analysis of HDAC levels in H1975R+ cells revealed elevated levels of HDACs 1, 2, 3, 4, and 6 compared with H1975R− cells (H1975R−) [13]. When a pan-HDAC inhibitor was employed, the H1975R− cells demonstrated tolerance to elevated doses of Vorinostat, whereas the H1975R+ cells exhibited significant vulnerability to Vorinostat (Fig. 5). This observation contradicts the findings from the western blot analysis, which suggests that the H1975R+ cells should have a higher tolerance to Vorinostat because of their elevated levels of HDACs compared to the non-treated cells (H1975R−) [13]. However, this elevation in HDACs may indicate inactivity, and a positive feedback mechanism may trigger their expression and promote protein abundance. The PI3K/mTOR pathway is known to induce phosphorylation of HDACs, thereby preventing their export to the nucleus and serving as transcriptional corepressors. We anticipated that treatment with Apitolisib would result in hypo-phosphorylation of HDACs. This hypo-phosphorylation of HDAC IV and V in H1975R+ cells was opposite to that in H1975R− cells, which indicates that these HDACs are inactive [13]. This finding indicates that HDACs play a significant role in the resistance of.

H1975 cells to Apitolisib and that their phosphorylation is essential for their functionality. As HDACs become more phosphorylated, their proliferation rates increase, and vice versa. This interpretation is supported by the higher proliferation rate observed in the H1975R− cells (Fig. 1).

In summary, Apitolisib reduced the survival of H1975R+ with exceedingly low levels of energy, resulting in a significant reduction in their proliferation rate (Fig. 1). This phenomenon may be attributed to a higher rate of apoptosis in comparison to the newly generated cells, which is caused by the compromise of both mitochondria and glycolysis by Apitolisib. This is indicated by heightened levels of ketone bodies and increased consumption of fatty acids. Additionally, Apitolisib has been observed to eliminate the phosphorylation of HDACs, thereby facilitating their entry into the nucleus to function as a co-repressor of genes essential for cell survival and proliferation. Upon discontinuation of Apitolisib, the phosphorylation of HDACs was restored and their expression diminished. This suggested the increased abundance of HDACs might stem from a feedback mechanism. Inhibitors of HDACs impede proliferation and induce apoptosis in tumour cells, indicating that they might affect post-translational modifications of molecules involved in cell proliferation and survival. Additionally, by interacting with checkpoint kinases and reducing the expression of cyclin-dependent kinase inhibitory genes, such as BAX and P21, which promote apoptosis through the activation of P53, these inhibitors further contribute to the higher proliferation of non-treated H1975 resistant cells. This finding was supported by western blot analysis [13]. Considering this scenario in the context of the in vivo resistance to Apitolisib observed in clinical settings, it can be inferred that if the medication is stopped, the elevated levels of HDACs become vulnerable to PI3K/AKT, resulting in a significant increase in phosphorylated HDACs, ultimately facilitating tumour relapse and enhancing aggressiveness. Therefore, it is highly recommended not to stop Apitolisib when no further clinical improvement obtained, and instead to combine with HDAC inhibitors. However, one limitation of this study is that we need to bear in mind that the specific effects of Apitolisib on cellular metabolism and the cell cycle may depend on several factors, such as cell type, genetic alterations, and the cell’s microenvironment. Therefore, further investigation of the metabolic effects of Apitolisib on resistant cells within panels of cell lines with varying oncogenicities would yield more comprehensive insights into these processes. The other limitation is that even with the use of a panel of cell lines, tumour tissues are highly heterogeneous and culturing cell suspension of patients’ tissues is not being used for testing drugs in translational medicine laboratories.