Abstract

Background

   The cellular tumor protein p53 (TP53) is a tumor suppressor gene that
   is frequently mutated in human cancers. Among various cancer types, the
   very aggressive high‐grade serous ovarian carcinoma (HGSOC) exhibits
   the highest prevalence of TP53 mutations, present in >96% of cases.
   Despite intensive efforts to reactivate p53, no clinical drug has been
   approved to rescue p53 function. In this study, our primary objective
   was to administer in vitro‐transcribed (IVT) wild‐type (WT) p53‐mRNA to
   HGSOC cell lines, primary cells, and orthotopic mouse models, with the
   aim of exploring its impact on inhibiting tumor growth and
   dissemination, both in vitro and in vivo.

Methods

   To restore the activity of p53, WT p53 was exogenously expressed in
   HGSOC cell lines using a mammalian vector system. Moreover, IVT WT p53
   mRNA was delivered into different HGSOC model systems (primary cells
   and patient‐derived organoids) using liposomes and studied for
   proliferation, cell cycle progression, apoptosis, colony formation, and
   chromosomal instability. Transcriptomic alterations induced by p53 mRNA
   were analyzed using RNA sequencing in OVCAR‐8 and primary HGSOC cells,
   followed by ingenuity pathway analysis. In vivo effects on tumor growth
   and metastasis were studied using orthotopic xenografts and metastatic
   intraperitoneal mouse models.

Results

   Reactivation of the TP53 tumor suppressor gene was explored in
   different HGSOC model systems using newly designed IVT mRNA‐based
   methods. The introduction of WT p53 mRNA triggered dose‐dependent
   apoptosis, cell cycle arrest, and potent long‐lasting inhibition of
   HGSOC cell proliferation. Transcriptome analysis of OVCAR‐8 cells upon
   mRNA‐based p53 reactivation revealed significant alterations in gene
   expression related to p53 signaling, such as apoptosis, cell cycle
   regulation, and DNA damage. Restoring p53 function concurrently reduces
   chromosomal instability within the HGSOC cells, underscoring its
   crucial contribution in safeguarding genomic integrity by moderating
   the baseline occurrence of double‐strand breaks arising from
   replication stress. Furthermore, in various mouse models, treatment
   with p53 mRNA reduced tumor growth and inhibited tumor cell
   dissemination in the peritoneal cavity in a dose‐dependent manner.

Conclusions

   The IVT mRNA‐based reactivation of p53 holds promise as a potential
   therapeutic strategy for HGSOC, providing valuable insights into the
   molecular mechanisms underlying p53 function and its relevance in
   ovarian cancer treatment.

   Keywords: apoptosis, cell cycle, chromosomal instability, HGSOC
   metastatic intraperitoneal mouse model, HGSOC orthotopic Xenograft
   model, high‐grade serous ovarian cancer, in vitro‐transcribed p53‐mRNA,
   liposomal IVT mRNA delivery system, patient‐derived organoid,
   patient‐derived primary cancer cell
     __________________________________________________________________

List of abbreviations

   APC
          Adenomatous polyposis coli protein

   BAK
          B‐cell lymphoma 2 (BCL‐2) antagonist/killer

   BAX
          BCL‐2 associated X‐protein

   BCL‐2
          B‐cell lymphoma 2

   Bid
          BH3 interacting domain death agonist

   Bub1
          Mitotic checkpoint serine/threonine kinase

   BUB1B
          BUB1 mitotic checkpoint serine/threonine kinase B

   CDC20
          Cell division protein 20 homolog

   CDK1
          Cyclin‐dependent Kinase 1

   EGFP
          Enhanced Green Fluorescence Protein

   FAS
          Tumor necrosis factor receptor superfamily member 6

   GSK3β
          Glycogen synthase kinase 3β

   H2AX
          H2A histone family member X

   HGSOC
          high‐grade serous ovarian cancer

   IVT
          in vitro transcription

   MDM2
          Double minute 2 homolog protein

   MYC
          Master regulator of cell cycle entry and proliferative
          metabolism

   Noxa
          NADPH oxidase activator 1

   p14ARF
          ARF tumor suppressor

   p16
          inhibitor of cyclin‐dependent kinase 4 (CDK4) and CDK6

   P21/(CDKN1A)
          Cyclin‐dependent kinase inhibitor 1

   p27
          Inhibitor of cyclin‐dependent kinase 4 (CDK4) and CDK6

   PARP
          Poly (ADP‐ribose) polymerase

   PLK1
          Polo‐like Kinase 1

   PLK3
          Polo‐like Kinase 3

   Puma
          The p53 upregulated modulator of apoptosis

   TP53
          Tumor Protein p53

   XIAP
          X‐linked inhibitor of apoptosis

1. BACKGROUND

   High‐grade serous ovarian cancer (HGSOC) is the most lethal
   gynecological malignancy. It interacts intensively with the tumor
   stroma, grows rapidly, metastasizes extensively, and shows an
   aggressive course [[42]1, [43]2]. Ovarian cancer (OC) cells remain
   within the peritoneal cavity and disseminate only to the
   mesothelium‐lined surface [[44]3]. The current standard of care
   includes cytoreductive surgery, chemotherapy (platinum compounds,
   paclitaxel [[45]4], and cyclophosphamide [[46]5]), bevacizumab
   (avastin) [[47]6], and poly‐ADP‐ribose polymerase (PARP) inhibitors
   [[48]7]. Nevertheless, approximately 70% of patients experience relapse
   within 3 years following surgery and platinum‐based chemotherapy and
   succumb to disease progression [[49]8].

   The tumor protein p53 (TP53) gene is the most frequently mutated human
   tumor suppressor gene, mostly through point mutations, whereby amino
   acid substitutions lead to the disruption of tumor protein p53 binding
   to DNA [[50]9, [51]10, [52]11]. Thus, p53 protects the human genome by
   counteracting cellular stress and DNA damage [[53]12]. Multiple reports
   have shown that tumor cells harboring functionally inactive p53 are
   resistant to chemotherapeutics because their mechanism of action
   involves DNA damage, leading to the activation of wild‐type (WT) p53.
   The Cancer Genome Atlas Research Network revealed that HGSOC is
   characterized by inactive or truncated TP53 in up to 96% of all cases
   [[54]13]. Moreover, mutations in TP53 seem to be early events in
   tumorigenesis, likely in OC precursor lesions, supporting the
   importance of mutated TP53 as a driver of this malignancy [[55]14].

   Approaches aimed at re‐establishing the WT function of p53 have the
   unique ability to improve outcomes in patients with many types of
   cancer. Several small molecules have been developed to counteract the
   loss of p53 function, mostly by stabilizing the native conformation of
   p53 [[56]15, [57]16], like PRIMA‐1, which reactivates mutant p53,
   induces apoptosis in cancer cells [[58]15, [59]17], and is currently in
   clinical trials (ClinicalTrials.gov). In our laboratory, silencing of
   Cyclin B1 upregulated the ARF tumor suppressor, p14ARF, which inhibits
   mouse double minute 2 homolog protein (MDM2) and reduces p53‐S315
   phosphorylation by Cyclin‐dependent kinase 1 (CDK1)/Cyclin B1, thereby
   modulating two mechanisms that stabilize and activate p53 [[60]18].
   Other small molecules have been developed to reinstate specific p53
   functions [[61]15, [62]19, [63]20], or nonsense mutations by promoting
   transcriptional readthrough [[64]21]. Nevertheless, different studies
   have provided evidence of off‐target effects associated with small
   molecules used for p53 reactivation [[65]22, [66]23, [67]24]. Thus,
   whether off‐target activities or restoration of p53 function are the
   origin of therapeutic responses is currently being investigated.

   With benefits such as scalable manufacturing, few adverse effects, and
   high efficiency [[68]25, [69]26], novel strategies based on mRNA
   therapeutics have emerged as effective and safe approaches for fighting
   predominantly infectious diseases. Although mRNA therapies for protein
   replacement are gaining attention in several clinical settings, their
   use in cancer remains limited [[70]27, [71]28]. To explore whether
   mutant TP53‐bearing HGSOC cells respond to p53 rescue, we tested an in
   vitro transcribed (IVT), codon‐optimized p53‐mRNA in human and murine
   models and achieved robust and persistent protein expression using
   diverse IVT‐mRNAs. We aimed to investigate the impact of prolonged
   p53‐mRNA treatment on the induction of cell cycle arrest, apoptosis,
   and chromosomal stability in HGSOC, a tumor type characterized by TP53
   loss and chromosomal instability [[72]29]. Finally, we assessed the
   ability of p53‐mRNA to inhibit tumor cell proliferation and
   dissemination in xenograft models, highlighting its potential for HGSOC
   therapy.

2. MATERIALS AND METHODS

2.1. Generation of mRNA

   Plasmid templates containing open reading frames of Enhanced Green
   Fluorescent Protein (EGFP), luciferase, or Flag‐p53 were amplified
   using PCR, restoring functionality to the T7 promoter sequence and
   adding a 120A tail via the reverse primer. The primer pair used for
   cloning TP53 into Flag vector are: 5’‐ATGGAGGAGCCGCAGTCAGATCC‐3’,
   forward primer; 5’‐TCAGTCTGAGTCAGGCCCTTCTG‐3’, reverse primer. These
   constructs were capped (Cap 1) using CleanCap® AG (Tebu‐Bio, Offenbach,
   Hessen, Germany) in a reaction where uridine‐5’‐triphosphate was fully
   substituted with Pseudo‐U. After the IVT, DNase (Tebu‐Bio, Offenbach,
   Hessen, Germany) was used to remove any residual DNA template from the
   reaction. Thorough purification of mRNAs included the use of DNase to
   degrade the DNA template and a spin column (silica membrane; Tebu‐Bio,
   Offenbach, Hessen, Germany) to remove the digested template, excess
   nucleoside triphosphates (NTPs), salts, and unused capping analogs. A
   Qiagen RNeasy (silica membrane) spin column was used to remove the
   digested template, excess NTPs, salts, and unused capping analogs.
   Antarctic phosphatase (Tebu‐Bio, Offenbach, Hessen, Germany) was then
   used to cleave residual immunogenic 5′ triphosphate from any uncapped
   material. A second RNeasy spin column was used to remove phosphatase
   from the reaction. Purified mRNA was purchased from TriLink (Tebu‐Bio).

2.2. Patients and tissue samples

   This study was conducted according to the “REporting recommendations
   for tumor MARKer prognostic studies” [[73]30]. Following the
   acquisition of informed consent, samples of ovarian carcinoma were
   obtained from individuals diagnosed with epithelial ovarian cancer at
   the Department of Gynecology at Goethe University Hospital in Frankfurt
   am Main, Germany from January 2015 to December 2022. The validation of
   these samples involved assessment by 3 pathologists, with exclusion
   criteria applied to patients with alternative tumor types. To establish
   primary patient‐derived OC cell cultures, we analyzed samples from 8
   patients who underwent surgical resection (Supplementary Table [74]S1).
   Sufficient archival materials for cell culture analyses were available
   for samples with validated diagnoses. The local research ethics
   committee approved the studies on human tissue (permit number:
   SGO‐1‐2017), and the samples were processed anonymously. The assessment
   of progression‐free survival (PFS) extended from the surgical procedure
   to the point of disease relapse or progression. Cases with a PFS
   duration of fewer than 6 months were designated as chemoresistant,
   whereas durations surpassing this threshold were characterized as
   chemosensitive.

2.3. Cell lines, primary cells, and transfection

   Human ovarian cancer cell lines SKOV3, OVCAR‐3, OVCAR‐4, OVCAR‐5, and
   OVCAR‐8 were purchased from the DCTD Tumor Repository (Bethesda, MD,
   USA), maintained in RPMI 1640 medium (Thermo Fisher Scientific,
   Dreieich, Hessen, Germany), containing 10% fetal bovine serum (FBS;
   Thermo Fisher Scientific) and 1% penicillin‐streptomycin (#17060063,
   Thermo Fisher Scientific). All cell lines were cultured in a humidified
   incubator of 5% CO[2] at 37°C.

   Primary cells were isolated from OC tissues derived from in‐house
   surgery using a tumor dissociation (Max Miltenyi 130‐095‐929, Bergisch
   Gladbach, NRW Germany) and tumor cell isolation (Max Miltenyi
   130‐108‐339) kits, according to the manufacturer's instructions.
   Briefly, the samples were stored in a solution comprising Advanced
   DMEM/F12 (Thermo Fisher Scientific), 2 mmol/L HEPES (Thermo Fisher
   Scientific), 1 × GlutaMAX‐I (Thermo Fisher Scientific), and 200 U/mL
   penicillin/streptomycin (Thermo Fisher Scientific). Upon arrival at the
   laboratory, the tissue specimens were put in a sterile petri dish on
   ice. Following this, the tumor was cut into a sterile 10 mm square. A
   section of the dissected tissue was stored at −80°C in 2 mL tubes using
   Recovery Cell Culture Freezing Medium (Thermo Fisher Scientific) for
   subsequent utilization. In the process of preparing organoids, tissue
   samples were collected in 50 mL tubes with HBSS 1 × (Thermo Fisher
   Scientific) and incubated on ice for of 5 minutes.

   We performed DNA transfections using jetPEI Transfection Reagent
   (Polyplus, Illkirch, Bas‐rhin, France) or LTX (Invitrogen, Waltham, MA,
   USA) (1 μg IVT mRNA: 2 μL Lipofectamine Messenger MAX Transfection
   Reagent [Invitrogen]), according to the previously described [[75]31].
   mRNA transfections were performed using the Lipofectamine Messenger MAX
   Transfection Reagent (Invitrogen), according to the manufacturer's
   instructions. Ovarian cancer cells were obtained by cell scraper and
   then lysed in radioimmunoprecipitation assay, RIPA buffer (Waltham, MA,
   USA). After centrifuging for 5 min at 15,000 ×g at 4°C, cytoplasmic
   protein was collected in the supernatant.

2.4. Colony formation assay

   Two thousand cells were seeded into 6‐well plates. The transfected
   cells were washed and incubated in fresh medium for 2 weeks. The
   colonies were fixed using 70% ethanol and stained with 0.5 μg Coomassie
   Brilliant Blue (Biorad, Feldkirchen, Bavaria, Germany). The number of
   grown colonies was counted, and images were captured using an
   AxioObserver Z1 microscope (Zeiss, Göttingen, Lower Saxony, Germany)
   and ChemiDoc MP system (BioRad, Hercules, CA, USA).

2.5. Organoids

   Following the in‐house surgery, tissues were collected in a storage
   medium (Dulbecco's Modified Eagle's Medium [DMEM; Thermo Fisher
   Scientific, 1% penicillin‐streptomycin [Thermo Fisher Scientific]) on
   ice and processed immediately. Fresh tissues were cut into small pieces
   (2‐4 mm) and rinsed 3 times with wash medium (500 mL DMEM, 1%
   penicillin‐streptomycin, and 2 g bovine serum albumin [BSA; Thermo
   Fisher Scientific]). Subsequently, the sample was left on ice for 2 min
   to settle, and the supernatant was removed. Finally, the tissue was
   kept in 5‐20 mL digestion medium (20 mL Advanced DMEM, 20 g BSA, and 10
   μmol/L Y‐27632), depending on the sample size, for up to 30 min at
   37°C. Every 5‐10 min, the sample was shaken and the stage of digestion
   was examined under a microscope. The digestion process was terminated
   with an equal volume of wash medium, as soon as the tissue was digested
   into single cells or small cell clusters. After that, the cell
   suspension was filtered through a 100 μm‐filter and spun down at 200 ×
   g for 5 min at 4°C, and the supernatant was then removed. Red blood
   cells as components of the pellet were lysed with
   ammonium‐chloride‐potassium buffer (Red lysis buffer: 0.15 M NH[4]Cl,
   10 mmol/L KHCO[3], and 0.1 mmol/L ethylenediaminetetraacetic acid) for
   2‐3 min at room temperature and transferred to an Eppendorf tube. After
   a second wash step (200 × g, 5 min, 4°C), the pellet was resuspended in
   a wash medium (#356231, Corning, Kaiserslautern, NRW, Germany) and
   mixed with a triple volume Base Membrane Extract (BME; cat. # BME
   001‐05, Cultrex UltiMatrix Reduced Growth Factor Basement Membrane
   Extract, R+D Systems, Wiesbaden, Hessen, Germany), or Corning Matrigel.
   Finally, 3D drops of the mixture (10‐20 μL) were seeded on a 12‐well
   suspension plate. The plate was flipped upside down and incubated in a
   cell culture incubator for 30 min at 37°C, 5% CO[2]. Finally, 1 mL
   growth medium (38.65 mL Advanced DMEM/F12 [Thermo Fischer, Dreieich,
   Hessen, Germany], 10 μmol/L Y‐27632 [Biozol, Eching, Bavaria, Germany],
   0.25 μmol/L A83‐01 [Tocris, Bristol, Somerset, UK], 1 mL 1 × B27
   supplement, 5 mmol/L Nicotinamide [Sigma, Frankfurt, Hessen, Germany],
   1.25 mmol/L N‐acetycysteine [Sigma], 50 ng/mL human Epidermal Growth
   Factor [Peprotech, Cranburry, NJ, USA), 50 ng/mL Neuregulin
   [Peprotech], 5 ng/mL FGF‐Basic [Peprotech, Cranburry, NJ, USA], 10
   ng/mL hFGF‐10 [Peprotech], 200 ng/mL primocin [InvivoGen, San Diego,
   CA, USA], 50 ng/mL R‐Spondin [Peprotech], and 100 ng/mL Noggin [R&D
   system, Mineapolis, MN, USA]), were added to each well. The medium was
   replaced twice weekly. The organoids were passaged according to their
   size and density.

2.6. Western blot (WB) and antibodies

   Cell protein extracts were prepared by lysis in the RIPA buffer (Sigma)
   supplemented with protease inhibitors (complete protease inhibitor
   cocktail, Roche, Mannheim, Rheinland Pfalz Germany). Protein extracts
   (25 μg) were separated using sodium dodecyl sulfate‐polyacrylamide gel
   electrophoresis and transferred onto polyvinylidene difluoride
   membranes using a TransBlot Turbo Transfer System (Bio‐Rad,
   Feldkirchen, Bavaria, Germany). The membranes were blocked with
   tris‐buffered saline with 0.05% Tween^® 20 (Sigma, Frankfurt, Hesen,
   Germany) containing 2% BSA Thermo Fisher Scientific.

   The following antibodies were used at the indicated concentrations:
   mouse monoclonal Plk1 (F‐8:sc‐17783) (1:1000; Santa Cruz, Biotechnology
   Heidelberg, Germany), GFP (B‐2) (1:1000; Santa Cruz), Cyclin A (H‐432)
   (1:1000; Santa Cruz), CDK1 (sc‐54) (1:1000; Santa Cruz), KIF4A
   (sc‐365144) (1:1000; Santa Cruz), c‐Myc (sc‐764) (1:1000; Santa Cruz),
   p53 (sc‐126) (1:1000; Santa Cruz), p73 (sc‐17823) (1:1000; Santa Cruz),
   MDM2 (sc‐965) (1:1000; Santa Cruz), H‐Ras (sc‐30) (1:1000; Santa Cruz),
   Luciferase‐HRP (sc‐57604) (1:100; Santa Cruz), K‐Ras (sc‐30) (1:1000;
   Santa Cruz), CDK4 (sc‐260; Santa Cruz), BRCA‐2 (sc‐293185) (1:1000;
   Santa Cruz), β‐Actin (AC‐15:A1978) (1:200; Sigma‐Aldrich, Taufkirchen,
   Germany), Flag (M2:A8592) (1:1000; Sigma‐Aldrich), PLK1 (35‐206,
   #05‐844) (1:1000; Millipore, Schwalbach, Germany), Cyclin B1 (#4138)
   (1:1000; Millipore), Aurora B (#3094) (1:1000; Millipore), GAPDH (ab
   9485) (1:1000; Abcam, Cambridge, Cambridgeshire, England), Bak (ab
   32371) (1:1000; Abcam), XIAP (ab 28151) (1:1000; Abcam), Aurora A
   (#14475) (1:1000; Cell Signaling, Frankfurt, Hessen, Germany), Cyclin E
   (#4129) (1:1000; Cell Signaling), p53‐pS20 (#9287) (1:1000; Cell
   Signaling), p21 (#2947) (1:1000; Cell Signaling), p27 (#3686) (1:1000;
   Cell Signaling), p16 (#18769) (1:1000; Cell Signaling), PARP (#9542)
   (1:1000; Cell Signaling), Casp.3 (#9665) (1:1000; Cell Signaling), Puma
   (#98672) (1:1000; Cell Signaling), Noxa (#14766) (1:1000; Cell
   Signaling), Fas (#8023) (1:1000; Cell Signaling), TOP2A (#12286)
   (1:1000; Cell Signaling), BIRC5 (#2802) (1:1000; Cell Signaling), Nup98
   (#2598) (1:1000; Cell Signaling), BID (#2002) (1:1000; Cell Signaling),
   PLK3 (#4896) (1:1000; Cell Signaling), Casp.9 (#65832) (1:1000; Cell
   Signaling), Bub1 (#4116) (1:1000; Cell Signaling), cdc25c (#4688)
   (1:1000; Cell Signaling), AKT3 (#14982) (1:1000; Cell Signaling),
   Cyclin D (#55506) (1:1000; Cell Signaling), GSKB (#12456) (1:1000; Cell
   Signaling), pGSK3B(Ser9) (#5558) (1:1000; Cell Signaling), β‐Catenin
   (#8480) (1:1000; Cell Signaling), p‐β‐Catenin(S33/37/T41) (#9561)
   (1:1000; Cell Signaling), Bub1B (A300‐386A) (1:1000; Biomol GmbH,
   Hamburg, Germany), cdc20 (Ls‐C357796) (1:1000; Biozol), Bcl‐2 (#610538)
   (1:1000; BD Biosciences, Heidelberg, Germany), Bax (AF 820) (1:1000;
   R&D, Abingdon, Oxfordshire, UK), and HRP‐conjugated secondary
   antibodies (1:5000; GE Healthcare and Jackson Laboratory, Bar Harbor,
   ME, USA. The ECL WB Substrate (Millipore) was used for detection.

2.7. Cell cycle, cell viability, and proliferation assays

   For cell cycle analysis, cells were harvested, washed with
   phosphate‐buffered saline (PBS), fixed in chilled 70% ethanol at 4°C
   for 30 min, treated with 1 mg/mL RNase A (Sigma‐Aldrich), and stained
   with 100 μg/mL propidium iodine for 30 min. Cell cycle quantification
   was performed using a FACSCalibur instrument and CellQuest Pro software
   (both BD Biosciences).

   The Caspase‐Glo 3/7 assay kit (Promega, Waldorf, Hessen, Germany) was
   used according to the manufacturer's instructions. The measured
   luminescence (RLU) was presented as the mean value ± standard deviation
   (n  =  3). Apoptotic loss of membrane asymmetry was analyzed by
   staining with PE Annexin V and 7‐AAD (BD Biosciences) and quantified
   using a FACSCalibur instrument. For the apoptosis assays,
   non‐synchronized cells were treated with the test compounds for the
   indicated times. Cell viability assays were conducted with the Cell
   Titer‐Blue^® Cell Viability assay (Promega) using fluorescence as a
   read‐out, according to the manufacturer's instructions
   (excitation/emission wavelengths: 562 nm/615 nm). The significance of
   differences between populations of data was assessed using the
   Student's two‐tailed test (*P ≤ 0.05; **P ≤ 0.01; ***P ≤ 0.001).
   Proliferation assays were conducted using the Cell Titer‐Blue® Cell
   Viability assay (Promega), as described [[76]32]. In brief, cells were
   seeded in 96‐well plates and transfected with P53‐mRNA or control mRNA.
   After, the fluorescence was measured at the indicated time points using
   a Victor X4 Multilabel Counter (Victor X4, PerkinElmer, Rodgau, Hessen,
   Germany).

2.8. Luciferase assay

   Luciferase assays were performed using the Luciferase Assay System
   (Promega). Briefly, the organoids were harvested and dissociated into
   single cells using TrypLE and mechanical dispersion. Luciferase
   activity in the cells was measured according to the manufacturer's
   instructions. Briefly, the cells were lysed in lysis buffer.
   Afterwards, 20 μL cell lysate were mixed with 100 μL luciferase assay
   reagent (Promega, Waldorf, Hessen, Germany) and the luminescence was
   immediately recorded using a plate‐reading luminometer (Victor X4,
   PerkinElmer, Rodgau, Germany).

2.9. Lactate dehydrogenase (LDH) cytotoxic assay

   The cell culture medium was collected and diluted (1:100) in an LDH
   storage buffer (Promega, Waldorf, Hessen, Germany). For a single
   reaction, 50 μL of diluted cell culture medium was mixed with 50 μL of
   LDH Detection Enzyme Mix (Promega) and 0.25 μL of reductase substrate
   (Promega). After a 60‐min incubation period at room temperature,
   luminescence was recorded using a plate‐reading luminometer (Victor X4,
   PerkinElmer, Rodgau, Germany).

2.10. Chromosome spreads

   The cells were treated overnight with 3.3 μmol/L Nocodazol (Sigma
   Aldrich). The next day, the cells were harvested by mitotic shake off
   and hypotonically swollen in 40% medium/60% tap water for 20 min at
   37°C. The cells were fixed with freshly prepared Carnoy's solution (75%
   methanol and 25% acetic acid), and the fixative was changed several
   times. For spreading, the cells in Carnoy's solution were dropped onto
   pre‐chilled glass slides. Slides were dried at room temperature for
   24 h and stained with 4’,6‐diamidino‐2‐phenylindole (Thermo Fisher
   Scientific). Chromosome number per condition was counted using an
   AxioObserver.Z1 microscope with an HCX PL APO CS 63.0 × 1.4 oil UV
   objective (Zeiss). Graphic representation of the results was performed
   using GraphPad Prism software (GraphPad.com).

2.11. RNA sequencing (RNA‐seq)

   The RNA samples were used for library preparation using the NEBNext
   Ultra RNA Library Prep Kit (Illumina, SanDiego, CA, USA). Indices were
   included for multiple samples. Briefly, mRNA was purified from the
   total RNA using poly T oligo‐attached magnetic beads (Thermo Fisher
   Scientific). After fragmentation, first‐strand cDNA was synthesized
   using random hexamer primers (Thermo Fisher Scientific), followed by
   second‐strand cDNA synthesis. The library was prepared by end‐repair,
   A‐tailing, adapter ligation, and size selection. After amplification
   and purification, insert size of the library was validated using
   Agilent 2100 (Agilent, Santa Calra, CA, USA) and quantified using
   quantitative PCR (qPCR). Libraries were then sequenced on an Illumina
   NovaSeq 6000 S4flow cell with PE150, according to the results of
   library quality control and expected data volume.

   Gene expression data were analyzed as previously reported [[77]33].
   Briefly, unpaired t‐tests were performed between cells transfected with
   p53‐mRNA and the EGFP control, considering group variances. A 2‐tailed
   Student's t‐test was performed for equal variances or Welch's test was
   used for unequal variances. All calculated P‐values were adjusted using
   the Benjamini and Hochberg false discovery rate (FDR) method to account
   for the possibility of false‐positive results due to multiple tests.
   Genes with an FDR q < 0.05 and an absolute fold change (FC) > 1.5 were
   further analyzed using Ingenuity Pathway Analysis (IPA) software
   (Qiagen, Hilden, Bavaria, Germany) to identify overrepresented
   canonical pathways. Heatmaps of RNA‐seq data were normalized to z‐score
   by subtracting the mean from each value and dividing by the standard
   deviation.

2.12. TCF/LEF reporter assay

   The TCF/LEF reporter assay was purchased from BPS Bioscience (#60500)
   (San Diego, CA, USA) and performed according to the manufacturer's
   instructions. Briefly, cells were seeded in a 96‐well plate one day
   prior to transfection to achieve a cellular confluence of 90% on the
   day of the transfection. The following day, a transfection mixture,
   consisting of Lipofectamine Messenger MAX Transfection Reagent (1 μL of
   the TCF/LEF reporter plasmid) and 1 μg of IVT mRNA, was prepared for
   each well. After a 25‐min incubation period at room temperature, the
   mixture was introduced into the cells. Subsequently, a 24‐h incubation
   at 37°C was carried out, and a dual luciferase assay was conducted
   using a plate‐reading luminometer (Victor X4, PerkinElmer, Rodgau,
   Hessen, Germany) to record the assay results.

2.13. Animal experiments

   All animal experiments were approved by the regional council of
   Darmstadt and performed under the supervision of the Division of
   Laboratory Animal Medicine (Goethe University, Medical School, and
   Georg‐Speyer‐Haus). For the OC orthotopic mouse model, 6‐8‐week‐old
   female nude mice (body weight: 25‐30 g) were purchased from Charles
   River Laboratories (Sulzfeld, Bavaria, Germany) and housed under a 12
   h‐light/dark schedule. Twenty minutes before the surgery, the mice
   received analgesia (butorphanol 3 mg/kg and carprofen 2 mg/kg,
   subcutaneously), were kept under isoflurane anesthesia, transferred
   onto a heating pad to prevent hypothermia, inspected with the dorsal
   side up under a microscope, and kept anesthetized during the whole
   procedure. To confirm that the animals were under deep anesthesia, the
   toes of the hind paws were pinched using forceps. The procedure was
   performed in the absence of a pedal reflex. The eyes were covered with
   ointment to avoid drying.

   The optimal point of incision to reach the ovary was located on the
   dorsal side, to the right of the middle, and under the ribs. The area
   was disinfected with 70% ethanol, anthethe skin was then gently lifted
   using forceps and cut into pieces. The skin was separated from the
   abdominal wall using blunt‐ended scissors. Subsequently, the abdominal
   wall was cut, and the abdominal cavity was opened. The ovarian fat pad
   was visible adjacent to the ovary and oviduct under the microscope. The
   fat pad was gently pulled out of the abdominal cavity with forceps,
   placed on a sterile gauze, soaked in saline, and held with forceps to
   maintain a stable position. Approximately 1 × 10^6 Luc‐OVCAR‐8 cells
   were resuspended in 5 μL PBS containing 2% Matrigel and injected into
   the space between the bursa and ovary with a Hamilton 10 μL‐syringe (30
   G needle). Once the needle was inserted into the bursa, a second person
   pushed the plunger, while the surgeon maintained the correct position
   of the needle under a microscope. While injecting the cell suspension,
   expansion of the bursa was observed, indicating that the tumor cells
   remained in the space between the bursa and ovary. The ovarian fat pad
   was then placed back into the abdominal cavity. Finally, both sites of
   the abdominal wall were gently placed together, and the skin was closed
   with wound clips. On days one and two after the surgery, the animals
   received analgesia (butorphanol 3 mg/kg and carprofen 2 mg/kg,
   subcutaneously). One week after the surgery, the wound clamps were
   removed. The surgery was associated with minimal pain, stress, and harm
   to the animals. All the mice survived the surgery and showed no
   behavioral abnormalities after the procedure. For the intraperitoneal
   OC, 6‐8‐week‐old female nude mice (body weight: 22‐28 g) were used. The
   injection area was disinfect infected with 70% ethanol. The skin was
   gently lifted with forceps, and 2 × 10^6 OVCAR‐8 cells were injected.

2.14. In vivo imaging system (IVIS)

   IVIS or bioluminescence imaging (BLI) was applied to monitor tumor
   growth and metastases. The mice were anesthetized with isoflurane.
   Approximately 100 μL of luciferin (Promega, Waldorf, Hessen, Germany)
   was administrated subcutaneously 15 min before measuring the
   bioluminescence. Emission of light signals was determined using an IVIS
   Lumina II Multispectral Imaging System (PerkinElmer, Rodgau, Hessen,
   Germany) [[78]34]. The first measurement was performed on the day of
   cell injection, followed by a second imaging seven days after the
   surgery. Thereafter, imaging was performed once a week.

2.15. Tissue processing for anatomical and histological examination

   For gross anatomical evaluation of the control and treated mice, the
   animals were anesthetized using isoflurane gas, injected with 100 μL
   luciferin, and kept in an isoflurane chamber for 15 min to allow the
   luciferin to dissociate in the organs. The animals were sacrificed by
   cervical dislocation, placed in dorsal recumbency, and fixed with
   needle pins. The skin and body walls were opened at the linea alba, and
   pictures of each animal's body cavities were taken. Subsequently, gross
   macroscopic analyses of the liver, kidneys, uterus and ovaries, spleen,
   omentum, small and large intestines, lungs, and peritoneal surfaces
   were performed. Noticeable changes in organ morphology have also been
   previously reported. The organs and structures were excised and placed
   in six‐well plates (all the organs except the intestine) or Petri
   dishes (small and large intestines) filled with PBS to keep the organs
   moist. IVIS measurements were performed, and all the organs were kept
   in PBS‐buffered formalin for histological and immunohistochemical
   analyses or on dry ice for further molecular analyses.

   Whole ovaries, including the ovarian bursa, and samples from the liver,
   spleen, lungs, and kidneys were carefully dissected. Intestinal tissue
   preparation and Swiss rolling of the small intestine and colon were
   performed as described [[79]35]. The specimens were fixed in 10%
   neutral‐buffered formalin for 48 h at room temperature. The formalin
   was removed by repeated washing with PBS. After dehydration in graded
   ethanol and exchange of ethanol with Xylol, the tissues were
   infiltrated with paraffin and embedded in blocks. Further, 5 μm
   sections taken from different parts of the blocks were stained with
   hematoxylin/eosin and microscopically evaluated with respect to overall
   organ histology and the presence of tumors.

2.16. Statistical methods

   All experiments were performed at least in triplicates. Standardization
   and statistical analysis were performed as previously described
   [[80]36]. Statistical analyses were performed using Microsoft Excel and
   GraphPad Prism. For paired t‐tests, all experimental groups were
   compared with their respective groups. The Student's t‐test and
   Wilcoxon test were used to determine statistical significance between
   two groups, unless otherwise indicated. P value < 0.05 was considered
   statistically different. Significant differences (*P ≤ 0.05, **P ≤
   0.01, and ***P ≤ 0.001) are indicated by asterisks.

3. RESULTS

3.1. Mammalian vector‐based expression of p53 induced apoptosis in HGSOC
cells

   As no sufficiently specific drug is available to rescue functional WT
   p53 in cancer cells bearing a loss‐of‐function p53 mutant, we
   considered reactivating the p53 gene using different experimental
   strategies in HGSOC cell lines and primary OC cells. First, 5 known OC
   cell lines were studied for p53 expression (Supplementary Figure
   [81]S1A). The expression of p53 was not detected in SKOV3 or OVCAR‐5
   cells, in line with previous observations that classified them as p53
   null [[82]37] (Supplementary Figure [83]S1A). In contrast, OVCAR‐3, ‐4,
   and ‐8 cells showed abundant p53 protein expression and low to
   undetectable levels of the classical p53‐target gene, p21. This was due
   to mutations within the DNA‐binding domain at R248Q and L130V or a
   single‐nucleotide mutation at the splice junction (c.376‐1G > A)
   [[84]38, [85]39, [86]40], which inhibited the expression of the major
   negative p53 regulator, MDM2 [[87]41], within a negative feedback loop.
   For further analysis, we selected different HGSOC cell lines, including
   OVCAR‐8 [[88]42], which harbor different genomic alterations that were
   confirmed by sequencing (Supplementary Figure [89]S1B). Exogenous
   expression of Flag‐tagged WT p53 driven by a mammalian vector system
   for 48 h in mutant TP53‐bearing OVCAR‐8 cells significantly reduced
   cell proliferation compared to that in the controls (Supplementary
   Figure [90]S1C). We examined cells transfected with increasing amounts
   of the WT p53‐expressing vector (Supplementary Figure [91]S1D) by FACS
   24 h after transfection and observed a moderate increase in the
   percentage of apoptotic cells in the sub‐G1 phase (Supplementary Figure
   [92]S1E), accompanied by increased levels of caspase 3/7 activity and
   annexin staining (Supplementary Figure [93]S1F). Western blot
   experiments for the detection of cleaved caspase‐3 and PARP
   corroborated increasing apoptosis (Supplementary Figure [94]S1D). To
   further assess the cell cycle progression, in the WB experiments, we
   observed an upregulation of CDK inhibitors (p16 and p27) and p21 upon a
   vector‐prone increase of p53 expression (Supplementary Figure [95]S1D).
   The expression of PLK1 and CDK1, two key players in cell cycle
   regulation, was reduced at higher amounts of transfected p53‐mRNA
   likely due to increased apoptosis (Supplementary Figure [96]S1D‐F).
   Next, we demonstrated the cytotoxicity of WT p53 in a dose‐dependent
   manner through in vitro assays assessing the colony‐forming ability
   compared to the controls (Supplementary Figure [97]S1G). Similar
   experiments in additional HGSOC cell lines (OVCAR‐3 and ‐5)
   corroborated growth inhibition, cell cycle arrest, induction of
   apoptosis, and reduction in colony‐forming ability upon WT p53
   expression (Supplementary Figure [98]S2). Collectively, these data
   indicate that rescuing p53 function by vector‐based re‐expression of WT
   p53 induced cell cycle dysregulation and death in HGSOC cells and
   occurs in a dose‐dependent manner.

3.2. Design, generation, and characterization of IVT mRNAs in HGSOC cell
lines

   Recently, the therapeutic use of IVT mRNA as a vaccine has fueled great
   hope in the fight against infectious diseases. However, whether this
   class of biological agents is suitable for improving precision medicine
   targeting cancer remains elusive. To address this unanswered question,
   we used an IVT system based on a bacterial vector including a T7 RNA
   Polymerase promoter for the transcription of mRNAs (human p53, firefly
   luciferase [Luc], and EGFP), considering aspects of translational
   efficacy and decay rates of mRNA [[99]26] (Supplementary Figure
   [100]S3A). In vitro‐transcribed mRNAs like IVT‐Flag‐tagged p53‐mRNA,
   dubbed here as p53‐mRNA, were analyzed using agarose gel
   electrophoresis for quality control (Supplementary Figure [101]S3B).

   Based on the clinical advantages of liposomes as carriers of
   chemotherapeutics over conventional chemotherapy due to reduced
   therapeutically side effects and increased anti‐cancer activity
   [[102]43, [103]44], we exploited a liposomal system using Lipofectamine
   Messenger MAX Transfection Reagent^® by testing its supportive quality
   for the delivery of IVT‐mRNA (EGFP‐mRNA, Luc‐mRNA) to the HGSOC cells.
   To quantitatively study the uptake of the control EGFP‐mRNA, EGFP
   expression was measured using FACS as a surrogate marker for
   transfection efficiency. At 24 h following transfection with EGFP‐mRNA
   (0.1‐2 μg), a dose‐dependent increase in EGFP expression was detected
   (Supplementary Figure [104]S3C). Although 0.25 μg EGFP‐mRNA was
   sufficient for expression in 90% of cells, at higher doses (0.5‐2 μg),
   97%‐100% of cells showed EGFP fluorescence, supporting an excellent
   transfection efficiency in the HGSOC cells (Supplementary Figure
   [105]S3C). This observation was confirmed by WB, which showed high
   levels of EGFP at doses of ≥ 0.25 μg EGFP‐mRNA (Supplementary Figure
   [106]S3C). In time kinetics, a high percentage of EGFP‐expressing cells
   (92.2%‐99.7%) persisted for 10 days following transfection and declined
   to baseline levels (≤5.5%) on day 15 (Supplementary Figure [107]S3D).
   The EGFP expression was corroborated for a period of 7 days by WB
   (Supplementary Figure [108]S3D). As a second control IVT‐mRNA, Luc‐mRNA
   transfection was monitored by WB, and activity measurements showed high
   enzymatic activity in cells transfected with 0.25‐2 μg Luc mRNA
   (Supplementary Figure [109]S3E).

   To further validate the usefulness of our approach for the expression
   of a protein of choice driven by IVT mRNA in the HGSOC, we tested
   additional cell lines (OVCAR‐3, ‐4, ‐5, and‐8) and primary cells using
   immunofluorescence (Supplementary Figure [110]S3F), FACS (Supplementary
   Figure [111]S3G), and WB (Supplementary Figure [112]S3H). Strong
   expression with 72% and 86% could be detected in OVCAR‐5 and primary
   HGSOC cells, respectively, and very strong expression (≥ 95%) in the
   OVCAR‐3, ‐4 and ‐8 cells (Supplementary Figure [113]S3G‐H) transfected
   with 1 μg EGFP‐mRNA. Collectively, different experiments demonstrated
   the high efficiency of liposomal‐based IVT mRNA transfection and
   expression, irrespective of the OC cell line, primary cells, or type of
   mRNA.

3.3. Characterization of mRNA expression in primary human cells and a mouse
model

   We investigated the time kinetics of EGFP and Luc mRNA expression in
   primary HGSOC cells. The EGFP expression using 1 μg mRNA could be
   detected in FACS studies on day one with 85% positive cells decreasing
   to 74% on day 11 and by WB for at least 7 days (Supplementary Figure
   [114]S4A). Upon transfection of Luc‐mRNA, enzymatic activity was
   detectable for 7 days, and WB signals were detectable for 4 days
   (Supplementary Figure [115]S4B). Organoids from primary HGSOC showed
   efficient Luc activity or EGFP expression for up to 5 days
   (Supplementary Figure [116]S4C‐D). To monitor the duration of Luc
   activity in the mice by in vivo BLI, we transfected the OVCAR‐8 cells
   with Luc mRNA (Supplementary Figure [117]S4E) for subsequent
   application to the mice: (I) orthotopic implantation of 1 × 10^6
   OVCAR‐8 cells transfected with 2 μg Luc mRNA; (II) intraperitoneal
   injection of 2 × 10^6 OVCAR‐8 cells transfected with 4 μg Luc mRNA, or
   (III) intraperitoneal injection of 10 μg liposomal Luc‐mRNA
   (Supplementary Figure [118]S4F). Luc activity was detectable for 2‐3
   days in all three experimental settings. Overall, the liposomal IVT
   mRNA was well suited to support efficient protein expression in primary
   human cells in cell culture and In vivo over several days.

3.4. mRNA‐based reactivation of p53 in HGSOC cells induces cell cycle arrest
and cell death

   We aimed to rescue p53 function in different preclinical models by
   transfecting HGSOC cell lines and primary cells with p53‐mRNA. Low
   amounts of p53‐mRNA (10‐250 ng) strongly induced p53 protein expression
   in OVCAR‐8 cells (Figure [119]1A). An increase in cells in G1 was
   observed at 100‐500 ng p53‐mRNA 24 h post‐transfection compared to the
   controls (Figure [120]1B), supporting the reactivation of p53 activity.
   Strong induction of the cell cycle inhibitors, p21, p16, and p27, was
   readily visible at low doses of p53‐mRNA (≥ 0.1 μg) in parallel with
   the downregulation of critical regulators, such as PLK1, Aurora A,
   CDK1, and Cyclin A/B, corroborating our results using a vector‐based
   re‐expression of p53 (Figure [121]1C). These observations correlated
   with a significant drop in cellular viability by ∼2‐fold after 48 h at
   a dose of 0.5 μg (Figure [122]1D), accompanied by the rise of different
   indicators of cell death, such as Puma, Noxa, and Fas, cleaved PARP,
   cleaved Caspase‐3 compared with untreated controls (Figure [123]1C),
   and increased caspase 3/7 activities (Figure [124]1E), and Annexin
   staining (Figure [125]1F). Loss of colony formation ability was induced
   by increasing the concentration of p53‐mRNA (Figure [126]1G). We
   additionally performed a cytotoxicity assay, which confirmed previous
   observations of the impact of IVT P53 mRNA on ovarian cancer cells
   using apoptosis and colony forming assays (Figure [127]1H). Moreover,
   we tested p53‐null HGSOC cells, such as OVCAR‐5, for their sensitivity
   to re‐expression of WT p53. Within 24 h following the p53‐mRNA
   transfection, prominent signs of cell death, including strong PARP
   cleavage (Supplementary Figure [128]S5A), high levels of cells in the
   sub‐G1 phase (Supplementary Figure [129]S5B), and high levels of
   caspase 3/7 (Supplementary Figure [130]S5C) were observed. The
   performed cytotoxicity assays also confirmed previous observations of
   the impact of IVT P53 mRNA on OVCAR‐5 cancer cells using the LDH‐Glo
   cytotoxicity Assay (Supplementary Figure [131]S5D). Overall, our
   observations indicated that rescuing p53 function by transfection of WT
   p53‐mRNA induced cell death in different HGSOC cell lines.

FIGURE 1.

   FIGURE 1
   [132]Open in a new tab

   Liposomal transfection of in vitro‐transcribed WT p53‐mRNA induces cell
   cycle arrest and apoptosis in the HGSOC cell line OVCAR‐8. (A) OVCAR‐8
   cells were transfected with increasing amounts of liposomal IVT
   Flag‐tagged WT p53‐mRNA. Cell lysates were subjected to WB using Flag‐
   and β‐Actin‐antibodies. (B) The dose‐dependent cell cycle distribution
   of Nocodazole‐synchronized cells is represented as a bar graph (upper
   panel; n = 3). Cell lysates were subjected to WB using Flag‐, Cyclin
   E1‐, Cyclin B1‐, PLK1‐, and β‐Actin‐antibodies (lower panel). (C) Cell
   lysates were subjected to WB for key mediators of cell cycle regulation
   using Flag‐, p53 (pS20)‐, p53‐, p21‐, p27‐, p16‐, PLK1‐, Cyclin A1‐,
   Cyclin B1‐, Cyclin E1‐, Aurora A‐, CDK1‐, and β‐Actin‐antibodies (upper
   panel). The same cell lysates were subjected to WB for regulators of
   apoptosis using PARP‐, Caspase‐3‐, Puma‐, Noxa‐, and Fas‐antibodies
   (lower panel). (D) Cells were mock‐ or 0.5 μg p53‐mRNA‐transfected, and
   proliferation was measured using a CellTiter‐Blue Cell Viability assay,
   24 h post‐treatment. ^** P < 0.01, ^*** P < 0.001, Student's t‐test,
   unpaired and two‐tailed. (E) The dose‐dependent Caspase‐3/7 activities
   were determined using a Caspase‐Glo 3/7 assay. ^*** P < 0.001,
   Student's t‐test, unpaired and two‐tailed. (F) Graphical representation
   of Annexin V positive (%) cells. FACS analysis of transfected cells,
   gated for Annexin V positive (early apoptotic), and Annexin V/7AAD
   positive (late apoptosis) populations is demonstrated. n  =  3 for each
   dose; ^** P < 0.01, ^*** P < 0.001. (G) Representative images show the
   number of 3D colonies (dose‐dependent) on day 10 (left panel). The
   dose‐dependent distribution of colonies is represented as bar graph
   (right panel; n = 3). ^* P < 0.05, ^*** P < 0.001, Student's t‐test,
   unpaired and two‐tailed. (H) Graphical representation of the cell
   cytotoxity caused by increasing concentrations of transfected p53 mRNA.
   n  =  3 for each concentration; ^* P < 0.5, ^**P < 0.01.Aurora‐A,
   Aurora Kinase A; CDK1, cyclin‐dependent kinase 1; FACS,
   Fluorescence‐Activated Cell SortingA; Fas, Tumor necrosis factor
   receptor superfamily member 6; HGSOC, High‐Grade Serous Ovarian Cancer;
   IVT, in vitro‐transcribed; Noxa, NADPH oxidase activator 1; p53, tumor
   protein p53; PARP, Poly (ADP‐ribose) polymerase; PLK1, Polo‐like kinase
   1; WB, western blot.

3.5. Reinstating p53 reduces numerical chromosomal instability (nCIN) in the
HGSOC cell line OVCAR‐8

   Chromosomal instability is a fundamental feature of OC and most human
   OC cells exhibiting aneuploidy [[133]45, [134]46, [135]47]. The OVCAR‐8
   cell line is classified as a hyperdiploid cell line with a high
   numerical complexity and modal number of chromosomes ranging between
   50‐59 [[136]45]. Thus, we aimed to determine whether restoring
   functional WT p53 could reduce the number of chromosomes in cells to a
   near‐diploid status. Therefore, the cells were treated every two days
   for 28 days with low doses of p53‐mRNA, and the distribution of
   chromosomes was assessed at the end of the incubation period
   (Figure [137]2A‐B). The mean chromosome number was reduced from 53 to
   47 upon restoration of WT p53 (Figure [138]2C‐D).

FIGURE 2.

   FIGURE 2
   [139]Open in a new tab

   Restoring p53 function reduces nCIN in OVCAR‐8 cells. (A) Treatment
   schedule. OVCAR‐8 cells were kept in culture for 28 d and treated with
   200 ng p53‐mRNA every 48 h. Metaphase spreading of untreated controls
   and treated OVCAR‐8 cells were prepared at the end of the incubation.
   Chromosomes were stained with Hoechst. (B) Western blot of p53‐mRNA
   expressing OVCAR‐8 using Flag‐ and β‐Actin‐antibodies. (C) Histogram
   plots of the distribution of chromosome numbers in controls and cells
   treated with 200 ng p53‐mRNA. (D) Quantification of the number of
   chromosomes in cells with/without p53‐mRNA‐treatment (upper panel). The
   results were analyzed statistically, mean ± SD, ^**** P < 0.001.
   Representative images of metaphase chromosome spreads (lower panel).
   Green, H2AX pSer139; blue, DAPI. Scale bar = 20 μm. (E) IF images
   showing OVCAR‐8 and OVCAR‐8 + mRNA p53 cells with H2AX pSer193 foci
   staining 15 days post‐treatment with mRNA p53 (upper left panel). Scale
   bar = 20 μm. Quantification of H2AX pSer139 foci 14 days post‐treatment
   with mRNA p53 (lower left panel). The results are presented as mean ±
   SEM, n = 85 cells, ^*** P ≤ 0.001. OVCAR‐8 cells were kept in culture
   for 14 days and treated with 200 ng p53‐mRNA every 48 h (lower right
   panel). Cell lysates were subjected to WB using Flag‐ and
   β‐Actin‐antibodies.

   d, day; H2AX pSer 139, H2A histone family member X phosphorylated at
   Ser 139; nCIN, numerical chromosomal instability; p53, tumor protein
   p53; SD, standard deviation; IF, Immunofluorescence; SEM, standard
   error of the mean.

   The HGSOC subtype is characterized by increased replicative stress
   [[140]48], resulting from various factors, including frequent loss of
   TP53 [[141]49]. This loss of TP53 function contributes to replicative
   stress by compromising the DNA damage response mechanisms [[142]49],
   which may partly explain the increased genetic instability observed in
   the OVCAR‐8 cells. To quantify the extent of replicative stress in the
   OVCAR‐8 cells before and after the restoration of p53 function, we
   assessed Ser139‐phosphorylated H2AX foci (γ‐H2AX) as a reliable marker
   of replicative stress and the presence of DNA double‐strand breaks
   [[143]50] after 7 and 14 days. We found that the baseline level of
   γ‐H2AX foci decreased significantly over the treatment period from an
   average of 8 to 3 foci after 14 days in the OVCAR‐8 and p53
   mRNA‐treated OVCAR‐8 cells (Figure [144]2E). This result provided
   further evidence for the critical role played by p53 in maintaining
   genomic integrity in the HGSOC cells, most likely by counteracting
   replicative stress.

3.6. mRNA‐based rescue of p53 function in primary HGSOC cells induces cell
cycle arrest and cell death

   To further validate the rescue of p53 expression in a preclinical
   scenario, we compared the effects of p53‐mRNA transfection in primary
   normal ovarian and HGSOC cells. The expression of p53 was detected in
   all the primary samples, except the keratinocytes (Figure [145]3A). The
   first comparison of HGSOC and corresponding normal ovarian tissue from
   the same patient revealed higher expression of p53 in tumor cells,
   despite equal amounts of transfected p53‐mRNA (Figure [146]3A).
   Analysis of additional patient‐derived tissues confirmed that the
   expression of p53 and caspase 3/7 activity in all the tumor samples
   (Figure [147]3B) were significantly higher than the corresponding
   levels in the normal ovarian cells. In contrast, nearly the same levels
   of EGFP expression were observed in the normal and tumor cells upon
   transfection with equal amounts of IVT‐EGFP mRNA (Supplementary Figure
   [148]S6A). After testing the transfection efficacy of fluorescent mRNA
   in the normal and HGSOC patient‐derived cells, no difference was
   detected (Supplementary Figure [149]S6B), suggesting that different
   transfection efficiencies did not seem to be the reason for the
   different p53 levels compared to the primary cells (HGSOC vs. normal).

FIGURE 3.

   FIGURE 3
   [150]Open in a new tab

   Liposomal transfection of p53‐mRNA results in higher p53 expression in
   primary HGSOC cells compared to their normal counterparts. (A) Primary
   human cells of different origins (tumor, normal) and OVCAR‐8 cells were
   transfected with p53‐mRNA. Cell lysates were subjected to WB using
   Flag‐ and β‐Actin‐antibodies. (B) Primary human HGSOC cells (tumor,
   normal) derived from patients P3‐8 were transfected with 0.5 μg
   p53‐mRNA. Cell lysates were subjected to WB using Flag‐ and
   β‐Actin‐antibodies (upper panel). (lower left) Expression efficacy was
   depicted as bar graphs (lower left panel). Caspase 3/7 activities were
   determined using a Caspase‐Glo 3/7 assay (lower right panel). ^** P <
   0.01, ^*** P < 0.001. Student's t‐test, unpaired and two‐tailed. (C)
   Lysates of mock‐ or p53‐mRNA‐transfected normal ovarian or tumor HGSOC
   (patient 2) cells were subjected to WB using Flag‐, p53‐, p21‐, CDK1‐,
   p27‐, p16‐, and β‐Actin‐antibodies (left panel). The normalized levels
   of p53 expression are represented as a bar graph (right panel). (D) The
   same cell lysates were subjected to WB using PARP‐, Puma‐, and
   Noxa‐antibodies (left panel). Caspase‐3/7 activities of mock‐ or
   p53‐mRNA‐transfected normal and tumor HGSOC cells were determined using
   a Caspase‐Glo 3/7 assay (right panel; n = 3). (E) Representative images
   show the 3D colonies of mock‐transfected cells on day 10 (left panel).
   Representative images show the 3D colonies of p53 mRNA‐transfected
   cells on day 10 (right panel; n = 3). ^* P < 0.05, ^** P < 0.01, ^*** P
   < 0.001. Wilcoxon test, unpaired and two‐tailed. (F) Graphical
   representation of the cell cytotoxicity caused by increasing
   concentrations of transfected p53 mRNA. n  =  3 for each concentration;
   ^*P < 0.05. (G) Representative images show organoids on day 4 after
   transfection with 2 μg p53 mRNA (left panel). Lysates of HGSOC
   organoids (patient 3) mock‐ or p53‐mRNA‐transfected cells were
   subjected to WB using Flag‐ and β‐actin‐antibodies (middle left panel).
   Organoid size was determined using ImageJ (middle right panel; n = 30).
   ^*** P < 0.001; Student's t‐test, unpaired and two‐tailed. Caspase‐3/7
   activities of HGSOC organoids mock‐ or p53‐mRNA‐transfected were
   determined using the Caspase‐Glo 3/7 assay (right panel; n = 3). ^* P <
   0.05; Student's t‐test, unpaired and two‐tailed. CDK1, cyclin‐dependent
   kinase 1; HUVEC, human umbilical vein endothelial cells; N, normal
   tissue; Noxa, NADPH oxidase activator 1: P (1‐8), patient‐derived
   material (n = 8 patients); p53, tumor protein p53; PARP, Poly
   (ADP‐ribose) polymerase; Puma, p53 upregulated modulator of apoptosis;
   T, tumor tissue.

   The WB examination revealed the upregulation of the inhibitors, p21
   [[151]51], p16, and p27 (Figure [152]3C) and apoptosis
   (Figure [153]3D), which was stronger in the p53‐mRNA‐transfected
   primary HGSOC cells compared to their normal counterparts, correlating
   with a more pronounced reduction in the number of HGSOC colonies
   (Figure [154]3E), which was corroborated by the LDH‐Glo cytotoxicity
   assay (Figure [155]3F). Analysis of different HGSOC patients‐derived
   organoids also demonstrated a significant decrease in volume and an
   increase in caspase 3/7 activity upon p53‐mRNA‐treatment under 3D cell
   culture conditions (Figure [156]3G). In summary, although the
   expression of p53 reduced the viability of both types of primary cells
   (normal vs. HGSOC), the effect was significantly stronger in the
   primary HGSOC cells compared to that in the normal cells, which could
   be due to elevated levels of p53 in the tumor cells.

3.7. Transcriptome alterations in OVCAR‐8 cells upon rescue of WT p53
functions

   Our observations of reduced proliferative activity and increased cell
   death in the cell lines and primary cells of HGSOC origin support the
   model of rescued p53 function upon p53 re‐expression by mRNA
   transfection. To deepen our understanding of the underlying biological
   processes and address this aspect in a relatively unbiased manner, we
   transfected the OVCAR‐8 cells with 1 μg p53‐mRNA or mock control (EGFP
   mRNA) in three independent experiments. In total, 25,540 human genes
   were profiled by RNA‐seq 24 h after the transfection (Supplementary
   Figure [157]S7A‐B). Both groups were compared using unpaired t‐tests,
   considering group variance. Transcriptomic analysis revealed that 3,061
   transcripts were upregulated, and 1,662 were downregulated
   (Figure [158]4A‐B). These significant genes were further analyzed with
   the IPA software to identify overrepresented canonical pathways
   associated with the p53‐mRNA transfection (Figure [159]4C,
   Supplementary Figure [160]S7C).

FIGURE 4.

   FIGURE 4
   [161]Open in a new tab

   Identification of differentially expressed genes in
   p53‐mRNA‐transfected OVCAR‐8 cells compared to mock‐transfected
   counterparts. (A) Heatmap of normalized gene expression represented by
   z scores of OVCAR‐8 cells treated with mock (EGFP mRNA) or p53‐mRNA
   from three independent RNA‐seq experiments (Student's t tests or Welch
   tests, depending on the variance, corrected by FDR q adjusted P‐value <
   0.05 and absolute FC > 1.5). (B) Volcano plot showing the DEGs in the
   RNA‐seq between OVCAR‐8 cells treated with EGFP vs. p53‐mRNA (red dots
   = upregulated genes, blue dots = down‐regulated genes). (C) Selected
   overrepresented Ingenuity pathways from pathway enrichment analysis of
   DEGs with FDR‐adjusted P‐values < 0.05 and FC > 1.5. Pathways with a
   positive z score, representing a global upregulation of the pathway
   genes, are depicted in orange; pathways with a negative z score,
   indicating a global gene downregulation in the pathway, are depicted in
   blue; and pathways with no specific pattern are depicted in gray. (D)
   Heatmap of all genes corresponding to the “p53 signaling pathway” from
   IPA, indicating the P‐value, FDR‐adjusted P‐value and FC for the
   comparison of EGFP with p53‐mRNA. Genes with validation by WB are
   highlighted in red. (E) Cell lysates were subjected to WB for p53
   signaling pathways using p21, MDM2, Puma, p73, Fas, and CDK1
   antibodies. CDK1, cyclin‐dependent kinase 1; DEG,
   differentially‐expressed gene; EGFP, enhanced green fluorescent
   protein; Fas, tumor necrosis factor receptor superfamily member 6; FC,
   fold change; FDR, false‐discovery rate; IPA, Ingenuity Pathway
   Analysis; MDM2, double minute 2 homolog protein; Puma, p53 upregulated
   modulator of apoptosis; Noxa, NADPH oxidase activator 1; p53, tumor
   protein p53; RNA‐seq, RNA sequencing.

   Consistent with the pro‐apoptotic functions of p53 as a tumor
   suppressor involved in the induction of cell cycle arrest and
   apoptosis, our IPA revealed an overrepresentation of several related
   pathways upon p53‐mRNA transfection, such as “p53 signaling”,
   “apoptosis signaling”, and several “cell cycle” pathways
   (Figure [162]4C). To confirm these findings, we investigated all genes
   involved in the IPA “p53 signaling” pathway and we realized that most
   of them (51/81, 63%) were differentially expressed with FDR q < 0.05
   and absolute FC >1.5 (Figure [163]4D). Notably, most known downstream
   targets of p53 signaling (96%) were coordinately upregulated by mRNA
   transfection of p53‐mRNA (Figure [164]4D). We further validated these
   results by performing WB of selected proteins from the p53 signaling
   pathway, including p21 (CDKN1A), MDM2, Puma (BBC3), p73, FAS (CD95),
   and CDK1 (Figure [165]4E). Their expression in the OVCAR‐8 cells was
   substantially upregulated or downregulated by p53‐mRNA in agreement
   with transcriptomic analysis. Most genes in the apoptosis signaling
   pathway (44/68, 65%) were significantly deregulated (Supplementary
   Figure [166]S8A). We extensively validated 12 of these proteins using
   WB (Supplementary Figure [167]S8B) to demonstrate that p53‐mRNA
   transfection induces apoptosis in the OVCAR‐8 cells.

   Several “cell cycle” pathways, including “cell cycle control of
   chromosomal replication”, “cell cycle: G2/M DNA damage checkpoint
   regulation”, and “cyclins and cell cycle regulation”, were
   significantly overrepresented, as well as “Mitotic roles of Polo‐like
   kinase” [[168]52, [169]53], and they showed a global downregulation
   (negative z score) (Figure [170]4C). Most of the genes from these
   pathways were differentially expressed (Supplementary Figure
   [171]S9A‐B) and validated using WB (Supplementary Figure [172]S9C),
   confirming the downregulation of the cell cycle and mitotic roles of
   PLK1. Remarkably, the cellular senescence pathway was also upregulated
   and may contribute to cell death along with the induction of apoptosis
   (Figure [173]4C).

   Next, we carried out a transcriptome comparison between the primary
   HGSOC cells from patient one and the OVCAR‐8 cells, of which both were
   p53‐mRNA‐transfected (Figure [174]5A). The high variability among the
   primary samples resulted in low significance after FDR correction for
   multiple testing. Thus, we considered the significance of the
   transcripts of primary HGSOC cells with P < 0.01 and FC > 1.5. A total
   of 156 transcripts were upregulated, and 115 were downregulated
   (Figure [175]5B‐C). Several genes from the primary HGSOC cells were
   significant and similar to the pathways identified in the OVCAR‐8 cells
   (Figure [176]5D), supporting the relevance of both datasets.

FIGURE 5.

   FIGURE 5
   [177]Open in a new tab

   Analysis of differentially expressed genes in primary HGSOC cells. (A)
   Primary HGSOC cells from patient 1 were transfected with p53‐mRNA or
   EGFP‐mRNA. Cell lysates were subjected to WB using EGFP, Flag, and
   β‐Actin antibodies. (B‐C) Heatmap (B) and volcano plot (C) showing the
   DEGs in the RNA‐seq between primary HGSOC cells treated with EGFP vs.
   p53‐mRNA (P value < 0.01 and absolute FC > 1.5; red dots = upregulated
   genes, blue dots = down‐regulated genes). (D) Transcriptome analysis of
   DEGs in common with OVCAR‐8 and primary HGSOC cells for selected
   significant overrepresented Ingenuity pathways. EGFP, enhanced green
   fluorescent protein; FC, fold change; DEG, differentially‐expressed
   gene; HGSOC, high‐grade serous ovarian cancer; RNA‐seq, RNA sequencing;
   TP53, tumor protein p53.

   Consistent with the paramount function of p53 in OC development, our
   IPA showed multiple transcriptional alterations in the “ovarian cancer
   signaling” pathway. Many genes in this pathway (55/76, 72%), including
   Wnt/β‐catenin signaling, were significantly deregulated
   (Figure [178]6A‐B). The Wnt/β‐catenin pathway promotes chemoresistance,
   metastasis, and cancer stem cell self‐renewal in OC by the aberrant
   activation of Wnt/β‐catenin signaling, leading to‐catenin hyperactivity
   [[179]54]. Treatment with p53‐mRNA resulted in upregulation of AXIN1,
   which functions as a scaffolding protein for components of the
   β‐catenin destruction complex, including APC, CK1PP2A, and GSK3β.
   Additionally, deregulation was observed in members of the frizzled
   receptor, Wnt, and TCF/LEF families (Figure [180]6A‐B). Overall, these
   transcriptional alterations led to a moderate reduction of β‐catenin
   (Figure [181]6B) and to the descending activity of Wnt signaling using
   a TCF/LEF reporter assay in the p53‐mRNA‐treated cells
   (Figure [182]6C).

FIGURE 6.

   FIGURE 6
   [183]Open in a new tab

   OC signaling pathway upregulation upon treatment of OVCAR‐8 cells with
   IVT p53‐mRNA.(A) Heatmap of all genes corresponding to the OC Signaling
   pathway from IPA, indicating the P value, FDR‐adjusted P value, and FC
   for the comparison of EGFP with p53‐mRNA in OVCAR‐8 cells. (B) WB
   analysis using Flag‐, EGFP‐, GSK3B‐, p‐GSK3B‐, β‐Catenin‐,
   p‐β‐Catenin‐, AKT3‐, Cyclin D‐, CDK4‐, BRCA2‐, K‐Ras‐, PLK1‐, and
   β‐Actin‐antibodies. (C) TCF/LEF luciferase reporter responses of cells
   transfected with increasing doses of p53‐mRNA. WB analysis using Flag‐
   and β‐Actin‐antibodies. (ser9), phosphorylated Glycogen synthase kinase
   3β AKT3, V‐Akt murine thymoma viral oncogene homolog 3; at serine 9;
   BRCA2, breast cancer gene 2; CDK4, cyclin‐dependent kinase 4; CTNNB1,
   β‐Catenin; EGFP, enhanced green fluorescence protein; FC, fold change;
   GSK3β, glycogen synthase kinase 3β; p‐ GSK3β IPA, Ingenuity Pathway
   Analysis; WB, western blot; IVT, in vitro transcribed; K‐ras, kristen
   murine sarcoma virus 2 homolog; OC, ovarian cancer; p53, tumor protein
   p53; PLK1, Polo‐like kinase 1.

3.8. Orthotopical application of p53‐mRNA‐treated OVCAR‐8 cells resulted in a
dose‐dependent reduction of tumor growth in an HGSOC Xenograft model

   To explore the viability of the p53‐mRNA‐treated OC cells in a
   physiological microenvironment, an orthotopic disease model was used.
   First, we tested whether stable Luc‐expressing OVCAR‐8 cells
   (OVCAR‐8/Luc) showed an altered response to p53‐reactivation. Treatment
   with 100 ng p53‐mRNA induced only moderate cell death, as evidenced by
   the low expression of apoptotic markers/activation, low caspase 3/7
   activity, and moderate inhibition of cell growth, whereas transfection
   with 500 ng p53‐mRNA induced a robust apoptotic response and strong
   drop in colony‐forming efficacy (Supplementary Figure [184]S10A‐C).
   Thus, the stable expression of Luc in OVCAR‐8 cells did not alter the
   response to mRNA‐dependent reactivation of p53 compared to that in WT
   cells.

   One day prior to the xenograft experiment, the OVCAR‐8/Luc cells were
   transfected with mock‐ or p53‐mRNA (100 ng or 500 ng) (Figure [185]7A).
   To establish an orthotopic model, a Matrigel/OVCAR‐8/Luc cell
   suspension was injected into the right ovary of each mouse. The
   determination of tumor volumes at 1× per week for an observation period
   of 10 weeks by BLI demonstrated growth retardation in tumors
   originating from cells treated with 100 ng p53‐mRNA and a complete
   block of cellular proliferation upon treatment with 500 ng p53‐mRNA
   (Figure [186]7B). To verify the luciferase activity in vivo, the mice
   were sacrificed at the end of the study and subjected to
   anatomical/histological examination of all organs by two independent
   veterinary histologists. The inspection of removed ovaries based on the
   determination of volume (Figure [187]7C) and BLI measurements
   (Figure [188]7D) confirmed the dose‐dependent size reduction of the
   ovary tumors by p53 mRNA treatment supporting the In vivo BLI
   measurements (Figure [189]7B). Histological examination revealed huge
   tumor masses and profoundly altered morphology due to the vast
   carcinoma tissue in the right ovaries injected with mock‐treated
   OVCAR‐8/Luc cells. In the 100 ng p53‐mRNA‐treated group, the right
   ovaries exhibited low to medium amounts of tumors (Figure [190]7E, d‐e)
   or normal histology. The right ovaries of the 500 ng p53‐mRNA‐treated
   group and all other organs showed normal histology (Figure [191]7E,
   g‐h, Supplementary Figure [192]S10D). In all treatment groups (mock,
   100 ng, and 500 ng p53‐mRNA), the left ovaries that did not receive
   OVCAR‐8/Luc cells showed normal histology with follicular and luteal
   structures, as well as stromal tissue (Figure [193]7E, c‐f‐i). In the
   mice injected with mock‐treated cells, metastases were detected in the
   intestine; metastatic activity in the 100‐ng group was low and absent
   in the 500‐ng group (Figure [194]7D‐F). Overall, these animal data
   revealed dose‐dependent growth retardation and blocked dissemination of
   p53‐mRNA‐treated OVCAR‐8/Luc cells under orthotopic conditions.

FIGURE 7.

   FIGURE 7
   [195]Open in a new tab

   Application of liposomal p53‐mRNA to OVCAR‐8 cells prevents intrabursal
   growth and dissemination in an orthotopic mouse model. (A) Schematic
   design of the orthotopic murine model. (B) Bioluminescence imaging. One
   day before, Luc‐expressing OVCAR‐8 cells were transfected (mock, 100 ng
   p53‐mRNA, or 500 ng p53‐mRNA) (left panel). Approximately 1 × 10^6
   cells per mouse were injected into the bursa (right ovary) of nude mice
   on day 0 and imaged using the IVIS Lumina system following D‐luciferin
   application. Eight weeks following orthotopic application of OVCAR‐8
   cells (mock‐ or p53‐mRNA‐treated) representative images of the mice
   (one per week) at baseline are shown. Photon flux is indicated in the
   pseudocolored heat map. BLI was performed at the indicated time points
   and data were presented in a graphic (right panel). Photon flux was
   normalized to baseline values obtained from the mice on day 0. ^* P <
   0.05, ^** P < 0.01. Wilcoxon test, unpaired and two‐tailed. (C)
   Representative photographs of removed ovaries 10 weeks after
   intrabursal application of mock‐ or p53‐mRNA‐transfected (100 ng, 500
   ng) OVCAR‐8 cells in mice (left panel, n = 7 for each group). Analysis
   of extracted ovaries. The dose‐dependent tumor volumes are represented
   as bar graph (right panel). ^* P < 0.05, ^** P < 0.01. Wilcoxon test,
   unpaired and two‐tailed. (D) BLI of removed, tumor‐bearing organs 10
   weeks after intrabursal application of mock‐ (n = 7), or
   p53‐mRNA‐transfected (100 ng, 500 ng) OVCAR‐8 cells (n = 7 mice for
   each group) represented as bar graph. ^** P < 0.01. Wilcoxon test,
   unpaired and two‐tailed. (E) Hematoxylin and eosin‐stained sections of
   ovaries injected with p53‐mRNA‐transfected (100 ng, 500 ng) (n = 7) or
   mock‐ (n = 7) OVCAR‐8 cells at the end of the experiment (10 weeks).
   Scale bars, 300 μm. (F) Representative hematoxylin and eosin‐stained
   sections of metastases at the small intestine and colon following
   intrabursal injection of mock‐, or p53‐mRNA‐transfected (100 ng, 500
   ng) OVCAR‐8 cells (after 10 weeks). Scale bars, 300 μm.BLI, Bio‐layer
   Interferometry; IVIS, In vivo imaging system; p53, tumor protein p53.

3.9. Intraperitoneal application of OVCAR‐8 cells followed by intraperitoneal
treatment with p53‐mRNA inhibited dissemination of tumor cells

   To further validate the clinical relevance of p53‐mRNA in metastatic
   HGSOC, we tested whether the scattered tumor cells within the
   peritoneum were accessible to liposomal p53‐mRNA. The intraperitoneal
   administration of drugs for OC was shown to be more efficacious than
   systemic administration and was recommended by the National Cancer
   Institute [[196]55]. Along these lines, we injected 2 × 10^6
   OVCAR‐8/Luc cells intraperitoneally and administered p53‐mRNA 4‐6 h
   later via the intraperitoneal route. The mice were treated for three
   weeks with 2 intraperitoneal injections per week with p53‐mRNA (0.16
   mg/kg) or control. Although the OVCAR‐8/Luc cells treated with the
   control grew exponentially, the intraperitoneal treatment with the
   liposomal p53‐mRNA completely blocked the tumor cell growth
   (Figure [197]8A). During the entire observation period, the body
   weights of the mice in both groups developed normally (Figure [198]8B).

FIGURE 8.

   FIGURE 8
   [199]Open in a new tab

   Intraperitoneal injection of OVCAR‐8 cells followed by intraperitoneal
   treatment of mice with liposomal p53‐mRNA prevents tumor formation and
   organ dissemination in a xenograft mouse model. (A) Bioluminescence
   imaging of luciferase‐expressing metastatic OVCAR‐8 cell‐bearing mice
   upon treatment with liposomal p53‐mRNA (upper panel). Approximately 2 ×
   10^6 OVCAR‐8 cells stably expressing luciferase were i.p. injected
   followed by treatment with liposomal p53‐mRNA for three weeks (2 × 5 μg
   liposomal p53‐mRNA per week). The color scale represents
   p/sec/cm^2/steradian. Tumor burden of control and liposomal
   p53‐mRNA‐treated nude mice determined weekly by bioluminescence over a
   period of 9 weeks (lower panel; n = 8). ^* P < 0.05, ^** P < 0.01, ^***
   P < 0.001. Wilcoxon test, unpaired and two‐tailed. (B) Determination of
   the body weight over the entire observation period. (C) Inspection of
   peritoneal structures and organs for tumor dissemination following
   laparotomy. (D) Bioluminescence imaging of removed murine organs
   (control vs. p53‐mRNA treatment). (E) Hematoxylin and eosin‐stained
   representative sections of ovary enclosed by bursal peritoneum and fat
   pad (a, e), small intestine (b, f), colon (c, g) and fat/peritoneum (d,
   h) of mice ip injected with OVCAR‐8 cells alone (control a‐d) or
   followed by i.p. application of p53‐mRNA (e‐h). In the control group
   animals, prominent tumor cell aggregates are visible associated with
   serosal surfaces and fat (arrows in a‐d) in contrast to the
   p53‐mRNA‐treated animals. Scale bars 300 μm.i.p., intraperitoneal.

   In the control mice receiving intraperitoneal OVCAR‐8/Luc cells and
   intraperitoneal mock treatment, gross anatomical analysis revealed
   tumor masses on the peritoneal surfaces (Figure [200]8C‐D). Most of
   these metastases were weakly attached to the organs. Large tumors were
   also detected in fatty tissues, such as the omentum, ovarian fat pad,
   and mesentery (Figure [201]8C‐D). In particular, larger tumor masses
   attached to the small and large intestines lead to adhesion between
   intestinal structures, implicating strictures of the gut. The IVIS
   measurements of the omentum and intestine revealed the most prominent
   signals, correlating with the histological results (Figure [202]8D‐E).
   In the mice treated with p53‐mRNA, gross anatomical and histological
   analyses did not reveal tumor masses in the peritoneal cavity, and all
   the organs appeared normal upon visual inspection. The IVIS
   measurements and histological inspection did not provide evidence of
   tumor cell dissemination (Figure [203]8D‐E).

4. DISCUSSION

   In this study, we sought to restore p53 functionality in HGSOC using
   IVT p53 mRNA. Re‐expression of WT p53 resulted in a significant
   reduction in cell proliferation, induction of apoptosis, and
   perturbations in cell cycle dynamics in the HGSOC cells. The liposomal
   delivery system demonstrated high transfection efficiency and sustained
   mRNA expression. Furthermore, the p53 IVT‐mRNA transfection in primary
   HGSOC cells yielded similar outcomes, including cell cycle arrest,
   apoptosis induction, and diminished cell viability, compared to normal
   ovarian cells. Transcriptome analysis revealed notable alterations in
   pathways related to p53 signaling, apoptosis, and cell cycle
   regulation. In an in vivo setting, p53 IVT mRNA treatment effectively
   inhibited tumor growth and prevented tumor cell dissemination within
   the peritoneal cavity. These findings underscore the potential utility
   of mRNA‐based strategies for reactivating p53 and combating HGSOC by
   inducing programmed cell death, orchestrating cell cycle checkpoints,
   and reducing tumor progression and metastasis.

   Mutations in p53 occur in up to 96% of patients with HGSOC [[204]13].
   Researchers have attempted to reactivate p53 and suppress OC growth
   using WT p53 gene therapy [[205]56]. However, therapies using WT Ad‐p53
   (Gendicine, Advexin, SCH‐58500), liposomal WT p53, or p53‐specific
   conditionally replicative adenoviral vectors have not yet been approved
   as p53‐based drugs for HGSOC therapy. Recent studies have demonstrated
   that novel IVT‐mRNA‐based therapeutics are powerful tools for
   developing novel vaccines [[206]57]. Here, we explored the potential
   role of IVT mRNA for the reactivation of tumor suppressor functions in
   cancer. We first examined the functional aspects of mRNA expression in
   HGSOC cells using different IVT mRNA species (EGFP, luciferase, and
   Flag‐p53). The translation of eukaryotic mRNA is a precisely controlled
   procedure in which the 5’ Cap, 3’, and 5’ untranslated regions (UTRs)
   contribute significantly to the stability of the mRNA [[207]26]. To
   reach this aim, we used here mRNA co‐transcriptionally capped with
   m7G(5’)ppp(5’) (2’OMeA)pG. The 5’ UTR contains an eIF4G aptamer that
   was previously shown to improve the translation efficacy of IVT mRNA.
   Additionally, in all the three mRNA used here, the four canonical bases
   (ACGU) were replaced by ACGPseudo‐U, which was found to improve the
   functionality of IVT mRNA [[208]58].

   Our IVT‐EGFP‐mRNA with optimized structural features achieved 72%‐99%
   expression in the HGSOC cell lines and up to 83% expression in the
   primary cells using liposomal transfection. High EGFP expression (≥
   90%) lasted over 10 days in the cell lines and 4‐11 days in 2D‐cultured
   primary cells and in tumor organoids for 5 days, suggesting that
   liposomal IVT‐mRNA transfection enables efficient, long‐lasting
   expression in HGSOC cells. Interestingly, comparison of the primary
   HGSOC samples with their normal counterparts, both transfected with
   equal doses, showed much higher p53 expression in the tumor cells,
   resulting in more robust anti‐proliferative effects in tumor cells. The
   p53 was previously found to be present at low levels in resting cells,
   but multiple cancer‐related stress stimuli, such as hyper‐induction of
   the HSP90 system, provoke post‐translational modifications of p53 to
   modulate its protein stability [[209]59]. Moreover, mutant p53 has been
   reported to be involved in the metabolic adaptation of cancer cells by
   sustaining the Warburg effect. By increasing glucose intake in tumor
   cells, mutant p53 hinders autophagy‐dependent degradation caused by
   glucose deprivation [[210]60]. Thus, the reprogramming of cancer cells
   increases the stability of mutant p53 through different mechanisms.
   Exogenous WT p53 expressed upon mRNA transfection may utilize these
   mechanisms for stabilization.

   Our findings support the idea that restoring WT p53 function induces
   cell cycle arrest, apoptosis, and tumor suppression [[211]61].
   Sequencing of RNA revealed coordinated upregulation of multiple p53
   target genes promoting cell death, including p21, MDM2, PUMA, GADD45B,
   FAS, BAX, and FBXW7, and downregulation of centromere‐related
   components such as BUB1, BUB1B, CDC20, APC, and PLK1, along with
   crucial regulators of cell cycle progression (CDK1/Cyclin B1,
   CDK4/Cyclin D, CDK2/Cyclin E, PLK1, and PLK4), providing evidence for
   deregulated cell cycle machinery and cell division. Importantly, these
   key targets agree with those of ongoing clinical trials investigating
   CDK1 and PLK1 inhibitors (ribociclib, rocovitine, palbociclib, and
   volasertib) alone or in combination with standard chemotherapy for OC
   treatment [[212]62]. Thus, p53 reactivation exhibits pleiotropic
   effects; among the most important effects is the simultaneous
   inhibition of multiple OC‐relevant kinases.

   Studies have linked functional loss of p53 to aneuploidy in vitro and
   In vivo [[213]29, [214]63–[215]65]. However, recent studies have
   suggested that the concurrent loss of p53 and p73 directly affects CIN
   in cancer cells [[216]66, [217]67, [218]68]. Notably, both
   transcription factors exhibit shared target genes, among which CDKN1A
   (p21) is critical for the inhibition of CDK1 and CDK2. However, p21
   contributes to cell cycle arrest in the G1 and G2 phases, and its
   absence introduces various mitotic aberrations in tumor cells [[219]51,
   [220]69, [221]70]. Schmidt et al. [[222]71] revealed that the
   inactivation of p53/p73 or the absence of p21 drives CIN in a
   CDK1‐dependent manner. The loss of p21 unleashes CDK1 activity, leading
   to abnormal rates of microtubule assembly and subsequent CIN in cancer
   cells, which typically maintain CIN. Interestingly, our transcriptomic
   data revealed that restoring functional p53 led to significant
   upregulation of p73 and CDKN1A gene expression. Restoration of the
   p53/p73 and p21 axes proved sufficient to mitigate or at least
   partially reverse the observed CIN in surviving parental OVCAR‐8 cancer
   cells and highlights the connection between these factors and CIN.
   Moreover, a recent report indicated that p53 activation enhances DNA
   replication efficiency, whereas its elimination impedes replication
   fork progression [[223]72]. Interestingly, MDM2, a gene product induced
   by p53, also supports DNA replication even in p53‐deficient cells,
   suggesting that sustained MDM2 expression is one of the mechanisms by
   which p53 prevents replicative stress [[224]72]. Thus, in addition to
   its well‐known role in eliminating damaged cells, p53 contributes to
   genome protection during S‐phase by preventing the appearance of
   stalled or collapsed replication forks by preventing DNA damage during
   replication. Furthermore, cisplatin‐induced apoptosis is enhanced by
   p73 overexpression [[225]73, [226]74]. Therefore, IVT‐p53‐mRNA delivery
   can be used with carboplatin, a standard therapeutic agent for HGSOC
   patients [[227]75]. Given the high rate of p53 mutations in HGSOC,
   restoring p53 function may make the cells more susceptible to
   carboplatin, providing an important treatment alternative for further
   investigation.

   We tested 2 xenograft models and demonstrated a dose‐dependent
   reduction in p53‐mRNA during tumor growth. Dissemination of
   mock‐transfected cells in our orthotopic model mirrored the
   dissemination in HGSOC cancer patients [[228]76]. The p53‐mRNA
   treatment prior to orthotopic injection hampered tumor cell growth in a
   dose‐dependent manner within the ovaries and spread to the organs of
   the peritoneal cavity, demonstrating the efficacy of p53‐mRNA under
   orthotopic conditions. The usefulness of p53‐mRNA treatment was also
   corroborated in a second xenograft experiment in which HGSOC cells were
   directly administered to the peritoneal cavity. A significant advantage
   of intraperitoneal chemotherapy is that it prevents hepatotoxicity and
   allows the metabolism of cancer drugs. Upon intraperitoneal
   administration of liposomal p53‐mRNA, a massive dose‐dependent
   reduction in tumor cell growth and dissemination was observed. The
   absence of weight loss and detailed necropsy revealed no tissue damage,
   supporting the nontoxic behavior of p53‐mRNA.

   No approved drugs exist to restore functional p53 in cancer cells
   expressing mutant p53. Overcoming this challenge is crucial,
   particularly for HGSOC. However, p53 gene therapy failed in OC clinical
   trials [[229]56] probably because of neutralizing antibodies against
   viral delivery systems and the dominant‐negative effect of the
   endogenous mutant p53. To explore this aspect, a detailed analysis of
   p53 tetramers in cells treated with IVT‐p53‐mRNA should be the subject
   of future biochemical investigations. New‐generation IVT‐mRNAs with
   optimized translational efficacy, delivered as liposomal IVT‐mRNA, show
   promise for bypassing these obstacles. The IVT‐mRNA offers several
   advantages over DNA‐based drugs, including a high transfection
   efficiency, no risk of insertional mutagenesis, and functional activity
   without nuclear entry. However, the reactivation of p53 in cancer
   therapy faces certain hurdles. Off‐target effects and immune responses
   pose risks requiring further detailed research. Long‐term safety
   concerns, potential resistance, and ethical considerations must be
   addressed. Combating the evolution of cancer cell resistance remains
   challenging. Additionally, the costs and clinical efficacy across
   diverse cancer types and stages are complex. Overcoming these hurdles
   demands ongoing research, technological advancements, and rigorous
   clinical trials to ensure the safe and effective reactivation of tumor
   suppressor genes in cancer treatment. Nevertheless, this study supports
   IVT‐mRNA as an attractive tool for reactivating p53 in susceptible
   cancers for effective cancer therapy.

5. CONCLUSION

   This study demonstrated the effectiveness of the reactivation of the
   TP53 tumor suppressor gene in HGSOC by IVT mRNA. Liposomal IVT mRNA
   resulted in efficient and sustained expression of WT p53 and precise
   rescue of p53 function; thus, inhibiting cell proliferation in HGSOC
   cells, reducing chromosomal instability, and triggering cell death. In
   different mouse models, p53‐mRNA treatment led to a dose‐dependent
   reduction in tumor growth and dissemination within the peritoneal
   cavity. These findings suggested that IVT‐mRNA‐based approaches hold
   promise for reactivating p53 in cancer cells and may offer a valuable
   therapeutic option for HGSOC.

DECLARATIONS

AUTHOR CONTRIBUTIONS

   Monika Raab and Klaus Strebhardt conceptualized the study. Monika Raab,
   Samuel Peña‐Llopis, Mourad Sanhaji, Daniela Fitz, Monika Kressin, Sven
   Becker, and Klaus Strebhardt coordinated the study and wrote the
   manuscript. Monika Raab, Izabela Kostova, Samuel Peña‐Llopis, Daniela
   Fitz, Monika Kressin, Seyed Mohsen Aberoumandi, Evelyn Ullrich, Mourad
   Sanhaji, and Klaus Strebhardt designed and performed experiments,
   interpreted results, and prepared figures. All the authors have read
   and approved the final version of the manuscript.

CONFLICT OF INTEREST STATEMENT

   The authors declare no potential conflict of interest.

CONSENT FOR PUBLICATION

   Not applicable.

ETHICS APPROVAL AND CONSENT TO PARTICIPATE

   All tissues were obtained from patients who had undergone surgery at
   the University Hospital of Goethe University, with the consent of the
   patients and approval of the Goethe University Committee for Ethical
   Review of Research involving Human Subjects (approval number:
   SGO‐1‐2017). All the patients signed an informed consent form. All
   animal experiments were approved by the regional council of Darmstadt
   (V 54‐19 c 18‐FK/1128).

Supporting information

   Supporting Information
   [230]Click here for additional data file.^ (12.8MB, docx)

ACKNOWLEDGEMENTS