Tumor Protection Following Vaccination With Low ... - CyberLeninka

© The American Society of Gene Therapy

original article

Tumor Protection Following Vaccination With Low Doses of Lentivirally Transduced DCs Expressing the Self-antigen erbB2 Miriam E Mossoba1, Jagdeep S Walia2, Vanessa I Rasaiah2, Nicole Buxhoeveden3, Renee Head2, Chuyan Ying4, Jason E Foley3, Jonathan L Bramson4, Daniel H Fowler3 and Jeffrey A Medin1,2,5 Department of Medical Biophysics, University of Toronto, Toronto, Ontario, Canada; 2Division of Stem Cell and Developmental Biology, Ontario Cancer Institute, University Health Network, Toronto, Ontario, Canada; 3Center for Cancer Research, Experimental Transplantation and Immunology Branch, National Cancer Institute, National Institutes of Health, Rockville, Maryland, USA; 4Centre for Gene Therapeutics, Department of Pathology and Molecular Medicine, McMaster University, Hamilton, Ontario, Canada; 5Institute of Medical Sciences, University of Toronto, Toronto, Ontario, Canada 1

Gene therapy strategies may accelerate the development of prophylactic immunotherapy against cancer. We synthesized a lentiviral (LV) vector encoding a kinase-deficient form of erbB2 (erbB2tr) to transduce murine dendritic cells (DCs) efficiently. Murine erbB2 models a clinically relevant tumor-associated self-antigen; its human homolog (HER-2/neu) is overexpressed in breast cancer and in 80% of metastatic prostate cancers. Following one infection, ~47% of DCs overexpressed erbB2tr. To determine whether low doses of transduced DCs could protect mice from prostate cancer cells, we performed prime/boost vaccinations with 2 × 103 or 2 × 105 erbB2tr-transduced DCs. Six weeks after vaccination, mice were simultaneously bilaterally challenged with the aggressive RM-1 prostate cancer cell line and an erbB2tr-expressing variant (RM-1-erbB2tr). Whereas control mice developed both tumors, all recipients of 2 × 105 erbB2tr-transduced DCs developed only wild-type RM-1 tumors. One-third of mice vaccinated with just 2 × 103 erbB2tr-transduced DCs also demonstrated erbB2tr-specific tumor protection. Protection against RM-1-erbB2tr tumors was associated with sustained levels of anti-erbB2tr antibody production and also correlated with erbB2tr-specific Th1 cytokine secretion. Depletion of CD4+, CD8+, or natural killer (NK) cells prior to tumor challenge underscored their role in mediating tumor protection. We conclude that administration of DCs expressing a self-antigen through efficient LV-based gene transfer activates cellular and humoral immunity, protecting host animals against specific tumor challenge. Received 8 May 2007; accepted 26 November 2007; published online 8 January 2008. doi:10.1038/sj.mt.6300390

Introduction Cancer immunotherapy aims to overcome the inability of the immune system to protect against the establishment of tumors or

reject established tumors efficiently. Dendritic cells (DCs) are potent antigen-presenting cells (APCs) that have been widely used to initiate or enhance tumor-associated antigen (TAA)-specific immune responses in animal models and clinical settings. Numerous reports show that modifying DCs via TAA peptide- or tumor lysatepulsing can induce antitumor immunity.1–5 Transfecting DCs with nucleic acid sequences encoding TAAs carries the advantage of inducing immunity toward a larger repertoire of naturally derived major histocompatibility complex class I and II compatible peptides. Comparative studies have shown that by transfecting DCs with RNA, stronger antitumor effects can be achieved than by pulsing DCs with peptides.6–9 Viral transduction of DCs offers similar advantages to RNA transfection, with the added potential benefit of more efficient transgene delivery and stable transgene expression, depending on the choice of virus. Retroviruses, including onco-retroviruses and lentiviruses (LVs), can also be used to transduce DCs with one or more genes. LVs are well suited to infecting DCs because they are capable of efficiently transducing slowly dividing cells. Their integration into the host genome provides a way to generate longterm stable transgene expression. In a few recent reports, LVs have been used to transduce murine and human DCs with TAAs.10–14 Overall, murine studies that have used virally transduced DCs for immunotherapy have demonstrated encouraging antitumor efficacy.15 However, the majority of murine studies published to date use human or rat homologs of TAAs instead of species-matched TAAs. In this study, we chose to use a naturally occurring splice variant of the murine erbB2 antigen (erbB2tr) as our target TAA in a mouse model of prostate cancer. In this setting, murine erbB2 represents a true self-antigen, which better mirrors the clinical setting. The human homolog of erbB2 (HER-2/neu) is overexpressed in 20% of primary prostate tumors and 80% of patients with metastatic prostate cancer, making this TAA a clinically relevant target for immunotherapy.16 HER-2/neu is also overexpressed in other malignancies including breast, ovarian, and lung tumors.17–19

Correspondence: Jeffrey A. Medin, University Health Network, 67 College Street, Room 406, Toronto, Ontario M5G 2M1, Canada. E-mail: [email protected] Molecular Therapy vol. 16 no. 3, 607–617 march 2008

607


Tumor Protection Using LV-transduced DCs

Results DCs are efficiently transduced with LVs A LV transfer vector encoding erbB2tr, a truncated (kinasedeficient) form of the murine self-antigen erbB2 (LV/erb) was constructed (Figure 1); an enhanced green fluorescent protein (enGFP) LV was previously described.21 Titers of produced LVs usually approximated between 5 × 106 and 3.6 × 108 functional infectious viral particles per ml. To determine the transduction efficiency of LV/erb, we infected bone marrow–derived murine pHR′EF-erbB2tr-W-SIN and pHR�EF-enGFP-W-SIN ∆ Gag

RRE

EF1-α

erbB2tr

WPRE

SIN LTR

EF1-α

enGFP

WPRE

SIN LTR

SA ∆ Gag

RRE

SD

SA

Figure 1 Schematic diagram of lentiviral vector constructs used in these studies. ψ, RNA packaging signal; EF1-α, elongation factor 1-α promoter; enGFP, enhanced green fluorescent protein; Gag, viral structural proteins; LTR, long-terminal repeat; RRE, Rev response element; SA, splice acceptor sites; SD, splice donor site; SIN LTR, self-inactivating LTR; WPRE, woodchuck hepatitis virus post-transcriptional regulatory element.

a 80

80 Day 7

70

30

50 40 30

20

20

10

10 100

101

102

103

0

104

Counts

Counts

Counts

32.6%

40

16.7%

60

101

102

103

47.4%

40 20

104

60

33.7%

40

101

102

103

0

104

100

101

102

103

104 enGFP

80 70 60 50 40 30 20 10 0

120 Day 9

100

101

22.3%

100

101

102

103

104

120

Day 5

100

102

103

0

104

2.7%

100

101

80 60

70.2%

40

102

103

104

60

20 0

101

102

103

104

79.7%

40

0

Day 9

100

80

20 100

120

Day 7

100 Counts

47.9%

40

erbB2tr

Counts

Counts

Counts

Day 7

60

20

20 100

Day 9

80

80

erbB2tr 80 70 60 50 40 30 20 10 0

100

Day 7

100

80

0 100

120 Day 5

100

60

50

0

b 120

Day 9

70

60

Counts

5′ LTR

ψ SD ψ

LV transduction does not alter DC phenotype or allostimulatory capacity To determine whether transducing DCs with our recombinant LVs at reasonable multiplicities of infection led to changes in phenotype, we first performed flow cytometry to compare the expression of typical surface molecules on mature transduced and control DCs. We assessed the percentage of cells expressing the myeloid marker CD11c, major histocompatibility complex II molecule I-Ab, along with co-stimulatory molecules CD80 and CD86. DCs used for the first scheduled vaccinations expressed similar levels of CD11c; 68.1% of the non-transduced (NT) DC cultures were CD11c+ compared to 75.7 and 70.2% for erbB2tr- and enGFPtransduced DC cultures, respectively (Figure 3a). Further comparisons revealed that the percentage of CD11c+ I-Ab+ DCs was nearly identical between NT and LV/erb-transduced DCs (39.3%

Counts

5′ LTR

DCs on day 3 of in vitro culture. In an initial pilot experiment, we determined that between 20 and 70% of a DC population was productively infected after one overnight incubation with LV/erb. Using these erbB2tr-transduced DCs, we performed an in vivo pilot study designed to test the efficacy of prime/boost vaccinations with LV-transduced DCs in mediating erbB2tr-specific antitumor immunity. After observing the potent antigen-specific effects of erbB2trexpressing DCs in vivo (data not shown), we initiated a second set of experiments to corroborate our pilot data. Freshly derived DCs were transduced and their expression levels of erbB2tr or enGFP were monitored over time. On culture day 7 for DCs used in the first immunization, we observed that 32.6% of transduced DCs were erbB2tr+ and 47.9% were enGFP+, respectively (Figure 2a). By day 9, when DCs were injected, erbB2tr+ and enGFP+ cells had decreased to 16.7 and 22.3%, respectively. For the second immunization, we also checked expression levels at day 5 and found that 47.4% of DCs were erbB2tr+ and 70.2% were enGFP+ (Figure 2b) at that time. The percentage of erbB2tr+ DCs decreased steadily to 33.7% on day 7 and 2.7% on day 9. The percentage of enGFP+ DCs was 79.7 and 73.7% on days 7 and 9, respectively.

Counts

A naturally occurring kinase-truncated variant of HER-2/neu has also been described in human tumor cells.20 Our aim was to generate immunity toward the self-antigen erbB2 in mice using DCs that were genetically engineered to express erbB2tr. We hypothesized that vaccinating mice with LVtransduced DCs could impart long-term erbB2-specific immunity and protection against subsequent challenge with erbB2-expressing tumors. In our model we used an aggressive RM-1 prostate tumor cell line that we have modified to express erbB2tr. We chose to focus on low-dose vaccination strategies, as limited availability of syngeneic DCs may be a factor in human vaccination protocols. This study provides a platform for the development of a low-dose DC immunotherapy strategy using LVs as gene transfer tools engineering expression of target TAAs.

80 60

73.7%

40 20

100

101

102

103

104

0

100

101

102

103

104

enGFP

Figure 2 Transduction efficiency of LV/erb and LV/enGFP on dendritic cells (DCs). Bone marrow cells were cultured in the presence of granulocyte–monocyte colony stimulating factor and interleukin-4. DC maturation was induced with tumor necrosis factor-α on culture day 8. Transductions were performed on culture day 3 using either erbB2tr- or enGFP-encoding lentiviruses. Transgene expression on DCs used for the (a) first and (b) second scheduled immunizations was monitored by flow cytometry. Numbers indicate the percentage of transgene-positive cells for each histogram. enGFP, enhanced green fluorescent protein.

608

www.moleculartherapy.org vol. 16 no. 3 march 2008



b

102 101

100 100 101 102 103 104

102

100 100 101 102 103 104

0

10

102 101

101

100 100 101 102 103 104

37.5%

103

100 101 102 103 104

104

103 102

91.1%

103

102

101

101

100

100

100

100 100 101 102 103 104

100 101 102 103 104

75.7%

101

102 101 100 100 101 102 103 104

100 100 101 102 103 104

43.8%

103

102

102

101

101

100 100 101 102 103 104

100 100 101 102 103 104

104

103

103

102

90.3%

101

100 100 101 102 103 104

102

104 59.0%

102

65.0%

103 102

101

101

101

100 100 101 102 103 104

100

100 100 101 102 103 104

CD11c

100 101 102 103 104

CD11c

mDC-enGFP

mDC-enGFP 104

104 18.9%

102 101

51.3% 100 100 101 102 103 104

101

CD11c

d

5

[H3]-thymidine incorporation (cpm)

57.3%

102 22.5% 100 100 101 102 103 104

CD11c

c

103

1.8×10

1.6×105 1.4×105 5

1.2×10

DC-erb + allo-T DC-NT + allo-T DC-NT + syn-T DC-erb + syn-T

1.0×105 5

0.8×10

0.6×105 5

0.4×10

0.2×105 0.0×105

0

0.05

0.1 0.15 0.2 Stimulator:effector ratio

0.25

0.3

[H3]-thymidine incorporation (cpm)

103

enGFP

enGFP

100 101 102 103 104

104

67.0%

103 CD80

102

CD80

10

104

45.5%

mDC-erbB2tr 104

I-Ab

103

102

101

100 101 102 103 104

No stain

104

38.1%

102

62.3%

103

CD11c

CD86

3

I-Ab

No stain

104

103

104

52.3%

101

CD11c mDC-erbB2tr 104

104

60.0%

103

CD86

68.1%

101

10

39.5%

CD86

102

3

I-Ab

103

No stain

103

mDC-NT 104

104

10

39.3%

CD86

104

4

CD80

104

I-Ab

No stain

mDC-NT

CD80

a

1.8×105 1.6×105 1.4×105 1.2×105

DC-erb + allo-T DC-NT + allo-T DC-NT + syn-T DC-erb + syn-T

1.0×105 5

0.8×10

0.6×105 0.4×105 0.2×105 0.0×105

0

0.05

0.1 0.15 0.2 Stimulator:effector ratio

0.25

0.3

Figure 3 Transducing dendritic cells (DCs) with LV/erb and LV/enGFP does not affect DC phenotype or allostimulatory capacity. Bone marrow-derived DCs were transduced on culture day 3 and matured on culture day 8. Transduced and non-transduced (NT) immature (day 5 and/or 7) and mature (day 9) DCs prepared for the (a) first and (b) second immunizations were stained with antibodies recognizing CD11c, I-Ab, CD80, and CD86 and analyzed by flow cytometry. Numbers indicate the percentage of positive cells in the indicated density plot quadrants. Mature day 9 DCs from C57BL/6 mice were also co-cultured with [3H]-thymidine-pulsed syngeneic (C57BL/6) or allogeneic (BALB/c) splenocytes to compare the allostimulatory capacity of NT DCs to either (c) erbB2tr- or (d) enGFP-transduced DCs. Co-cultures were plated in triplicate and the mean ± SD values are shown. LV, lentivirus; enGFP, enhanced green fluorescent protein.

versus 38.1%, respectively). A minor difference in the percentage of CD11c+CD80+ DCs was measured from NT compared to erbB2tr-transduced cultures (39.5% versus 45.5%, respectively). Similarly, 37.5% of NT DCs and 43.8% of erbB2tr-transduced DCs were CD11c+CD86+ (Figure 3a). The DCs generated for the second vaccinations exhibited similar trends (Figure 3b). The percentage of CD11c+ cells in the control cultures was 91.1%, compared to 90.3% for erbB2tr-transduced DCs, and 79.8% for enGFP-transduced DCs. The percentages of CD11c+I-Ab+ were similar for control and erbB2tr-transduced DCs (60.0% versus 67.0%, respectively). Comparing the percentages of DCs expressing costimulatory molecules, we found that 52.3% of NT DCs and 59.0% of erbB2tr-transduced DCs were CD11c+CD80+. A minor difference in the CD11c+CD86+ percentage was also detected between the control and transduced DCs (62.3% versus 65.0%, respectively). To determine whether transduction with LV/enGFP or LV/erb affected DC function, we compared the ability of NT and transduced DCs to induce an allogeneic mixed lymphocyte reaction. Molecular Therapy vol. 16 no. 3 march 2008

We cultured H-2b-expressing DCs (transduced and control) with either H-2d splenocytes from BALB/c mice or H-2b splenocytes from C57BL/6 mice and measured splenocyte proliferation by thymidine incorporation (Figure 3c and d). No significant differences were found between the allostimulatory capacities of NT DCs and either LV/enGFP- or LV/erb-transduced DCs.

Vaccination with low doses of LV-modified DCs generates antigen-specific tumor protection Although many studies employing DC-vaccination strategies15 evaluate tumor protection ~1–2 weeks after vaccination, we chose to investigate the long-term benefits of a prime/boost vaccination strategy by challenging mice ectopically with RM-1 prostate tumors cells 6 weeks after the second vaccination. RM-1 cells grow aggressively in vivo, providing a stringent model for assessing tumor growth after vaccination; subcutaneously implanting just 104 RM-1 cells will yield palpable tumors within 1 week and can compromise mouse survival by 10 days after implantation.22 The RM-1 tumor cell line lacks endogenous erbB2 expres609



sion according to our fluorescence-activated cell sorting analysis (Figure 4a). Therefore, we generated a clonal cell population of erbB2tr-expressing RM-1 cells (RM-1-erbB2tr) by onco-retroviral transduction followed by clonal isolation (Figure 4a). In our above-mentioned pilot study, we first tested the efficacy of three immunizations using doses of 2 × 105 and 2 × 103 erbB2trtransduced DCs to protect against subsequent challenge with erbB2-expressing tumors. We vaccinated mice three times with either erbB2tr-transduced or NT DCs. Two weeks after the third vaccination, we injected mice with NT RM-1 cells (RM-1-NT) on one dorsal flank and RM-1-erbB2tr cells on the opposite dorsal flank in order to generate a bilateral tumor model in the same individual animal. Whereas many tumor protection studies utilizing virally transduced DCs typically inject between 0.5 × 106 and 1 × 106 DCs per immunization,15 the focus of this pilot study was to investigate the possible benefits of using markedly lower doses of transduced DCs. In that initial pilot study, we observed that both erbB2 immunization regimens offered considerable protection against erbB2tr-expressing RM-1 tumors specifically, compared to that obtained using NT DCs (data not shown). To examine these low-dose outcomes in more detail, we next performed a larger study. We again used the dose of 2 × 105 DCs

a 300

RM-1-NT

240 Counts

for immunizing one cohort of mice, and the 100-fold lower dose of 2 × 103 DCs for another. Importantly, however, we also reduced the number of vaccinations per animal to just two. We injected mice twice with either control, erbB2-transduced, or enGFPtransduced DCs, 2 weeks apart. We next inoculated animals with the same tumor challenge described in our pilot study above. As a positive control, one group of mice was immunized twice with 2 × 105 erbB2tr-transduced DCs mixed with the complete Freund’s adjuvant emulsion. Using this prime/boost strategy, none of the six mice that were immunized with 2 × 105 erbB2trtransduced DCs showed RM-1-erbB2tr tumor growth, whereas RM-1-NT tumors grew rapidly in each animal (Figure 4b). By contrast, the control naïve mice and mice immunized with 2 × 105 NT or enGFP-transduced DCs developed both RM-1-NT and RM-1-erbB2tr tumors with an aggressive growth profile that necessitated killing within 2 weeks. Strikingly, significant tumor protection from RM-1-erbB2tr tumors was also observed in mice that were immunized with the 100-fold lower dose of 2 × 103 erbB2-transduced DCs (Figure 4c). In this group, two of six mice displayed complete tumor protection until the point of killing at 2 weeks post-tumor challenge, and two of six mice showed reduced RM-1-erbB2tr growth compared to control cohorts.

RM-1-erbB2tr

180 120 60 0

100 101 102 103 104 erbB2tr

Tumor volume (mm3)

800 600 400

**

* *

200 7

9 11 Days post-tumor challenge

Naïve DC-NT DC-enGFP DC-erbB2tr +ve control

1,000 800 600

**

200 7


1,600 1,200 800 400 7

13


13

RM-1-NT tumor growth

2,400

400

0


2,000

0

13

RM-1-erbB2tr tumor growth

1,200

RM-1-NT tumor growth

2,400 Tumor volume (mm3)


1,000

0

c

RM-1-erbB2tr tumor growth

1,200

Tumor volume (mm3)

Tumor volume (mm3)

b


2,000 1,600 1,200 800 400 0

7


13

Figure 4 Immunization with low doses of DC-erbB2tr generates potent antitumor responses. (a) Cell lines RM-1-erbB2tr and non-transduced RM-1 (RM-1-NT) were stained with anti-erbB2 antibody and analyzed by flow cytometry. Mice were immunized two times with either (b) 2 × 105 or (c) 2 × 103 DCs and then inoculated with RM-1-erbB2tr and RM-1-NT cells on opposite flanks. Plotted values are the mean ± SEM of five or six mice per group, except for the positive control group (n = 3). *P < 0.05, versus DC-erbB2tr vaccination; **P < 0.005, versus DC-erbB2 vaccination. DC, dendritic cell; enGFP, enhanced green fluorescent protein.

610




Plasma levels of anti-erbB2tr antibodies

10 9 8 7 6 5 4 3 2 1 0


300

20

30

40 Days

*

50

60

70

80

Plasma levels of anti-erbB2tr antibodies Naïve DC-NT DC-enGFP DC-erbB2tr +ve control

0

10

20

DC-GFP

200

DC-erb No DCs

150 100 50

1,200

* *

10

DC-NT

250

2 × 105 2 × 105 2 × 105 2 × 103 2 × 103 2 × 103 DC-NT DC-GFP DC-erb DC-NT DC-GFP DC-erb

Naïve

Immunization cohort

*

0

IL-2 splenocyte secretion

350

0

30

40 Days

50

60

70

DC-NT DC-GFP

600

DC-erb No DCs

400 200 2 × 105 2 × 105 2 × 105 2 × 103 2 × 103 2 × 103 DC-NT DC-GFP DC-erb DC-NT DC-GFP DC-erb

Naïve

Immunization cohort 700

Figure 5 DC-erbB2tr immunization dose of 2 × 105 cells causes a sustained erbB2-specific humoral response. Plasma samples from naïve and immunized mice were pooled according to cohort and quantitatively tested for the presence of anti-erbB2 antibodies by a flow cytometry– based enzyme-linked immunosorbent assay. Graphs show a comparison between antierbB2tr levels from control groups and either (a) 2 × 105 or (b) 2 × 103 DC dose groups. Plotted values are the mean ± SEM of three independent assays. *P < 0.01, versus vaccination with DC-NT, DC-enGFP, or no vaccination. DC, dendritic cell; enGFP, enhanced green fluorescent protein; NT, non-transduced.

Molecular Therapy vol. 16 no. 3 march 2008

IFN- splenocyte secretion

800

0

80

2 × 105 DC-erb +CFA

1,000

erbB2-Specificity index

Anti-erbB2tr titer (mg/ml)

400

*

b 10 9 8 7 6 5 4 3 2 1 0

Analysis of cytokine secretion from splenocytes To further evaluate mechanisms, we harvested the spleens from naïve and immunized mice 6 weeks after the second vaccination. We re-stimulated splenocytes in vitro for 24 hours with freshly prepared transduced or control DCs and analyzed culture supernatants for production of Th1 [interleukin-2 (IL-2), interferon-γ (IFN-γ), and tumor necrosis factor-α (TNF-α)] and Th2 (IL-4 and IL-10) cytokines. We found that recipients of 2 × 105 erbB2trtransduced DCs produced greater levels of IL-2, IFN-γ, and TNF-α following in vitro re-stimulation with erbB2tr-transduced DCs relative to controls (Figure 6). By contrast, this erbB2trspecific cellular response was absent from the supernatants of all other mouse cohorts. To quantify the levels of antigenspecificity, we calculated a “specificity index” by normalizing


Anti-erbB2tr titer (mg/ml)

a

ground immunogenicity of these tumors. Injecting the lower dose of 2 × 103-transduced DCs did not lead to detectable anti-erbB2tr antibodies at least within the sensitivity limits of this assay, despite the antitumor effects observed above (Figure 5b).


Mice vaccinated with DC-erbB2tr show strong antigen-specific humoral immunity To begin investigating potential mechanisms by which DCerbB2tr immunizations could break tolerance against erbB2tr, we collected blood from each mouse on a weekly basis and measured the plasma levels of anti-erbB2 antibodies. As shown in Figure 5a, mice immunized with complete Freund’s adjuvant + erbB2tr-transduced DCs (positive control) began producing modest levels of anti-erbB2tr antibodies. Following the second vaccination, these mice showed steadily increasing titers that peaked at ~45 days after the first DC injection. Relatively high antibody levels were detected for up to 70 days after the prime vaccination when the mice were killed. In our experimental groups, mice injected twice with 2 × 105 erbB2tr-transduced DCs showed a rapid increase in antibody titer after the boost vaccination. Indeed, within 10 days, the average anti-erbB2tr titer rose to over five times the average level in control mice. After this peak, a steady decline was measured, but specific anti-erbB2tr antibodies were still detectable at day 50, when mice were inoculated with tumors. In the nonvaccinated group of mice, challenge with RM-1-erbB2tr tumors caused a slight increase in antibody titers, revealing weak back-


TNF- splenocyte secretion

600 500

DC-NT

400

DC-GFP

300

DC-erb No DCs

200 100 0

2 × 105 2 × 105 2 × 105 2 × 103 2 × 103 2 × 103 DC-NT DC-GFP DC-erb DC-NT DC-GFP DC-erb

Immunization cohort

Naïve


Figure 6 Immunization using dendritic cells (DCs) transduced with LV-erb offers antigen-specific Th1 immunity. Splenocytes from naïve and immunized mice were co-cultured in triplicate with LV-transduced or non-transduced DCs (NT DCs) for 24 hours and supernatants were analyzed for interleukin-2 (IL-2), interferon-γ (IFN-γ), and tumor necrosis factor-α (TNF-α) by Bio-Plex multiplex sandwich immunoassays. Specifi city index values were calculated by normalizing cytokine concentration values from each 24-hour re-stimulation condition to the values obtained from re-stimulation with non-transduced DCs. LV, lentivirus.

611



60

b

CD8+ cytokine production

Percentage of CD4+IL-2+splenocytes

Percentage of CD8+IL-2+splenocytes

a

50 40 30 20 10 0

2 × 105 2 × 105 2 × 105 DC-NT DC-GFP DC-erb

2 × 103 2 × 103 2 × 103 DC-NT DC-GFP DC-erb

Naïve

60 50 40 30 20 10 0


Immunization cohort

Percentage of CD4+IFN-+splenocytes

Percentage of CD8+IFN-+splenocytes

10 8 6 4 2 2 × 105 2 × 105 2 × 105 DC-NT DC-GFP DC-erb

2 × 103 2 × 103 2 × 103 DC-NT DC-GFP DC-erb

Naïve

Naïve


8 6 4 2 2 × 105 2 × 105 2 × 105 DC-NT DC-GFP DC-erb

2 × 103 2 × 103 2 × 103 DC-NT DC-GFP DC-erb

Immunization cohort 7 Percentage of CD4+TNF-�+splenocytes

Percentage of CD8+TNF-�+splenocytes


10

0


7 6 5 4 3 2 1

2 × 105 2 × 105 2 × 105 DC-NT DC-GFP DC-erb

2 × 103 2 × 103 2 × 103 DC-NT DC-GFP DC-erb

Naïve

6 5 4 3 2 1 0


2 × 105 2 × 105 2 × 105 DC-NT DC-GFP DC-erb

2 × 103 DC-NT

2 × 103 2 × 103 DC-GFP DC-erb

Naïve


Naïve


Immunization cohort

Immunization cohort 5 Percentage of CD4+IL-4+splenocytes

5 Percentage of CD8+IL-4+splenocytes

Naïve

12

Immunization cohort

4 3 2 1 0

2 × 103 2 × 103 2 × 103 DC-NT DC-GFP DC-erb

14

12

0

2 × 105 2 × 105 2 × 105 DC-NT DC-GFP DC-erb

Immunization cohort

14

0

CD4+ cytokine production

2 × 105 2 × 105 2 × 105 DC-NT DC-GFP DC-erb

2 × 103 2 × 103 2 × 103 DC-NT DC-GFP DC-erb

Immunization cohort

Naïve


4

3

2

1

0

2 × 105 2 × 105 2 × 105 DC-NT DC-GFP DC-erb

2 × 103 2 × 103 2 × 103 DC-NT DC-GFP DC-erb

Immunization cohort

Figure 7 Cytokine production analysis of fractionated splenocytes shows erbB2tr-specific Th1 immunity. Mice were immunized two times with non-transduced, enhanced green fluorescent protein (enGFP)-transduced, or erbB2tr-transduced dendritic cells (DCs) (2 weeks apart) and then 6 weeks later, 4 of 10 mice were killed from each group to check the cytokine levels from their splenocytes. Percentages of (a) CD8+ or (b) CD4+ T cells producing interferon-γ (IFN-γ), interleukin-2 (IL-2), tumor necrosis factor-α (TNF-α), and IL-4 were measured by intracellular flow cytometry.

612




of CD4+ or CD8+ T cells producing IL-2, IFN-γ, TNF-α, or IL-4 were measured by intracellular flow cytometry (Figure 7). Splenocytes from mice that received 2 × 105 erbB2tr-transduced DCs produced high levels of IL-2 and IFN-γ. In this immunization cohort, ~57% of CD8+ cells and 48% of CD4+ cells were IL-2+, compared to 1,100-fold more IFN-γ was produced after erbB2tr-specific re-stimulation. We also found a large (635-fold) increase in TNF-α production following in vitro re-stimulation with erbB2tr-transduced DCs relative to controls. Levels of IL-4 and IL-10 in the co-culture supernatants were generally very low and specificity toward erbB2tr was not observed (data not shown). We confirmed the specificity of our cytokine responses toward erbB2tr antigen using fractionated splenocytes. Six weeks after immunizing mice with erbB2tr- or control-DCs, percentages

2,000 1,500 1,000 500 0

No CD4+ CD8+ NK depletion depletion depletion depletion 2×

No depletion CD4+ depletion CD8+ depletion NK depletion

500 400 300 200 100 0

0

2

4 6 8 10 12 Days post-challenge

500 400 300 200 100 0

0

2


400 300 200 100 0

2


14

600


500 400 300 200 100 0

0

2


14

DC-enGFP immunization

No CD4+ CD8+ NK depletion depletion depletion depletion

No depletion

2 × 103 DC-enGFP immunization

Naïve

IL-4 production

2,500 2,000 1,500 1,000 500 0

14

RM-1-erbB2tr tumor growth after 3 immunization with 2 × 10 DC-enGFP Tumor volume (mm3)

Tumor volume (mm3)



500

0

14

RM-1-erbB2tr tumor growth after 5 immunization with 2 × 10 DC-enGFP 600

600

d



No depletion

2 × 105 DC-erbB2tr immunization

2 × 103 DC-erbB2tr immunization

Naïve

IL-4 production

2,500 IL-4 secretion (pg/ml)

600

RM-1-erbB2tr tumor growth after 3 immunization with 2 × 10 DC-erbB2tr

IL-4 secretion (pg/ml)

RM-1-erbB2tr tumor growth after 5 immunization with 2 × 10 DC-erbB2tr Tumor volume (mm3)

Tumor volume (mm3)

b

105

2,000 1,500 1,000 500 0



No depletion



Naïve

Figure 8 Involvement of CD4+, CD8+, and natural killer (NK) cells in mediating RM-1-erbB2tr tumor protection. Five mice per group were first vaccinated with 2 × 103 or 2 × 105 erbB2tr- or enhanced green fluorescent protein (enGFP)-transduced dendritic cells (DCs) two times, 2 weeks apart. Approximately 6 weeks after the last immunization, mice were depleted of CD4+, CD8+, or NK cells, or left non-depleted, and injected bilaterally with RM-1-NT and RM-1-erbB2tr tumors. (a) RM-1-NT and (b) RM-1-erbB2tr (Tumor volumes were calculated based on caliper measurements of length × width × height. Graphed values are the mean ± SEM of four or five mice per group. Splenocytes were prepared from each mouse at 2 weeks post-tumor challenge and cultured at a concentration of 2 × 106 cells/ml for 48 hours. Analysis of (c) interferon-γ (IFN-γ) and (d) interleukin-4 (IL-4) production was done in triplicate by enzyme-linked immunosorbent assay. NT, non-transduced.


613


Role of CD4+, CD8+, and natural killer cells in mediating tumor protection To evaluate the importance of CD4+, CD8+, and natural killer (NK) cells in determining tumor protection, we repeated our study using mice depleted of these cell lineages. First, mice were immunized two times (2 weeks apart) with either erbB2tr- or enGFP-transduced DCs at the dose of 2 × 105 or 2 × 103 cells per immunization. Approximately 6 weeks after immunization, mice were depleted of CD4+, CD8+, or NK cells by antibody administrations and bilaterally challenged with RM-1-NT and RM-1erbB2tr tumors. As expected, RM-1-NT tumor growth was not significantly affected by cell depletion (Figure 8a). Whereas RM1-erbB2tr tumor growth was inhibited in non-depleted recipients of 2 × 105 erbB2tr-transduced DCs, there was some enhanced growth following all three depletion conditions (Figure 8b). Interestingly, mice that were immunized with 2 × 103 erbB2trtransduced DCs relied on a protection mechanism that appeared to be highly dependent on CD4+ T cells. In this cell depletion study, the use of enGFP-transduced DCs as control vaccines did not yield clear differences in tumor growth between depleted and non-depleted conditions, but overall rates of tumor growth in this iteration of the experiment were slower than expected. To further understand the mechanism of tumor protection, we used enzyme-linked immunosorbent assays to analyze the production of IFN-γ and IL-4 in the supernatants of cultured splenocytes harvested from depleted and non-depleted immunized mice (Figure 8c and d). Although cytokine levels were generally difficult to detect in all mice that were immunized with 2 × 103 erbB2tr-transduced DCs, we were able to measure cytokines produced by splenocytes from mice immunized with the higher dose of DCs. Production of IFN-γ was reduced by 44.5-fold in NK cell–depleted mice and 3.3-fold in CD4+ T-cell depleted mice compared to non-depleted mice (Figure 8c). CD8+ T-cell depletion also yielded a relatively small reduction in IFN-γ splenocyte secretion on average (1.2-fold), but was associated with a high level of variability (Figure 8c). The depletion of each of the three lineages also correlated with an increase in IL-4 production in mice that received 2 × 105 erbB2tr-transduced DCs (Figure 8d). This finding reflects how Th1/Th2 cross-regulation can upregulate Th2 immunity when Th1 cytokine production is low. To evaluate the role of a fourth cell type in mediating antitumor immunity, we studied the impact of gene-modified DC immunization on the behavior of regulatory T cells, defined as CD4+CD25+FoxP3+cells. We used flow cytometry to investigate whether there were changes in regulatory T–cell levels in the spleens of all immunized and naïve mice. However, significant differences in regulatory T–cell levels between experimental groups were not detected (data not shown).

Discussion This is the first study to demonstrate the use of LV-transduced DCs in an immuno-gene therapy cancer model targeting the self-antigen erbB2 in mice. Despite the growing number of DC-mediated immuno-gene therapy studies, complete protection against cancer has been mostly elusive.15 One limitation might be related to the level of antigen presentation by DCs. Thus, the method of engineering DCs to present TAAs may play an important role in the 614


potency of immunotherapy schemas. Our use of recombinant retroviral vectors encoding full-length or large portions of TAAs may permit transduced DCs to present a broad repertoire of natural immunogenic tumor antigen peptides in stable major histocompatibility complexes. Indeed, we have previously demonstrated the utility of DCs transduced with onco-retroviruses to express xenogeneic human prostate TAAs in mediating antitumor immunity in mice.23 In our current study, we employed an LV-based system and found that DCs could be even more efficiently transduced to express antigens, without compromising their phenotype or function. This finding is especially important given that the functional effects of LV transduction of murine DCs with erbB2 have not been previously investigated and our LV/GFP transduction efficiency is somewhat higher than those reported by other groups.10,24 To address the possibility of oncogenic effects due to the use of integrating vectors, many groups have already extensively assessed the safety of LV transduction.25–28 In addition, although genetically modifiying DCs to express murine erbB2 is original, data from other DC studies employing human or rat homologs have shown that this type of manipulation does not lead to oncogenic effects in DCs.29–32 To further decrease the possibility of affecting the target myeloid DC population by overexpressing a heterologous signaling molecule, we employed a kinase-deficient version of erbB2 for our LV construct. Although recent evidence suggests that some cases of acute myeloid leukemia may have a myeloid DC origin,33,34 myeloid DCs may be less amenable to oncogenesis than plasmacytoid DCs, whose malignant counterparts are well documented.35–38 In the future, gene therapy protocols may also incorporate additional safety mechanisms, such as the tmpk/ azidothymidine suicide gene therapy strategy we developed.39 While monitoring enGFP and erbB2tr levels in transduced DC cultures, we observed a decrease in transgene-positive DCs between culture day 5 and 9. The decrease was more pronounced in DC populations transduced with erbB2tr as compared to enGFP. Some researchers have attributed this phenomenon to the elimination of genetically modified DCs by T cells that contaminate DC cultures.40 The possibility that this elimination occurred in our cultures cannot be ruled out, since our DC generation protocol allows for the persistence of low levels of lymphocyte contaminants.41 However, the finding that enGFP levels can remain stable over time does not fully support this notion. Instead, it may simply point to the possibility that peptide formation from protein degradation within DCs occurs more readily for some antigens than for others. Our flow cytometry data, which shows decreased whole antigen expression, may be capturing the transition from whole protein expression to peptide formation as DCs develop into mature APCs. It is important to consider the number of DCs being employed for immunization strategies, as an effort to minimize the DC dose could improve the feasibility of employing such strategies in clinical settings. The availability of effective patient DCs may be a limiting factor. We used only two immunizations, in order to decrease the number of DCs required, and were still able to induce specific immune-mediated antitumor responses. As stated previously, typical in vivo immunotherapy approaches involving virally transduced DCs administer between 0.5 × 106 and 1 × 106 DCs per immunization.15 We found that a dose as low as 2 × 103 DCs www.moleculartherapy.org vol. 16 no. 3 march 2008


offered partial protection against erbB2tr-expressing tumors, suggesting that the minimum effective DC dose falls in the range between our tested doses. In addition, these results were obtained in a relatively long-term setting, as mice were challenged with tumors 6 weeks after the last immunization. As a direct result of testing these low-dose DC vaccinations, we also gain the ability to establish a workable threshold that can be used as a sensitive read-out in future studies to determine if other co-manipulations further enhance the immunotherapy effect. In agreement with the recent results of Sakai et al. (2004),31 we also observed a humoral response in vaccinated mice that correlated with tumor protection in the 2 × 105 DC-dose group. Mice vaccinated with 2 × 103 erbB2tr-transduced DCs had comparatively low antibody titers. In addition, long-term Th1 immunity was observed in the 2 × 105 DC-dose mice. It was not detected in the 2 × 103 DC dose cohort despite the finding that two of six of these mice were fully protected from developing RM-1-erbB2tr tumors and an additional two mice had markedly reduced tumor volumes compared to controls. This may reflect the detection limit of the assays, and may also point to the very sensitive nature of this system. It may be that specific antitumor immune responses can be induced with very subtle changes to the immune marker profiles. In our Th1 assays, the erbB2tr specificity of cytokine production in the 2 × 105 DC-erbB2tr cohort is clear. It is worth noting that we did detect low levels of Th1 reactivity toward enGFP, probably because enGFP is a xeno-antigen and therefore inherently foreign to the mouse species. That we did not observe significant levels of Th2 cytokines at 6 weeks after vaccination is consistent with the waning of anti-erbB2tr antibody levels over time. Our DC vaccination strategy using erbB2-transduced cells was well tolerated. Although the self-antigen erbB2 is naturally expressed at varying levels throughout the mouse body including the lungs, intestines, and brain,42 manifestations of autoimmune toxicity were not observed in the 10 weeks after initiation of the vaccination schedule. Other groups have also reported that antitumor immunity in mouse models can occur without damaging normal tissues, even when self-antigens are targeted.43–45 In conclusion, our results show that vaccination using relatively low doses of DCs transduced with a recombinant LV to express a truncated form of erbB2 can safely and effectively protect mice against tumor development in an antigen-specific manner. This is the first implementation of this stable gene transfer method for this broadly important antigen in tumor biology. Tumor protection was associated with antigen-specific cellular and humoral immunity. The recombinant LV system we utilized served as an efficient gene transfer vehicle, which did not adversely affect DCs. Efforts to develop low dose DC-immunotherapy strategies will be of value in clinical situations where patient DCs may be scarce. Such cancer immunotherapy vaccines may be particularly applicable before tumors are established and in early stage disease, and could reduce the need for more intensive treatments with systemic toxicity such as chemotherapy or radiation therapy.

Materials and Methods LV vector construction and preparation of high titer stocks. The enGFP-

containing LV pHR′EF-GW-SIN (LV/enGFP) was described previously.46 Molecular Therapy vol. 16 no. 3 march 2008


To construct an erbB2tr-containing recombinant LV (LV/erb), the enGFP complementary DNA sequence was excised from pHR′EF-GW-SIN by EcoRI (New England Biolabs, Beverly, MA) digestion and replaced with the complementary DNA sequence for erbB2tr (GI:28386210). This erbB2tr sequence was amplified by polymerase chain reaction from the Invitrogen pYX-Asc plasmid (IMAGE 5702040) with Taq polymerase (both Invitrogen, Burlington, Canada), ligated into PCR-Script Amp(+) SK(+) (Stratagene, La Jolla, CA), and excised by EcoRI digestion. LV particles were generated by calcium–phosphate transfection of 293T cells (kindly provided by Dr. Michele Calos, Stanford University, CA) with the plasmids pCMVDR8.91, pMD.G,47 and either LV/enGFP or LV/erb. Viral supernatants were collected at 24 and 48 hours posttransfection, filtered using a 0.45-µm filter, and concentrated at 19,000g for 2 hours using an Optima L-100 XP Ultracentrifuge (Beckman Coulter Canada, Mississauga, Canada). Concentrated virus preparations were serially diluted and titered on 293T cells by fluorescence-activated cell sorting analysis as previously described.21 Mice and cell lines. C57BL/6 (Jackson Laboratories, Bar Harbor, ME) and

BALB/c (Charles River, Wilmington, MA) mice were bred and housed under specific pathogen-free conditions at the University Health Network Animal Resource Centre. RM-1 cells, a murine prostate cancer cell line syngeneic to the C57BL/6 strain, were kindly provided by Dr. Timothy Thomson (Baylor College of Medicine, Houston, TX). The clonal RM-1erbB2tr cell line was generated by transducing RM-1 cells to overexpress a kinase-truncated form of erbB2 (erbB2tr) and then isolating single cell clones. For these transductions, an onco-retroviral pUMFG-erbB2tr vector was constructed (M.E.M. and J.A.M., unpublished data) and transfected into the E86 packaging cell line to generate virus-producing E86 cells, as previously described.48 In vitro growth characteristics of RM-1-erbB2tr versus wild-type RM-1 cells were nearly identical (data not shown). All animal experiments were performed under a protocol approved by the Animal Care Committee at the University Health Network. Murine DC generation and transduction. DCs were generated accord ing to Lutz et al. (1999)41 with slight modifications. Briefly, bone marrow was flushed from femurs and tibiae of C57BL/6 mice using a 25-G needle. Red blood cells were lysed using red blood cell lysing buffer (Sigma, St. Louis, MO). Remaining cells were plated in 10-cm Petri dishes at a concentration of 2 × 105 cells/ml in a total volume of 10 ml/ dish. DC media consisted of RPMI medium with 10% fetal bovine serum (PAA Laboratories, Etobicoke, Canada), 1% penicillin/streptomycin, 5 × 10–5 mol/l 2-mercaptoethanol (both from Sigma St. Louis, MO), 40 ng/ ml rmGM-CSF and 5 ng/ml rmIL-4 (both from Peprotech, Rocky Hill, NJ). Cells were infected on day 3 of culture with either LV/erb or LV/enGFP, or left uninfected. Half-volume media changes were done every other day starting on day 4. On day 7, 50 ng/ml of rmTNF-α (Peprotech Rocky Hill, NJ) was added for 24–48 hours of DC maturation. Flow cytometric analysis of DCs, tumor cells, and splenocytes. DCs

and tumor cells were stained with an anti-erbB2 primary antibody (Ab4, Oncogene Science, Tarzana, CA) and a phycoerythrin (PE)-conjugated poly-adsorption goat anti-mouse immunoglobulin secondary antibody (BD Biosciences Canada, Mississauga, Canada) and cell-surface expression of erbB2tr was measured using a FACS Calibur (BD Biosciences, Franklin Lakes, NJ). For phenotypic analysis of DCs, the following BD antibodies were used with appropriate isotype controls: PE- or fluorescein isothiocyanate (FITC)-conjugated anti-CD11c (clone HL3), purified anti-CD80 (clone 1G10), and FITC-conjugated anti-CD86 (clone GL1), FITC-conjugated anti-I-Ab (clone AF6-120.1). For analysis of splenocytes, the following antibodies were used with appropriate isotype controls: anti-CD8-FITC, anti-TNF-α-PE, anti-CD4-PerCP, anti-IFN-γ-APC, anti-IL-4-PE, anti-IL2-APC. Levels of regulatory T cells were measured by cell-surface staining with anti-CD4-FITC, anti-CD25-APC, and intracellular staining with

615



anti-FoxP3-PE according to manufacturer’s instructions (BD Biosciences Canada, Mississauga, Canada). Allogeneic mixed lymphocyte reaction. Transduced and control DCs were

harvested on day 9 of culture and dosed with 30 cGy in a Gammacell 3000 Elan 137Co irradiator (Nordion International, Ottawa, Canada). Freshly isolated splenocytes from C57BL/6 and BALB/c mice were B-cell depleted using goat anti-mouse immunoglobulin magnetic beads (Dynal, Brown Deer, WI). The remaining T-cell enriched population was plated in triplicate in a 96-well U-bottom plate (BD Biosciences, San Jose, CA) at 2 × 105 cells per well in T cell media. Next, serially diluted, irradiated DCs (range of 0–0.6 × 105 cells/well) were added. Following 4 days of co-incubation, 1 µCi of [3H]methyl-thymidine was added to each well for 20 hours. Thymidine incorporation was measured using a Beckman LS 1801 Liquid Scintillation Counter (Beckman Coulter Canada, Mississauga, Canada). Immunizations and tumor inoculations. C57BL/6 mice were injected

intraperitoneally with 2 × 105 or 2 × 103 DCs transduced with LV/erbtr, LV/enGFP, or NT controls in 200 ml of phosphate-buffered saline (PBS). As a positive control, one group of five mice was injected with 2 × 105 erbB2tr-transduced DCs along with complete Freund’s adjuvant (Sigma St. Louis, MO). These immunizations were repeated 2 weeks later. Six weeks after the second immunization, 6 of 10 mice in each cohort were challenged with bilateral tumors and the remaining mice were killed for splenocyte cytokine secretion analyses (see below). For the tumor challenge, each mouse was injected subcutaneously with 2 × 105 RM-1-NT and RM-1-erbB2tr cells (in 200 ml of PBS) in the dorsal left and right flanks, respectively. Starting 6 days later, the length (l), width (w), and height (h) of each tumor was measured by caliper on a daily basis. Tumor volume was calculated by multiplying l × w × h. Measurement

of

anti-erbB2tr

antibody

from

mouse

plasma.

Approximately 200 µl of blood was collected weekly from the tail vein of each mouse into EDTA-coated tubes (Sarstedt, Montreal, Canada). Plasma was isolated by centrifugation at 18,000g at 4 °C for 20 minutes. Plasma anti-erbB2 measurements were performed using a flow cytometry–based enzyme-linked immunosorbent assay method we developed that was based on that described by Piechocki et al. (2002).49 Briefly, RM-1-erbB2tr and wild-type RM-1 cells were first stained with diluted plasma samples or primary Ab4 antibody (above) for 1 hour on ice followed by 2 washes with PBS. Secondary staining with PE-conjugated poly-adsorption goat anti-mouse antibody was done for 1 hour on ice, again followed by 2 PBS washes. 7-Amino-actinomycin D was added to each sample to exclude dead cells from flow cytometric analysis. The mean fluorescence intensity value in the FL2 channel was measured on a FACS Calibur for each sample. A standard curve was generated by plotting Ab4 antibody concentration versus the mean fluorescence intensity values of the Ab4-stained RM-1erbB2tr cells. This curve was used to convert mean fluorescence intensity values of plasma anti-erbB2 levels from each mouse cohort into antibody concentration values. Each experiment was performed three times and the SD of the means was calculated. Cytokine secretion assays. Spleens from immunized and naïve control

C57BL/6 mice were dissociated into single-cell suspensions and treated with red blood cell lysis buffer. Red blood cell–depleted splenocytes were cryopreserved in freezing medium (90% fetal calf serum, 10% dimethyl sulfoxide), then thawed when needed using a method described by Maecker et al. (2005).50 Briefly, cryovials were warmed to 37 °C in a waterbath and the contents diluted drop by drop with an equal volume of warm media. Diluted cells were transferred to a 50-ml tube containing 8 ml of warm media per cryovial of added cells and centrifuged at 290g for 7 minutes. Cell pellets were resuspended and brought to a final concentration of 5 × 106 cells/ml in RPMI medium containing 10% fetal calf serum, 1% penicillin/streptomycin, 1% minimal non-essential amino acids (Invitrogen

616

Burlington, Canada), and 5 × 10–5 mol/l 2-mercaptoethanol. Next, 200 ml of cell suspensions were transferred to each well of 96-well round-bottom plates (BD) and incubated at 37 °C for 18 hours. Splenocytes were then collected from each well, counted, and plated in 24-well plates at 3 × 106 cells per well in 1 ml. Approximately 2 × 105 freshly prepared DCs that were left NT or that were transduced with LV/erb, LV/enGFP were added to each well. Co-cultures were incubated at 37 °C for 24 hours and supernatants were collected and stored at –20 °C. IFN-γ, IL-2, TNF-α, IL-4, and IL-10 levels were measured from thawed supernatant samples by BioPlex multiplex sandwich immunoassays according to the manufacturer’s protocol (Bio-Rad Laboratories, Hercules, CA). CD4+, CD8+, and NK cell depletion. Five mice per group were immunized

two times (2 weeks apart) with either erbB2tr- or enGFP-transduced DCs at the dose of 2 × 105 or 2 × 103 cells per immunization. Approximately 6 weeks after immunization, mice were regularly depleted of CD4+, CD8+, or NK cells by antibody administrations and bilaterally challenged with ~0.1 × 105 RM-1-NT and RM-1-erbB2tr tumors (subcutaneously). For CD8+ T-cell depletion, mice were injected intraperitoneally with 200 µg of anti-CD8 (clone 2.43) 3 days and 1 day before bilateral tumor inoculations. Mice were then given 200 µg of antibody once per week. For CD4 depletion, mice were injected with 400 µg of anti-CD4 (clone GKI.5) 3 days and 1 day before bilateral tumor challenge. Thereafter, the dose of anti-CD4 was reduced to 200 µg per mouse and anti-CD4 was administered every 3–4 days. For NK-cell depletion, mice were given 200 µg of anti-NK1.1 (clone PK136) 2 days and 1 day before bilateral tumor inoculation. Mice were then injected with 200 µg of anti-NK1.1 per mouse every 3–4 days. All intraperitoneal injections were prepared in PBS in a final volume of 400 µl per mouse. Depletion of all the three cell lineages from the peripheral blood of recipients was verified by flow cytometry in seven animals per depletion group (data not shown). Statistical analysis. Student’s t-tests were used to perform pairwise com-

parisons. Differences in means were considered statistically significant at P < 0.05.

Acknowledgments This work was supported by the Prostate Cancer Research Foundation of Canada. We thank Thomas Calzascia (Ontario, Cancer Institute, Toronto, Canada) and David Bateman (Department of Material Biophysics, University of Toronto, Canada) for helpful discussions.

References

1. Vari, F and Hart, DN (2004). Loading DCs with Ag. Cytotherapy 6: 111–121. 2. Banchereau, J, Ueno, H, Dhodapkar, M, Connolly, J, Finholt, JP, Klechevsky, E et al. (2005). Immune and clinical outcomes in patients with stage IV melanoma vaccinated with peptide-pulsed dendritic cells derived from CD34+ progenitors and activated with type I interferon. J Immunother 28: 505–516. 3. Lee, WC, Wang, HC, Hung, CF, Huang, PF, Lia, CR and Chen, MF (2005). Vaccination of advanced hepatocellular carcinoma patients with tumor lysate-pulsed dendritic cells: a clinical trial. J Immunother 28: 496–504. 4. Lesterhuis, WJ, de Vries, IJ, Schuurhuis, DH, Boullart, AC, Jacobs, JF, de Boer, AJ et al. (2006). Vaccination of colorectal cancer patients with CEA-loaded dendritic cells: antigen-specific T cell responses in DTH skin tests. Ann Oncol 17: 974–980. 5. Liu, KJ, Wang, CC, Chen, LT, Cheng, AL, Lin, DT, Wu, YC et al. (2004). Generation of carcinoembryonic antigen (CEA)-specific T-cell responses in HLA-A*0201 and HLA-A*2402 late-stage colorectal cancer patients after vaccination with dendritic cells loaded with CEA peptides. Clin Cancer Res 10: 2645–2651. 6. Boczkowski, D, Nair, SK, Nam, JH, Lyerly, HK and Gilboa, E (2000). Induction of tumor immunity and cytotoxic T lymphocyte responses using dendritic cells transfected with messenger RNA amplified from tumor cells. Cancer Res 60: 1028–1034. 7. Boczkowski, D, Nair, SK, Snyder, D and Gilboa, E (1996). Dendritic cells pulsed with RNA are potent antigen-presenting cells in vitro and in vivo. J Exp Med 184: 465–472. 8. Nair, SK, Boczkowski, D, Morse, M, Cumming, RI, Lyerly, HK and Gilboa, E (1998). Induction of primary carcinoembryonic antigen (CEA)-specific cytotoxic T lymphocytes in vitro using human dendritic cells transfected with RNA. Nat Biotechnol 16: 364–369. 9. Zhou, Y, Bosch, ML and Salgaller, ML (2002). Current methods for loading dendritic cells with tumor antigen for the induction of antitumor immunity. J Immunother 25: 289–303. 10. He, Y, Zhang, J, Mi, Z, Robbins, P and Falo, LD, Jr. (2005). Immunization with lentiviral vector-transduced dendritic cells induces strong and long-lasting T cell responses and therapeutic immunity. J Immunol 174: 3808–3817.



11. He, Y, Zhang, J, Donahue, C and Falo, LD, Jr. (2006). Skin-derived dendritic cells induce potent CD8+ T cell immunity in recombinant lentivector-mediated genetic immunization. Immunity 24: 643–656. 12. Firat, H, Zennou, V, Garcia-Pons, F, Ginhoux, F, Cochet, M, Danos, O et al. (2002). Use of a lentiviral flap vector for induction of CTL immunity against melanoma. Perspectives for immunotherapy. J Gene Med 4: 38–45. 13. Breckpot, K, Dullaers, M, Bonehill, A, van Meirvenne, S, Heirman, C, de Greef, C et al. (2003). Lentivirally transduced dendritic cells as a tool for cancer immunotherapy. J Gene Med 5: 654–667. 14. Metharom, P, Ellem, KA, Schmidt, C and Wei, MQ (2001). Lentiviral vector-mediated tyrosinase-related protein 2 gene transfer to dendritic cells for the therapy of melanoma. Hum Gene Ther 12: 2203–2213. 15. Mossoba, ME and Medin, JA (2006). Cancer immunotherapy using virally-transduced DCs: a review of animal studies and human clinical trials. Expert Rev Vaccines 5: 717–732. 16. Osman, I, Scher, HI, Drobnjak, M, Verbel, D, Morris, M, Agus, D et al. (2001). HER-2/neu (p185neu) protein expression in the natural or treated history of prostate cancer. Clin Cancer Res 7: 2643–2647. 17. Cobleigh, MA, Vogel, CL, Tripathy, D, Robert, NJ, Scholl, S, Fehrenbacher, L et al. (1999). Multinational study of the efficacy and safety of humanized anti-HER2 monoclonal antibody in women who have HER2-overexpressing metastatic breast cancer that has progressed after chemotherapy for metastatic disease. J Clin Oncol 17: 2639–2648. 18. Hirsch, FR, Franklin, WA, Veve, R, Varella-Garcia, M and Bunn, PA, Jr. (2002). HER2/neu expression in malignant lung tumors. Semin Oncol 29: 51–58. 19. Hellstrom, I, Goodman, G, Pullman, J, Yang, Y and Hellstrom, KE (2001). Overexpression of HER-2 in ovarian carcinomas. Cancer Res 61: 2420–2423. 20. Scott, GK, Robles, R, Park, JW, Montgomery, PA, Daniel, J, Holmes, WE et al. (1993). A truncated intracellular HER2/neu receptor produced by alternative RNA processing affects growth of human carcinoma cells. Mol Cell Biol 13: 2247–2257. 21. Yoshimitsu, M, Sato, T, Tao, K, Walia, JS, Rasaiah, VI, Sleep, GT et al. (2004). Bioluminescent imaging of a marking transgene and correction of Fabry mice by neonatal injection of recombinant lentiviral vectors. Proc Natl Acad Sci USA 101: 16909–16914. 22. Chhikara, M, Huang, H, Vlachaki, MT, Zhu, X, Teh, B, Chiu, KJ et al. (2001). Enhanced therapeutic effect of HSV-tk+GCV gene therapy and ionizing radiation for prostate cancer. Mol Ther 3: 536–542. 23. Medin, JA, Liang, SB, Hou, JW, Kelley, LS, Peace, DJ and Fowler, DH (2005). Efficient transfer of PSA and PSMA cDNAs into DCs generates antibody and T cell antitumor responses in vivo. Cancer Gene Ther 12: 540–551. 24. Dullaers, M, Breckpot, K, Van Meirvenne, S, Bonehill, A, Tuyaerts, S, Michiels, A et al. (2004). Side-by-side comparison of lentivirally transduced and mRNA-electroporated dendritic cells: implications for cancer immunotherapy protocols. Mol Ther 10: 768–779. 25. Worsham, DN, Schuesler, T, von Kalle, C and Pan, D (2006). In vivo gene transfer into adult stem cells in unconditioned mice by in situ delivery of a lentiviral vector. Mol Ther 14: 514–524. 26. Will, E, Bailey, J, Schuesler, T, Modlich, U, Balcik, B, Burzynski, B et al. (2007). Importance of murine study design for testing toxicity of retroviral vectors in support of phase I trials. Mol Ther 15: 782–791. 27. Tarantal, AF, McDonald, RJ, Jimenez, DF, Lee, CC, O’Shea, CE, Leapley, AC et al. (2005). Intrapulmonary and intramyocardial gene transfer in rhesus monkeys (Macaca mulatta): safety and efficiency of HIV-1-derived lentiviral vectors for fetal gene delivery. Mol Ther 12: 87–98. 28. Levine, BL, Humeau, LM, Boyer, J, MacGregor, RR, Rebello, T, Lu, X et al. (2006). Gene transfer in humans using a conditionally replicating lentiviral vector. Proc Natl Acad Sci USA 103: 17372–17377. 29. Chan, T, Sami, A, El-Gayed, A, Guo, X and Xiang, J (2006). HER-2/neu-gene engineered dendritic cell vaccine stimulates stronger HER-2/neu-specific immune responses compared to DNA vaccination. Gene Ther 13: 1391–1402. 30. Chen, Z, Huang, H, Chang, T, Carlsen, S, Saxena, A, Marr, R et al. (2002). Enhanced HER-2/neu-specific antitumor immunity by cotransduction of mouse dendritic cells with two genes encoding HER-2/neu and α tumor necrosis factor. Cancer Gene Ther 9: 778–786.



31. Sakai, Y, Morrison, BJ, Burke, JD, Park, JM, Terabe, M, Janik, JE et al. (2004). Vaccination by genetically modified dendritic cells expressing a truncated neu oncogene prevents development of breast cancer in transgenic mice. Cancer Res 64: 8022–8028. 32. zum Büschenfelde, CM, Metzger, J, Hermann, C, Nicklisch, N, Peschel, C, Bernhard, H (2001). The generation of both T killer and Th cell clones specific for the tumor-associated antigen HER2 using retrovirally transduced dendritic cells. J Immunol 167: 1712–1719. 33. Garnache-Ottou, F, Chaperot, L, Biichle, S, Ferrand, C, Remy-Martin, JP, Deconinck, E et al. (2005). Expression of the myeloid-associated marker CD33 is not an exclusive factor for leukemic plasmacytoid dendritic cells. Blood 105: 1256–1264. 34. Nigro, LL, Sainati, L, Leszl, A, Mirabile, E, Spinelli, M, Consarino, C et al. (2007). Acute differentiated dendritic cell leukemia: a variant form of pediatric acute myeloid leukemia with MLL translocation. Leukemia 21: 360–362. 35. Reichard, KK, Burks, EJ, Foucar, MK, Wilson, CS, Viswanatha, DS, Hozier, JC et al. (2005). CD4+ CD56+ lineage-negative malignancies are rare tumors of plasmacytoid dendritic cells. Am J Surg Pathol 29: 1274–1283. 36. Rossi, JG, Felice, MS, Bernasconi, AR, Ribas, AE, Gallego, MS, Somardzic, AE et al. (2006). Acute leukemia of dendritic cell lineage in childhood: incidence, biological characteristics and outcome. Leuk Lymphoma 47: 715–725. 37. Garnache-Ottou, F, Feuillard, J and Saas, P (2007). Plasmacytoid dendritic cell leukaemia/lymphoma: towards a well defined entity? Br J Haematol 136: 539–548. 38. Bueno, C, Almeida, J, Lucio, P, Marco, J, Garcia, R, de Pablos, JM et al. (2004). Incidence and characteristics of CD4+/HLA DRhi dendritic cell malignancies. Haematologica 89: 58–69. 39. Sato, T, Neschadim, A, Konrad, M, Fowler, DH, Lavie, A and Medin, JA (2007). Engineered human tmpk/AZT as a novel enzyme/prodrug axis for suicide gene therapy. Mol Ther 15: 962–970. 40. Chinnasamy, N, Treisman, JS, Oaks, MK, Hanson, JP, Jr. and Chinnasamy, D (2005). Ex vivo generation of genetically modified dendritic cells for immunotherapy: implications of lymphocyte contamination. Gene Ther 12: 259–271. 41. Lutz, MB, Kukutsch, N, Ogilvie, AL, Rossner, S, Koch, F, Romani, N et al. (1999). An advanced culture method for generating large quantities of highly pure dendritic cells from mouse bone marrow. J Immunol Methods 223: 77–92. 42. Freeman, TC, Dixon, AK, Campbell, EA, Tait, TM, Richardson, PJ, Rice, KM et al. (1998). Expression mapping of mouse genes. MGI Direct Data Submission. 43. Morgan, DJ, Kreuwel, HT, Fleck, S, Levitsky, HI, Pardoll, DM and Sherman, LA (1998). Activation of low avidity CTL specific for a self epitope results in tumor rejection but not autoimmunity. J Immunol 160: 643–651. 44. Schreurs, MW, Eggert, AA, de Boer, AJ, Vissers, JL, van Hall, T, Offringa, R et al. (2000). Dendritic cells break tolerance and induce protective immunity against a melanocyte differentiation antigen in an autologous melanoma model. Cancer Res 60: 6995–7001. 45. Bronte, V, Apolloni, E, Ronca, R, Zamboni, P, Overwijk, WW, Surman, DR et al. (2000). Genetic vaccination with “self” tyrosinase-related protein 2 causes melanoma eradication but not vitiligo. Cancer Res 60: 253–258. 46. Siatskas, C, Underwood, J, Ramezani, A, Hawley, RG and Medin, JA (2005). Specific pharmacological dimerization of KDR in lentivirally transduced human hematopoietic cells activates anti-apoptotic and proliferative mechanisms. FASEB J 19: 1752–1754. 47. Naldini, L, Blomer, U, Gage, FH, Trono, D and Verma, IM (1996). Efficient transfer, integration, and sustained long-term expression of the transgene in adult rat brains injected with a lentiviral vector. Proc Natl Acad Sci USA 93: 11382–11388. 48. Takenaka, T, Qin, G, Brady, RO and Medin, JA (1999). Circulating α-galactosidase A derived from transduced bone marrow cells: relevance for corrective gene transfer for Fabry disease. Hum Gene Ther 10: 1931–1939. 49. Piechocki, MP, Pilon, SA and Wei, WZ (2002). Quantitative measurement of anti-ErbB-2 antibody by flow cytometry and ELISA. J Immunol Methods 259: 33–42. 50. Maecker, HT, Moon, J, Bhatia, S, Ghanekar, SA, Maino, VC, Payne, JK et al. (2005). Impact of cryopreservation on tetramer, cytokine flow cytometry, and ELISPOT. BMC Immunol 6: 17.

617