In Vivo Evaluation of the Combined Anticancer Effects of Cisplatin and SAHA in Nonsmall Cell Lung Carcinoma Using [18F]FAHA and [18F]FDG PET/CT Imaging

Combining standard drugs with low doses of histone deacetylase inhibitors (HDACIs) is a promising strategy to increase the efficacy of chemotherapy. The ability of well-tolerated doses of HDACIs that act as chemosensitizers for platinum-based chemotherapeutics has recently been proven in many types and stages of cancer in vitro and in vivo. Detection of changes in HDAC activity/expression may provide important prognostic and predictive information and influence treatment decision-making. Use of [18F] FAHA, a HDAC IIa-specific radionuclide, for molecular imaging may enable longitudinal, noninvasive assessment of HDAC activity/expression in metastatic cancer. We evaluated the synergistic anticancer effects of cisplatin and the histone deacetylase inhibitor suberoylanilide hydroxamic acid (SAHA) in xenograft models of nonsmall cell lung cancer (NSCLC) using [18F] FAHA and [18F] FDG PET/CT imaging. Cisplatin alone significantly increased [18F] FAHA accumulation and reduced [18F] FDG accumulation in H441 and PC14 xenografts; coadministration of cisplatin and SAHA resulted in the opposite effects. Immunochemical staining for acetyl-histone H3 confirmed the PET/CT imaging findings. Moreover, SAHA had a more significant effect on the acetylome in PC14 (EGFR exon 19 deletion mutation) xenografts than H441 (wild-type EGFR and KRAS codon 12 mutant) xenografts. In conclusion, [18F] FAHA enables quantitative visualization of HDAC activity/expression in vivo, thus, may represent a clinically useful, noninvasive tool for the management of patients who may benefit from synergistic anticancer therapy.


Introduction
Lung cancer is the most common cancer and the leading cause of cancer-related deaths worldwide and is more com-mon in developing countries [1]. The majority of cases of lung cancer are nonsmall cell lung cancer (NSCLC), which includes squamous cell carcinoma, adenocarcinoma, and large cell carcinoma [1]. The current classification system for advanced NSCLC includes several histological and molecular subtypes and is a vital component of therapeutic decision-making [2,3].
Platinum-based agents such as cisplatin are highly active cytotoxic drugs used to treat NSCLC and represent an essential component of neoadjuvant, adjuvant, and palliative chemotherapy regimens [4][5][6][7]. Extensive research has demonstrated that changes in various aspects of the use of cisplatin, such as the administration schedule and methods and frequency of monitoring toxicity, have incrementally improved the outcomes and quality of life of patients with NSCLC [5,8]. Cisplatin is thought to activate DNA damage recognition proteins that transmit DNA damage signals to downstream signaling cascades that involve p53, MAPK, and p73, which ultimately induce apoptosis [9][10][11].
Several key oncogenic events have been identified in NSCLC. The incidence of epidermal growth factor receptor (EGFR) mutations in the Caucasian population is approximately 10%, but is higher among never-smokers, patients with adenocarcinoma, females, and individuals from East Asia [12]. Moreover, the EML4-ALK fusion gene is present in approximately 4% of lung tumors and encountered more frequently in the tumors of never-smokers, younger patients, and patients with adenocarcinoma [12]. Thus, only a small proportion of patients with advanced NSCLC are candidates for existing molecular-targeted therapies. For the 85-90% of patients with NSCLC who do not have mutations associated with drug sensitivity (i.e., in genes targeted by EGFR kinase inhibitors), platinum-based chemotherapy remains the standard first-line chemotherapy [4].
The roles of the members of the HDAC family have recently been elucidated in several human malignancies [13]. Histone acetylation and deacetylation of the lysine residues within histone tails occur as a dynamic, reversible process catalyzed by two classes of enzymes, histone acetyltransferase (HAT), and histone deacetylase (HDAC). In general, histone acetylation correlates with transcriptional activation and histone deacetylation correlates with transcriptional repression. Histone acetylation, one of the first epigenetic mechanisms of transcriptional regulation to be studied, is involved in numerous, diverse cellular processes including cell-cycle progression, DNA repair, and gene silencing. Additionally, disruption of the balance between histone acetylation and deacetylation has been implicated as a causative factor in tumor cell proliferation, migration, angiogenesis, differentiation, invasion, and metastasis [14][15][16].
SAHA is a potent, reversible pan-HDAC inhibitor. SAHA inhibits both class I and class II HDACs and, thus, alters gene transcription and induces cell cycle arrest and/or apoptosis in a wide variety of transformed cells [23]. SAHA has been clinically approved for the treatment of cutaneous T cell lymphoma [24] and has been shown to exert antitumor activity in other solid tumors, including NSCLC [25][26][27][28], breast cancer [29][30][31], and ovarian cancer [32,33].
However, HDACIs may block the DNA damage responses induced by cisplatin-mediated toxicity [34,35]. Through detailed knowledge of the mechanisms underlying the sensitivity of tumors to cisplatin and HDACIs exists, the effects of altered HDAC activity/expression on combination therapy are poorly understood. Therefore, a reliable and quantitative biomarker for HDAC activity is urgently required.
We previously developed 6-[ 18 F]fluoroacetamido)-1hexanoicanilide ([ 18 F]FAHA) as a potential PET imaging agent and highly selective radiotracer for quantitative imaging of HDAC class IIa enzyme expression and activity in vivo using PET/CT/(MRI) [36,37]. More recently, we demonstrated that [ 18 F] FAHA PET could be used to monitor alterations in HDAC activity/expression in a rat model of chemotherapy-induced neurotoxicity in the brain [38].
Here, we aimed to assess the efficacy of PET/CT using [ 18 F] FAHA to image HDAC class IIa activity/expression in a mouse model of NSCLC. We noninvasively monitored the effects of cisplatin in the presence and absence of SAHA on HDAC class IIa activity in H441 (wild-type EGFR and KRAS codon 12 mutant) and PC14 (EGFR exon 19 deletion mutation) NSCLC xenograft tumors. Furthermore, we investigated whether the responses to cisplatin and cisplatin/-SAHA were related to the presence of an EGFR mutation. F] FDG PET images were summed using a rigid transformation algorithm and normalized mutual method after running a reslicing process (PMOD Technologies Ltd., Zurich, Switzerland). The regional radioactivity concentrations (KBq/mL) of [ 18 F] FAHA or [ 18 F] FDG PET were estimated from the maximum pixel values within each ROI and expressed as a percentage of injected dose/tissue g (%ID/g).

Materials and Methods
The posttherapy radioactive accumulation rate was calculated using: Post-Tx radioactive accumulation rate = Suv max of Posttherapy Suv max of Pretherapy where SUV mean is the maximum standardized uptake value, and drug inhibition rate is the percentage increase or reduction in accumulation of the PET radiotracer based on the volume accumulated in the tumor. The SAHA effect rate between Group B and Group C was calculated using: where SUV mean is the maximum standardized uptake value, and the SAHA effect rate is the percentage increase or reduction in accumulation of the PET radiotracer based on the volume accumulated in the tumor.   [36,38], as follows: where t is time, R ðtÞ is the mean count of the tumor, C ðtÞ is the mean count of the blood, Ki is the clearance that determines the rate of entry into the tumor, and V 0 is the distribution volume. The time between injection and the start of the linear phase in the Patlak plot was 4-6 min. Based on data from the start of the linear phase, an accurate linear fit was observed from 6-8 min up to 18-20 min. The slope of the Patlak plot represents the influx rate constant Ki.
2.8. Immunohistochemistry. After imaging, the mice were humanely euthanized, and the tumors were excised for immunohistochemical analysis (IHC) of acetyl-histone H3 (AH3). Paraffin-embedded sections (5 μm-thick) were incubated in antigen retrieval solution (10 mM citrate buffer, pH 6.0) at 100°C for 10 min, washed, incubated in 3% hydrogen peroxidase solution for 15 min at room temperature, and then placed in blocking solution for 60 min at room temperature. Then, the sections were incubated with a primary acetyl-histone A3 (Lys9) antibody (Cell Signaling Technology, Danvers, MA, USA) at 1 : 50 dilution in blocking solution overnight at 4°C, followed by secondary biotinylated horse anti-mouse IgG (Vector Laboratories, Inc., Burlingame, CA, USA), and developed via avidin-peroxidase conjugation and the chromophore 3 ′ 3-diaminobenzidine using a Vectastain ELITE kit (Vector Laboratories, Burlingame, CA, USA) according to the manufacturer's protocol. The tissue sections were either counterstained with hematoxylin or not counterstained for densitometric analysis of the intensity of AH3 immunostaining.
Microscopic evaluation of the immunostained sections was performed using a BX51 microscope equipped with a DP71 digital camera (Olympus, Tokyo, Japan). Protein expression was semiquantitatively assessed based on the number of cells showing nuclear expression of AH3 in five nonoverlapping ×100 microscopic fields as: 0 = absent, less than 5% immunopositive cells; 1 = rare, 10-20% immunopositive cells; 2 = mild, 20-40% mildly or moderately positive cells; 3 = moderate, 40-60% moderately or strongly positive cells; or 4 = strong, more than 80% strongly positive cells per field of view. The percentage score for each tumor was calculated as follows: actual rating × 100/maximal score (i.e., a rating value of 4).
Percentage of positive signal = Sum of score of the group Number of case × maximal score 4 × 100%:  (Figure 2(a), p < 0:05 and p < 0:005, respectively), but not after administration of cisplatin with SAHA in Group C (Figures 3 upper panel and 2(a)).     (Figure 5(c)).  In the PC14 tumor sections, cisplatin significantly reduced AH3 immunoactivity in Groups A (p < 0:005) and B (p < 0:005; Figures 6 lower panel and 7(b)). However, the combination of SAHA with cisplatin (Group C) reversed the cisplatin-induced reduction in AH3 immunoactivity compared to mice injected with cisplatin alone (Group B; p < 0:05, Figure 7(b)).   TRIB1 and HDACs play crucial roles in cisplatin-induced enrichment of cancer stem cells (CSCs) and drug resistance in NSCLC, and patients with high levels of TRIB1 had a significantly poorer response to cisplatin and a poorer prognosis [39]. Furthermore, cisplatin and SAHA suppressed the viability and growth of NSCLC tumor cells in vitro and in vivo [39].

Discussion
Moreover, Sun et al. (2019) reported that cotreatment with S11-which simultaneously inhibits HDACs and Pglycoprotein (P-gp)-and cisplatin suppressed colony formation and blocked the migration of cisplatin-resistant NSCLC cells [40]. Subsequently, the combination of SAHA with cis-platin was shown to induce miR-149 expression, which directly targets the excision repair cross-complementation group (ERCC1). Inhibition of ERCC1 expression correlated positively with DNA repair capacity, thus, miR-149 and ERCC1 may represent a target to increase the sensitivity of tumor cells to cisplatin [41,42]. Combined administration of SAHA and cisplatin also induced apoptosis, promoted cell cycle arrest, and increased the levels of acetylated histone H3 and α-tubulin in a xenograft model of small cell lung cancer (SCLC) in nude mice [43]. Additionally, previous reports demonstrated various HDACIs exert synergistic antitumor effects with fluoropyrimidines in several tumor types, including breast cancer, colorectal cancer [44][45][46][47], and head and neck squamous cell carcinoma. Overall, these studies indicate H441 PC14 Cisplatin 2 mg/kg Cisplatin 4 mg/kg Cisplatin 4 mg/kg+ SAHA 300 mg/kg Control Figure 6: Immunohistochemical staining for AH3 in xenograft tumors. In H441 xenograft tumors, AH3 immunoreactivity was lower in Group A (2 mg/kg cisplatin) and Group B (4 mg/kg cisplatin) than the control group and was almost equal in Group C (4 mg/kg +300 mg/kg SAHA) and the control group. Similar trends were observed in PC14 tumors; however, Group C exhibited a significantly higher increase in AH3 immunoactivity than Group B. Scale bar, 100 μm.  HDACIs promote CSC expansion by reprogramming differentiated cancer cells into stem-like cells that exhibit enhanced Pentose phosphate pathway (PPP) metabolism. PPP plays a substantial role in the production of cellular NADPH, which is required for fatty acid synthesis and intracellular ROS detoxification [52]. Interestingly, we found that cisplatin alone significantly reduced [ 18 F] FDG uptake in the NSCLC xenografts. However, this effect was reversed by coadministration of SAHA, consistent with our previous findings in a rat model of cisplatin-induced neurotoxicity [38]. Therefore, the significant enhancement in the [ 18 F] FDG PET signals in NSCLC tumors implies that the combination of cisplatin and SAHA may-at least partially-enrich CSCs, since glucose is one of the main energy sources for both CSCs and differentiated tumor cells [53].
Accumulating evidence indicates HDACIs modulate the epigenetic regulation of CSCs, specifically the CSC subpopulation, in solid cancers [54]. HDACIs have been suggested to modulate stemness and enable tumor cells to overcome drug resistance [55,56]. Our results demonstrate that the changes in HDAC activity/expression and metabolic adaptation of CSCs included by combined treatment with chemotherapy and HDACIs could be monitored in vivo using coupled [ 18 F] FDG and [ 18 F]FAHA PET/CT imaging.
Cisplatin-based chemotherapy remains the first-line strategy for wild-type EGFR in NSCLC; however, cisplatin often becomes ineffective as most tumors acquire drug resistance over time. Cell lines expressing mutant EGFR are mostly resistant to cisplatin, and KRAS mutant cell lines exhibit varied sensitivity to cisplatin, depending on E-cadherin mRNA expression [57]. In the present study, we conducted followup [ 18 F] FDG PET/CT scans after completion of treatment, which is a standardized imaging procedure for monitoring the response to therapy [58]. [ 18 F] FDG uptake is proportional to the metabolic rate of viable tumor cells, which have a higher demand for glucose than normal cells [59].
Similarly to a previous report [57], cisplatin alone led to a relatively poor cisplatin-inhibition rate (lower [ 18 [60].
Immunostaining for AH3 confirmed the PET/CT imaging findings, in that cisplatin increased HDAC activity/expression by reducing the expression of AH3, whereas coadministration of SAHA reversed these effects. Moreover, the immunostaining also supported the PET/CT finding that the magnitude of the changes in HDAC activity/expression was greater in the PC14 xenografts than H441 xenografts.
Overall These results suggest that SAHA blocks excess HDAC deacetylation and would thus inhibit the antitumor effects of cisplatin in NSCLC. In addition, the IHC also supported our hypothesis that SAHA blocks cisplatin-induced decreases in HAT acetylation and increases in HDAC deacetylation in NSCLC.
The effects of HDACIs in cancer have been examined in preclinical and early clinical studies. HDACIs are expected to be used in combination with other anticancer drugs. HDACIs synergistically enhance the anticancer effects of chemotherapy drugs in several tumor types [23,[61][62][63]. However, HDACIs do not cooperate with anticancer drugs to synergistically inhibit cell proliferation in all tumor types. For example, Chai et al. (2008) reported the HDACIs depsipepside and Trichostatin A enhanced cytarabineinduced inhibition of cell proliferation, but synergistic inhibition of cell proliferation was not observed for other DNA damage inducers-including cisplatin. Chai et al. concluded HDACIs selectively act with nucleoside analogs to inhibit cell proliferation [64]. Differences between the tumor models and anticancer drugs may be one explanation for these discrepancies.
Platinum-based therapy still represents a major therapeutic strategy in several solid tumors, including colorectal, breast, and pancreatic cancer. However, chemo-resistance remains a major unresolved issue. As discussed above, HDA-CIs modulate gene expression and usually function as sensitizers to act synergistically with chemotherapeutics and molecular targeted agents. Although numerous studies have demonstrated the benefit of coadministration of cisplatin with SAHA as mentioned before, the mechanisms by which HDA-CIs interact with cisplatin in cancer cells have not yet been elucidated. Furthermore, we have no knowledge of the metabolic adaptations that take place during the transition from normal stem cells to CSCs induced by such cancer therapy, and only a handful of studies have explored the transition of CSCs to differentiated tumor cells [39]. Moreover, additional studies are needed to evaluate the clinical applicability of SAHA or other HDACIs as a component of chemoradiotherapy regimens. Thus, repetitive, noninvasive PET/CT imaging with [ 18 F]FAHA may facilitate future preclinical or clinical studies to elucidate the roles of class IIa HDAC enzymes in tumor progression, chemoresistance, and the expansion of CSCs, 8 Molecular Imaging and may help to optimize therapeutic doses of novel class IIa HDACIs for combined chemoradiotherapy.

Conclusion
Combining traditional chemotherapy drugs with HDACIs may improve therapeutic efficacy in solid cancers. Molecular imaging using [ 18 F] FAHA, a novel HDAC IIa-specific radiotracer, provided unique insight into the location of and quantitative changes in HDAC activity/expression in tumors in vivo in response to treatment with cisplatin alone or cisplatin combined with a HDACI. Additional PET imaging may help to determine the mechanistic, therapeutic, and prognostic roles of HDACs in various diseases and enable monitoring of HDAC-targeted therapies. Further clinical and preclinical investigations are necessary to identify the mechanisms by which HDACIs modulate signaling pathways in different tumor types.

Data Availability
All data generated or analyzed during this study are included in this published article.

Conflicts of Interest
The authors declare that they have no conflicts of interest.