Journal of Biotechnology 300 (2019) 70–77

Contents lists available at ScienceDirect

Journal of Biotechnology
journal homepage: www.elsevier.com/locate/jbiotec

Evaluation of AAV-mediated delivery of shRNA to target basal-like breast T

cancer genetic vulnerabilities
Catarina Pintoa,b, Gabriela Silvaa, Ana S. Ribeiroc,d, Mónica Oliveirac,d, Manuel Garridoa,
Vanessa S. Bandeiraa, André Nascimentoa, Ana Sofia Coroadinhaa,b, Cristina Peixotoa,b,
Ana Barbasa,c, Joana Paredesd,e, Catarina Britoa,b, Paula M. Alvesa,b,
⁎

a
iBET, Instituto de Biologia Experimental e Tecnológica, Apartado 12, 2780-901 Oeiras, Portugal
b
Instituto de Tecnologia Química e Biológica António Xavier, Universidade Nova de Lisboa, Av. da República, 2780-157 Oeiras, Portugal
c
Bayer Portugal, Carnaxide, Portugal
d
i3S, Instituto de Investigação e Inovação em Saúde, Universidade do Porto, Rua Alfredo Allen, 208, 4200-135 Porto, Portugal
e
IPATIMUP, Institute of Molecular Pathology and Immunology of the University of Porto, Rua Dr Roberto Frias s/n, Porto, Portugal

ARTICLE INFO ABSTRACT

Keywords: Adeno-associated viral vectors (AAV) for gene therapy applications are gaining momentum, with more therapies
Gene therapy moving into later stages of clinical development and towards market approval, namely for cancer therapy. The
Adeno-associated viral vectors development of cytotoxic vectors is often hampered by side effects arising when non-target cells are infected, and
Basal-like breast cancer their production can be hindered by toxic effects of the transgene on the producing cell lines. In this study, we
AAV production
evaluated the potential of rAAV-mediated delivery of short hairpin RNAs (shRNA) to target basal-like breast
cancer genetic vulnerabilities. Our results show that by optimizing the stoichiometry of the plasmids upon
transfection and time of harvest, it is possible to increase the viral titers and quality. All rAAV-shRNA vectors
obtained efficiently transduced the BLBC cell lines MDA-MB-468 and HCC1954. In MDA-MB-468, transduction
with rAAV-shRNA vector targeting PSMA2 was associated with significant decrease in cell viability and apop-
tosis induction. Importantly, rAAV2-PSMA2 also slowed tumor growth in a BLBC mouse xenograft model, thus
potentially representing a therapeutic strategy against this type of cancer.

1. Introduction as the only treatment option(Caldas-lopes et al., 2009; Schneider et al.,

2008). Standard treatment for these patients, namely surgery, radiation
Gene therapy has the opportunity to expand beyond conventional and chemotherapy, often fails to completely eradicate the disease, is
drug based therapeutics and deliver a targeted treatment through reg- incapable of maintaining sustained protection, and is accompanied by
ulation of endogenous genes(Naldini, 2015). As our knowledge on serious side effects, such as cytotoxicity to non-cancerous cells(Luo
cancer´s onset and progression evolves, tumor-specific phenotypic et al., 2015; Santiago-Ortiz and Schaffer, 2016). Therefore, there is a
characteristics are uncovered that can be explored in this context, to strong unmet medical need for targeted therapies for BLBC with im-
achieve higher specificity and potency of the developed therapies(Miest proved clinical efficacy, targeted effect with lower side effects, and
and Cattaneo, 2013). Basal-like breast cancer (BLBC) patients can longer patient survival times(Carey et al., 2010; Santiago-Ortiz and
greatly benefit from the development of such therapies. Although this Schaffer, 2016). A lot of effort has been conducted towards the dis-
subtype represents up to 20% of worldwide breast cancer (BC) in- covery of actionable molecular targets for this disease(Bianchini et al.,
cidence, it is responsible for a disproportionate number of deaths as it 2016). Recently, Petrocca et al conducted a genome-wide small inter-
often leads to relapse with distant metastasis(Schneider et al., 2008). fering RNA (siRNA) screen to identify selective genetic vulnerabilities
The lack of molecular markers and deficient characterization, impaired causing lethality in this breast cancer subtype(Petrocca et al., 2013).
by an heterogeneous molecular presentation, leads to lack of targeted Although promising, these leads still face considerable challenges
therapies and leaves clinicians with cytotoxic chemotherapeutic drugs reaching the clinic due to ineffective delivery platforms(Kacsinta and

Corresponding author at: iBET, Apartado 12, 2780-901 Oeiras, Portugal.

⁎

E-mail addresses: [email protected] (C. Pinto), [email protected] (G. Silva), [email protected] (A.S. Ribeiro), [email protected] (M. Oliveira),
[email protected] (M. Garrido), [email protected] (V.S. Bandeira), [email protected] (A. Nascimento), [email protected] (A.S. Coroadinha),
[email protected] (C. Peixoto), [email protected] (A. Barbas), [email protected] (J. Paredes), [email protected] (C. Brito), [email protected] (P.M. Alves).

https://doi.org/10.1016/j.jbiotec.2019.05.016
Received 1 August 2018; Received in revised form 24 May 2019; Accepted 27 May 2019
Available online 28 May 2019
0168-1656/ © 2019 Published by Elsevier B.V.
C. Pinto, et al. Journal of Biotechnology 300 (2019) 70–77

Dowdy, 2016). and orientation were confirmed by restriction analysis and sequencing.
Recombinant adeno-associated viral (rAAV) vectors have been in- Gene specific shRNA sequences were obtained from the RNAi con-
creasingly successful as gene therapy delivery platforms and have sortium database (http://www.broadinstitute.org/rnai/public/). The
proven to be a promising approach to treat a variety of human diseases scramble control shRNA sequences, not targeting any known human or
(Brimble et al., 2016; Kaplitt et al., 2007; Testa et al., 2013). The in- mouse genes, were from Sigma. The shRNA catalog numbers are:
creasing number of preclinical studies and human clinical trials has PLK1shRNA: TRCN0000006247; MCL1shRNA: TRCN0000196390;
demonstrated their favorable safety profile, efficacious gene delivery, PSMA2shRNA: TRCN0000003879; PSMB4shRNA: TRCN0000003914;
and robust and persistent transgene expression in vivo, with manageable Ctrl1shRNA: SHC202; and Ctrl2shRNA: SHC016.
immune response and minor adverse effects upon injection(Samulski
and Muzyczka, 2014; Santiago-Ortiz and Schaffer, 2016; Snyder and 2.3. Production, purification and titer determination of rAAV-shRNA
Moullier, 2011). In fact, an rAAV1-based vector for the treatment of a vectors
lipoprotein lipase deficiency became the first commercially-approved
gene therapy product for clinical applications by the European Com- rAAV-shRNA vectors were produced in HEK293 T cells by 2- or 3-
mission(Bryant et al., 2013). plasmid transfection system using 5 μg DNA/1 × 106 cells, with linear
rAAV vectors’ potential for cancer applications has also been de- polyethylenimine (PEI; Polysciences) or CaPO4 as transfection reagents.
monstrated in several in vitro cancer studies, in vivo pre-clinical cancer In the 2-plasmid system HEK293 T cells (seeded at 4–5 × 104 cell/cm2
models, and clinical trials, with a wide variety of approaches, such as 20–24 h before) were co-transfected at 70–80% confluency with in-
the delivery of anti-angiogenesis factors (PEDF, endostatin 34, Kringle dividual pAAV-shRNA plasmids, generated above, plus the pDG
5, Kallistatin), suicide genes (HSV-TK system and polyomavirus middle plasmid(Grimm et al., 2003), encoding the rep and cap genes and the
T antigen), immunostimulatory molecules (TRAIL, Interferons, adenovirus encoding products essential for AAV vector production,
Interleukins), DNA-encoded small RNA molecules for post-transcrip- using different plasmid ratios as indicated below. 3-plasmid transfec-
tional regulation of oncogenes, immunogenic cell surface molecules and tion was performed with plasmids: pAAV-shRNA, pAAV-Helper - en-
tumor antigens(Ledford, 2015; Luo et al., 2015; Santiago-Ortiz and coding adenovirus gene products required for the production of in-
Schaffer, 2016). fective AAV particles, and pAAV-RC - encoding the rep and cap genes
Herein we describe a rAAV-based therapeutic approach to target (necessary trans-acting elements for replication and packaging of the
BLBC. Signature genetic dependencies previously identified for this viral genome) (all from Stratagene/Agilent), at a plasmid molar ratio of
breast cancer subtype (i.e. MCL1, PSMA2 and PSMB4)(Petrocca et al., 1:1:1. All plasmids, except pAAV-shRNA, were amplified in E. coli
2013) were targeted using rAAV-mediated delivery of shRNA. Con- DH5α cells (Invitrogen). Plasmid DNA was purified with the EndoFree
sequent downregulation of the target genes led to a decrease in cell plasmid kit from Qiagen, as per supplier´s instructions. Transfections
viability and induced apoptosis in BLBC cell lines. Furthermore, in- with the CaPO4 method were performed using the Calcium Phosphate
tratumoral injections of rAAV vectors targeting PSMA2 resulted in Transfection Kit from Sigma (CAPHOS) as per kit´s instructions.
slowed tumor growth in a BLBC xenograft model. Moreover, our results Transfections of HEK293 T cells with PEI were performed as per stan-
indicate that optimization of plasmid stoichiometry upon transfection dard protocols using optimized DNA:PEI ratio of 1:1.8 (μg), as pre-
and time of harvest should be done in a transgene dependent manner, viously described(Merten and Al-Rubeai, 2011). 48–72 h post trans-
with the goal of circumventing potential cytotoxic effects of the trans- fection, cells were lysed with 0.1% Triton-X100 (Sigma), releasing the
gene in the producer cell line and yield rAAV batches of higher quality. intracellular vectors into the supernatant. Media with lysed cells was
collected and incubated with 30 unit/ml Benzonase® nuclease (Merck
2. Materials and methods Millipore) and sterile 2 mM MgCl2 (Sigma) for 1 h at 37C, for removal of
host cell and unpacked DNA. The solution was clarified by centrifuga-
2.1. Cell culture tion (4000xg/30 min/40C) and the vector-containing supernatant (pri-
mary viral stock) was filtered through 0.45-μm pore size filters (Merck
HEK293 T (ATCC CRL-3216™) and HT1080 cells (ATCC® CCL121) Millipore). Purification of the different rAAV-shRNA vectors was per-
cells were cultured in high glucose (4.5 g/L) DMEM supplemented with formed by affinity chromatography using the AVB Sepharose High
10% (v/v) fetal bovine serum (FBS). The basal breast cancer cell lines Performance resin (GE Healthcare), according to manufacturer’s in-
HCC1954 and MDA-MB-468, kindly provided by professor Judy structions. Purity was monitored by SDS-PAGE followed by In-
Lieberman (Harvard Medical Scholl, USA) were cultured in RPMI 1640 stantBlue™ and endotoxin content assessed with the chromogenic LAL
medium supplemented with 10% (v/v) FBS, 6 mM HEPES and 5 μM 2- test kit (Endosafe-PTS). All vectors used in this study passed the en-
Mercaptoethanol. Media and cell culture reagents were from Gibco Life dotoxin test.
Technologies. Cells were incubated at 37 °C in a humidified atmosphere Viral productivities were assessed in primary viral stocks and pur-
with 5% CO2 in air. ified vectors. Viral particle (vp) titration was performed using the AAV
titration kit ELISA (PROGEN Biotechnik GmbH), according to manu-
2.2. Construction of pAAV-shRNA plasmids facturer’s instructions. For rAAV-shRNA vector genome (vg) quantifi-
cation, viral DNA was extracted and purified with the high pure viral
An AAV2 vector plasmid carrying a CMV-promoter driven GFP gene nucleic acid kit (Roche). Vg titer was quantified by real-time PCR
flanked by ITRs, and a puromycin resistance gene under the control of a quantification (qPCR) of encapsidated CMV promoter DNA fragment
hEF1 promoter outside the AAV2 expression cassette with no shRNA driving the expression of the GFP transgene, in at least 3 different serial
(pAAV-Puro) was generated based on the AAV2 vector construct pAAV- dilutions of extracted viral DNA stocks, against a linearized pAAV-
MSC-CMV-eGFP-CytbAS (kind gift from Prof. U. Michel, University Ctrl1shRNA plasmid DNA standard curve. Eight serial dilutions of the
Medicine Göttingen, Germany), after removal of the extra CMV pro- plasmid standard (containing 108, 107, 106, 105, 104, 103, 102, and 101
moter. To construct the pAAV2-shRNA vectors, DNA fragments con- copies of plasmid DNA) were used to generate a standard curve for
taining the H1 promoter, gene specific or control scramble shRNA se- absolute quantification of vector samples. All reactions were performed
quences, and the BstBI and HindIII restriction sites, at the 5´and in triplicate using CMV specific primers (Forward: 5´-GCGCCTCTTAT
3´extremities, respectively, were chemically synthesized (GenScript) ACCCACGTAG-3′ and reverse: 5´-TAACACCGCCCCGGTTT-3´) and the
and cloned into the BstBI/HindIII site of pAAV-Puro inside the AAV2 SYBR Green I Master mix (Roche). Analysis of qPCR results was carried
expression cassette. The generated pAAV-shRNA plasmids were am- out with LC480 software. Infectious particle (ip) titration was per-
plified in E. coli Stbl3 cells (Invitrogen), and their sequence integrity formed in HT1080 fibrosarcoma cells as previously described(Merten

71
C. Pinto, et al. Journal of Biotechnology 300 (2019) 70–77

and Al-Rubeai, 2011). Briefly, cells were transduced at 50% confluency instructions.
with serial dilutions of the viral stocks for 2 h. After 48 h incubation in
complete media cells were harvested and the percentage of GFP+/in- 2.8. Mouse xenografts and vector delivery
fected cells was quantified by flow cytometry (Cyflow Space, Partec).
All animal experiments were carried out in accordance with the
2.4. pAAV-shRNA plasmid transfection Guidelines for the Care and Use of Laboratory Animals, directive 86/
609/EEC. Human BLBC model was established in N:NIH(s)II:nu/nu
HEK293 T cells at 60–70% confluency were transiently transfected nude mice with MDA-MB-468 cells. Briefly, 2 × 106 MDA-MB-468 cells
with individual pAAV-shRNA constructs at 5 μg DNA per 1 × 106 cells were injected into the mammary fat pad of Female N:NIH(s)II:nu/nu
using PEI as transfection reagent, at a DNA:PEI ratio of 1:2 (μg). After nude mice. Three to four mice were kept per cage, with food and water
8 h incubation, medium was replaced, and cells further cultured. After ad libitum, and examined daily. Mice weight and tumor volume were
24–48 h, cells were imaged by fluorescence microscopy for the ex- measured twice a week until tumors reached an average volume of 100-
pression of GFP transgene and harvested for RNA extraction for gene 150 mm3. Mice were then distributed into different experimental
expression studies and GFP+ cell quantification by flow cytometry groups, with equal average tumor volume, for treatment initiation.
(Cyflow Space, Partec). rAAV2-shRNA intratumoral injections were performed twice a week,
for 3–4 weeks. Tumors were measured in two perpendicular dimensions
2.5. rAAV-shRNA vector transduction and Fluorescence activated cell and the volume was estimated by the formula [volume = (length) x
sorting (width)2/2] for approximating the volume of an ellipsoid.

HCC1954 and MDA-MB-468 cells at 60–70% confluency were 3. Results

transduced with individual rAAV2-shRNA vectors at a multiplicity of
infection (MOI) of 4 × 104 vg/cell in 2% serum containing media. After 3.1. Development of rAAV vectors expressing eGFP and an shRNA targeting
8 h incubation, medium was replaced by complete media and trans- BLBC genetic vulnerabilities
duced cells were FACS sorted for GFP expression 24 h and 48 h there-
after. Following FACS isolation, cells were immediately used for RNA 3.1.1. Design and validation of rAAV-shRNA plasmids
extraction for gene expression studies or further cultured for viability Building on the overrepresentation of proteasome machinery leads
and apoptosis assays. Sorting was conducted at the Instituto Gulbenkian identified as BLBC intrinsic vulnerabilities in a previous study(Petrocca
de Ciência Flow Cytometric facility in Moflo (Beckman Coulter) and et al., 2013), two proteasome subunits were chosen as potential targets
FACSAria (Becton Dickinson) cytometers. – proteasome subunit alpha 2 (PSMA2) and proteasome subunit beta 4
(PSMB4). Another persistent susceptibility identified was MCL1, BCL2
2.6. Gene expression analysis family apoptosis regulator (MCL1), an anti-apoptotic protein that was
the third target chosen for the present study. shRNA sequences tar-
Total RNA from cells was extracted with Qiagen RNeasy Mini Kit as geting these genes (MCL1sh, PSMA2sh and PSMB4sh) were obtained
per manufacturer´s instructions with DNAse digestion. cDNA was syn- from a publicly available database (https://www.broadinstitute.org/
thesized, from equal amounts of RNA, by reverse transcription using the rnai/trc). An shRNA targeting polo like kinase 1 (PLK1), a Ser/Thr
Advantage RT-for-PCR kit (Clontech Laboratories, Mountain View, CA), protein kinase highly expressed during mitosis(Zitouni et al., 2014),
or the High Fidelity cDNA Synthesis Kit (Roche), as per the manufac- was also obtained. Inhibition of this protein has been previously shown
turer's Instructions. Gene expression was quantified on Roche to decrease proliferation and induce apoptosis in cancer cells(Bhola
LightCycler 480 (LC480) using gene specific primers and the SYBR et al., 2015; Liu et al., 2017), thus PLK1sh was used as positive control.
Green I Master mix (Roche). All experiments were performed in tri- Additionally, two scramble shRNA (Ctrl1sh and Ctrl2sh), with no
plicate and the melting-curve data were collected to check product known targets in human or mouse cells, served as negative controls. All
specificity. mRNA transcripts were normalized to both hypoxanthine shRNA sequences were cloned into a pAAV expression vector, co-ex-
phosphoribosyltransferase 1 (HPRT1) and ribosomal protein L22 pressing an eGFP reporter gene to facilitate the monitoring of target cell
(RPL22) mRNA levels in the same sample, and the results were calcu- transduction. Validation of shRNA knockdown efficiency of the dif-
lated as fold change relative to control cells using the advanced relative ferent plasmid constructs was assessed by transient transfection of
quantification method from the LC480 software. The primers used were HEK293 T cells. Target gene knockdown efficiencies ranging from 40 to
(forward and reverse, respectively): HPRT1: 5´-CCTGGCGTCGTGATT 90% were obtained, depending on the shRNA (Fig. 1), showing that the
AGTGAT-3´ and 5´- AGACGTTCAGTCCTGTCCATAA-3′; MCL1: 5´-TGC designed sequences caused potent and specific target gene silencing.
TTCGGAAACTGGACATCA-3´ and 5´- TAGCCACAAAGGCACCAAAAG-
3´; PLK1: 5´-AAAGAGATCCCGGAGGTCCTA-3´ and 5´- GGCTGCGGTG 3.1.2. Platform implementation for endotoxin-free rAAV-shRNA vector
AATGGATATTTC-3´; PSMA2: 5´-GAGCGCGGGTACAGCTTTT-3´ and 5´- production
ACCACACCATTTGCAGCTTTA-3; PSMB4: 5´-GAAGCGTTTTTGGGGT Production processes based on transient transfection of HEK293 T
CGC-3´ and 5´-GAGTGGACGGAATGCGGTA-3; RPL22: 5´-CACGAAGG cells constitute a versatile platform, that enables a rapid performance
AGGAGTGACTGG-3´ and 5´-TGTGGCACACCACTGACATT-3´. screen of multiple rAAV vectors for the most promising lead(Snyder,
2016; Snyder and Moullier, 2011; van der Loo and Wright, 2016).
2.7. Cell viability and apoptosis Therefore, for production of rAAV-shRNA vectors, 2 and 3-plasmid
transfection using CaPO4 or PEI as transfection agent, were tested.
To investigate the effects of rAAV-shRNA vectors on cell viability, Initial assessment of vector productivity was conducted using rAAV-
sorted-transduced cells were plated in 96-well plates at different cell Ctrl1sh to circumvent effects arising from target gene knockdown on
densities and incubated for 48–120 h. After each 24 h incubation the production cell line (HEK293 T), which could mask the results. As
period, cell viability was assessed using the PrestoBlue™ cell viability observed in Fig. 2 – A, there were no significant differences in the
reagent (Life Technologies), according to the manufacturer’s instruc- productivity (vg/cell) obtained with the two transfection reagents
tions. For apoptosis assays transduced cells were plated in 24-well tested. While low cost and high transfection rates can make CaPO4 the
plates, at different cell densities, and the number of apoptotic cells was obvious choice for small scale productions, the similar yields obtained
determined by flow cytometry using the FLICA apoptosis kit in these preparations led us to pursue with PEI transfection, for easier
(Immunochemistry Technologies) as per the manufacturer’s scalability and more consistent results (van der Loo and Wright, 2016).

72
C. Pinto, et al. Journal of Biotechnology 300 (2019) 70–77

Table 1
Optimization of plasmid ratio and harvesting time of rAAV-shRNA vectors.
Harvesting time pDG:pAAVshRNA vg/cell ip/cell vp/vg

48 h post TF 1:1 1.3E+04 7.3E+01 94

1:3 2.9E+04 9.9E+01 29
1:9 2.1E+04 6.0E+01 26
72 h post TF 1:1 1.2E+04 5.2E+01 130
1:3 1.4E+04 8.3E+01 105
1:9 9.1E+03 6.0E+01 87

Fig. 1. Validation of AAV-shRNA plasmids for downregulation of BLBC genetic

vulnerabilities. qRT-PCR analysis of PLK1, MCL1, PSMA2, PSMB4 gene ex-
pression 24 h post-transfection of HEK293 T cells with the different plasmid
constructs. mRNA levels were normalized against the housekeeping RPL22 Fig. 3. Purified rAAV-shRNA vector characterization. Purification of different
mRNA levels within the same sample and are shown as fold change relative to rAAV-shRNA vectors was performed by affinity chromatography using the AVB
the respective levels in control cells transfected with scramble negative control Sepharose high performance resin. Six rAAV-shRNA vectors (two negative
Ctrl1sh plasmid. The graph represents mean values + standard deviation (SD) scramble controls, one positive control, and three rAAV-shRNA vectors tar-
of two independent experiments, with three technical replicates each. geting three different BLBC dependency genes) were obtained. Purity of rAAV-
shRNA vectors was monitored by SDS-PAGE followed by Instant Blue staining.
M: protein standard; VP1, VP2, VP3: AAV structural proteins.
Regarding the transfection systems, as observed in Fig. 2 – B, 2-plasmid
transfection system delivered significantly higher production yields. On
average, the viral vector titer, as per vg/cell, was up to 5-fold higher consistently decreased vp/vg ratio. Also, increasing the proportion of
with the 2-plasmid system relative to the 3-plasmid. pAAV in the transfection mix further decreased empty capsid produc-
Quality control of each AAV vector preparation in terms of total tion. The final optimized protocol was set at pDG:pAAV = 1:3 and
viral particles, packaged viral genomes and infectious particles is cri- toh = 48 hpt for the higher ip yield.
tical before potency assays are conducted, since different rAAV variants Six rAAV-shRNA vectors, two encoding control scramble shRNA
produced potentially have different ratios of the referred parameters. (rAAV-Ctrl1sh, rAAV-Ctrl2sh) and four gene target shRNA (rAAV-
Standard upstream methods for rAAV vector production combine the PK1sh, rAAV-MCL1sh, rAAV-PSMA2sh, rAAV-PSMB4sh) were produced
plasmid carrying the transgene (pAAV) and the helper/packaging and purified by affinity chromatography. SDS-PAGE followed by
plasmid (pDG) in a 1:1 equimolar ratio, and set time of harvest (TOH) at InstantBlue™ staining revealed high purity levels for all batches (Fig. 3).
72 h post transfection (hpt). As observed in Table 1, this combination The titers obtained after purification (1013 vp/ml; 0.4–1 × 1012 vg/ml;
led to a high vp/vg ratio. 1010 ip/ml – Table 2) are within reported ranges(Kotin, 2011; Samulski
In order to decrease this ratio and, consequently, decrease empty and Muzyczka, 2014). Batch quality, in terms of vp/vg/ip, was main-
capsids produced, different combinations of the referred parameters tained within a small range, suggesting that the optimized protocol
were tested. Table 1 shows that decreasing time of harvest to 48 hpt overcome possible side effects arising from RNAi machinery activation

Fig. 2. Platform implementation for rAAV-shRNA vector produc-

tion. Evaluation of rAAV vector productivity following HEK293 T
cell transfection using A) different transfection agents (CaPO4 vs.
PEI) and B) different transfection conditions (2- vs. 3-plasmid
system). The graphs show data (mean + SD) from at least two
independent experiments. *, p < 0.05, statistical significance was
determined using two-tailed Student´s t-test.

73
C. Pinto, et al. Journal of Biotechnology 300 (2019) 70–77

Table 2 impact in BLBC tumorigenic potential, the effects of their knockdown

Purified rAAV-shRNA vector yields. were evaluated in terms of impact on cell viability and apoptosis in-
rAAV-shRNA vp/mL vg/mL ip/mL duction in HCC1954 and MDA-MB-468 cell lines. In MDA-MB-468, a
generalized decrease in viability of around 30% was observed after
scramble control Ctrl1sh 1.3E+13 1.1E+12 2.1E+10 rAAV-mediated delivery of PLK1sh, MCL1sh, PSMA2sh and PSMB4sh,
Ctrl2sh 4.0E+13 9.3E+11 3.4E+10 when compared to cells transduced with scramble control (Fig. 6 – A).
positive control PLK1sh 2.7E+13 3.6E+11 2.0E+10
BLBC genetic vulnerabilities MCL1sh 3.3E+13 4.3E+11 1.7E+10
These results indicate that even a moderate level of gene knockdown
PSMA2sh 4.3E+13 1.8E+12 4.3E+10 may deliver the desired effect. Therefore, further optimization of the
PSMB4sh 3.6E+13 4.7E+11 1.6E+10 shRNA sequences for the genetic vulnerabilities here exploited could be
warranted, in an attempt to increase gene knockdown to potentiate the
functional effect. In terms of apoptosis induction in these cells, a sta-
in HEK293 T cells. Nevertheless, rAAV-shRNA targeting genetic vul- tistically significant increase was achieved with rAAV-PSMA2sh, with
nerabilities showed a tendency to have higher vp/vg ratios. up to 2-fold increase in the percentage of apoptotic cells compared with
the scramble control (Fig. 6 – B). This is in line with the displayed
3.2. PSMA2 downregulation decreases cell viability and induces apoptosis knockdown efficiency of the referred vector. In HCC1954 cells, sig-
in BLBC cell lines nificant decrease of cell viability was observed only for rAAV-PLK1sh
viral vector, and there was no significant effect on apoptosis induction
3.2.1. rAAV-shRNA target gene knockdown efficiency by any of the viral vectors tested in these cells (data not shown). The
rAAV-shRNA vector transduction efficiency was assessed in two different biologic activities of the viral vectors between MDA-MB-468
BLBC cell lines −HCC1954, a poorly differentiated epithelial cell line and HCC1954 might be due to their different molecular signatures and
derived from a ductal carcinoma with no lymph node invasion; and oncogenic status(Grigoriadis et al., 2012).
MDA-MB-468, a cell line with epithelial morphology derived from a
metastatic site. All vectors developed had comparable levels of trans- 3.3. Effect of rAAV-PSMA2sh intratumoral injections in BLBC mouse
duction efficiency for each of the cell lines – around 60% for MDA-MB- xenografts
468 and approximately 80% for HCC1954 (Fig. 4).
Target gene knockdown after rAAV-mediated delivery of shRNA to To evaluate whether the predicted genetic vulnerabilities can be
HCC1954 and MDA-MB-468 cell lines was also assessed. rAAV- harnessed for breast cancer therapy using the developed technology, we
PSMA2sh showed a consistent downregulation of the target gene in assessed whether PSMA2 knockdown could suppress tumor growth in
both cell lines, of around 80% knockdown efficiency (Fig. 5). However, vivo using BLBC mouse xenografts. When tumors were circa 100 mm3,
the other rAAV-shRNA vectors assessed showed either no detectable mice were treated with rAAV-PSMA2sh, rAAV-PLK1sh, rAAV-Ctrl1sh or
gene knockdown (rAAV-MCL1sh), or levels that did not surpass 50% PBS, using two vector dosages, by intratumoral injections. Tumor pro-
knockdown efficiency (rAAV-PLK1sh, rAAV-PSMB4sh). These results gression was monitored every 3 days by measuring the tumor volume in
significantly differ from the levels obtained after HEK293 T transient each condition.
transfection, likely reflecting differences in the shRNA delivery system As an aggressive tumor model, growth of tumor mass was rapid,
and target cell. Nevertheless, since the level of gene knockdown that especially for control groups treated with PBS or with the scramble
translates into an observable phenotype is likely to depend on the gene controls, in both vector dosages tested (Fig. 7). However, mice treated
of interest and on the expected outcome, it is difficult to predict if the with rAAV-PSMA2 showed a reduction in tumor growth over time,
knockdown efficiency observed will have a functional effect. apparent after day 13 of treatment (Fig. 7). Interestingly, for rAAV-
PLK1sh, a reduction in tumor growth is only apparent for the highest
3.2.2. In vitro biological activity dose, possibly reflecting the lower gene downregulation achieved with
To evaluate whether downregulation of the target genes has an this construct and consequent lower impact on cell viability.

Fig. 4. Transduction efficiency of rAAV-shRNA viral vectors in BLBC cell lines. Exponentially growing MDA-MB-468 and HCC1954 cells were not transduced (NT) or
transduced with the indicated viral vectors (4.6E + 4 vg/cell each). rAAV vector transduction efficiency was monitored 48 h post transduction by flow cytometry
detection of eGFP. Plots show representative data obtained for each cell line. Graphics represent percentage of transduced cells (mean ± SD) from three independent
experiments.

74
C. Pinto, et al. Journal of Biotechnology 300 (2019) 70–77

Fig. 5. Target gene knockdown efficiency of

rAAV-shRNA vectors in BLBC cell lines.
Exponentially growing A) MDA-MB-468 and B)
HCC1954 cells were transduced with the in-
dicated viral vectors (4.6E + 4 vg/cell each).
Twenty-four or forty-eight hours post trans-
duction cells were sorted and processed. Total
RNA was extracted and the mRNA levels of
PLK1, MCL1, PSMA2 and PSMB4 were de-
termined by qRT-PCR. mRNA levels were nor-
malized against housekeeping HPRT1 and
RPL22 genes within the same sample and are
shown as fold change relative to the respective
levels in cells transduced with rAAV-Ctrl1sh
vector (mean + SD) of three independent ex-
periments.

Fig. 6. Functional effects of rAAV-shRNA vec-

tors in BLBC cell lines. Exponentially growing
MDA-MB-468 cells were transduced with the
indicated viral vectors (4.6E + 4 vg/cell each).
Twenty-four and forty-eight hours thereafter,
transduced cells were sorted and further cul-
tured. A) Cell viability was measured using
Presto Blue fluorometric assay at different
culture time points. Cellular viability was cal-
culated as a percentage relative to cells trans-
duced with rAAV-Ctrl1sh vector. Shown is
mean percent viability of transduced cells at
72 h post-transduction + SD from 3 in-
dependent experiments. B) Sorted cells were
seeded, and apoptosis determined by FLICA
apoptosis kit by flow cytometry. The graphics
show average cell apoptosis (+ SD) of three
independent experiments, at day 5 post-trans-
duction. *, p < 0.05 vs. values in cells trans-
duced with scramble control (Ctrl1sh).
Statistical significance was determined using
the two-tailed Student´s t-test.

No significant differences were observed in the gain or loss of mice delivery for the downregulation of genes necessary for the survival of
weight. Moreover, no major macroscopic alterations were observed in BLBC cells. Our results indicate that standard upstream methods for
the liver of mice treated with either vector, suggesting that the rAAV2- rAAV production, based on transient transfection of HEK293 T cells,
shRNA viral injections did not prompt significant side-effects nor he- lead to high empty capsid titer in the final formulation of rAAV vectors
patotoxicity. encoding shRNA constructs. We showed that by increasing the trans-
gene/packaging plasmid ratio in the transfection mix, which likely re-
duces the probability of empty capsid formation, the vp/vg ratio was
4. Discussion improved and ip titers were slightly increased. Additionally, the com-
bination of the optimal ratio with a decrease in the time of harvest to
This study evaluates the potential of rAAV-mediated shRNA

Fig. 7. Effect of rAAV-shRNA vectors on tumor

growth rate of BLBC xenografts. A) and B) Fold
change in tumor volume progression in mice
injected with the indicated viral vectors at
2 × 109 and 2 × 108 vg/tumor/injection, re-
spectively. Data are mean ± standard error of
mean of 10 animals injected with PBS, of 5
animals injected with 2 × 108 virus/tumor,
and of 7 animals injected with 2 × 109 virus/
tumor. The statistical difference between the
groups was determined by Two-way ANOVA
using the Graphpad 6 (Prism®) software
(**p < 0.01, *p < 0.05 compared with con-
trol mice injected with the scramble control
rAA-Ctrl1sh, for each vector dosage tested).

75
C. Pinto, et al. Journal of Biotechnology 300 (2019) 70–77

48 hpt, further lowered empty viral vector titers. A lower vp/vg ratio is posttranscriptional gene silencing via shRNA can target each gene’s
instrumental for the successful employment of gene therapy approaches mRNA, as well as the considered undruggable genome(Kacsinta and
as empty capsids will decrease therapy potency and increase immune Dowdy, 2016). In contrast to proteasome inhibitors, this approach of-
responses to capsid proteins. fers the possibility to evolve in a sequence specific manner, which could
The success of anticancer therapeutics using the referred approach, overcome resistance mechanisms arising from genetic mutations. Also,
or even the correct interrogation as to possible clinical utility, is de- it possesses superior advantages over other strategies with proven
pendent on the development of suitable delivery systems for safe and clinical efficacy, including small-molecule inhibitors, such as imatinib
efficient introduction of shRNA molecules into tumor cells. AAV was for the treatment of chronic myeloid leukemia(Druker et al., 2006), and
shown to be a viable option, as rAAV vectors are among the most monoclonal antibodies, such as Trastuzumab for HER2-positive early
promising gene therapy products developed to date. After numerous breast cancer(Smith et al., 2007). Small-molecule inhibitors have low
successful applications in the treatment of monogenic diseases by gene degree of specificity when it comes to target modulation, which can
replacement(Cideciyan et al., 2009; Nathwani et al., 2014; Valori et al., lead to adverse side effects, and often do not restrict the entire function
2010), AAV arsenal has been broadened to comprise several anti-cancer of a protein, leading to inefficient anti-oncogenic effects(Kikani et al.,
therapies(Luo et al., 2015) and different clinical modalities that include 2005; Lim et al., 2008; Pecot et al., 2011; Weihua et al., 2008). On the
RNAi and genome engineering strategies such as CRISPR/ other hand, monoclonal antibodies are restricted to membrane or cir-
Cas9(Valdmanis and Kay, 2017). The low transgene loading capacity, culating proteins(Pecot et al., 2011).
often impairing rAAV-mediated gene therapy, is not an issue with these Taken together, this study showed evidence that further optimiza-
approaches, and makes such small DNA-encoded RNA molecules ideal tion of standard upstream rAAV production platforms, based on tran-
transgenes for these vectors. sient transfection of HEK293 T cells, has the potential to overcome
Also pivotal for successful therapy is maximizing vector production transgene cytotoxicity in vector-producing cells. Moreover, our results
while maintaining batch consistency. However, reaching high titers provide proof-of-concept that rAAV-mediated delivery of shRNA tar-
during upstream processing entails high expression of the transgene in geting BLBC dependency genes can be a promising approach for the
producer cells. While few concerns arise when dealing with inert gene treatment of this disease. The strategy here developed represents an
products, that generally set the gold standard for vector titer produc- improvement on traditional employment of suicide gene therapy for
tion, the biological activity of cytotoxic transgenes for cancer therapy cancer. Classical suicide gene approaches will kill cells indiscriminately
on producer cell lines may impact its protein synthesis capacity, me- and rely solely on targeted delivery for specificity. Directing therapy to
tabolism and viability, or impact vector assembly itself(Maunder et al., cancer specific dependency genes will confer a higher degree of speci-
2017). Concerning shRNA applications, the activation of the RNAi ficity and safety to the treatment. Nevertheless, it still maintains the
machinery in HEK293 T cells during rAAV production, combined with possibility to be combined with capsid engineering approaches for an
the extra burden of producing heterologous proteins, decreased cell additional layer of specificity(Chen et al., 2017; Sayroo et al., 2016).
viability over time (which was lower at 72 hpt – data not shown), and This strategy might also be suitable for treatment of other malignancies.
impaired the quality of vector preparations. Our findings add to a The shRNA can be customized to any tumor dependency and tailored to
growing body of literature on strategies to circumvent such side effects. different subtypes or different tumors. Also, a combination of rAAV
These include the addition of extra proteins and extensive genetic batches encoding different transgenes can be used to decrease the
manipulation of the transgene(Maunder et al., 2017), addition of ap- emergence of tumor resistance.
tamers(Strobel et al., 2015), the inhibition of transgene expression on Finally, given the indication for therapeutic potential of these tar-
producer cells via siRNA or by activation of toxic genes only through gets, evaluation of clinical efficacy in more relevant models of tumor-
the excision of spacer DNA by Cre-recombinase delivered via a second igenesis should be warranted. Although a decrease in tumor growth
vector(Lee and Jameson, 2002). The simpler strategy applied here is over time was observed in BLBC mouse xenograft models, these mice
flexible enough to allow transgene-dependent optimization if required, are immunocompromised and, thus, do not portray the immunogenicity
while reaching yields comparable to current rAAV production platforms of the therapy developed. It is now established that several anticancer
(Kotin, 2011; Snyder, 2016). therapeutics that induce apoptosis in tumor cells can trigger an antigen-
To assess the potential of this strategy to be used as a BLBC ther- specific immune response(Galluzzi et al., 2016). This, in turn, re-
apeutic approach, we transduced BLBC cell lines with rAAV vectors activates the immune system against tumor cells, enhancing therapeutic
expressing shRNA against previously identified genetic vulnerabilities response. Consequently, the assessment of anticancer drugs is now
of this breast cancer subtype. rAAV-PSMA2 transduction caused a shifting towards the use of immunocompetent mice for proper evalua-
specific and potent decrease in PSMA2 expression in both cell lines tion of the therapeutic outcome(Galluzzi et al., 2016). Furthermore,
tested, as compared to cells transduced with rAAV carrying a scramble with the development of in vitro human models that incorporate im-
shRNA. The reduced expression of PSMA2 resulted in a decrease of mune cells, testing promising leads in a human setting may also be
viability and increased apoptosis in MDA-MB-468 cells. Furthermore, possible(Nyga et al., 2016; Rebelo et al., 2018).
rAAV2-PSMA2 intratumoral injections in BLBC mouse xenografts led to
a decrease in tumor growth over time, when compared to PBS or
scramble controls. These results are consistent with Petrocca et al, Declarations of interest
which indicated that proteins from the proteasome machinery are
fundamental for BLBC cell viability(Petrocca et al., 2013), and suggest None.
that this could be a potential therapeutic vector against this type of
cancer, upon further studies.
Although previously conducted clinical trials have led to regulatory Acknowledgments
approval of proteasome inhibitors for the treatment of various diseases,
including multiple myeloma and mantle-cell lymphoma, acquired re- The authors acknowledge Dr Judy Lieberman for scientific support.
sistance mechanisms commonly arise and undermine effectiveness of The work was supported by FCT grant PTDC/BBB-BIO/1240/2012 and
the therapy(Manasanch and Orlowski, 2017). Also, preclinical efficacy PhD fellowship PD/BD/52202/2013 to C. Pinto. iNOVA4Health - UID/
of these inhibitors for solid tumors could not be translated into the Multi/04462/2019, a program financially supported by Fundação para
clinic, indicating the importance for development of suitable delivery a Ciência e Tecnologia/ Ministério da Educação e Ciência, through
platforms, as even the most potent therapeutic can fail in the clinic due national funds and co-funded by FEDER under the PT2020 Partnership
to inefficient delivery(Manasanch and Orlowski, 2017). Therapeutic Agreement is acknowledged.

76
C. Pinto, et al. Journal of Biotechnology 300 (2019) 70–77

References K.A., Farley, D.C., 2017. Enhancing titres of therapeutic viral vectors using the
transgene repression in vector production (TRiP) system. Nat. Commun. 8, 14834.
https://doi.org/10.1038/ncomms14834.
Bhola, N.E., Jansen, V.M., Bafna, S., Giltnane, J.M., Balko, J.M., Estrada, M.V., Meszoely, Viral vectors for Gene therapy, methods in molecular biology. In: Merten, O.-W., Al-
I., Mayer, I., Abramson, V., Ye, F., Sanders, M., Dugger, T.C., Allen, E.V., Arteaga, Rubeai, M. (Eds.), Methods in Molecular Biology. Humana Press, Totowa, NJ. https://
C.L., 2015. Kinome-wide functional screen identifies role of PLK1 in hormone-in- doi.org/10.1007/978-1-61779-095-9.
dependent, ER-positive breast cancer. Cancer Res. 75, 405–414. https://doi.org/10. Miest, T.S., Cattaneo, R., 2013. New viruses for cancer therapy: meeting clinical needs.
1158/0008-5472.CAN-14-2475. Nat. Rev. Microbiol. 12, 23–34. https://doi.org/10.1038/nrmicro3140.
Bianchini, G., Balko, J.M., Mayer, I.A., Sanders, M.E., Gianni, L., 2016. Triple-negative Naldini, L., 2015. Gene therapy returns to centre stage. Nature 526, 351–360. https://doi.
breast cancer: challenges and opportunities of a heterogeneous disease. Nat. Rev. org/10.1038/nature15818.
Clin. Oncol. 13, 674–690. https://doi.org/10.1038/nrclinonc.2016.66. Nathwani, A.C., Reiss, U.M., Tuddenham, E.G.D., Rosales, C., Chowdary, P., McIntosh, J.,
Brimble, M.A., Reiss, U.M., Nathwani, A.C., Davidoff, A.M., 2016. New and improved Della Peruta, M., Lheriteau, E., Patel, N., Raj, D., Riddell, A., Pie, J., Rangarajan, S.,
AAVenues: current status of hemophilia B gene therapy. Expert Opin. Biol. Ther. 16, Bevan, D., Recht, M., Shen, Y.-M., Halka, K.G., Basner-Tschakarjan, E., Mingozzi, F.,
79–92. https://doi.org/10.1517/14712598.2015.1106475. High, K.A., Allay, J., Kay, M.A., Ng, C.Y.C., Zhou, J., Cancio, M., Morton, C.L., Gray,
Bryant, L.M., Christopher, D.M., Giles, A.R., Hinderer, C., Rodriguez, J.L., Smith, J.B., J.T., Srivastava, D., Nienhuis, A.W., Davidoff, A.M., 2014. Long-term safety and ef-
Traxler, Ea, Tycko, J., Wojno, A.P., Wilson, J.M., 2013. Lessons learned from the ficacy of factor IX gene therapy in hemophilia B. N. New England J. Med. Surg. Collat.
clinical development and market authorization of glybera. Hum. Gene Ther. Clin. Branches Sci. 371, 1994–2004. https://doi.org/10.1056/NEJMoa1407309.
Dev. 24, 55–64. https://doi.org/10.1089/humc.2013.087. Nyga, A., Neves, J., Stamati, K., Loizidou, M., Emberton, M., Cheema, U., 2016. The next
Caldas-lopes, E., Cerchietti, L., Ahn, J.H., Clement, C.C., Robles, A.I., Rodina, A., Moulick, level of 3D tumour models: immunocompetence. Drug Discov. Today 21, 1421–1428.
K., Taldone, T., Gozman, A., Guo, Y., Wu, N., Stanchina De, E., White, J., Gross, S.S., https://doi.org/10.1016/j.drudis.2016.04.010.
Ma, Y., Varticovski, L., Melnick, A., Chiosis, G., 2009. Hsp90 inhibitor PU-H71, a Pecot, C.V., Calin, G.A., Coleman, R.L., Lopez-Berestein, G., Sood, A.K., 2011. RNA in-
multimodal inhibitor of malignancy, induces complete responses in triple-negative terference in the clinic: challenges and future directions. Nat. Rev. Cancer 11, 59–67.
breast cancer models. PNAS 106, 8368–8373. https://doi.org/10.1038/nrc2966.
Carey, L., Winer, E., Viale, G., Cameron, D., Gianni, L., 2010. Triple-negative breast Petrocca, F., Altschuler, G., Tan, S.M., Mendillo, M.L., Yan, H., Jerry, D.J., Kung, A.L.,
cancer: disease entity or title of convenience? Nat. Rev. Clin. Oncol. 7, 683–692. Hide, W., Ince, Ta, Lieberman, J., 2013. A genome-wide siRNA screen identifies
https://doi.org/10.1038/nrclinonc.2010.154. proteasome addiction as a vulnerability of basal-like triple-negative breast cancer
Chen, M., Maeng, K., Nawab, A., Francois, R.A., Bray, J.K., Reinhard, M.K., Boye, S.L., cells. Cancer Cell 24, 182–196. https://doi.org/10.1016/j.ccr.2013.07.008.
Hauswirth, W.W., Kaye, F.J., Aslanidi, G., Srivastava, A., Zajac-Kaye, M., 2017. Rebelo, S.P., Pinto, C., Martins, T.R., Harrer, N., Estrada, M.F., Loza-Alvarez, P.,
Efficient gene delivery and expression in pancreas and pancreatic tumors by capsid- Cabeçadas, J., Alves, P.M., Gualda, E.J., Sommergruber, W., Brito, C., 2018. 3D-3-
optimized AAV8 vectors. Hum. Gene Ther. Methods 28, 49–59. https://doi.org/10. culture: a tool to unveil macrophage plasticity in the tumour microenvironment.
1089/hgtb.2016.089. Biomaterials 163, 185–197. https://doi.org/10.1016/j.biomaterials.2018.02.030.
Cideciyan, A.V., Hauswirth, W.W., Aleman, T.S., Kaushal, S., Schwartz, S.B., Boye, S.L., Samulski, R.J., Muzyczka, N., 2014. AAV-mediated gene therapy for research and ther-
Windsor, E.A.M., Conlon, T.J., Sumaroka, A., Roman, A.J., Byrne, B.J., Jacobson, apeutic purposes. Annu. Rev. Virol. 1, 427–451. https://doi.org/10.1146/annurev-
S.G., 2009. Vision 1 year after gene therapy for Leber’s congenital amaurosis. N. Engl. virology-031413-085355.
J. Med. 361, 725–727. https://doi.org/10.1056/NEJMc0903652. Santiago-Ortiz, J.L., Schaffer, D.V., 2016. Adeno-associated virus (AAV) vectors in cancer
Druker, B.J., Guilhot, F., O’Brien, S.G., Gathmann, I., Kantarjian, H., Gattermann, N., gene therapy. J. Control. Release 240, 287–301. https://doi.org/10.1016/j.jconrel.
Deininger, M.W.N., Silver, R.T., Goldman, J.M., Stone, R.M., Cervantes, F., Hochhaus, 2016.01.001.
A., Powell, B.L., Gabrilove, J.L., Rousselot, P., Reiffers, J., Cornelissen, J.J., Hughes, Sayroo, R., Nolasco, D., Yin, Z., Colon-Cortes, Y., Pandya, M., Ling, C., Aslanidi, G., 2016.
T., Agis, H., Fischer, T., Verhoef, G., Shepherd, J., Saglio, G., Gratwohl, A., Nielsen, Development of novel AAV serotype 6 based vectors with selective tropism for human
J.L., Radich, J.P., Simonsson, B., Taylor, K., Baccarani, M., So, C., Letvak, L., Larson, cancer cells. Gene Ther. 23, 18–25. https://doi.org/10.1038/gt.2015.89.
R.A., 2006. Five-year follow-up of patients receiving imatinib for chronic myeloid Schneider, B.P., Winer, E.P., Foulkes, W.D., Garber, J., Perou, C.M., Richardson, A.,
leukemia. N. Engl. J. Med. 355, 2408–2417. https://doi.org/10.1056/ Sledge, G.W., Carey, L.A., 2008. Triple-negative breast cancer: risk factors to po-
NEJMoa062867. tential targets. Clin. Cancer Res. 14, 8010–8018. https://doi.org/10.1158/1078-
Galluzzi, L., Buqué, A., Kepp, O., Zitvogel, L., Kroemer, G., 2016. Immunogenic cell death 0432.CCR-08-1208.
in cancer and infectious disease. Nat. Rev. Immunol. 2–5. https://doi.org/10.1038/ Smith, I., Procter, M., Gelber, R.D., Guillaume, S., Feyereislova, A., Dowsett, M.,
nri.2016.107. Goldhirsch, A., Untch, M., Mariani, G., Baselga, J., Kaufmann, M., Cameron, D., Bell,
Grigoriadis, A., Mackay, A., Noel, E., Wu, P., Natrajan, R., Frankum, J., Reis-Filho, J.S., R., Bergh, J., Coleman, R., Wardley, A., Harbeck, N., Lopez, R.I., Mallmann, P.,
Tutt, A., 2012. Molecular characterisation of cell line models for triple-negative Gelmon, K., Wilcken, N., Wist, E., Sánchez Rovira, P., Piccart-Gebhart, M.J., 2007. 2-
breast cancers. BMC Genomics 13, 619. https://doi.org/10.1186/1471-2164-13-619. year follow-up of trastuzumab after adjuvant chemotherapy in HER2-positive breast
Grimm, D., Kay, M.A., Kleinschmidt, J.A., 2003. Helper virus-free, optically controllable, cancer: a randomised controlled trial. Lancet 369, 29–36. https://doi.org/10.1016/
and two-plasmid-based production of adeno-associated virus vectors of serotypes 1 to S0140-6736(07)60028-2.
6. Mol. Ther. 7, 839–850. https://doi.org/10.1016/S1525-0016(03)00095-9. Snyder, R.O., 2016. An overview of rAAV vector product development for Gene therapy.
Kacsinta, A.D., Dowdy, S.F., 2016. Current views on inducing synthetic lethal RNAi re- In: Childers, M.K. (Ed.), Stem Cell Biology and Regenerative Medicine. Springer, New
sponses in the treatment of cancer. Expert Opin. Biol. Ther. 16, 161–172. https://doi. York, New York, NY, pp. 21–37. https://doi.org/10.1007/978-1-4939-3228-3_2.
org/10.1517/14712598.2016.1110141. Adeno-associated virus. In: Snyder, R.O., Moullier, P. (Eds.), Methods in Molecular
Kaplitt, M.G., Feigin, A., Tang, C., Fitzsimons, H.L., Mattis, P., Lawlor, P.A., Bland, R.J., Biology. Humana Press, Totowa, NJ. https://doi.org/10.1007/978-1-61779-370-7.
Young, D., Strybing, K., Eidelberg, D., During, M.J., 2007. Safety and tolerability of Strobel, B., Klauser, B., Hartig, J.S., Lamla, T., Gantner, F., Kreuz, S., 2015. Riboswitch-
gene therapy with an adeno-associated virus (AAV) borne GAD gene for Parkinson’s mediated attenuation of transgene cytotoxicity increases adeno-associated virus
disease: an open label, phase I trial. Lancet 369, 2097–2105. https://doi.org/10. vector yields in HEK-293 cells. Mol. Ther. 23, 1582–1591. https://doi.org/10.1038/
1016/S0140-6736(07)60982-9. mt.2015.123.
Kikani, C.K., Dong, L.Q., Liu, F., 2005. “New”-clear functions of PDK1: beyond a master Testa, F., Maguire, A.M., Rossi, S., Pierce, E.A., Melillo, P., Marshall, K., Banfi, S., Surace,
kinase? J. Cell. Biochem. 96, 1157–1162. https://doi.org/10.1002/jcb.20651. E.M., Sun, J., Acerra, C., Wright, J.F., Wellman, J., High, K.A., Auricchio, A., Bennett,
Kotin, R.M., 2011. Large-scale recombinant adeno-associated virus production. Hum. J., Simonelli, F., 2013. Three-Year Follow-up after Unilateral Subretinal Delivery of
Mol. Genet. 20, R2–R6. https://doi.org/10.1093/hmg/ddr141. Adeno-Associated Virus in Patients with Leber Congenital Amaurosis Type 2.
Ledford, H., 2015. Cancer-fighting viruses win approval. Nature 526, 622–623. https:// Ophthalmology 120, 1283–1291. https://doi.org/10.1016/j.ophtha.2012.11.048.
doi.org/10.1038/526622a. Valdmanis, P.N., Kay, M.A., 2017. Future of rAAV gene therapy: platform for RNAi, gene
Lee, E.J., Jameson, J.L., 2002. Cell-specific cre-mediated activation of the diphtheria editing, and beyond. Hum. Gene Ther. 28, 361–372. https://doi.org/10.1089/hum.
toxin gene in pituitary tumor cells: potential for cytotoxic gene therapy. Hum. Gene 2016.171.
Ther. 13, 533–542. https://doi.org/10.1089/10430340252809829. Valori, C.F., Ning, K., Wyles, M., Mead, R.J., Grierson, A.J., Shaw, P.J., Azzouz, M., 2010.
Lim, S.T., Mikolon, D., Stupack, D.G., Schlaepfer, D.D., 2008. FERM control of FAK Systemic delivery of scAAV9 expressing SMN prolongs survival in a model of spinal
function: implications for cancer therapy. Cell Cycle 7, 2306–2314. https://doi.org/ muscular atrophy. Sci. Transl. Med. 2https://doi.org/10.1126/scitranslmed.
10.4161/cc.6367. 3000830. 35ra42-35ra42.
Liu, Z., Sun, Q., Wang, X., 2017. PLK1, a potential target for Cancer therapy. Transl. van der Loo, J.C.M., Wright, J.F., 2016. Progress and challenges in viral vector manu-
Oncol. 10, 22–32. https://doi.org/10.1016/j.tranon.2016.10.003. facturing. Hum. Mol. Genet. 25, R42–R52. https://doi.org/10.1093/hmg/ddv451.
Luo, J., Luo, Y., Sun, J., Zhou, Y., Zhang, Y., Yang, X., 2015. Adeno-associated virus- Weihua, Z., Tsan, R., Huang, W.C., Wu, Q., Chiu, C.H., Fidler, I.J., Hung, M.C., 2008.
mediated cancer gene therapy: current status. Cancer Lett. 356, 347–356. https://doi. Survival of Cancer cells is maintained by EGFR independent of its kinase activity.
org/10.1016/j.canlet.2014.10.045. Cancer Cell 13, 385–393. https://doi.org/10.1016/j.ccr.2008.03.015.
Manasanch, E.E., Orlowski, R.Z., 2017. Proteasome inhibitors in cancer therapy. Nat. Rev. Zitouni, S., Nabais, C., Jana, S.C., Guerrero, A., Bettencourt-Dias, M., 2014. Polo-like
Clin. Oncol. 5, 101–110. https://doi.org/10.1038/nrclinonc.2016.206. kinases: structural variations lead to multiple functions. Nat. Rev. Mol. Cell Biol. 15,
Maunder, H.E., Wright, J., Kolli, B.R., Vieira, C.R., Mkandawire, T.T., Tatoris, S., 433–452. https://doi.org/10.1038/nrm3819.
Kennedy, V., Iqball, S., Devarajan, G., Ellis, S., Lad, Y., Clarkson, N.G., Mitrophanous,

77