Co-expression of p21Waf1/Cip1 in adenovirus vectors improves expression of a second transgene


First-generation adenoviral (Ad) vectors are frequently used vectors for experimental and clinical gene transfer. Earlier it has been shown that parallel overexpression of the cell cycle regulator p21Waf1/Cip1 (p21) or antiapoptotic bcl-2 from a second vector reduces cytotoxicity and improves transgene expression. Here, we investigate whether the co-expression of p21 and α1-antitrypsin from a single vector improves vector safety and α1-antitrypsin expression. Cell lines (A549 and HeLa) and primary cells (small airway epithelial cells and hepatocytes) were infected with adenovirus vectors transducing α1-antitrypsin with (AdCMV.p21-RSV.hAAT) or without (AdRSV.hAAT) p21. α1-Antitrypsin expression and cytotoxicity were analyzed using western blot/ELISA and LDH/ALT/AST assays, respectively. Cell cycle profiles were determined by flow cytometry. Co-expression of p21 strongly increased the α1-antitrypsin expression in all cell types and at all doses tested. No changes in ALT/AST from hepatocytes and only minor increases in the LDH release in A549 and HeLa were observed with either vector. Cell cycle profiles were also not affected adversely. Incorporation of p21 in Ad vectors together with a gene of interest improves the vector performance; such vectors will allow the application of lower doses and thereby reduce immunological side effects.


Adenoviral (Ad) vectors are one of the most widely used vector systems for gene transfer studies, owing to their broad tropism and efficient infection of resting cells. Consequently, Ad vectors have been used in one-quarter of all clinical gene therapy trials to date and in over 50% of trials with pulmonary indications (Gene Therapy Clinical Trials Worldwide database, update March 2008; Depending on the application their use is, however, limited by their immunogenicity, as well as their cytotoxic effects and induction of apoptosis.1, 2, 3 Approaches for reducing these detrimental effects include deletion of further or all coding sequences of the virus or incorporation of antiapoptotic genes like bcl-2 or p21Waf1/Cip1 (p21).4, 5

Although bcl-2 is classified as an oncogene, p21 is not regarded as such and under certain circumstances even functions as a tumor suppressor.6 p21 is a member of the Cip/Kip family of cyclin-dependent kinase inhibitors and, as such, acts as a cell cycle regulator at the G1/S and G2/M boundaries.7, 8 In addition, cytoplasmic p21 associates with procaspase 3 and inhibits Fas-mediated apoptosis and stress-induced apoptosis by inhibiting ASK1 (apoptosis signal-regulating kinase 1) and JNK (c-Jun N-terminal kinase).9, 10, 11, 12 Furthermore, p21 acts as a transcriptional regulator by either directly inhibiting the transcription factors like E2F-1, c-myc and STAT3 or by activating transcription by relieving repression mediated by the multifunctional p300 and CBP (CREB-binding protein) transcriptional regulators.13, 14, 15, 16, 17 This co-activation affects mainly, although not exclusively, NFκB (nuclear factor-kappaB) responsive promoters.18

Co-infection with two Ad vectors expressing p21 together with a reporter gene has been shown recently to result in reduced cytotoxicity with a concomitant increased expression of the reporter gene.5 This co-infection strategy also allowed improved and prolonged reporter gene expression in vivo.5 However, co-infection results in a mosaic of infected cells and requires increased virus doses, which may compound immunologic consequences in in vivo applications.

Therefore, in this study, we combined the expression cassettes for p21 and human α1-antitrypsin in a single vector and compared its safety and expression profile with that of a control vector lacking p21.

To determine the effect of co-expression of p21 from a single vector on Ad vector gene transfer, HeLa and A549 cells were transduced with either AdCMV.p21-RSV.hAAT or AdRSV.hAAT. Forty-eight hours after gene transfer, protein expression was analyzed using western blot. In both cell lines, expressions of p21 and α1-AT are detected readily in cultures infected with AdCMV.p21-RSV.hAAT, whereas cultures infected with the control Ad vector did not express α1-antitrypsin (α1-AT) (HeLa) or expressed very weakly only at higher multiplicities of infection (MOI) (A549) (Figure 1a). It is also apparent that doubling the vector dose results in a stronger increase in p21 expression in HeLa than in A549, whereas the concomitant increase of α1-AT expression is less pronounced than in A549. These results were corroborated by dose-response analyses after 48 and 72 h (Figure 1b, data not shown). Furthermore, the co-expression of p21 markedly improved α1-AT expression in both cell lines at all tested MOIs. In A549, higher α1-AT concentrations were obtained from both vectors, but in HeLa, the fold induction by co-expressing p21 was more pronounced with a factor of 3.4 at 1000 MOI versus 1.7 in A549 for the same dose.

Figure 1

Analysis of transgene expression. (a) Western blot analysis of α1-antitrypsin and p21 expression 48 h after transduction with AdRSV.hAAT (α1-AT) or AdCMV.p21-RSV.hAAT (p21+α1-AT) at 50 or 100 MOI. C denotes mock-infected controls. β-Actin serves as a loading control. (b) ELISA quantitation of the dose-dependent α1-AT expression 48 h after transduction with AdRSV.hAAT (•) or AdCMV.p21-RSV.hAAT (). A549 and HeLa cells were maintained in RPMI1640 supplemented with 10% FCS. First-generation adenoviral vectors AdRSV.hAAT and AdCMV.p21-RSV.hAAT encode human α1-antitrypsin alone or together with the cdk inhibitor p21Waf1/Cip1, respectively (for details of the construction, see Supplementary information). Gene transfer was carried out using Ad vectors at the MOIs indicated, the day after seeding the cells at 2 × 104 cells per cm2. The vectors were incubated with the cells in serum-free medium (1.5 ml per 10 cm dish, 300 μl per 24-well plate; for western blot and ELISA) for 90 min in the CO2-incubator with occasional agitation. Complete medium was added (to 10 ml and 600 μl, respectively) and the cultures were further incubated for 48 h. For western blot, cells were harvested and solubilized in lysis buffer containing protease inhibitors. 10 μg protein was separated by 10% SDS gel electrophoresis and transferred to a PVDF membrane by semi-dry blotting. Mouse monoclonal antibodies were used to detect p21 (#554228, BD Biosciences, Heidelberg, Germany), α1-antitrypsin (#BM2152, Acris Antibodies, Hiddenhausen, Germany) and β-actin (#A5316, Sigma-Aldrich, Taufkirchen, Germany) followed by the addition of biotinylated goat anti-mouse IgG and streptavidin horseradish peroxidase (both Vector Laboratories, Burlingame, CA, USA). Membranes were developed using the ECL Detection Reagents Kit from GE Healthcare (Freiburg, Germany). α1-Antitrypsin was quantitated, in the culture supernatant, using an ELISA kit (Immundiagnostik, Bensheim, Germany). Data for all assays are given as mean±s.d. from two to three biological replicates. ELISA, enzyme-linked immunosorbent assay; MOI, multiplicity of infection.

Next, we investigated whether the beneficial effect of co-expression of p21 was also present in primary cells. To this end, HSAEPCs (human primary small airway epithelial cells), which do not express α1-AT, and human hepatocytes, which are the primary source of α1-AT in the body, were infected. The HSAEPC cells infected with AdCMV.p21-RSV.hAAT produced considerably more α1-AT in 48 and 72 h, than cells infected with AdRSV.hAAT (Figure 2). Infection with 100 instead of 50 MOI showed the same pattern with approximately twice the α1-AT levels (data not shown). For hepatocytes, we obtained comparable results, although the relative differences were smaller due to the basal expression of the endogenous α1-AT. Co-expression of p21 led to an increased overexpression of the α1-AT transgene after 48 and 72 h when compared with the expression of α1-AT alone (Figure 2). In summary, an improved expression of α1-AT by simultaneous overexpression of p21 could be confirmed in both cell lines and primary cells.

Figure 2

ELISA analysis of α1-antitrypsin expression in primary human cells. HSAEPCs and hepatocytes were mock infected, infected with AdRSV.hAAT and AdCMV.p21-RSV.hAAT (100 and 50 MOI, respectively) and the supernatant was harvested at the indicated time points. Primary HSAEPCs were kept in a serum-free airway epithelial cell growth medium with supplements (all from provitro, Berlin, Germany) and were passaged before reaching 75% confluence. For experiments, HSAEPCs were seeded at 5 × 104 cells per well in 24-well plates. For primary human hepatocytes, Inducible Cryopreserved Human Hepatocytes (Cat. no. 454551; BD Biosciences) were purified using the One Step Purification Kit (BD Biosciences) according to the manufacturer's protocol and plated at 3.7–4.0 × 105 cells per well in collagen I-coated 24-well plates in the supplemented ISOM's seeding medium (BD Biosciences). After 4 h, non-attached cells were aspirated and the medium was changed to serum-free hepatocyte maintenance medium with supplements (provitro). Infection and ELISA were carried out as given in Figure 1. ELISA, enzyme-linked immunosorbent assay; HSAEPC, human small airway epithelial cell.

To assess Ad vector-induced cytotoxicity, we assayed the LDH release from HeLa and A549 cultures infected with increasing vector doses (HSAEPC could not be analyzed due to pyruvate in the culture medium, which interferes with the test). In A549 cultures, the LDH release after 48 and 72 h was not changed up to 500 MOI, but was changed at the highest vector doses (1000 MOI) and increased by 50% with no differences between the Ad vectors (Figure 3, data not shown). In HeLa, a stronger effect was observed in LDH increasing up to eightfold at 1000 MOI, but results were more variable (Figure 3, data not shown). In primary hepatocytes, no cytotoxicity was detected by ALT or AST (Figure 3).

Figure 3

Analysis of Ad vector-induced cytotoxicity. LDH was assayed from cell lines (A549 and HeLa) and ALT and AST from hepatocytes that were mock infected, infected with AdRSV.hAAT and AdCMV.p21-RSV.hAAT. Cell lines and hepatocytes were infected with 100 and 50 MOI, respectively, and the supernatant was collected at the indicated time points. The cytotoxic effects on A549 and HeLa were assessed by determining LDH from the culture medium with a Cytotoxicity Detection Kit (Roche, Mannheim, Germany) using L-lactate dehydrogenase (Sigma-Aldrich) as standard. For hepatocytes, ALT and AST concentrations from the medium were measured by automated analysers in the central laboratory. Cultivation and infection was carried out as given in Figures 1 and 2. ALT, alanine aminotransferase; AST, aspartate aminotransferase; LDH, lactate dehydrogenase.

To further examine the vector-mediated adverse effects, cell cycle profiles of A549 cells were assessed by flow cytometry. Forty-eight hours after infection with 250 MOI of AdRSV.hAAT, no gross cell cycle perturbation was visible and no clear-cut differences existed when compared with the cell cycle profile of mock-infected cells (Table 1). For AdCMV.p21-RSV.hAAT, a similar profile was obtained with a further shift from G1 to G2/M. No signs of aneuploid cells or polyploidization were observed with any of the Ad vectors.

Table 1 Cell cycle profiles

This study shows that the co-expression of p21 together with a gene of interest from the same Ad vector improves the expression of this gene in both established cell lines and primary cells. These data corroborate and extend earlier results, in which co-infection of an Ad vector expressing p21 with a second vector coding for a gene of interest augmented the expression from this second Ad vector.5 In that study, the beneficial effect of p21 was ascribed to its function as a cell cycle regulator that counteracts Ad vector-mediated cell cycle dysregulation and cytotoxicity in infected cells. In this study, however, no cytotoxicity and vector-induced aberrant cell cycle profiles were observed. This discrepancy is most likely due to the different vector constructs employed in the two studies. While in our previous study the expression cassettes were oriented rightward with respect to the viral genome placing the promoter next to the residual Ad enhancer, which overlaps with the packaging signal, this orientation is inverted in the present constructs. Juxtaposition of the promoter with the enhancer has been shown to interfere with a controlled transcription of the transgene and is suspected to lead to a low-level expression of the remaining Ad early genes with detrimental effects.19 Recently, Nakai et al.20 have verified read-through from the transgene into the adjacent Ad pIX gene even with the production of aberrantly spliced transgene-pIX chimeras, effects that were absent in identical vectors but with leftward orientation of the expression cassette.

As no cell cycle disturbance was evident using the actual constructs, the function of p21 as a cell cycle regulator cannot be responsible for the observed improved expression of α1-AT. An explanation could be the recently described function of p21 as a transcriptional co-activator of p300/CBP-dependent/regulated promoters.13, 17, 18, 21 p300/CBP interact with a wide range of DNA-binding proteins, including YY1, C/EBP, the NFκB subunit RelA and many more.18, 22 p21 has been shown to interfere with the transcriptional repression mediated by the CRD1 domain of these transcriptional co-activators, thereby relieving the repression and activating transcription. As NFκB is one of the transcription factors identified to be activated by p21 and as the RSV promoter driving α1-AT expression contains, besides C/EBP and YY1 binding sites, RelA/NFκB responsive elements, it appears likely that the observed overexpression occurs through this mechanism.23, 24 This view is supported by the fact that the co-expression of p21 had an equally positive effect on RSV- and CMV-driven β-galactosidase expression (see Figure 5 in Wolff et al.5) and CMV has been proven to be activated by p2118. Further backing is provided by the varying efficiency in the α1-AT expression in HeLa and A549 despite similar p21 expressions (see Figure 1). This is most likely the result of the interference from the E7 protein of the enogenous HPV 18 provirus in HeLa, which is known to inhibit the p21 function.25, 26

Together, these data suggest that a parallel overexpression of p21 with a second gene from a single Ad vector can augment the expression of that gene by a dual mechanism. First, by activating the transcription through responsive promoters of which RSV and CMV are two of the most widely used but not the only ones. Second, independently but additively, by maintaining a normal cell cycle of the host cell when interference from the expressed transgene or viral factors occurs.


  1. 1

    Yang Y, Jooss KU, Su Q, Ertl HC, Wilson JM . Immune responses to viral antigens versus transgene product in the elimination of recombinant adenovirus-infected hepatocytes in vivo. Gene Therapy 1996; 3: 137–144.

    PubMed  Google Scholar 

  2. 2

    Teodoro JG, Branton PE . Regulation of apoptosis by viral gene products. J Virol 1997; 71: 1739–1746.

    CAS  PubMed  PubMed Central  Google Scholar 

  3. 3

    Muruve DA . The innate immune response to adenovirus vectors. Hum Gene Ther 2004; 15: 1157–1166.

    CAS  Article  Google Scholar 

  4. 4

    Bilbao G, Contreras JL, Zhan H-G, Pike MJ, Overturf K, Mikheeva G et al. Adenovirus-mediated gene expression in vivo is enhanced by the antiapoptotic bcl-2 gene. J Virol 1999; 73: 6992–7000.

    CAS  PubMed  PubMed Central  Google Scholar 

  5. 5

    Wolff G, Schumacher A, Nüssler AK, Ruppert V, Karawajew L, Wehnes E et al. Coexpression of p21(WAF1/CIP1) in adenovirus vector transfected human primary hepatocytes prevents apoptosis resulting in improved transgene expression. Gene Therapy 2003; 10: 668–677.

    CAS  Article  Google Scholar 

  6. 6

    Yang ZY, Perkins ND, Ohno T, Nabel EG, Nabel GJ . The p21 cyclin-dependent kinase inhibitor suppresses tumorigenicity in vivo. Nat Med 1995; 1: 1052–1056.

    CAS  Article  Google Scholar 

  7. 7

    Medema RH, Klompmaker R, Smits VAJ, Rijksen G . p21waf1 can block cell cycle at two points in the cell cycle, but does not interfere with processive DNA-replication or stress-activated kinases. Oncogene 1998; 16: 431–441.

    CAS  Article  Google Scholar 

  8. 8

    Niculescu 3rd AB, Chen X, Smeets M, Hengst L, Prives C, Reed SI . Effects of p21Cip1/Waf1 at both the G1/S and the G2/M cell cycle transitions: pRb is a critical determinant in blocking DNA replication and in preventing endoreduplication. Mol Cell Biol 1998; 18: 629–643.

    CAS  Article  Google Scholar 

  9. 9

    Asada M, Yamada T, Ichijo H, Delia D, Miyazono K, Fukumuro K et al. Apoptosis inhibitory activity of cytoplasmic p21Cip1/Waf1 in monocytic differentiation. EMBO J 1999; 18: 1223–1234.

    CAS  Article  Google Scholar 

  10. 10

    Suzuki A, Tsutomi Y, Yamamoto N, Shibutani T, Akahane K . Mitochondrial regulation of cell death: mitochondria are essential for procaspase 3-p21 complex formation to resist Fas-mediated cell death. Mol Cell Biol 1999; 19: 3842–3847.

    CAS  Article  Google Scholar 

  11. 11

    Huang S, Shu L, Dilling MB, Easton J, Harwood FC, Ichijo H et al. Sustained activation of the JNK cascade and rapamycin-induced apoptosis are suppressed by p53/p21Cip1. Mol Cell 2003; 11: 1491–1501.

    CAS  Article  Google Scholar 

  12. 12

    Shim J, Lee H, Park J, Kim H, Choi EJ . A non-enzymatic p21 protein inhibitor of stress-activated protein kinases. Nature 1996; 381: 804–806.

    CAS  Article  Google Scholar 

  13. 13

    Gregory DJ, Garcia-Wilson E, Poole JC, Snowden AW, Roninson IB, Perkins ND . Induction of transcription through the p300 CRD1 motif by p21WAF1/CIP1 is core promoter specific and cyclin dependent kinase independent. Cell Cycle 2002; 1: 343–350.

    CAS  Article  Google Scholar 

  14. 14

    Delavaine L, La Thangue NB . Control of E2F activity by p21Waf1/Cip1. Oncogene 1999; 18: 5381–5392.

    CAS  Article  Google Scholar 

  15. 15

    Kitaura H, Shinshi M, Uchikoshi Y, Ono T, Iguchi-Ariga SMM, Ariga H . Reciprocal regulation via protein-protein interaction between c-Myc and p21cip1/waf1/sdi1 in DNA replication and transcription. J Biol Chem 2000; 275: 10477–10483.

    CAS  Article  Google Scholar 

  16. 16

    Coqueret O, Gascan H . Functional interaction of STAT3 transcription factor with the cell cycle inhibitor p21WAF1/CIP1/SDI1. J Biol Chem 2000; 275: 18794–18800.

    CAS  Article  Google Scholar 

  17. 17

    Snowden AW, Anderson LA, Webster GA, Perkins ND . A novel transcriptional repression domain mediates p21(WAF1/CIP1) induction of p300 transactivation. Mol Cell Biol 2000; 20: 2676–2686.

    CAS  Article  Google Scholar 

  18. 18

    Poole JC, Thain A, Perkins ND, Roninson IB . Induction of transcription by p21Waf1/Cip1/Sdi1: role of NFkappaB and effect of non-steroidal anti-inflammatory drugs. Cell Cycle 2004; 3: 931–940.

    CAS  Article  Google Scholar 

  19. 19

    Steinwaerder DS, Lieber A . Insulation from viral transcriptional regulatory elements improves inducible transgene expression from adenovirus vectors in vitro and in vivo. Gene Therapy 2000; 7: 556–567.

    CAS  Article  Google Scholar 

  20. 20

    Nakai M, Komiya K, Murata M, Kimura T, Kanaoka M, Kanegae Y et al. Expression of pIX gene induced by transgene promoter: possible cause of host immune response in first-generation adenoviral vectors. Hum Gene Ther 2007; 18: 925–936.

    CAS  Article  Google Scholar 

  21. 21

    Garcia-Wilson E, Perkins ND . p21WAF1/CIP1 regulates the p300 sumoylation motif CRD1 through a C-terminal domain independently of cyclin/CDK binding. Cell Cycle 2005; 4: 1113–1119.

    CAS  Article  Google Scholar 

  22. 22

    Shikama N, Lyon J, La Thangue NB . The p300/CBP family: integrating signals with transcription factors and chromatin. Trends Cell Biol 1997; 7: 230–236.

    CAS  Article  Google Scholar 

  23. 23

    Bowers WJ, Baglia LA, Ruddel A . Regulation of avian leukosis virus long terminal repeat-enhanced transcription by C/EBP-Rel interactions. J Virol 1996; 70: 3051–3059.

    CAS  PubMed  PubMed Central  Google Scholar 

  24. 24

    Mobley CM, Sealy L . Role of the transcription start site core region and transcription factor YY1 in Rous sarcoma virus long terminal repeat promoter activity. J Virol 1998; 72: 6592–6601.

    CAS  PubMed  PubMed Central  Google Scholar 

  25. 25

    Funk JO, Waga S, Harry JB, Espling E, Stillman B, Galloway DA . Inhibition of CDK activity and PCNA-dependent DNA replication by p21 is blocked by interaction with the HPV-16 E7 oncoprotein. Genes Dev 1997; 11: 2090–2100.

    CAS  Article  Google Scholar 

  26. 26

    Jones DL, Alani RM, Münger K . The human papillomavirus E7 oncoprotein can uncouple cellular differentiation and proliferation in human keratinocytes by abrogating p21Cip1-mediated inhibition of cdk2. Genes Dev 1997; 11: 2101–2111.

    CAS  Article  Google Scholar 

Download references


We gratefully acknowledge the technical assistance of Margitta Schüman. This work was funded by a Grant (10134771) from the ProFIT program of the Investitionsbank Berlin.

Author information



Corresponding author

Correspondence to A Schumacher.

Additional information

Supplementary Information accompanies the paper on Gene Therapy website (

Supplementary information

Rights and permissions

Reprints and Permissions

About this article

Cite this article

Schumacher, A., Horvat, S., Woischwill, C. et al. Co-expression of p21Waf1/Cip1 in adenovirus vectors improves expression of a second transgene. Gene Ther 16, 574–578 (2009).

Download citation


  • adenovirus
  • gene expression
  • p21
  • vector design

Further reading


Quick links