Bicistronic transfer of CDKN2A and p53 culminates in collaborative killing of human lung cancer cells in vitro and in vivo


Cancer therapies that target a single protein or pathway may be limited by their specificity, thus missing key players that control cellular proliferation and contributing to the failure of the treatment. We propose that approaches to cancer therapy that hit multiple targets would limit the chances of escape. To this end, we have developed a bicistronic adenoviral vector encoding both the CDKN2A and p53 tumor suppressor genes. The bicistronic vector, AdCDKN2A-I-p53, supports the translation of both gene products from a single transcript, assuring that all transduced cells will express both proteins. We show that combined, but not single, gene transfer results in markedly reduced proliferation and increased cell death correlated with reduced levels of phosphorylated pRB, induction of CDKN1A and caspase 3 activity, yet avoiding the induction of senescence. Using isogenic cell lines, we show that these effects were not impeded by the presence of mutant p53. In a mouse model of in situ gene therapy, a single intratumoral treatment with the bicistronic vector conferred markedly inhibited tumor progression while the treatment with either CDKN2A or p53 alone only partially controlled tumor growth. Histologic analysis revealed widespread transduction, yet reduced proliferation and increased cell death was associated only with the simultaneous transfer of CDKN2A and p53. We propose that restoration of two of the most frequently altered genes in human cancer, mediated by AdCDKN2A-I-p53, is beneficial since multiple targets are reached, thus increasing the efficacy of the treatment.


The complexity of cancer continues to thwart the many attempts to develop effective treatments. Much attention has been given to understanding of genetic alterations in cancer in order to guide the development of highly specific, targeted therapies and immunotherapies [1]. Although critical advances have been made, the fact remains that cancer stems from a combination of alterations that vary not only between tissue types, but also within the tumor itself. Treatments that target a single alteration or pathway may be effective in the short term, but the selective pressure created by the therapy may favor the survival of resistant sub-populations and, in the long run, the therapy fails. However, strategic targeting of multiple pathways may subvert the selective process and bring about effective tumor cell killing [2,3,4].

The contribution of the p53 and pRB pathways in tumor suppression has been clearly delineated. Pharmacologic agents and gene transfer strategies directed at either p53 or CDKN2A (p16INK4A), an important inhibitor of the pRB pathway, have been individually explored. However, the simultaneous targeting of both pathways is an under-represented approach for the control of tumor progression. pRB is known for its central role in coordinating progression through the cell cycle [5]. When hyperphosphorylated (inactivated), pRB releases a variety of sequestered proteins, including E2F, which then signal DNA replication and proliferation. In the G1 phase of the cell cycle, the first step in phosphorylation of pRB is mediated by the CDK4/6-cyclin D complex. Inhibition of this complex by the CDKN2A gene product is a key control point initiated upon replicative and oncogenic stress [5]. In the absence of CDKN2A, activity of pRB may go unchecked and promote cell proliferation.

Treatments specifically targeting the CDKN2A/pRB pathway are gaining ground, but are not numerous. For example, small molecule drugs that inhibit CDK4/6 in order to restore pRB’s tumor suppressor activity have recently been approved by the FDA [5]. Cancer gene therapy involving the CDKN2A/pRB pathway seems a logical choice but has not been studied extensively in the clinic. As CDKN2A activity is geared more towards arrest/senescence rather than death, perhaps its utility in cancer gene therapy has been overlooked. Our previous studies have shown that CDKN2A gene transfer can effectively inhibit the progression of glioma [6] and glioblastoma [7] cell lines, but only when endogenous pRB is intact. We interpret this situation as an indication that CDKN2A is a powerful means to arrest proliferation, but an additional factor may be necessary in order to bring about cell death.

The p53 pathway has an essential role as a sensor of cellular stress and coordinator of the response to genomic anomalies, regulator of metabolism, inhibitor of angiogenesis, and promoter of innate immunity, among other functions [8,9,10]. In the simplest interpretation, p53 exerts its influence by transactivating or repressing a large number of target genes that play a part in cell cycle arrest (CDKN1A or p21) and apoptosis (Bax, Puma, Noxa, Bcl2), to name just a few examples [8, 11]. The loss of p53 can lessen sensitivity to chemo and radiotherapies [12].

Inactivation of p53 may come from the overexpression of MDM2, the E3 ubiquitin ligase that directs p53 for proteasomal degradation, or loss of p14Arf, an alternative gene product of the CDKN2A locus, which blocks the interaction of MDM2 with p53 [13, 14]. This example also points to the dynamic coordination between the pRB and p53 pathways, as CDKN2A locus is essential for both. On the other hand, p53 uses CDKN1A, a pan inhibitor of pRB’s kinases, to block the progress of the cell cycle. The myriad post-translational modifications of p53 required for its proper function, and the proteins which mediate these events, can also be targets for deregulation [15]. Mutation of p53 may lead to the simple loss of activity, but may also confer a gain of function to the mutated protein, including the ability to promote survival, migration, and target gene expression [16]. Dominant negative forms of mutant p53 can inhibit the wild-type protein, suggesting that restoration of p53 activity may be hampered in the presence of a dominant negative mutant [17].

Treatments targeting the p53 pathway are quite numerous and include both pharmacological and gene transfer approaches [18, 19]. Clinical investigation of drugs targeting p53 includes RG7112 and R05503781 that liberate p53 from the grips of MDM2/MDM4 and PRIMA-1MET, which reactivates mutant p53 [19]. Gene therapy with p53 was originally proposed for the treatment of lung carcinoma [20]. Since then, p53 gene therapy has progressed and shown clear signs of benefit, especially for the treatment of head and neck cancers, an indication that, in 2003, was approved for commercialization in China [21].

Mutations in the p53 and CDKN2A genes are among the most common events in cancer [22, 23] and this scenario certainly applies to lung cancer [24]. Since the mechanisms of tumor suppression converge to control both cell cycle progression and cell death it is logical to consider that the reestablishment of both p53 and CDKN2A functions would represent an effective therapeutic tactic.

Even with the crucial roles of p53 and CDKN2A, only a few examples exist where both have been targeted simultaneously in order to induce tumor cell death [25, 26]. Here, we have transferred the CDKN2A and p53 cDNAs to the H1299 human lung cancer cell line utilizing a bicistronic adenoviral vector. Although transfer of either cDNA alone impedes proliferation, the simultaneous transfer of both the CDKN2A and p53 cDNAs acts in a collaborative and rapid manner to halt growth and induce cell death, both in vitro and in vivo.

Materials and methods

Cell culture

The human lung carcinoma cell lines H1299 (-CDKN2A/-p53/+pRB), H358 (-CDKN2A/-p53/+pRB), and A549 (-CDKN2A/+p53/+pRB) were obtained from the Rio de Janeiro Cell Bank and 293 A (human embryo kidney cells transformed with the E1 region of adenovirus) were obtained commercially (Life Technologies, Carlsbad, CA, USA). The cells were maintained in Dulbecco’s Modified Eagle’s Medium (Life Technologies) supplemented with 10% fetal calf serum (Hyclone, UH, USA), 1% penicillin/streptomycin mix (Life Technologies) and incubated at 37 °C in a humidified atmosphere containing 5% CO2.

For the isogenic cell lines derived from H1299, cells were plated and transduced with either the pCLp53(R175H), pCLp53(R248Q), or pCLH2B-GFP retroviral vectors at a multiplicity of infection (MOI) < 1. The next day, antibiotic selection was initiated using medium containing 800 µg/ml G418 (Life Technologies) and cells were incubated until non-transduced controls had died, ~7 days. Cell line derivatives were then expanded and maintained in medium containing 400 µg/ml G418.

Construction of vectors

For the adenoviral vectors, the cDNAs of interest were cloned into the pENTR-2B vector (Life Technologies) before LR recombination with the pAd/CMV/V5-DEST Gateway vector (Life Technologies) as per the manufacturer’s instructions. The enhanced green fluorescent protein cDNA (Clonetech) was isolated from the pCLeGFP retrovirus [27] and inserted in the pENTR vector. Similarly, the monocistronic CDKN2A and p53 vectors were derived upon isolating the respective cDNAs from pCLp16 (CDKN2A) and pCLp53 [7] and inserting these individually in pENTR-2B. For the bicistronic vector, the encephamyelocarditis virus IRES sequence [28] was isolated from pLZIN (kindly provided by John Majors, Washington University) and subcloned in pBluescript II KS(-) (Stratagene). The p53 cDNA was isolated from pCLp53 using NcoI and BamH1 restriction enzymes before subcloning into these sites in pBluescript-IRES. In this way, the p53 cDNA was placed in frame utilizing the start codon present in the IRES element. The IRES-p53 sequence was then placed in pENTR-2B and then the CDKN2A cDNA placed upstream of the IRES element. The established pENTR vectors were then recombined with the adenoviral destiny construct, positive clones identified by restriction mapping and verified by sequence analysis (data not shown). Further details of the cloning procedure are available upon request.

The retroviral vectors encoding mutant p53 were cloned by isolating the p53R175H and p53R248Q cDNAs from pCIS expression vectors [29] and inserting these individually in the pCLXSN retrovirus [30] as per our previous study [7]. The cDNA encoding the H2BeGFP fusion protein (kindly provided by Geoffrey M. Wahl, Salk Instutite, La Jolla, CA) was inserted in the pCLXSN retrovirus, used here as a control vector. Amphotropic retroviral stocks were produced and titers determined by end-point dilution, as described [7, 27]. These resulting cell derivatives were termed H1299H2BeGFP, H1299p53R175H, H1299p53R248Q.

Production and titration of adenovirus

For adenovirus, the constructs were first digested with PacI before transfection in 293 A cells followed by amplification of the viral stock and purification through two rounds of CsCl gradient ultracentrifugation followed by at least three rounds of dialysis against 10 mm trisHCl pH 8.0/2 mm MgCl2/5% sucrose. The purified virus was then aliquoted and stored at −80 °C. Biological and physical titers were determined as per Nyberg-Hoffman et al. [31]. The biological titer (infectious virus particles) was used to calculate MOI. For in vitro experiments, a MOI of 10 was used for the H1299 cell line and a MOI of 30 for the H358 and A549 cell lines.

Immunofluorescence (in vitro)

Ten thousand H1299 cells were plated on coverslips and transduced the next day. After 48 h, the cells were washed with PBS, and fixed with 4% paraformaldehyde for 10 min at 4 °C. Next, the cells were washed with PBS and then permeabilized with 0.1% NP-40 in PBS for 30 min at 37 °C. After washing with PBS, cells were blocked with 1% albumin for 30 min at 37 °C then washed. Cells were then incubated with the primary antibody in a humidified chamber, 4 °C, overnight. The antibodies used were DO-1 monoclonal α-p53, polyclonal α-CDKN2A (p16), H-121 polyclonal α-CDKN1A (p21), polyclonal α-pRB, monoclonal α-Bax, and polyclonal anti-Ki 67 (all from Santa Cruz Biotechnology, Inc. Santa Cruz, CA) and monoclonal α-phospho-RB (Cell Signaling, Danvers, MA, USA). All primary antibodies were diluted 1:100 in PBS containing 1% bovine serum albumin. For the secondary antibodies, anti-mouse Alexa 488, anti-rabbit Alexa 594 and anti-rabbit Alexa 350 (Life Technologies) were diluted 1:500 in PBS before incubation with the cells during 90 min at 4 °C in the dark. Next the cells were incubated for 30 min in 100 μg/mL solution of DAPI (Life Technologies) in PBS. After washing, the coverslips were placed on slides using Vectashield mounting solution (Vector Labroatories, Inc, Burlingame, CA, USA) and visualized by confocal microscopy (LSM510 Meta Zeiss/UV, Zeiss, Jena, Germany).

Growth curve, MTT, and cell cycle

These assays were performed as described previously [32].

Senescence-associated beta-galactosidase (SA-βgal) activity

Cytochemical SA-βgal was detectable at pH 6.0 following a published protocol [33]. In brief, 2 × 103 cells/cm2 were transduced with the mono and the bicistronic tumor suppressor adenovirus and also with the control AdeGFP virus or non-transduced (mock). Nine sixty hours later, the plates were stained for detection of SA-βgal activity. Large flat cells (minimum five times bigger than the controls) with intense indigo blue staining were scored as senescent cells. The number of cells was counted in a minimum of 10 fields totaling > 500 cells per treatment. The percentage of senescent cells was then calculated.

Caspase 3/7 activity

Activation of caspases 3/7 was revealed upon staining with CellEvent caspase 3/7 green detection reagent (Life Technologies) and Hoechst 33342 (Life Technologies) before automated imaging and quantification using an In Cell analyzer 2200 (GE Lifescience, Freiberg, Germany) and the data analyzed by In Carta software, version 1.8.3257347 (GE Lifescience).


Total RNA was isolated using the RNAeasy Kit (Qiagen) 24 h after transduction with adenoviral vectors AdLacZ, AdpCDKN2A, Adp53, or AdpCDKN2A-I-p53. Random primers were used to synthesize cDNA. Specific primers were designed using the software Primer [34] and had their specificity confirmed through GenBank’s BLAST program ( The sequence (5′–3′) of the forward primers were: CCCTTTTGCTTCAGGGTTTC, GAGGATTGTGGCCTTCTTTG and TGCCCAAGCTCTACCTTCC, respectively, for Bax, Bcl2 and CDKN1A. The sequence (5′–3′) of reverse primers were: TCTTCTTCCAGATGGTGAGTG, ACAGTTCCACAAAGGCATCC, and CCACATGGTCTTCCTCTGCT, respectively for Bax, Bcl2 and CDKN1A. The qPCR reactions were run with SYBR Green PCR Master Mix (Applied Biosystems) using the following program: 50 °C/2 min, 95 °C/10 min. and 40 cycles: 95 °C/15 s, 60 °C/min. All assays were done in triplicate. The quantification (ddCT method) was expressed in relation with AdLacZ, where the value of AdLacZ received the reference value of 1.

Western blot

Cells were plated and transduced the next day followed by 24 h incubation before collecting total cellular protein in RIPA buffer (1% NP-40, 0.1% SDS, 0.5% deoxycholic acid in 1 × PBS) supplemented with protease inhibitors. The protein concentration was determined and 100 µg of each was submitted to sodium dodecyl sulfate-polyacrylamide gel electrophoresis before transfer to membrane and probing with specific antibodies (identified above). Detection was performed as described previously [32].

In situ gene therapy

All assays involving animals were performed in accordance with the guidelines of our Institute, including evaluation by the Committee on Ethical use of Animals, ICB-USP. One million H1299 cells were injected s.c. in the flank of 6–8-week-old female Balb/C nude mice, which were then monitored; N = 10 animals/group. About 2 weeks after injection of cells, tumors measured 52.42 ± 17.65 mm3 and virus treatment was initiated. Virus administration was performed while animals were anesthetized with 0.012% ketamine chloral hydrate (Vetbrands)/0.2% xylazine (Bayer) in PBS. Intratumoral virus injection was performed on one occasion using a total of 5 × 108 IVP in a volume of 50 µl, distributed in several small injections throughout the tumor. Although there was no predetermined selection of which animals would be included in each experimental condition, the distribution of animals/cage was maintained. Tumors were measured with calipers on the indicated days and tumor volume was calculated using the formula 1/2 × LW2, as per Tomayko et al. [35]. No animals were excluded from analysis, no investigator blinding was performed. For normalization of tumor growth, tumor volume at the time of treatment was normalized within each group, defined as 1, and this value used to calculate the relative tumor volume at each time point.


Tumors were recovered from three animals per group 48 h after virus treatment, formalin fixed and included in paraffin. Histologic sections, 5 μm, were mounted on slides before performing immunohistochemistry. Primary antibodies (CDKN2A, p53, and Ki 67) were described above. Secondary antibodies conjugated with alkaline phosphatase and the ABC staining sytem were obtained from Vector Labs. Tunel staining was performed using the In Situ Cell Death Detection Kit POD (Sigma-Aldrich, St. Louis, MO, USA).

Statistical analysis

The quantitative measurements from replicate samples and independent assays, as indicated in each figure legend, were used to derive average ± standard error of the mean (SEM) then analyzed for statistical significance. For all in vitro assays, unless otherwise noted, ANOVA One-Way with Tukey’s multiple comparison post-test was performed using Prism 8 (GraphPad Software, San Diego, CA, USA). For the in vivo assay, the Mixed-effects Model and Sidak’s multiple comparison post-test was performed using Prism 8. These tests were chosen since they best match the assumptions of the experiments. For all, p < 0.05 was considered statistically significant. *p < 0.05; **p < 0.01; ***p < 0.001; ****p < 0.0001.


For the simultaneous transfer of CDKN2A and p53, we constructed a bicistronic adenoviral vector (non-replicating, Ad5, CMV promoter used to drive transgene expression) including an internal ribosome entry site (IRES) element in order to translate the two proteins from a single transcript (Fig. S1). Monocistronic vectors encoding only the CDKN2A or p53 cDNA as well as control vectors encoding either eGFP (enhanced green fluorescent protein) or LacZ were also used. The AdeGFP vector was used to determine that a MOI of 10 transduced essentially 100% of H1299 cells without inducing toxicity (Fig. S2) and this MOI was used in all further in vitro assays involving H1299 cells in vitro. Simultaneous expression of virus-encoded CDKN2A and p53 was confirmed by immunofluorescence staining in transduced H1299 cells (Fig. 1), where simultaneous expression of both CDKN2A and p53 was associated only with the bicistronic adenovirus.

Fig. 1

Simultaneous expression of the CDKN2A and p53 transgenes in transduced H1299 cells. For this immunofluorescence assay, the cells were plated on coverslips and the next day transduced with the indicated vectors at a MOI of 10. After 48 h, the cells were fixed, stained, and observed by confocal microscopy. All photomicrographs were acquired using identical conditions

The impact of CDKN2A plus p53 gene transfer was assessed first in a growth curve, where we see that combined transfer of CDKN2A and p53 drastically reduced the rate of cell proliferation, yet the effect of the monocistronic CDKN2A or p53 vectors was quite moderate (Fig. 2a). As expected, the transfer of CDKN2A alone induced senescence in a greater number of cells than was seen for p53 (Fig. 2b), yet cells transduced with the bicistronic vector generally detached from the dish and those that remained presented altered morphology indicating cell death and atypical SA-βGal staining (Fig. S3). The cell viability assay (MTT) shows reduction in mitochondrial activity in the presence of the bicistronic virus at all time points, including an 80% reduction by 72 h post transduction, yet the monocistronic vectors only reached 20–30% reduction at this time point (Fig. 2c).

Fig. 2

CDKN2A and p53 act collaboratively to reduce proliferation and viability of H1299 cells. a Growth curve assay where cells were plated, transduced, and viable cells counted manually at the indicated times. The data represent the average and SEM from three independent experiments each performed with technical duplicates. b Senescence-associated β-galactosidase (SA-βGal) assay where replicate wells were transduced and, 96 h later, stained with X-gal under acidic conditions. The enlarged, flat, blue cells were counted, whereas the cells with a pyknotic morphology were excluded. The data represent the average and SEM from three independent experiments each performed with technical duplicates. c Cellular response as measured by MTT staining. Cells were transduced, incubated, and subjected to a standard MTT assay where 100% viability was defined as the signal measured in the AdLacZ (control) condition. The data represent the average and SEM from three independent experiments each performed with six technical replicates

Cell cycle analysis revealed that combined gene transfer to H1299 cells induced cell death at 48 h and more than double the number of sub-G1 (hypodiploid) cells as compared with p53 alone at the 72 h time point, yet CDKN2A conferred G1 arrest associated with a reduced number of cells in S phase but without induction of cell death (Fig. 3a, Tables S1 and S2). Similar results were seen when gene transfer was performed in the H358 and A549 cell lines, revealing superior cell killing by the bicistronic vector (Fig. S4, Tables S3 and S4). Bicistronic gene transfer resulted in faster and more efficient induction of caspase 3/7 activity and cell death as compared to p53 alone, culminating in the loss of cells by 72 h (Fig. 3b). These results indicate that proliferation and viability were reduced by the combined gene transfer to a greater degree and more quickly than with either CDKN2A or p53 alone. The loss in viability may be related to cell death by a mechanism consistent with apoptosis.

Fig. 3

Combined gene transfer correlates with accumulation of sub-G1 cells and elevated caspase 3/7 activity. a Cell cycle distribution was determined by propidium iodide staining of fixed cells collected at the indicated time points and observed by flow cytometry. The data represent the average and SEM from three independent experiments each performed with technical duplicates. b Caspase 3/7 activity presented as the percentage of cells staining positive (green fluorescence) as compared with the total number of Hoechst 33342 stained cells. The data represent the average and SEM from three independent experiments each performed with technical triplicates

Expression of critical p53 target genes was examined by quantitative real time PCR (qPCR), revealing that transduction with the bicistronic vector was especially effective for the induction of CDKN1A (p21) and Bax, while reducing Bcl2 levels (Fig. 4a). Immunofluorescence confirmed increased Bax at the protein level (Fig. S5). Using western blot analysis, not only were the protein products of the CDKN2A and p53 transgenes detected, but also their impact on cellular targets was revealed (Fig. 4b). The presence of exogenous CDKN2A led to decreased detection of the phosphorylated form of pRB. As active pRB is known to inhibit proliferation, we also show reduced Ki 67 staining in the presence of exogenous CDKN2A (Fig. 4c). Combined gene transfer was correlated with increased CDKN1A protein levels (Fig. 4b) with nuclear protein accumulation (Fig. S6). As shown here, CDKN2A gene transfer did indeed impact pRB activity, whereas p53 gene transfer increased CDKN1A protein levels.

Fig. 4

Combined gene transfer induces a cell death-associated gene expression profile while CDKN2A activated pRB and promoted arrest. a qPCR assessment of p53 target gene expression 36 h after transduction. The data represent the average and SEM from three independent experiments each performed with technical triplicates. b Western blot analysis was performed using total cellular protein collected 36 h after transduction. c Immunofluorescence detection of Ki 67 performed 48 h after transduction. All images were obtained by confocal microscopy using identical parameters

As the presence of endogenous, mutant p53 may hamper the activity of exogenous, wild-type p53, we developed isogenic cells lines with forced expression of hot-spot p53 mutations, R175H or R248Q. The in vitro growth kinetics of the H1299p53R175H and H1299p53R248Q cell lines were compared with the H1299H2BeGFP cells (Fig. S7), revealing a slight proliferative advantage in the presence of the mutant proteins. The presence of endogenous mutant p53 did not impede the activity of the bicistronic gene transfer approach, yet exogenous p53 alone was less effective in the presence of the mutants in a growth curve assay (Fig. S7). Strikingly, cell viability (MTT) was reduced when the isogenic cell lines were treated with the bicistronic vector (Fig. 5a), a finding that correlated with accumulation of sub-G1 cells (Fig. 5b). Cell cycle analysis (Table S2), revealed G1 arrest upon CDKN2A gene transfer even in the presence of mutant p53. Western blot analysis revealed that exogenous p53 continued to activate CDKN1A expression and that reduced levels of phosphorylated pRB were correlated with CDKN2A gene transfer (Fig. S8). The use of these isogenic cell lines revealed that endogenous mutant p53 is not a barrier for the bicistronic vector to affect growth inhibition.

Fig. 5

Presence of mutant p53 did not hinder the performance of the gene transfer approach. Isogenic H1299 cell lines with stable expression of mutant p53 proteins or, as a control, a fusion protein of histone H2B and eGFP (H2BeGFP) were established using retrovirus-mediated gene transfer. a Cellular response to the adenoviral gene transfer approach was assessed using a standard MTT assay where the AdLacZ condition was considered as 100% viable cells. The data represent the average and SEM from three independent experiments each performed with technical quadruplicates. b Accumulation of sub-G1 (hypodiploid) cells was determined by staining fixed cells with propidium iodide and analyzed for flow cytometry. The data represent the average and SEM from three independent experiments each performed with technical duplicates

The bicistronic transfer of CDKN2A and p53 was next evaluated in a mouse model of in situ gene therapy. For this assay, H1299 cells were injected s.c. in one flank of each Balb/c nude mouse. When tumors reached a volume of ~ 52 mm3, intratumoral injection of the indicated vectors, 5 × 108 IVP/tumor, was performed once on day 0. The tumor volumes, shown in Fig. 6, indicate that the bicistronic vector greatly inhibited tumor progression, whereas the monocistronic CDKN2A or p53 vectors also inhibited progression, but not completely. Histologic analysis of treated tumors revealed strong staining of exogenous CDKN2A in both the cytoplasm and nucleus, yet p53 staining was predominantly nuclear (Fig. 6). We also show histologic evidence for reduced proliferation in the presence of CDKN2A as revealed by reduced Ki 67 staining, whereas increased TUNEL staining indicated a greater number of dead cells in the presence of the bicistronic vector. Together, our data show a clear collaborative effect mediated by the simultaneous transfer of CDKN2A e p53 in the human lung carcinoma cell line, H1299.

Fig. 6

In situ gene therapy reveals superior tumor inhibition by the bicistronic adenovirus. H1299 cells were implanted s.c. in Balb/c nude mice and when tumors reach ~52 mm3, they were treated with intratumoral injection of 5 × 108 IVP on a single occasion (day 0). a Tumors were monitored and volumes calculated, presented here as relative tumor size as compared with day 0. N = 10 animals/group. The data represent the average and SEM. b Alternatively, mice were euthanized 72 h after the adenoviral treatment, then tumors were recovered, fixed, included in paraffin and histologic sections were made. Immunohistochemistry was performed using specific antibodies for CDKN2A, p53, Ki 67 or by Tunel and detected by staining with alkaline phosphatase-conjugated secondary antibodies and reaction with substrate


Here, we have shown that combined, but not individual, transfer of the CDKN2A and p53 genes led to collaborative killing of H1299 lung carcinoma cells both in vitro and in vivo. CDKN2A alone activated pRB, seen as reduced phosphorylation, provoked loss of Ki 67 staining, reduction in cellular proliferation and increased the number of senescent cells in vitro. When applied alone, p53 activated CDKN1A expression, induced caspase 3/7 activity and reduced proliferation. Interestingly, the bicistronic approach was clearly associated with rapid cell killing, but not induction of senescence. In addition, combined gene transfer was associated with drastically reduced proliferation in vitro, even when the H1299 cells expressed hot-spot p53 mutants. In vivo, the simultaneous transfer of CDKN2A and p53 resulted in near elimination of tumor progression, correlating with reduced proliferation and increased cell death. In all, we demonstrate that the bicistronic gene therapy approach provides antitumor activities at levels not seen with either CDKN2A or p53 alone.

Although the results presented here are promising, there is still more development necessary to ready this approach for pre-clinical studies. One area of consideration is the cellular genotype required for CDKN2A and p53 activity. For CDKN2A to have an impact, the cell must carry wild-type pRB, since this is the downstream target. Since 50% of non-small cell lung carcinomas maintain pRB [36], a great many cases would be candidates for a therapy involving CDKN2A gene transfer. The importance of CDK4/6 amplification or cyclin D overexpression could limit the effectiveness of CDKN2A, but we argue that supra-physiologic levels of CDKN2A expression would support effective cell cycle control. Also, the use of CDK4/6 inhibitors, such as palbociclib, could be an ally to the gene transfer approach.

The transfer of the p53 gene should be effective in cells that are deprived of p53. The presence of endogenous, wild-type p53 would imply that the cell has gained tolerance by controlling p53 activity, perhaps by loss of p14ARF, overexpression of MDM2 or deregulation of post-translational modifications of p53. In the case of non-small cell lung carcinoma, we would expect about half of all cases to lose wild-type p53, suggesting that thousands of cases per year would be candidates for approaches involving p53 replacement [37]. Previous studies have suggested that p53 gene therapy is actually aided when patient cells maintain the endogenous, wild-type protein [38]. The R175H and R248Q mutations have each been reported to confer gain of function to the p53 protein, including increased proliferation, invasion or migration [39], whereas dominant negative activity has been associated with the R175H allele [40]. As shown here, the presence of endogenous mutant p53 was not an impediment to the effectiveness of our bi-tumor suppressor gene transfer approach. Taken together, we interpret that endogenous p53R175H or R248Q is not a limiting factor when considering the introduction of exogenous p53, especially when combined with CDKN2A.

The study presented here explored only the gene transfer aspect of our novel approach but does not exclude the possibility of associating gene therapy with pharmacological agents. As suggested by our approach, we do not believe that monotherapy will be effective, but instead propose strategic combinations of gene transfer with chemotherapy, targeted therapy or immunotherapy. Such studies are underway, but for now we can only speculate that the effective induction of cell death could also benefit from additional approaches that will broaden the effects seen here. As it stands, our gene transfer approach may debulk the primary treatment site but would not be expected to offer benefit for the reduction of distant, non-transduced tumor cells. Especially with the inclusion of an immune-activating component, the benefit of the localized gene therapy could be expanded.

The application of multiple genes in gene therapy has been only partially explored. For example, HSV-TK has been paired with IL-2 or IL-4 [41, 42] or with cytosine deaminase [43]. Also, combined TK and Flt3 are being tested for the treatment of glioblastoma, though this approach involves the simultaneous transfer of two adenoviral vectors [44]. An additional parameter of our AdCDKN2A-I-p53 vector that may deserve further attention is the technology for expressing two proteins from a single vector. The use of the IRES element was quite reliable here, but other options may be explored, including the 2A peptide [45, 46]. The bicistronic arrangement is consistent with the notion that cancer is a multifactorial disease and its treatment with gene therapy may require the targeting of multiple pathways. Plus our data show that hitting multiple targets using the bicistronic vector resulted in improved control over tumor progression, while single gene transfer was inferior.

Using this multigene approach, we have renewed our interest in gene therapy using tumor suppressor genes. In summary, we found that cancer gene therapy mediated by bicistronic adenovirus AdCDKN2A-I-p53 is advantageous as compared to monocistronic AdCDKN2A or Adp53 for the induction of cell death in this model of non-small cell lung carcinoma and warrants further development for the treatment of this and other tumor types that maintain pRB expression.


  1. 1.

    Carr TH, McEwen R, Dougherty B, Johnson JH, Dry JR, Lai Z, et al. Defining actionable mutations for oncology therapeutic development. Nat Rev Cancer. 2016;16:319–29.

  2. 2.

    Vanneman M, Dranoff G. Combining immunotherapy and targeted therapies in cancer treatment. Nat Rev Cancer. 2012;12:237–51.

  3. 3.

    Friedman AA, Letai A, Fisher DE, Flaherty KT. Precision medicine for cancer with next-generation functional diagnostics. Nat Rev Cancer. 2015;15:747–56.

  4. 4.

    Melero I, Berman DM, Aznar MA, Korman AJ, Perez Gracia JL, Haanen J. Evolving synergistic combinations of targeted immunotherapies to combat cancer. Nat Rev Cancer. 2015;15:457–72.

  5. 5.

    Sherr CJ, Beach D, Shapiro GI. Targeting CDK4 and CDK6: from discovery to therapy. Cancer Discov. 2016;6:353–67.

  6. 6.

    Strauss BE, Fontes RB, Lotfi CF, Skorupa A, Bartol I, Cipolla-Neto J, et al. Retroviral transfer of the p16INK4a cDNA inhibits C6 glioma formation in Wistar rats. Cancer Cell Int. 2002;2:2.

  7. 7.

    Costanzi-Strauss E, Strauss BE, Naviaux RK, Haas M. Restoration of growth arrest by p16INK4, p21WAF1, pRB, and p53 is dependent on the integrity of the endogenous cell-cycle control pathways in human glioblastoma cell lines. Exp Cell Res. 1998;238:51–62.

  8. 8.

    Chen J. The cell-cycle arrest and apoptotic functions of p53 in tumor initiation and progression. Cold Spring Harb Perspect Med. 2016;6:a026104.

  9. 9.

    Wang X, Simpson ER, Brown KA. p53: protection against tumor growth beyond effects on cell cycle and apoptosis. Cancer Res. 2015;75:5001–7.

  10. 10.

    Miciak J, Bunz F. Long story short: p53 mediates innate immunity. Biochim Biophys Acta. 2016;1865:220–7.

  11. 11.

    Beckerman R, Prives C. Transcriptional regulation byp53. Cold Spring Harb Perspect Biol. 2010;2:a000935.

  12. 12.

    Ablain J, Poirot B, Esnault C, Lehmann-Che J, De The H. p53 as an effector or inhibitor of therapy response. Cold Spring Harb Perspect Med. 2016;6:a026260.

  13. 13.

    Deben C, Deschoolmeester V, Lardon F, Rolfo C, Pauwels P. TP53 and MDM2 genetic alterations in non-small cell lung cancer: Evaluating their prognostic and predictive value. Crit Rev Oncol Hematol. 2016;99:63–73.

  14. 14.

    Lu X. Tied up in loops: positive and negative autoregulation of p53. Cold Spring Harb Perspect Biol. 2010;2:a000984.

  15. 15.

    Meek DW, Anderson CW. Posttranslational modification of p53: cooperative integrators of function. Cold Spring Harb Perspect Biol. 2009;1:a000950.

  16. 16.

    Oren M, Rotter V. Mutant p53 gain-of-function in cancer. Cold Spring Harb Perspect Biol. 2010;2:a001107.

  17. 17.

    Haupt S, Raghu D, Haupt Y. Mutant p53 drives cancer by subverting multiple tumor suppression pathways. Front Oncol 2016;6:12.

  18. 18.

    Lane DP, Cheok CF, Lain S. p53-based cancer therapy. Cold Spring Harb Perspect Biol. 2010;2:a001222.

  19. 19.

    Khoo KH, Verma CS, Lane DP. Drugging the p53 pathway: understanding the route to clinical efficacy. Nat Rev Drug Discov. 2014;13:217–36.

  20. 20.

    Nguyen DM, Spitz FR, Yen N, Cristiano RJ, Roth JA. Gene therapy for lung cancer: enhancement of tumor suppression by a combination of sequential systemic cisplatin and adenovirus-mediated p53 gene transfer. J Thorac Cardiovasc Surg. 1996;112:1372–6. discussion1376-7

  21. 21.

    Chen GX, Zhang S, He XH, Liu SY, Ma C, Zou XP. Clinical utility of recombinant adenoviral human p53 gene therapy: current perspectives. Onco Targets Ther. 2014;7:1901–9.

  22. 22.

    Futreal PA, Coin L, Marshall M, Down T, Hubbard T, Wooster R, et al. A census of human cancer genes. Nat Rev Cancer. 2004;4:177–83.

  23. 23.

    Helsten T, Kato S, Schwaederle M, Tomson BN, Buys TP, Elkin SK, et al. Cell-cycle gene alterations in 4,864 tumors analyzed by next-generation sequencing: implications for targeted therapeutics. Mol Cancer Ther. 2016;15:1682–90.

  24. 24.

    Jamal-Hanjani M, Wilson GA, McGranahan N, Birkbak NJ, Watkins TBK, Veeriah S, et al. Tracking the evolution of non-amall-cell Lung cancer. N Engl J Med. 2017;376:2109–21.

  25. 25.

    Sandig V, Brand K, Herwig S, Lukas J, Bartek J, Strauss M. Adenovirally transferred p16INK4/CDKN2 and p53 genes cooperate to induce apoptotic tumor cell death. Nat Med. 1997;3:313–9.

  26. 26.

    Ghaneh P, Greenhalf W, Humphreys M, Wilson D, Zumstein L, Lemoine NR, et al. Adenovirus-mediated transfer of p53 and p16(INK4a) results in pancreatic cancer regression in vitro and in vivo. Gene Ther. 2001;8:199–208.

  27. 27.

    Bajgelman MC, Costanzi-Strauss E, Strauss BE. Exploration of critical parameters for transient retrovirus production. J Biotechnol. 2003;103:97–106.

  28. 28.

    Ghattas IR, Sanes JR, Majors JE. The encephalomyocarditis virus internal ribosome entry site allows efficient coexpression of two genes from a recombinant provirus in cultured cells and in embryos. Mol Cell Biol. 1991;11:5848–59.

  29. 29.

    Strauss BE, Haas M. The region 3′ to the major transcriptional start site of the MDR1 downstream promoter mediates activation by a subset of mutant P53 proteins. Biochem Biophys Res Commun. 1995;217:333–40.

  30. 30.

    Naviaux RK, Costanzi E, Haas M, Verma IM. The pCL vector system: rapid production of helper-free, high-titer, recombinant retroviruses. J Virol. 1996;70:5701–5.

  31. 31.

    Nyberg-Hoffman C, Shabram P, Li W, Giroux D, Aguilar-Cordova E. Sensitivity and reproducibility in adenoviral infectious titer determination. Nat Med. 1997;3:808–11.

  32. 32.

    Merkel CA, da Silva Soares RB, de Carvalho AC, Zanatta DB, Bajgelman MC, Fratini P, et al. Activation of endogenous p53 by combined p19Arf gene transfer and nutlin-3 drug treatment modalities in the murine cell lines B16 and C6. BMC Cancer 2010;10.

  33. 33.

    Debacq-Chainiaux F, Erusalimsky JD, Campisi J, Toussaint O. Protocols to detect senescence-associated beta-galactosidase (SA-betagal) activity, a biomarker of senescent cells in culture and in vivo. Nat Protoc. 2009;4:1798–806.

  34. 34.

    Rozen S, Skaletsky H. Primer3 on the WWW for general users and for biologist programmers. Methods Mol Biol. 2000;132:365–86.

  35. 35.

    Tomayko MM, Reynolds CP. Determination of subcutaneous tumor size in athymic (nude) mice. Cancer Chemother Pharmacol. 1989;24:148–54.

  36. 36.

    Kaye FJ. RB and cyclin dependent kinase pathways: defining a distinction between RB and p16 loss in lung cancer. Oncogene. 2002;21:6908–14.

  37. 37.

    Robles AI, Harris CC. Clinical outcomes and correlates of TP53 mutations and cancer. Cold Spring Harb Perspect Biol. 2010;2:a001016.

  38. 38.

    Roth JA. Adenovirus p53 gene therapy. Expert Opin Biol Ther. 2006;6:55–61.

  39. 39.

    Muller PA, Vousden KH. Mutant p53 in cancer: new functions and therapeutic opportunities. Cancer Cell. 2014;25:304–17.

  40. 40.

    Freed-Pastor WA, Prives C. Mutantp53: one name, many proteins. Genes Dev. 2012;26:1268–86.

  41. 41.

    Colombo F, Barzon L, Franchin E, Pacenti M, Pinna V, Danieli D, et al. Combined HSV-TK/IL-2 gene therapy in patients with recurrent glioblastoma multiforme: biological and clinical results. Cancer Gene Ther. 2005;12:835–48.

  42. 42.

    Okada H, Pollack IF, Lotze MT, Lunsford LD, Kondziolka D, Lieberman F, et al. Gene therapy of malignant gliomas: a phase I study of IL-4-HSV-TK gene-modified autologous tumor to elicit an immune response. Hum Gene Ther. 2000;11:637–53.

  43. 43.

    Freytag SO, Stricker H, Pegg J, Paielli D, Pradhan DG, Peabody J, et al. Phase I study of replication-competent adenovirus-mediated double-suicide gene therapy in combination with conventional-dose three-dimensional conformal radiation therapy for the treatment of newly diagnosed, intermediate- to high-risk prostate cancer. Cancer Res. 2003;63:7497–506.

  44. 44.

    VanderVeen N, Raja N, Yi E, Appelman H, Ng P, Palmer D, et al. Preclinical efficacy and safety profile of allometrically scaled doses of doxycycline used to turn “on” therapeutic transgene expression from high-capacity adenoviral vectors in a glioma model. Hum Gene Ther Methods. 2016;27:98–111.

  45. 45.

    Kim JH, Lee SR, Li LH, Park HJ, Park JH, Lee KY, et al. High cleavage efficiency of a 2A peptide derived from porcine teschovirus-1 in human cell lines, zebrafish and mice. PLoS ONE. 2011;6:e18556.

  46. 46.

    Szymczak-Workman AL, Vignali KM, Vignali DA. Design and construction of 2A peptide-linked multicistronic vectors. Cold Spring Harb Protoc. 2012;2012:199–204.

Download references


We are grateful to Patrícia Léo and Juliana C. Gregório for the initial vector constructions. Financial support from the Sao Paulo Research Foundation (FAPESP): grant 98/15120-3 (ECS), grant 2015/26580-9 (BES), and fellowships 14/12322-5 (JGX), 11/21256-8 (RET).

Author information

Correspondence to Eugenia Costanzi-Strauss.

Ethics declarations

Conflict of interest

The authors declare that they have no conflict of interest.

Additional information

Publisher’s note: Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Supplementary information

Rights and permissions

Reprints and Permissions

About this article

Verify currency and authenticity via CrossMark

Cite this article

Xande, J.G., Dias, A.P., Tamura, R.E. et al. Bicistronic transfer of CDKN2A and p53 culminates in collaborative killing of human lung cancer cells in vitro and in vivo. Gene Ther 27, 51–61 (2020).

Download citation