Just Accepted
“Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just
Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscriptsis published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036
Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Bioconjugate Chemistry

Suicide gene therapy in AML with receptor targeted AAV vectors is effective both in vitro and in vivo.
254x190mm (300 x 300 DPI) ACS Paragon Plus Environment

A CD33 antigen targeted AAV6 vector expressing an inducible caspase-9 suicide gene is
therapeutic in a xenotransplantation model of acute myeloid leukemia.

Nusrat Khan1, Sridhar Bammidi1, Giridhara R. Jayandharan1*
1Department of Biological Sciences and Bioengineering, Indian Institute of Technology,
Kanpur, 208016, UP, India;

*Correspondence:
GR Jayandharan, PhD
Associate Professor,
Department of Biological Sciences and Bioengineering,
Kanpur 208 016 (U.P), India
Email: [email protected]
Ph: +91 512 2594086
Fax: +91 512 2594010
Short title: Targeted AAV6 vectors for AML gene therapy

Keywords: suicide gene delivery, AAV, targeted vector, AML, xenotransplantation.

ABSTRACT:
4
Current chemotherapeutic regimens for acute myeloid leukemia (AML) has been modestly
effective in patients, and are associated with poor long term survival (<30% at 5 years). Viral
8 9
7 vector-based suicide gene therapy is an attractive option, if these vectors can target the AML
10 cells with high specificity and efficiency. In this study, we have developed a receptor specific
Adeno-associated virus (AAV) based vector to target the CD33 antigen which is over-
13 14
12 expressed in leukemic cells. A targeting peptide was rationally designed from the antigen-
binding regions of a CD33 monoclonal antibody. This peptide was further expressed on thecapsid of AAV6 vector, since this serotype was most
efficient among AAV1-rh10 vectors to infect the pro-monocytic, human myeloid leukemia cells (U937). AAV6-CD33 vectors
20 21
19 expressing a suicide gene, the inducible caspase 9 (iCasp9) and its prodrug AP20187
22 significantly reduced (~59%) the viability of U937 cells. To further test its efficacy andspecificity in vivo, AAV6-CD33 vectors were
administered into a xenotransplantation model
25 of AML in zebrafish through systemic delivery. We observed a significant anti-leukemic
27 28
26 effect with AAV6-CD33 vectors, with a markedly higher survival (100% for AAV6-CD33
29 vectors vs. 15% for mock-treated) and higher number of TUNEL positive apoptotic cells after
31 systemic vector delivery. Taken together, our work demonstrates the efficacy and translational
32 33
31 potential of CD33-targeted AAV6 vectors for cytotoxic gene therapy in AML.

INTRODUCTION

5 Acute myeloid leukemia (AML) is a hematological malignancy characterized by presence of
6
7 undifferentiated myeloid blasts and is associated with poor prognosis in both adults and
8 9
7 children 1-4. The heterogeneity of AML with its different clinical manifestations, the genetic
10 abnormalities as well as the performance status of the patient strongly influences the choice
11
12 of therapy. Conventional chemotherapy fails to address the presence of leukemia initiating
13 14
12 compartment in the bulk AML population that are generally resistant to chemotherapy. Even
15 though, ~65 to 75% of AML patients (<60 years) achieve complete hematological remission
16
17 in response to currently available drugs for treatment, the overall long-term (5 year) survival
18
19 is < 30%, due to a high relapse rate 5. There is no effective treatment for late stage disease and
20 21
19 the cytotoxic induction chemotherapy is accompanied by significant mortality (~30%) 6, 7 or
22 systemic toxicity 8. Thus, novel treatment strategies are warranted.
23
24 Suicide gene therapy may be a valuable tool to treat AML, as it can offer a high
25
26 therapeutic index achieved by selectively introducing pro-apoptotic genes into cancer cells,
27 28
26 rendering them susceptible to specific enzyme pro-drug combination 9. A widely used suicide
29 gene system is the Herpes simplex virus thymidine kinase (HSV-TK) and its prodrug,
30
31 ganciclovir (GCV) 10, 11. A significant limitation however, is the immunogenicity of the viral
32 33
31 construct, the HSV-TK gene and their specific requirement of the activated drug that can
34 efficiently target only the dividing cells 11. To overcome this limitation, recently an inducible
35
36 caspase (iCasp9) gene and a chemical inducer of dimerization (AP20187) has been evaluated
37
38 successfully for suicide gene therapy in hepatocellular carcinoma (HCC) 12. While this
39 40
38 approach is effective in solid tumors such as HCC due to the localized nature of gene delivery,
41 its efficacy in blood cancers like AML may be limited and accompanied by off-target effects.
42
43 Thus, the critical step for the success of suicide gene therapy in AML, is the availability of a
44
45 potent and specific gene delivery system that lead to a sustained and targeted increase of the
46 47
45 pro-apoptotic gene product such as iCasp9 in only the cancer cells, with no apparent vector-
48 related toxicities in the recipient.
49
50 Among the currently available viral vectors, Adeno-associated virus (AAV) based
51 52
50 vectors (serotypes 1-rh10) are known to be potentially safe and efficacious for cancer gene
53 therapy 13-15. The utility of AAV1 serotype for muscle-directed systemic cancer gene therapy
54
55 using anti-angiogenic agents together with MDA-7/IL24 has been established 16. Further, an
56
57 AAV serotype 2 mediated HSV-TK/GCV gene therapy with a doxycycline based inducible
58 59
57 enzyme activity achieved by direct intra-tumoral injections has been efficacious in a murine
60 model of breast cancer 17. Recombinant AAV is also known to be efficient in transducing

3 primary B-CLL cells with the co-stimulatory CD40 ligand molecules thus enabling
4
5 vaccination strategies 18. While these studies underscore the utility of multiple AAV serotypes
6
7 for cancer gene therapy, it must be noted that the crucial element of success of a therapeutic
8 9
7 AAV vector in AML is primarily based on its preferential and specific activity on leukemic
10 cells to accentuate its suicide gene activity. However, the availability of AAV vectors
11
12 sufficiently and specifically transducing hematopoietic or myeloid cell type is very limited.
13 14
12 In fact, hematopoietic cells are particularly intransigent to AAV transduction 19, 20.
15 Different strategies have been attempted to develop targeted AAV vectors mostly in
16
17 context of AAV2 serotype 21, 22. Strategies such as coupling of chemical conjugates or
18
19 bispecific ligands to AAV have shown significant efficacy in vitro 22. However, further in
20 21
19 vivo application of these systems may be constrained by the relative instability of the vector-
22 conjugate complex and the requirement of a significant amount of antibodies to generate these
23
24 complexes 22. The strategy we have proposed here, involves the genetic fusion of a specific
25
26 targeting sequence in defined regions of AAV capsid 20, 23-26. This allows for the direct and
27 28
26 stable expression of such receptor targeting ligands on AAV vectors, thereby greatly
29 improving the efficiency of vector delivery 27. Several reports have demonstrated that the loop
30
31 region of AAV VP1 capsid (amino acids 584-590), which is surface exposed, can tolerate a
32 33
31 peptide insertion of ~7 to 14-mer length without affecting their packaging ability or infectivity
34 25, 26, 28-30. An RGD motif inserted adjacent to an arginine at amino acid position 588 (R588)
35
36 in AAV2 serotype, has demonstrated preferential transduction of integrin-expressing 26 or
37
38 CD13 or Myc-expressing cells in various in vitro cancer models 29. In the context of AML,
39 40
38 such receptor-targeted vectors towards leukemic cells are limited, with only a single in vitro
41 analysis of a peptide selected from a random library and its display in an AAV2 serotype 20.
42
43 Interestingly, CD33 is a myeloid-specific sialic acid-binding receptor over-expressed in ~90%
44
45 of AML blasts 31, 32. The specificity of CD33 targeting in AML is clinically tested through
46 47
45 Gemtuzumab ozogamicinâ (GO), a humanized anti-CD33 monoclonal antibody based
48 immunotherapy 33. Further, the expression of CD33 is restricted to early progenitor cells but
49
50 absent in hematopoietic stem cells 34, 35, rendering them an ideal target to direct suicide gene
51
52 delivery vectors in the context of AML. Thus, we reasoned that synthetic engineering of AAV
53 54
52 capsid by incorporation of a CD33 binding peptide is likely to facilitate the specific and
55 efficient targeting of cytotoxic genes to AML cells both in vitro and in vivo.

RESULTS AND DISCUSSION
4
5 Analysis of CD33 expression in leukemic cells.
6
7 In the first set of studies, we wished to determine the level of expression of CD33 surface
8 9
7 antigen in leukemic cells in anticipation of using them further as suitable models to test the
10 AAV vectors. We used a panel of human cells of leukemic origin, human lymphoblastic
11
12 myeloid leukemic cells (U937), human erythroleukemia cells (K562), human T lymphoblastic
13 14
12 leukemia cell line (CEM) along with a control cell line, human hepatocellular carcinoma
15 (Huh7). Approximately, 3x104 cells in synchronous phase were first incubated with
16
17 unconjugated anti-human CD33 antibody and further detected by an AlexaFlour488
18
19 conjugated secondary antibody. Immunophenotypic analysis revealed that the expression of
20 21
19 CD33 surface marker varies significantly among leukemic cell lines. As can be seen in
22 Supporting information Table S1, our data revealed that the expression of CD33 antigen
23
24 was significantly high in U-937 cells (85.7 + 9.1%), but was negligible in K562 cells (5.7 +
25
26 2.6%) cells. Furthermore, the expression of CD33 was negligible in either CEM or Huh7 cells
27 28
26 (Supporting information Table S1). These findings are further confirmed by other studies
29 for CD33 expression on common leukemic cell lines 36, 37. Based on these data, we
30
31 subsequently employed U937 cells to test the AAV vectors developed both in vitro and in
32 33
31 vivo. Furthermore, U937 cells was specifically utilized, since we have recently established
34 and characterized the AML xenograft model of zebrafish using this cell line 38.

38 Characterization of AAV serotype 1-rh10 transduction in leukemic cells.
39 40
38 Multiple AAV serotypes exhibit a wide tissue tropism particularly in the context of cells
41 derived from post-mitotic tissues such as the liver and eye 39. However hematopoietic cells
42
43 and leukemic cells in particular, are often so refractory to transduction that viral vectors that
44
45 can specifically infect this cell type are scarce. Many of the human megakaryocytic leukemia
46 47
45 cell lines such as MB-02 and MO7e 19¸ UT-7/Epo 40 and Dami 22 are non-permissive to AAV
48 infection. In addition, the transduction efficiency of naturally occurring AAV serotypes in
49
50 hematopoietic cells is extremely low, ranging up to 7 % for AAV1 and AAV2 serotypes 41.
51 52
50 In case of leukemic cells, data on the efficacy of alternate AAV serotypes is only available in
53 the context of erythroleukemia cells (K562) 41. Therefore, in our studies, we evaluated the
54
55 ability of multiple AAV serotypes for their gene transfer efficiency in an in vitro model of
56
57 CD33+ myeloid leukemia cells (U937). Cells were either mock-infected or infected with self-
58 59
57 complementary (sc) AAV 1-rh10 vectors that contained a reporter gene (GFP) driven by a

3 chicken β-actin (CBA) promoter. Our results obtained by flow cytometric analysis (Figure
4
5 1), demonstrate that AAV6 serotype had a significantly higher GFP expression in U937 cells

27 Figure 1. Transduction efficiency of AAV1 through AAV rh10 serotypes vectors in CD33+
28 human lymphoblastic myeloid leukemia cells. (A) About 1.5 x 104 U937 cells were seeded in
29 30
27 a 48 well plate. Twelve hours later, cells were either mock-infected or infected with 5x103 vgs.
31 per cell of self-complementary AAV vectors (serotypes 1-rh10) encoding a EGFP transgene.
32 Forty-eight hours later, EGFP-positive cells were enumerated by flow cytometric analysis. The
33 34
31 data presented is representative of a single biological replicate. (B) Quantitative analyses of
35 the data (mean+ SD) from two independent experiments (N=6 replicates). **p<0.01 and
36 ***p<0.001.
40 (64.6 + 15.2 %) followed by AAV5 vectors (11.9 + 4.7 %). Interestingly, AAV1 serotype
41 42
40 differs from AAV6 by only six amino acids 42-44, but failed to transduce human leukemic cells
43 in vitro. This is possibly due to differences in their ability to bind to unique primary cell
44
45 receptors ( a2–3/ a 2–6 N linked sialic acid for AAV1 vs. a2–3/ a 2–6 N linked sialic acid
46
47 and heparin sulfate proteoglycan for AAV6) 44, 45. Our data underscore the utility of AAV6
48 49
47 serotype vectors to target myeloid leukemic cells. It further highlights that the rate-limiting
50 step for efficient AAV transduction into leukemic cells is their recognition of specific cell
surface receptors.

55 Development and validation of CD33 receptor targeted AAV6 vectors.
57 While AAV6 serotype is efficient in transducing AML cells, the critical step for its success in
58
59 suicide gene therapy will be to further refine its properties to target the CD33 antigen over-
3 expressed in a majority of myeloid leukemic cells 31, 35. In addition, CD33 is a sialic acid
4
5 binding receptor 46 and AAV6 serotype is known to utilize a sialic acid receptors for entry
6
7 into their target cells 45. Thus, further engineering of an AAV6 capsid with a peptide that can
8 9
7 specifically and with high affinity, bind to cell surface CD33 is likely to enhance its
10 transduction into CD33 positive leukemic cells.
11
12 In order to achieve this goal, two parameters were crucial. This includes the site of
13 14
12 vector capsid engineering and the designing of CD33 binding peptide. Several studies on
15 AAV2, have identified surface-exposed regions of the capsid proteins that are largely tolerant
16
17 to insertion of peptides 23, 24, 27. For example in case of AAV2, Girod et al. introduced a 14-
18
19 amino-acid peptide (L14) specific to integrin receptor, adjacent to R588 amino acid and re-
20 21
19 targeted AAV2 to cells that are normally resistant to AAV2 infection 25. In case of AAV6, a
22 homologous site at R585 site has been used to insert an L15 oligomer containing the RGD
23
24 motif, for targeting of integrin-overexpressing malignant cells in different types of human
25
26 carcinoma 47. Based on these data, we selected R585 site in AAV6 capsid to display the CD33
27 28
26 targeting peptide.
29 For designing a targeting peptide, we have used bioinformatic tools to identify
30
31 complementarity determining regions (CDRs) from a monoclonal antibody (M195) available
32 33
31 against CD33 48. A consensus heptamer sequence (AASNQGS) was selected after analysis
34 and identification of the CDRs of antibody M195 by abYsis (http://www.abysis.org/), a web-
35
36 based antibody research system 49. abYsis integrates information on antibody sequence and
37
38 structure and automatically numbers with schemes such as Kabat and Chothia 50, 51. We further
39 40
38 used Paratome software (http://38 www.ofranlab.org/paratome) that identifies antigen-binding
41 regions (ABRs) of an antibody 52. This software was used for verifying the previously selected
42
43 peptide insert (AASNQGS) referred to as “CD33 targeting peptide” and to confirm that the
44
45 peptide sequence comprising of CDRs predicted by abYsis, is indeed having the residues that
46 47
45 mediate antigen binding. For an antigen-antibody interaction, there should be a high affinity
48 and specificity between the antibody and their cognate antigen that facilitates either the
49
50 destruction of antigen or block its activity. Thus peptides selected using this platform would
51 52
50 more correctly identify/predict the residues that mediate these interactions, since the algorithm
53 used by the software has been trained on set of all known antibody–antigen complexes 53. This
54
55 has been shown earlier by Payandeh et al., where the authors used Paratome, for improving
56
57 the affinity of Ofatumumab, an anti-CD20 mAb used in the treatment of chronic lymphocytic
58 59
57 leukemia 54.
3 The consensus heptameric sequence selected above was flanked with 2 amino acids
4
5 (GAASNQGSA) as linkers 55 and inserted into the loop IV region of AAV6 capsid at R585
6
7 site by standard cloning techniques. A schematic representation of the strategy followed for
8 9
7 insertion of the 9-mer CD33 specific peptide is depicted in Figure 2A. The cloned construct,
10 AAV6-CD33 rep/cap was further verified by DNA sequencing (Figure 2B).

Figure 2. Designing of CD33 targeting AAV6 vectors. (A) CD33 targeting peptides were
59 shortlisted from complementarity determining regions (CDRs) of a M195 monoclonal
60 antibody against CD33. After validation with multiple software’s (CHOTHIA, ABM, KABAT,

3 CONTACT, abYsis, Paratome) as described in the methods section, a heptameric consensus
4 5
3 flanked with 2 amino acids (GAASNQGSA) as linkers was selected for cloning into AAV6
6 capsid at residue 585. The wild type AAV6 capsid region open reading frame (ORF) is shown.
7 Position at which insertion is tolerated is marked by red line and magnified to show the
8 9
6 sequence. Modified AAV6 vector with CD33 targeting peptide inserted at 585 position is
10 marked by dashed frame. (B) Sequencing verification of the modified AAV6 vector is shown
11 in the electropherogram, harbouring the CD33 insert depicted by red dashed box, introduced
12 13
10 adjacent to the Q585 amino acid (VP1 numbering) in the capsid of AAV6.
14
15 The AAV6-CD33 vectors thus developed, was packaged with a suicide gene (iCasp9) as
16
17 described in the methods section. The packaging titre of the novel AAV6-CD33 vectors were
18 19
17 lower (3.35+1.06x1010 vs.4.13+1.27x1011 vgs/ml) in comparison to AAV6-wild type (WT)
20 vectors. To further investigate if the AAV6-CD33 capsids are intact and are able to
21
22 encapsidate the genome, we analysed the vectors using transmission electron microscopy
23
24 (TEM). As can be seen in Figure 3, capsids which are intact and contain an encapsulated
25 26
24 genome appear as ~23nm white spheres (Figure 3, panel A and B; marked by white arrow).
ds section. The empty AAV particles are denoted by bold arrow while the full particle
4 5
3 containing the genome are denoted by dashed arrow (C) Immunoblotting of AAV6 vectors.
6 About 1x 109 vgs. of AAV6 vectors were resolved in a 10% denaturing SDS-PAGE gel and the
7 proteins transferred onto a PVDF membrane (Pall Corporation, New York, USA). After
8 9
6 adequate blocking, membranes were probed with anti-AAV antibody (B1, Fitzgerald, North
10 Acton, MA, USA) primary antibody and detected with an anti-mouse HRP conjugated
11 secondary antibody (Abcam, Milton, Cambridge, UK). The bands were developed by a
12 13
10 chemiluminescent detection method (SuperSignal™ West Pico PLUS, Thermo Scientific). A
14 different lot of AAV6-EGFP vectors were used for immunoblotting since a high dose was
15 required for probing.

18 In contrast, capsids which are empty are permeable to uranyl acetate staining and therefore
19
20 appear as dark rings (Figure 3, panel A and B; marked by black arrow). In case of AAV6-
21
22 CD33 vectors (Figure 3, panel B) the density of empty particles in a given field was higher
23 24
22 in comparison to AAV6-WT vectors which corroborates with their packaging titres. To
25 further understand of the AAV6-CD33 vectors had a similar VP1:VP2:VP3 stoichiometry to
26
27 wild type AAV6 vectors, we performed an immunoblot analysis of capsid protein. The data
28 29
27 shown in Figure 3C, demonstrates that synthesis and assembly of capsids were unaltered due
30 to the insertion of the 9-mer peptide in AAV6 capsid. Taken together, our analysis indicates
31
32 that insertion of CD33 targeting peptide did not alter the conformation of the viral capsid but
33
34 had a modest effect on packaging ability of AAV6 chimeric vectors and are in agreement with
35 36
34 previous studies where a 5-10 fold reduction packaging titres were observed for AAV6
37 incorporated with a RGD motif 47. While alternative approaches such as attachment of
38
39 chemical conjugates or antibodies to AAV 22, 56 can be explored to target the CD33 antigen,
40
41 the molecular engineering strategy proposed here can overcome certain disadvantages such
42 as the relative instability of the adaptor-vector complex or the increased particle/vector size
43
44 and their potential immunogenicity during systemic administration in vivo 23.

47 Receptor targeted AAV6 vectors demonstrate enhanced transduction in AML cells.
49 In the next set of studies, we wished to determine if the AAV6-CD33 vectors can enhance
50
51 suicide gene transfer in AML cells. An inducible caspase 9 suicide gene (iCasp9) was
52
53 packaged into both AAV6-WT and AAV6-CD33 capsids and the vectors were further
54 55
53 quantified [as vector genomes (vg)/ml] 57. We then evaluated their cytotoxic effect, in a panel
56 of cell lines which were of hematopoietic origin, that either over-express the CD33 antigen
57
58 (U937) or that do not express CD33 antigen (CEM). In addition, we used a non-hematopoietic,
59 60
58 hepatic cell line Huh7, as control cells. These cells were transduced with AAV6 vectors at a

3 multiplicity of infection (MOI) of 5x104 vgs/cell. Twenty four hours later, cells were treated
4
5 with a iCasp9 specific pro-drug AP20187 at a concentration of 10nM as described earlier 58.
6
7 Forty-eight hours later, cytotoxicity was measured in vector treated and control cells by a
8 9
7 luminescence based ATP assay (CellTiter-Gloâ, Promega, Madison, WI, USA). In U937
10 cells, the cell viability assay revealed that mock-infected cells were resistant to prodrug
11
12 treatment (AP20187), whereas cells that had been infected with AAV6-WT iCasp9 vectors
13
14 demonstrated a strong cytotoxic effect upon AP20187 treatment in comparison to mock-
15 16
14 treated cells (67 vs. 100 %, p < 0.001) (Figure 4A). Interestingly, the targeted AAV6 vectors

3 Figure 4. Efficiency of CD33 targeted AAV6 vectors for iCasp9 suicide gene delivery in vitro.
4 5
3 To assess the cytotoxicity of AAV6-iCasp9 vectors, U937 (A), CEM (B) and Huh7 (C) cells
6 were either mock-infected or infected with AAV6-WT or CD33 targeted vectors at a
7 multiplicity of infection (MOI) of 5x104 vgs/cell. A day later, cells were treated with 10nm of
8 9
6 the dimerizer pro-drug AP20187 specifically required for inducible caspase 9 activity. To
10 document the pro-apoptotic activity of vector treated cells in comparison to control cells, we
11 measured the cellular ATP activity forty eight hours later, by a luminescence based assay
12 13
10 (CellTiter-Glo®, Promega, Madison, WI, USA). Cells treated with Triton X 100, were used
14 as a positive control. Values are % cell viability in comparison to mock-treated cells. Data
15 represents mean values ± SD from 2 independent experiments (n=6 replicates). ***P<0.001
16 17
14 vs. mock treated cells.
18
19
20 were highly cytotoxic (1.6 fold, 41 vs. 67 % viability, p< 0.001) when compared to AAV6-
21
22 WT vectors in the CD33 positive U937 cells. However, in cells that do not express CD33
23
24 antigen, such as CEM or Huh7, AAV6-CD33 vectors vector showed no advantage over wild-
25 26
24 type AAV6 vectors in terms of their cytotoxicity (Figure 4B and 4C). The mean survival of
27 AAV6-CD33 vector treated non-hematopoietic cells (Huh7) or hematopoietic (CEM) cells
28
29 ranged between 27 to 33 % respectively, which was similar to the survival of cells treated
30
31 with AAV6-WT vectors (16 to 27 %, respectively). It must be noted that the varied levels of
32 cytotoxicity in between the three cell lines tested here, could be due to differences in their
33
34 permissivity to AAV6 vectors as noted in several other studies 39, 41. For example in a study
35
36 by Ellis and group, where they compared the transduction efficiency of nine naturally
37 38
36 occurring AAV serotypes (AAV 1-9) in multiple human derived cell lines, identified that
39 AAV6 serotype transduced Jurkat cells (T-cell leukemia cell line) modestly (20%) whereas
40
41 other cell lines such as HEK 293 (human embryonic kidney cell line) and HeLa (human
42
43 cervical carcinoma cell line) were transduced at a greater efficiency (> 80%) 41. Nonetheless,
44 45
43 our data underscores the specific and potent cytotoxic activity of CD33 receptor targeted
46 AAV6 vectors in AML cells in vitro.
50 Suicide gene delivery with CD33 targeted AAV6 vectors enhances cytotoxicity in an AML
51
52 model of Zebrafish in vivo.
53
55
54 To further validate the novel AAV6 vectors developed, we utilized a Zebrafish xenograft model
56 reported recently 38. Briefly, ~1×105 fluorescently labelled U937 cells per recipient were
57
58 transplanted into busulfan (20mg/kg bodyweight) conditioned wild-type Zebrafish (Figure 5).
26 Figure 5. Schematic for in vivo testing of wild type and CD33 targeted AAV6-iCasp9 vectors.
27 28
26 Experimental outline for xenotransplantation of U937 cell in Zebrafish is presented. After
29 engraftment of the tumors in the recipient fish (4 days post transplantation), they were
30 randomized into four groups. Control group included fishes that developed tumors but were
31 32
29 mock-treated (n=13). In the treatment arm, fishes were administered with wild type AAV6-
33 iCasp9 vectors either by local injections (intra-tumorally, IT, n=5), or by a systemic retro-
34 orbital injection (RO, n=13). The receptor targeted AAV6-CD33 iCasp9 vectors were
35 36
33 administered by RO route (n=10). All the vectors were administered at a dose of 3×109 vgs/fish.
37 The fishes were subsequently administered with the pro-drug AP20187 and followed up for ten
38 days after vector administration.
42 Four days after engraftment of leukemic cells, fishes were randomised into four groups: mock-
43 44
42 treated (n=13), AAV6-iCasp9 vector administered either systemically (retro-orbital; RO, n=13)
45 or locally (intra-tumoral; IT, n=5) and the receptor targeted AAV vector (AAV6-CD33, n=10)
46
47 administered systemically through the retro-orbital plexus. AAV vectors were administered at
48
49 a dose of 3×109 vgs/fish. Ten days later, the rate of tumor growth in vector treated fish was
50 51
49 compared to mock-treated Zebrafish. As shown in Figure 6A, the fluorescence of harvested
52 tumours from the AAV6 vector administered groups reveal a qualitative reduction in their
53
54 intensity in comparison to tumors from the control group. To further understand and compare
55 56
54 the therapeutic benefit of suicide gene delivery in the model, we evaluated additional
57 parameters such as the survival of fish after suicide gene transfer and histological analysis of
58
59 tumor tissues. Kaplan-Meir analysis (Figure 6B) performed ten days after vector

3 administration, revealed that the survival was significantly higher in the treatment group that
4
5 received AAV6-iCasp9 vectors as compared to mock-treated control fishes (~80% vs. 15%,
6
7 p<0.001). Moreover, Zebrafish in the AAV6-CD33 treatment arm showed better survival as
8 9
7 compared to untreated fishes (~100% vs. 15%, p<0.001) or to those treated with wild type
10 AAV6 vectors administered either by RO or IT route (100% vs. 77% or 80%). We then
11
12 performed a morphological characterisation, on the tissue sections obtained from tumor
13 14
12 samples by hematoxylin and eosin staining. As can be seen in Figure 6C, the tumor sections
15 from the mock-treated Zebrafish, revealed a large number of mitotically active cells. However,
16
17 in case of tumor sections from vector treated fish, a higher number of apoptotic cells with a
18
19 characteristic pyknosis were observed. To further quantify the efficacy of AAV6-iCasp9 based
20 21
19 suicide gene therapy, we performed a terminal deoxynucleotidyl transferase-mediated dUTP
22 nick-end labelling assay (Click-iT® TUNEL Alexa Fluor® 647 Imaging Assay, Thermo
23
24 Fisher, Waltham, Massachusetts, USA) (Figure 6D and E). This test has a high sensitivity in
25
26 detecting the DNA strand break in apoptotic cells and has been widely used as a biomarker to
27 28
26 evaluate the efficacy of various therapeutic interventions 59, 60. Our analysis revealed a low
29 baseline number (4.2 + 2.3) of TUNEL positive cells in tumors harvested from mock-treated
30
31 fish. Zebrafish treated with AAV6-iCasp9 vectors intra-tumorally or retro-orbitally, had a
32 33
31 relatively higher proportion (~12 fold vs. control group, p<0.05) of apoptotic cells. These data
34 are in agreement with the cytotoxic effect seen with AAV6-WT-iCasp9 vectors on the
35
36 engrafted tumors and the resultant tumor regression seen in these groups (Figure 6A).
37
38 Interestingly, tumors that received CD33 targeted AAV6-iCasp9 vectors by systemic
39 40
38 administration, had a significantly higher (~39 fold) increase in TUNEL positive cells in
41 comparison to the control group. Furthermore, the receptor targeted AAV6 vectors augmented
42
43 the in vivo cytotoxicity of U937 tumors by at least 3-fold (p<0.001) when administered retro-
44
45 orbitally in comparison to AAV6-WT vector administered group. Our findings are significant
46 47
45 as previous studies targeting AML have only reported the in vitro efficacy. An AAV2 vector
48 containing a heptamer [NQVGSWS] identified from a random peptide library and harbouring
49
50 a suicide gene (HSV-TK) conferred selective killing to Kasumi-1 cells in vitro but it’s in vivo
51 52
50 efficacy and specificity is not known 20. Taken together, our data suggests that the insertion of
53 CD33 targeting peptide augments the infectivity and selectivity of AAV6 vectors to AML cells
in vitro and in vivo.
26 Figure 6. Therapeutic efficiency of suicide gene delivery with wild type and CD33 targeted
27 AAV6 vectors in a Zebrafish model of AML. (A) Representative images of tumors harvested
28 29
26 from each group 10 days after vector administration are shown. The exposure settings in a
30 Leica microscope were uniform (Exposure: 400ms; Gain: 2.5; Original magnification – x16)
31 (B) Kaplein Meier survival analysis of Zebrafish treated with AAV6-iCasp9 vectors. Data are
32 mean ± SD. ***p < 0.001 by Log Rank Mantel-Cox test. (C) Hematoxylin and eosin staining
33
34 of tumor tissue harvested from the experimental fishes 10 days after vector administration.
35 Tumors were fixed in formalin, sectioned and were stained with hematoxylin and eosin.
36 Microscopic analysis revealed marked reduction in mitotically active cells in the treatment
37
38 groups and accumulation of pyknotic nuclei (200X magnification). (D) TUNEL staining of
39 cryosections of tumors harvested from experimental Zebrafish. The images demonstrate the
40 apoptotic TUNEL positive cells in green while the nucleus is stained with DAPI and appears
41
42 blue. The imaging settings were constant for all groups (Exposure: 400ms Gain: 5; Original
43 magnification – x200) (E) Quantification data from at least 9 representative sections by Image
44 J analysis software (NIH Image, Bethesda, MD). Data demonstrates a significant higher
45 46
42 number of TUNEL positive cells between the AAV6-CD33 targeted and AAV6-WT treatment
47 groups. #p value < 0.001, **p<0.01 and ***p<0.001; *p value compared to untreated tumor
48 control fish (mock group); #p value compared to AAV6-iCasp9 treated group.

52 § CONCLUSIONS
53
54 This study demonstrates for the first time, the therapeutic benefit of AAV6-iCasp9 vectors in
55
56 general, and CD33-targeted AAV6 vectors in particular, for cytotoxic gene therapy in AML.
57
58 The knowledge gained from these studies is likely to facilitate the use of AAV6-CD33 hybrid
59 system as an effective adjunct for AML therapy, thereby enabling the administration of reduced
dose of conventional chemotherapeutic agents that are currently associated with significant
4
5 toxicity in patients8. Our study also has certain limitations. While we have demonstrated the
6
7 efficiency of receptor targeted AAV6 vectors in a xenotransplantation model of Zebrafish, their
8 9
7 further validation in murine model of AML61 or primary human leukemic cells may be
10 necessary before scaling up this strategy for suicide gene therapy in humans. Furthermore,
11
12 since suicide gene therapy for conditions such as AML requires a systemic vector delivery, the
13 14
12 effect of pre-existing neutralizing antibodies to these modified AAV vectors needs to be studied
15 in detail.

20 § EXPERIMENTAL PROCEDURES
21
22 Analysis of CD33 expression by flow cytometry.
23 24
22 Human lymphoblastic myeloid leukemia cells (U937) cell line was a kind gift from Dr.
25 Vikram Mathews, Christian Medical College, Vellore, India. Human hepatocellular
26
27 carcinoma (Huh7) cell line was a kind gift from Dr. Saumitra Das, Indian Institute of Science,
28
29 Bangalore. AAV-293 packaging cells were purchased from Stratagene (San Diego, CA,
30 31
29 USA). Human erythroleukemia cells (K562) and the human T lymphoblastic leukemia cell
32 line (CEM) was purchased from American Type culture collection (ATCC, Manassas, USA).
33
34 Cells were maintained in Iscove's Modified Dulbecco's growth medium (IMDM, Gibco, Life
35
36 Technologies, Carlsbad, USA) supplemented with 10% fetal bovine serum (Gibco), 10µg/ml
37 2
36 each of piperacillin and ciprofloxacin at 37 °C and 5% CO .
38
39 To assess the CD33 surface marker expression, ~3x104 cells (U937, K562, CEM and
40
41 Huh7) were first incubated for 30 mins, with an optimal concentration (1:400) of
42 43
41 unconjugated anti-human CD33 antibody (Abcam, ab199432, Cambridge, UK).
44 Subsequently, these cells were probed with a flourochrome conjugated secondary antibody
45
46 for 30 mins in dark (Goat anti-rabbit Alexa Fluor 488, Molecular probes, Thermo Fisher).
47
48 Subsequently, the cells were assessed by flow cytometry. Mean of percentage CD33
49 50
48 positivity (FITC positive cells) from two independent experiments and 3 replicates each, were
51 used to obtain the CD33 levels in a BD Accuri C6 Plus flow cytometer (BD Biosciences,
52
53 Franklin Lakes, NJ, USA).

56 Design and assembly of CD33 targeted AAV6 vectors.
57
58 The CD33 targeting peptide was rationally designed using the antibody sequence of M195, a
59
60 mouse monoclonal antibody against CD33 [Accession number: AAA38370.1 (Heavy chain);

1
2
3 AAA39015.1 (Light chain)] were obtained from NCBI- GenBank database
4
5 (https://www.ncbi.nlm.nih.gov/). The sequence was then analysed in abYsis, a software to
6
7 identify complementarity determining regions (CDRs), given an antibody sequence. abYsis
8 9
7 is an antibody analysis database that integrates sequence data from the European Molecular
10 Biology Laboratory European Nucleotide Archive (EMBL-ENA) and Kabat as well as
11
12 structure data from the Protein Data Bank (PDB). Three CDRs from both heavy and light
13 14
12 chain were identified. Heavy chain CDRs: CDR-H1: GYTFTDYNMH; CDR-H2:
15 YIYPYNGGTG; CDR-H3: GRPAMDY. Light chain CDRs: CDR-L1:
16
17 RASESVDNYGISFMN; CDR-L2: AASNQGS; CDR-L3: QQSKEVPWT. It is well
18
19 understood that the residues that actually bind the antigen fall outside the traditionally defined
20 21
19 CDRs. Thus concurrently, the sequence was also analysed in Paratome, a software to identify
22 antigen-binding regions (ABRs), given an antibody sequence. The Paratome web server
23
24 predicted two ABRs from both heavy as well as light chain - heavy chain ABRs: ABR2:
25
26 WIGYIYPYNGGTGY; ABR3: RGRPAMDY and light chain ABRs: ABR2:
27 28
26 LLIYAASNQGS; ABR3: QQSKEVPW. Finally a consensus heptamer sequence
29 (AASNQGS) was selected, which encompassed the ABR2 of light chain based on the fact
30
31 that L2 has been identified as the unique region for antigen binding by previous studies 53. We
32 33
31 further flanked this by a terminal amino acid at both the 5’ and 3’ end, as linkers. These GA
34 linkers were introduced to increase the length of the peptide and potentially enable sufficient
35
36 conformational flexibility. The 9-mer ‘GAASNQGSA’ peptide was in vitro synthesized and
37
38 inserted in AAV6 capsid by directional cloning (Genscript Technologies, Piscataway, NJ,
39 40
38 USA). A detailed outline of the cloning strategy is provided in Supporting information
41 Figure S1.

45 Production of AAV vectors.
46 47
45 Recombinant AAV vectors were generated by a triple transfection method in AAV-293
48 producer cells as previously described 62. Briefly, producer cells were first expanded in twenty
49
50 15cm2 dishes followed by their transfection with AAV1-rh10 rep/cap plasmid (p.AAVR2/C1-
51 52
50 p.AAVR2/rh.10 or p.AAVR2/C6-CD33), an enhanced green fluorescent protein or inducible
53 caspase 9 transgene (p.AAV-CBa-EGFP; p.AAV-CBa-iCasp9) and helper plasmids
54
55 (p.helper) at equimolar ratio using polyethylenimine (PEI, Polysciences, Warrington, PA).
56
57 After 68 hours, cells were collected, benzonase treated (25 units/ml; Sigma-Aldrich, St. Louis,
58 59
57 MI, USA) and purified by iodixanol gradient ultracentrifugation (OptiPrep, Sigma-Aldrich)
60 followed by column chromatography (HiTrap SP column; GE Healthcare Life Sciences,

Chicago, IL, USA) and concentration with Amicon Ultra 10K centrifugal filters. (Millipore,
4
5 Burlington, MA, USA). The physical particle titres were estimated by a quantitative PCR
6
7 based method and expressed as vector genomes (vg)/ml 57.
8
9
10 Transduction efficiency of AAV1-rh10 vectors in leukemic cells.
11
12 To identify the best AAV vector to infect leukemic cells, U937 cells were mock (PBS)
13 14
12 infected or infected with 5x103vgs/cell of AAV serotypes 1 to rh10 vectors containing EGFP
15 as the transgene. Forty-eight hours post-transduction, GFP expression was measured by flow
16
17 cytometry (Accuri C6 Plus, BD). For flow cytometric analysis, cells were trypsinized (0.05%
18
19 Trypsin, Gibco) and rinsed twice with PBS. A total of 1x104 events were analyzed for each
20 21
19 sample. Mean of percentage GFP positivity from two independent experiments and a total of
22 six replicates were used for comparison between mock and AAV1-rh10 vector infected U937
Transmission electron microscopy.
29 To assess the integrity of capsids in the vectors generated, ~ 10 µl viral suspension of AAV6-
30
31 WT or AAV6-CD33 vectors in 1x PBS were adsorbed onto 300-μm mesh carbon-coated
32 33
31 copper transmission electron microscopy (TEM) grids (Ted Pella, California, USA) for 5
34 mins. Excess solution was drained by blotting followed by washing twice with 0.2-μm-filtered
35
36 distilled water. After washing, the grids were then stained with freshly prepared 2% uranyl
37
38 acetate for 30 sec. After drying, grids were imaged using FEI Technai G2 12 Twin TEM 120
39 40
38 kV transmission electron microscope (FEI company, Hillsboro, OR, USA). Ten to twenty
41 images of each grid were captured and assessed for the number of full /empty capsid particles
43 as described previously 63.
44
AAV6 mediated suicide gene delivery in vitro.
48 To evaluate and compare the cytotoxic potential of CD33 targeted and WT- AAV6 vectors,
49
50 we used three different cell lines: U937, CEM and Huh7. About ~5 x 103 cells per well in a
51 52
50 96-well plate was either mock (PBS) infected or infected with AAV6-WT or AAV6-CD33
53 vectors, encoding iCasp9 gene at an multiplicity of infection (MOI) of 5 x 104 vgs/cell. A day
54
55 later, cells were treated with 10nM concentration of a dimerizer drug AP20187 (ARIAD
56
57 Pharmaceuticals, Cambridge, MA, USA). The cell viability was assessed forty-eight hours
58 59
57 later, by a CellTiter-Glo luminescence based ATP assay (Promega Corporation, Madison,
60 USA). The percentage cell viability was estimated as per manufacturer’ instruction, where the

cell viability is a ratio of normalised luminescence of vector infected cells / normalised
4
5 luminescence of control cells x 100.
6 Leukemic cell (U937) xenotransplantation in adult Zebrafish.
10 A wild type strain of adult Zebrafish (Tübingen, Danio rerio a kind gift from Dr. Sonawane’s
11
12 Lab, TIFR, Mumbai) was used for transplantation procedure as described earlier 64. For
13 14
12 transplantation, the 6-mpf Zebrafish were administered with busulfan (Sigma Aldrich, St.
15 Louis, Missouri, USA) two days prior to cell transplantation at a dose of 20 mg/kg body weight
16
17 intraperitoneally 61, 65. On the day of transplantation, Zebrafish were anaesthetized in tricaine
18
19 solution (150 mg/L). We then administered the Zebrafish with 1×105 fluorescently labelled
20 21
19 U937 cell suspension near the dorsal aorta using a 5μl hamilton syringe (Hamilton, Nevada
22 USA). The fishes were then transferred to a recovery tank (1 hour at 28 °C) and then to a normal
23
24 tank at 34 °C for serial monitoring in the presence of water containing 1% penicillin and
25
26 streptomycin (Gibco) for an initial time point of 24 hours after cell administration.
27 28
26 Transplanted fishes were kept in isolated maintenance tanks, removed from the circulating
29 system to avoid risk of infection and fed with brine shrimp meal, twice a day.

33 Administration of AAV6-CD33 vectors in an AML model of Zebrafish.
34
35 Transplanted Zebrafish were subsequently monitored on a daily basis. After four days of
36
37 transplantation, Zebrafish were anaesthetized with tricaine and subsequently imaged with a
38 39
37 Leica M205FA fluorescent stereoscopic microscope equipped with a Leica DFC310FX camera
40 (Leica Microsystems, Wetzlar, Germany). The images were acquired in the same focal plane
41
42 in bright field and in transmitted light passing through the RFP filter. Subsequently, fishes were
43 44
42 randomised into four groups and were either mock-treated (n=13) or AAV6-WT iCasp9 vector
45 administered either by retro-orbital (n=13) or intra-tumoral (n=5) route or administered with
46
47 AAV6-CD33 vectors by retro-orbital route (n=10). Twenty four hours after vector
48
49 administration, Zebrafish that were injected with AAV6-iCasp9 vectors, received AP20187 at
50 51
49 a dose of 75µg/kg of body weight in 5mM citrate buffer (pH 5.0) as described previously 66.
52 Three doses of AP20187 were administered, 24 hrs apart. The fishes were followed-up for a
5 period of ten days.

3 Histopathological studies.
4
5 To ascertain the effect of suicide gene delivery on the transplanted cells, mock-treated and
6
7 AAV6 treated Zebrafish were anesthetized in tricaine solution and humanely euthanized 67.
8 9
7 The fishes were dissected and the tumor was harvested based on published protocols 68.
10 Subsequently, tumors were washed (1X PBS) and fixed in 10% neutral buffered formalin. After
11
12 overnight fixation, samples were washed rigorously in 1X PBS and were infiltrated with
13 14
12 sucrose for cryoprotection. The cryoprotected tissue was then mounted in OCT medium (Sigma
15 Aldrich). After embedding, the frozen tissue block was then sliced in ~ 8 µm thin sections on
a cryotome (Leica, Wetzlar, Germany). These tissue sections were then stained with
18
19 hematoxylin and eosin as described earlier 69, 70.
Apoptosis assay.
23
24 Tumor tissue harvested from euthanized Zebrafish were cryosectioned (8 µm sections) and
25
26 were probed by an in situ TUNEL assay (Thermo Fisher). Samples were further
27 28
26 counterstained with 4′, 6-diamidino-2-phenylindole (DAPI, Thermo Fisher) and the images
29 were captured (Leica DM5000B). At least three tissue sections/tumor/fish (n=3 fishes) were
30
31 analysed for each condition. The data was further analysed by ImageJ software 71.
Statistical Analysis.
35
36 Data are expressed as mean +/- standard deviation, unless otherwise specified. Multiple
37
38 comparisons between groups were performed by either unpaired student’s t-test or by one way
39 40
38 ANOVA as applicable using GraphPad Prism version 7.0.0 (GraphPad Software, San Diego,
41 California USA). Comparison between the test and control groups were plotted as significant
42
43 if p value< 0.05.

ASSOCIATED CONTENT

6 Supporting Information

9 The supporting information is available free of charge on the ACS Publications website at
10
11 DOI:
12
13 CD33 expression levels in various cell lines of hematopoietic and non-hematopoietic
14
15 origin (Table S1); Generation of AAV6 vectors displaying CD33 targeted peptides in capsid.
16
17 (Figures S1).
18
19 § AUTHOR INFORMATION
20
22
21 Corresponding Author
23
24 *E-mail: [email protected]. Phone: +91 512 2594086. Fax: +91 512 2594010.
25
26 ORCID

29 Nusrat Khan: 0000-0002-6676-7269
30
31 Sridhar Bammidi: 0000-0002-3189-6140
32
34
33 Giridhara R. Jayandharan: 0000-0001-9353-9053
35
36 Author Contributions
37
38 NK, SB performed the experiments; GRJ conceived and designed the experiments and
39
40 interpreted the data. NK and GRJ wrote the manuscript. The datasets generated during the
41
42 study is available from the corresponding author on reasonable request.
43
44 Notes
45
47
46 The authors declare that they have no competing financial interests.

ACKNOWLEDGMENTS
52
53
54 This research was supported through a Nanomission grant SR/NM/NS-1084/2016 and an in-
55 56
54 part Wellcome Trust DBT India Alliance Senior fellowship to GRJ. NK was supported by a
57 PhD fellowship from an MHRD grant to IIT-Kanpur. The authors thank Dr Pradip Sinha, IIT
58
59 Kanpur for Zebrafish facility and Dr. Jonaki Sen, IIT Kanpur for imaging facility.

2
3 § ABBREVIATIONS
4
5 ABR, Antigen-binding region; AAV, Adeno-associated virus; AML, Acute myeloid leukemia;
6
7 CID or AP20187, CDR, complementarity determining regions; Chemical inducer of dimerization;
8 9
7 GCV, Ganciclovir; HSV-TK, Herpes simplex virus thymidine kinase; iCasp9, inducible caspase
10 9; TUNEL, Terminal deoxynucleotidyl transferase-mediated dUTP nick-end labelling assay.

§ REFERENCES
4
5
6 1. Estey, E.; Döhner, H. (2006) Acute myeloid leukaemia. The Lancet, 368 (9550),
7 1894-1907.
8
9
10 2. Grimwade, D.; Ivey, A.; Huntly, B. J. (2016) Molecular landscape of acute myeloid
11 12
10 leukemia in younger adults and its clinical relevance. Blood, 127 (1), 29-41.
13 3. De Kouchkovsky, I.; Abdul-Hay, M. (2016) 'Acute myeloid leukemia: a
14
15 comprehensive review and 2016 update'. Blood Cancer J, 6 (7), e441.
16
17 4. Johnston, D. L.; Alonzo, T. A.; Gerbing, R. B.; Hirsch, B.; Heerema, N. A.;
18 19
17 Ravindranath, Y.; Woods, W. G.; Lange, B. J.; Gamis, A. S.; Raimondi, S. C. (2013)
20 Outcome of pediatric patients with acute myeloid leukemia (AML) and −5/5q− abnormalities
21
22 from five pediatric AML treatment protocols: a report from the children's oncology group.
24
23 Pediatr. Blood Cancer, 60 (12), 2073-2078.
25 5. El-Aneed, A. (2004) Current strategies in cancer gene therapy. Eur. J.Pharmacol.,
26
27 498 (1-3), 1-8.
28
29 6. Walter, R. B.; Othus, M.; Borthakur, G.; Ravandi, F.; Cortes, J. E.; Pierce, S. A.;
30 31
29 Appelbaum, F. R.; Kantarjian, H. A.; Estey, E. H. (2011) Prediction of early death after
32 induction therapy for newly diagnosed acute myeloid leukemia with pretreatment risk scores:
33
34 a novel paradigm for treatment assignment. J. Clin. Oncol., 29 (33), 4417-4424.
35
36 7. Othus, M.; Kantarjian, H.; Petersdorf, S.; Ravandi, F.; Godwin, J.; Cortes, J.;
37 Pierce, S.; Erba, H.; Faderl, S.; Appelbaum, F. R.; et al. (2014) Declining rates of
38
39 treatment-related mortality in patients with newly diagnosed AML given 'intense' induction
40
41 regimens: a report from SWOG and MD Anderson. Leukemia, 28 (2), 289-292.
42 43
41 8. Duarte, S.; Carle, G.; Faneca, H.; de Lima, M. C.; Pierrefite-Carle, V. (2012)
44 Suicide gene therapy in cancer: where do we stand now? Cancer Lett., 324 (2), 160-170.
45
46 9. Kuriyama, S.; Kikukawa, M.; Masui, K.; Okuda, H.; Nakatani, T.; Akahane, T.;
47
48 Mitoro, A.; Tominaga, K.; Tsujinoue, H.; Yoshiji, H.; et al. (1999) Cancer gene therapy
49 50
48 with HSV-tk/GCV system depends on T-cell-mediated immune responses and causes
51 apoptotic death of tumor cells in vivo. Int. J. Cancer, 83 (3), 374-380.
52
53 10. Luo, J.; Luo, Y.; Sun, J.; Zhou, Y.; Zhang, Y.; Yang, X. (2016) Adeno-associated
54 55
53 virus-mediated cancer gene therapy: current status. Cancer Lett., 356, 347-356.
56 11. Morgan, R. A. (2012) Live and let die: a new suicide gene AP20187 therapy moves to the
57
58 clinic. Mol. Ther., 20 (1), 11-13.

3 12. Khan, N.; Bammidi, S.; Chattopadhyay, S.; Jayandharan, G. R. (2019) Combination
4
5 suicide gene delivery with an adeno-associated virus vector encoding inducible caspase-9 and
6
7 a chemical inducer of dimerization is effective in a xenotransplantation model of
8 9
7 hepatocellular carcinoma. Bioconjug. Chem., 30 (6), 1754-1762.
10 13. Xiao, P. J.; Samulski, R. J. (2012) Cytoplasmic trafficking, endosomal escape, and
11
12 perinuclear accumulation of adeno-associated virus type 2 particles are facilitated by
13 14
12 microtubule network. J. Virol., 86 (19), 10462-10473.
15 14. Bartlett, J. S.; Wilcher, R.; Samulski, R. J. (2000) Infectious entry pathway of adeno-
16
17 associated virus and adeno-associated virus vectors. J. Virol., 74 (6), 2777-2785.
18
19 15. Ponnazhagan, S.; Curiel, D. T.; Shaw, D. R.; Alvarez, R. D.; Siegal, G. P. (2001)
20 21
19 Adeno-associated virus for cancer gene therapy. Cancer Res., 61 (17), 6313-6321.
22 16. Tahara, I.; Miyake, K.; Hanawa, H.; Kurai, T.; Hirai, Y.; Ishizaki, M.; Uchida, E.;
23
24 Tajiri, T.; Shimada, T. (2007) Systemic cancer gene therapy using adeno-associated virus
25
26 type 1 vector expressing MDA-7/IL24. Mol. Ther., 15 (10), 1805-1811.
27 28
26 17. Li, Z.-B.; Zeng, Z.-J.; Chen, Q.; Luo, S.-Q.; Hu, W.-X. (2006) Recombinant AAV-
29 mediated HSVtk gene transfer with direct intratumoral injections and Tet-On regulation for
30
31 implanted human breast cancer. BMC Cancer, 6, 66.
32 33
31 18. Wendtner, C. M.; Kofler, D. M.; Theiss, H. D.; Kurzeder, C.; Buhmann, R.;
34 Schweighofer, C.; Perabo, L.; Danhauser-Riedl, S.; Baumert, J.; Hiddemann, W.; et al.
35
36 (2002) Efficient gene transfer of CD40 ligand into primary B-CLL cells using recombinant
37
38 adeno-associated virus (rAAV) vectors. Blood, 100 (5), 1655-1661.
39 40
38 19. Ponnazhagan, S.; Wang, X. S.; Woody, M. J.; Luo, F.; Kang, L. Y.; Nallari, M. L.;
41 Munshi, N. C.; Zhou, S. Z.; Srivastava, A. (1996) Differential expression in human cells
42
43 from the p6 promoter of human parvovirus B19 following plasmid transfection and
44
45 recombinant adeno-associated virus 2 (AAV) infection: human megakaryocytic leukaemia
46 47
45 cells are non-permissive for AAV infection. J. Gen. Virol., 77 ( Pt 6), 1111-1122.
48 20. Michelfelder, S.; Lee, M. K.; deLima-Hahn, E.; Wilmes, T.; Kaul, F.; Muller, O.;
49
50 Kleinschmidt, J. A.; Trepel, M. (2007) Vectors selected from adeno-associated viral display
51 52
50 peptide libraries for leukemia cell-targeted cytotoxic gene therapy. Exp. Hematol., 35 (12),
53 1766-1776.
54
55 21. Shi, W.; Bartlett, J. S. (2003) RGD inclusion in vp3 provides adeno-associated virus
56
57 type 2 (aav2)-based vectors with a heparan sulfate-independent cell entry mechanism. Mol.
59
58 Ther., 7 (4), 515-525.
60
22. Bartlett, J. S.; Kleinschmidt, J.; Boucher, R. C.; Samulski, R. J. (1999) Targeted
4
5 adeno-associated virus vector transduction of nonpermissive cells mediated by a bispecific
6
7 f(ab’gamma)2 antibody. Nat. Biotechnol., 17 (2), 181-186.
8 9
7 23. Muller, O. J.; Kaul, F.; Weitzman, M. D.; Pasqualini, R.; Arap, W.; Kleinschmidt,
10 J. A.; Trepel, M. (2003) Random peptide libraries displayed on adeno-associated virus to
11
12 select for targeted gene therapy vectors. Nat. Biotechnol., 21 (9), 1040-1046.
13 14
12 24. Munch, R. C.; Janicki, H.; Volker, I.; Rasbach, A.; Hallek, M.; Buning, H.;
15 Buchholz, C. J. (2013) Displaying high-affinity ligands on adeno-associated viral vectors
16
17 enables tumor cell-specific and safe gene transfer. Mol. Ther., 21 (1), 109-118.
18
19 25. Girod, A.; Ried, M.; Wobus, C.; Lahm, H.; Leike, K.; Kleinschmidt, J.; Deleage,
20 21
19 G.; Hallek, M. (1999) Genetic capsid modifications allow efficient re-targeting of adeno-
22 associated virus type 2. Nat. Med., 5 (9), 1052-1056.
23
24 26. Perabo, L.; Buning, H.; Kofler, D. M.; Ried, M. U.; Girod, A.; Wendtner, C. M.;
25
26 Enssle, J.; Hallek, M. (2003) In vitro selection of viral vectors with modified tropism: the
27 28
26 adeno-associated virus display. Mol. Ther., 8 (1), 151-157.
29 27. Buchholz, C. J.; Friedel, T.; Buning, H. (2015) Surface-engineered viral vectors for
30
31 selective and cell type-specific gene delivery. Trends Biotechnol., 33 (12), 777-790.
32 33
31 28. Varadi, K.; Michelfelder, S.; Korff, T.; Hecker, M.; Trepel, M.; Katus, H. A.;
34 Kleinschmidt, J. A.; Muller, O. J. (2012) Novel random peptide libraries displayed on AAV
35
36 serotype 9 for selection of endothelial cell-directed gene transfer vectors. Gene Ther., 19 (8),
37
38 800-809.
40
39 29. Grifman, M.; Trepel, M.; Speece, P.; Gilbert, L. B.; Arap, W.; Pasqualini, R.;
41 Weitzman, M. D. (2001) Incorporation of tumor-targeting peptides into recombinant adeno-
42
43 associated virus capsids. Mol. Ther., 3 (6), 964-975.
44
45 30. Wu, P.; Xiao, W.; Conlon, T.; Hughes, J.; Agbandje-McKenna, M.; Ferkol, T.;
46 47
45 Flotte, T.; Muzyczka, N. (2000) Mutational analysis of the adeno-associated virus type 2
48 (AAV2) capsid gene and construction of AAV2 vectors with altered tropism, J. Virol., 74
49
50 (18), 8635-8647.
52
51 31. De Propris, M. S.; Raponi, S.; Diverio, D.; Milani, M. L.; Meloni, G.; Falini, B.;
53 Foa, R.; Guarini, A. (2011) High CD33 expression levels in acute myeloid leukemia cells
54
55 carrying the nucleophosmin (NPM1) mutation. Haematologica, 96 (10), 1548-1551.
56
57 32. Miller, B. G.; Stamatoyannopoulos, J. A. (2010) Integrative meta-analysis of
58 59
57 differential gene expression in acute myeloid leukemia. PLoS One, 5 (3), e9466.

3 33. Hamann, P. R.; Hinman, L. M.; Hollander, I.; Beyer, C. F.; Lindh, D.; Holcomb,
4
5 R.; Hallett, W.; Tsou, H.-R.; Upeslacis, J.; Shochat, D.; et al. (2002) Gemtuzumab
6
7 ozogamicin, a potent and selective anti-CD33antibody−calicheamicin conjugate for treatment
8 9
7 of acute myeloid leukemia. Bioconjug. Chem., 13 (1), 47-58.
10 34. Dutour, A.; Marin, V.; Pizzitola, I.; Valsesia-Wittmann, S.; Lee, D.; Yvon, E.;
11
12 Finney, H.; Lawson, A.; Brenner, M.; Biondi, A.; et al. (2012) In vitro and in vivo
13 14
12 antitumor effect of anti-CD33 chimeric receptor-expressing EBV-CTL against CD33 acute
15 myeloid leukemia. Adv. Hematol., 683065.
16
17 35. Ehninger, A.; Kramer, M.; Rollig, C.; Thiede, C.; Bornhauser, M.; von Bonin, M.;
18
19 Wermke, M.; Feldmann, A.; Bachmann, M.; Ehninger, G.; et al. (2014) Distribution and
20 21
19 levels of cell surface expression of CD33 and CD123 in acute myeloid leukemia. Blood
22 Cancer J., 4, e218.
23
24 36. Inoue, T.; Swain, A.; Nakanishi, Y.; Sugiyama, D. (2014) Multicolor analysis of cell
25
26 surface marker of human leukemia cell lines using flow cytometry. Anticancer Res., 34 (8),
28
27 4539-4550.
29 37. Perez-Oliva, A. B.; Martinez-Esparza, M.; Vicente-Fernandez, J. J.; Corral-San
30
31 Miguel, R.; Garcia-Penarrubia, P.; Hernandez-Caselles, T. (2011) Epitope mapping,
32 33
31 expression and post-translational modifications of two isoforms of CD33 (CD33M and
34 CD33m) on lymphoid and myeloid human cells. Glycobiology, 21 (6), 757-770.
35
36 38. Khan, N.; Mahajan, N. K.; Sinha, P.; Jayandharan, G. R. (2019) An efficient method
37
38 to generate xenograft tumor models of acute myeloid leukemia and hepatocellular carcinoma
39 40
38 in adult zebrafish. Blood Cells Mol. Dis., 75, 48-55.
41 39. Srivastava, A. (2016) In vivo tissue-tropism of adeno-associated viral vectors. Curr.
42
43 Opin. Virol., 21, 75-80.
44
45 40. Mizukami, H.; Young, N. S.; Brown, K. E. (1996) Adeno-associated virus type 2
46 47
45 binds to a 150-kilodalton cell membrane glycoprotein. Virology, 217 (1), 124-130.
48 41. Ellis, B. L.; Hirsch, M. L.; Barker, J. C.; Connelly, J. P.; Steininger, R. J., 3rd;
49
50 Porteus, M. H. (2013) A survey of ex vivo/in vitro transduction efficiency of mammalian
51 52
50 primary cells and cell lines with nine natural adeno-associated virus (AAV1-9) and one
53 engineered adeno-associated virus serotype. Virol. J., 10, 74.
54
55 42. Qiao, C.; Zhang, W.; Yuan, Z.; Shin, J.-H.; Li, J.; Jayandharan, G. R.; Zhong, L.;
56
57 Srivastava, A.; Xiao, X.; Duan, D. (2010) Adeno-associated virus serotype 6 capsid tyrosine-
58 59
57 to-phenylalanine mutations improve gene transfer to skeletal muscle. Hum. Gene Ther., 21
60 (10), 1343-1348.

3 43. Ussher, J. E.; Taylor, J. A. (2010) Optimized transduction of human monocyte-
4
5 derived dendritic cells by recombinant adeno-associated virus serotype 6. Hum. Gene Ther.,
6
7 21 (12), 1675-1686.
9
8 44. Huang, L.-Y.; Patel, A.; Ng, R.; Miller, E. B.; Halder, S.; Mckenna, R.; Asokan,
10 A.; Agbandje-Mckenna, M. (2016) Characterization of the adeno-associated virus 1 and 6
11
12 sialic acid binding site. J. Virol., 90 (11), 5219-5230.
13 14
12 45. Wu, Z.; Miller, E.; Agbandje-Mckenna, M.; Samulski, R. J. (2006) Alpha 2,3 and
15 alpha2,6 n-linked sialic acids facilitate efficient binding and transduction by adeno-associated
16
17 virus types 1 and 6. J. Virol., 80 (18), 9093-9103.
18
19 46. Crocker, P. R.; Varki, A. (2001) Siglecs, sialic acids and innate immunity. Trends
21
20 Immunol., 22 (6), 337-342.
22 47. Sayroo, R.; Nolasco, D.; Yin, Z.; Colon-Cortes, Y.; Pandya, M.; Ling, C.;
23
24 Aslanidi, G. (2016) Development of novel aav serotype 6 based vectors with selective
25
26 tropism for human cancer cells. Gene Ther., 23 (1), 18-25.
27 28
26 48. Scheinberg, D. A.; Tanimoto, M.; Mckenzie, S.; Strife, A.; Old, L. J.; Clarkson, B.
29 D. (1989) Monoclonal Antibody M195: a diagnostic marker for acute myelogenous leukemia.
30
31 Leukemia, 3 (6), 440-445.
33
32 49. Swindells, M. B.; Porter, C. T.; Couch, M.; Hurst, J.; Abhinandan, K. R.; Nielsen,
34 J. H.; Macindoe, G.; Hetherington, J.; Martin, A. C. R. (2017) Abysis: integrated antibody
35
36 sequence and structure—management, analysis, and prediction. J. Mol. Biol., 429 (3), 356-
37
38 364.
40
39 50. Wu, T. T.; Kabat, E. A. (1970) An analysis of the sequences of the variable regions of
41 bence jones proteins and myeloma light chains and their implications for antibody
42
43 complementarity. J. Exp. Med., 132 (2), 211-250.
44
45 51. Chothia, C.; Lesk, A. M. (1987) Canonical structures for the hypervariable regions of
46 47
45 immunoglobulins. J. Mol. Biol., 196 (4), 901-917.
48 52. Kunik, V.; Ashkenazi, S.; Ofran, Y. (2012) Paratome: an online tool for systematic
49
50 identification of antigen-binding regions in antibodies based on sequence or structure.
51 52
50 Nucleic Acids Res., 40 (Web Server Issue), W521-524.
53 53. Kunik, V.; Peters, B.; Ofran, Y. (2012) Structural consensus among antibodies
54
55 defines the antigen binding site. Plos Comput. Biol., 8 (2), E1002388.
56
57 54. Payandeh, Z.; Rajabibazl, M.; Mortazavi, Y.; Rahimpour, A.; Taromchi, A. H.
58 59
57 (2018) Ofatumumab monoclonal antibody affinity maturation through in silico modeling.
60 Iran. Biomed. J., 22 (3), 180-192.

1
2
3 55. Chen, X.; Zaro, J. L.; Shen, W.-C. (2013) Fusion protein linkers: property, design
4
5 and functionality. Adv. Drug Deliv. Rev., 65 (10), 1357-1369.
6
7 56. Ponnazhagan, S.; Mahendra, G.; Kumar, S.; Thompson, J. A.; Castillas, J. M.
8 9
7 (2002) Conjugate-based targeting of recombinant adeno-associated virus type 2 vectors by
10 using avidin-linked ligands. J. Virol., 76 (24), 12900-12907.
11
12 57. Aurnhammer, C.; Haase, M.; Muether, N.; Hausl, M.; Rauschhuber, C.; Huber, I.;
13 14
12 Nitschko, H.; Busch, U.; Sing, A.; Ehrhardt, A.; et al. (2012) Universal real-time PCR for
15 the detection and quantification of adeno-associated virus serotype 2-derived inverted
16
17 terminal repeat sequences. Hum. Gene Ther. Methods, 23 (1), 18-28.
18
19 58. Straathof, K. C.; Pulè, M. A.; Yotnda, P.; Dotti, G.; Vanin, E. F.; Brenner, M. K.;
20 21
19 Heslop, H. E.; Spencer, D. M.; Rooney, C. M. (2005) An inducible caspase 9 safety switch
22 for T-cell therapy. Blood, 105 (11), 4247-4254.
23
24 59. Singh, S. S.; Mehedint, D. C.; Ford, O. H., 3rd; Jeyaraj, D. A.; Pop, E. A.;
25
26 Maygarden, S. J.; Ivanova, A.; Chandrasekhar, R.; Wilding, G. E.; Mohler, J. L. (2009)
27 28
26 Comparison of ACINUS, caspase-3, and TUNEL as apoptotic markers in determination of
29 tumor growth rates of clinically localized prostate cancer using image analysis. Prostate, 69
30
31 (15), 1603-1610.
33
32 60. Duan, W. R.; Garner, D. S.; Williams, S. D.; Funckes-Shippy, C. L.; Spath, I. S.;
34 Blomme, E. A. (2003) Comparison of immunohistochemistry for activated caspase-3 and
35
36 cleaved cytokeratin 18 with the TUNEL method for quantification of apoptosis in histological
37
38 sections of PC-3 subcutaneous xenografts. J. Pathol., 199 (2), 221-228.
39 40
38 61. Saland, E.; Boutzen, H.; Castellano, R.; Pouyet, L.; Griessinger, E.; Larrue, C.; De
41 Toni, F.; Scotland, S.; David, M.; Danet-Desnoyers, G. et al. (2015) A robust and rapid
42
43 xenograft model to assess efficacy of chemotherapeutic agents for human acute myeloid
44
45 leukemia. Blood Cancer J., 5, E297.
46 47
45 62. Xiao, X.; Li, J.; Samulski, R. J. (1998) Production of high-titer recombinant adeno-
48 associated virus vectors in the absence of helper adenovirus. J. Virol., 72 (3), 2224-2232.
49
50 63. Horowitz, E. D.; Rahman, K. S.; Bower, B. D.; Dismuke, D. J.; Falvo, M. R.;
51 52
50 Griffith, J. D.; Harvey, S. C.; Asokan, A. (2013) Biophysical and ultrastructural
53 characterization of adeno-associated virus capsid uncoating and genome release. J. Virol., 87
54
55 (6), 2994-3002.
56
57 64. Guo, M.; Wei, H.; Hu, J.; Sun, S.; Long, J.; Wang, X. (2015) U0126 inhibits
58 59
57 pancreatic cancer progression via the KRAS signaling pathway in a zebrafish
60 xenotransplantation model. Oncol. Rep., 34 (2), 699-706.

1
2
3 65. Nobrega, R. H.; Greebe, C. D.; Van De Kant, H.; Bogerd, J.; De Franca, L. R.;
4
5 Schulz, R. W. (2010) Spermatogonial stem cell niche and spermatogonial stem cell
6
7 transplantation in zebrafish. Plos One, 5 (9), 1-16.
8 9
7 66. Schmitner, N.; Kohno, K.; Meyer, D. (2017) Ptf1a+, Ela3l− Cells are
10 developmentally maintained progenitors for exocrine regeneration following extreme loss of
11
12 acinar cells in zebrafish larvae. Dis. Model. Mech., 10 (3), 307-321.
13 14
12 67. Wilson, J. M.; Bunte, R. M.; Carty, A. J. (2009) Evaluation of rapid cooling and
15 tricaine methanesulfonate (MS222) as methods of euthanasia in zebrafish (Danio rerio). J.
16
17 Am. Assoc. Lab. Anim. Sci., 48 (6), 785-9.
18
19 68. Gupta, T.; Mullins, M. C. (2010) Dissection of organs from the adult zebrafish. J. Vis.
21
20 Exp., (37).
22 69. Nuckels, R. J.; Gross, J. M. (2007) Histological preparation of embryonic and adult
23
24 aebrafish eyes. CSH Protoc., Pdb Prot4846.
25
26 70. Copper, J. E.; Budgeon, L. R.; Foutz, C. A.; Van Rossum, D. B.; Vanselow, D. J.;
27 28
26 Hubley, M. J.; Clark, D. P.; Mandrell, D. T.; Cheng, K. C. (2018) Comparative analysis of
29 fixation and embedding techniques for optimized histological preparation of zebrafish. Comp.
30
31 Biochem. Physiol. C Toxicol. Pharmacol., 208, 38-46.
32 33
31 71. Choi, H. K.; Yessayan, D.; Choi, H. J.; Schellenberger, E.; Bogdanov, A.; Josephson,
34 L.; Weissleder, R.; Ntziachristos, V. (2005) Quantitative analysis of chemotherapeutic effects
35
36 in tumors using in vivo staining and correlative histology. Cell. Oncol., 27 (3), 183-190.