Insertion of the DNA for the 163–171 peptide of IL1β enables a DNA vaccine encoding p185neu to inhibit mammary carcinogenesis in Her-2/neu transgenic BALB/c mice

An assessment was made of the effectiveness of DNA vaccination in prevention of the mammary adenocarcinomas of BALB/c female mice transgenic for the activated rat Her-2/neu oncogene. Atypical hyperplasia is evident in their mammary glands when they are 6 weeks old and in situ carcinoma by the 13th week. Palpable invasive carcinomas appear around the 17th week and are always evident in all 10 glands by the 33rd week. Intramuscular vaccinations with 100 μg plasmid DNA encoding the extracellular domain of the Her-2/neu p185 (ECD) performed at the 6th, 12th, 18th and 24th week provided no significant protection, whereas those ECD plasmids in which the DNA coding for the immunomodulatory 163–171 (VQGEESNDK) nonapeptide of human IL1β (ECD-IL1βp) had been inserted both delayed carcinogenesis and reduced tumor multiplicity. This reduction was associated with a marked immune-inflammatory reaction and a conspicuous leukocyte infiltrate located in the stroma surrounding the hyperplastic mammary ductul-alveolar structures. It was also directly correlated with a high anti-p185neu antibody production and an immunoglobulin switch to IgG2a and IgA. No anti-p185neu cytotoxic response was found. No significant protection was obtained when the DNA coding for the non-active peptide 189–197 of IL1β was inserted.


Introduction
DNA vaccines are molecularly defined reagents that are easy to construct and elicit long-lasting cellular and humoral immune responses to a variety of antigens. Clinical trials have shown that they are nontoxic and well tolerated, though the responses their vaccination induce are low and vary from one individual to another, 1-3 while their efficacy is often limited by low levels of gene and protein expression and the complex requirements for protein presentation and lymphocyte activation. 4 Enhancement of the potency of DNA vaccines has been sought through the employment of costimulatory molecules and cytokines as adjuvants. 5,6 Vaccines encoding antigens fused with immunological molecules and cytokines elicit more effective responses 6,7 and the ability of cytokines to enhance the immune recognition of tumor antigens has been extensively exploited. 7 Cytokine-gene engineered tumor cells, 8 and DNA encoding fusion proteins between cytokines and tumor antigens 9 induce marked immune responses, even against poorly immunogenic tumors.
Interleukin 1 (IL-1) is a particularly effective adju-immunomodulatory 163-171 (VQGEESNDK) nonapeptide of human IL1␤ (ECD-IL1␤p) had been inserted both delayed carcinogenesis and reduced tumor multiplicity. This reduction was associated with a marked immune-inflammatory reaction and a conspicuous leukocyte infiltrate located in the stroma surrounding the hyperplastic mammary ductulalveolar structures. It was also directly correlated with a high anti-p185 neu antibody production and an immunoglobulin switch to IgG2a and IgA. No anti-p185 neu cytotoxic response was found. No significant protection was obtained when the DNA coding for the non-active peptide 189-197 of IL1␤ was inserted. Gene Therapy (2001) 8, 447-452.
vant, 10 but its potent pyrogenic and/or proinflammatory properties drastically limit its use. The nonapeptide sequence VQGEESNDK corresponding to the amino acid stretch between the positions 163-171 of human IL-1␤, on the other hand, is free from these properties and retains the immunostimulatory capability of the entire molecule. 11,12 Its local administration, in fact, markedly increase the immunogenicity of poorly immunogenic tumors in syngeneic mice. 12 Insertion of the DNA sequence encoding this nonapeptide in recombinant antigens enhances their immunogencity. 13 DNA vaccination with plasmids encoding a fusion protein between idiotypic determinants of B cell lymphomas and this peptide induced a protective immune response against a subsequent lymphoma challenge. 14 In this paper, we compare the ability of DNA vaccination with plasmids coding for the extracellular domain of product of rat Her-2/neu (p185 neu ) alone (ECD) or fused with the DNA coding for this IL-1␤ peptide (ECD-IL1␤p) to block the progression of Her-2/neu carcinogenesis in female BALB/c mice transgenic for the activated rat Her-2/neu oncogene under the control of the MMTV promoter (BALB-neuT). 15 All the mammary glands of these mice independently undergo a very aggressive carcinogenesis that mirrors some features of the formation of lobular carcinoma in women. 16 Vaccination with plasmids coding for ECD alone did not block this carcinogenesis, whereas vaccination with ECD-IL-1␤p was followed by a significant delay.

Results
Immunization of BALB-neuT mice Owing to expression of the activated rat Her-2/neu gene under the control of the MMTV promoter, high membrane overexpression of rat p185 neu is already evident in the terminal ductular-lobular units of all the mammary glands of BALB-neuT female at the 3rd week of age. Atypical mammary hyperplasia is evident between week 3 and 6. Palpable invasive carcinomas appear around the 17th week in one or two glands of 40% of mice and are always evident in all 10 glands by the 33rd week. 15,16 To assess the ability of DNA vaccination to hamper this progression, mice were immunized at the 6th, 12th, 18th and 24th week with plasmids coding for the ECD alone, ECD and IL-1␤p (ECD-IL␤p), or ECD and the non-active peptide 189-197 of IL-1␤ (ECD-IL1␤na). Vaccination with these plasmids did not hamper tumor growth, whereas immunization with ECD-IL1␤p both delayed carcinogenesis and reduced tumor multiplicity ( Figure 1). The mean number of mammary glands with palpable tumors at this time was six only in the ECD-IL1␤p group, whereas all mice in the other groups displayed a palpable tumor in all their mammary glands ( Figure 1, lower panel). More- over, one of the 10 mice in the ECD-IL1␤p group was still completely tumor free at week 33 ( Figure 1, upper panel).

Pathological analysis of mammary glands
Pathological observations performed at 7 weeks, 1 week after the first immunization showed that both mice injected with saline and immunized with ECD-IL1␤p displayed foci of epithelial hyperplasia of the terminal ductular-lobular units (TDLU) ( Figure 2). However, in immunized mice the TDLU were surrounded by a reactive leukocyte infiltrate (Figure 2b). At 13 weeks, 1 week after the second immunization mice injected with saline displayed a well-developed atypical epithelial hyperplasia. Epithelial cells were atypical and their growth inside the lumens distended and expanded the alveoli and lobules (Figure 2c). By contrast, mice immunized twice with ECD-IL␤p showed reduced hyperplasia and infiltrating reactive cells in the surrounding fibrotic stroma (Figure 2d). At 25 weeks, 1 week after the third immunization, a well-developed invasive lobular carcinoma was present in most mammary glands from control mice (Figure 2e), while only several foci of atypical hyperplasia and in situ carcinoma were found in some mammary glands of immunized mice ( Figure 2f). These foci were bordered by a dense stroma markedly infiltrated by reactive leukocytes.
Cytotoxic response to p185 neu positive target cells The infiltrate and inhibition of the progression of carcinogenesis did not correlate with the induction of a detectable cytotoxic response in Spc collected 7 days after each vaccination and tested immediately or after 6 days in vitro restimulation with distinct p185 neu + target cells, as evaluated in 4 and 18 h 51 Cr release assays and 48 and 72 h [ 3 H]TdR release assays (data not shown).
Antibody response associated with the inhibition of natural carcinogenesis The ability of these treatments to induce anti-p185 neu antibodies was evaluated in the sera collected at 33 weeks when all the mice vaccinated with saline, ECD or ECD-IL1␤na displayed 10 large tumors. The antibody titer in mice with 10 large tumors was similar, irrespective of the vaccination with ECD, ECD-IL1␤na or ECD-IL1␤p. By contrast, the titer was much higher in the sera from mice vaccinated with ECD-IL1␤p with only 1-4 tumors ( Figure 3).
The distribution of immunoglobulin isotypes was also evaluated in these sera. IgM and IgG3 were increased in all immunized mice ( Figure 4). In addition, mice with 1-4 tumors after ECD-IL1␤p immunization presented an increase of several isotypes, especially IgG2a and IgA.

Discussion
Present data show that insertion of IL-1␤p DNA dramatically increases the protective efficacy of vaccination with plasmids coding for ECD. Only mice immunized with ECD-IL1␤p plasmid display a delay in the appearance of the first tumor and a strong decrease of the number of mammary glands with a palpable carcinoma, whereas the natural consequence of activated rat Her-2/neu gene overexpression is that a large, fast-growing lobular carcinoma is palpable at week 33 in all 10 glands of control BALB-neuT mice injected with saline only. Neither vaccination with ECD nor with ECD-IL1␤na plasmids significantly counteracts this aggressive carcinogenesis, whereas in a less aggressive model of Her-2/neu carcinogenesis, vaccination with the same ECD vector used in the present study provided a significant protection 17 that was further increased by coinjection of the ECD plasmid with a bicistronic vector coding for IL-12. 18 This critical immunomodulatory role played by IL1␤p fits in well with our previous experience using this peptide as a systemic adjuvant. 12 Moreover, its sustained local presence in the tumor area activates an effective antitumor reaction against a poorly immunogenic transplantable mammary tumor. 12 Insertion of IL1␤p augments the antitumor immune response induced by protein and DNA vaccines. 14 Moreover, vaccination with plasmids containing IL1␤p DNA sequence 13,14 or protein antigen fused with IL1␤p 13 increases the immunogenicity of many antigens.
In BALB-neuT mice, rat p185 neu is a self-protein already overexpressed by the mammary gland by the 3rd week of life. 16 Vaccination with ECD is unable to break this natural tolerance. Insertion of IL1␤p DNA in the construct inhibits Her-2/neu carcinogenesis and is accompanied by lymphocyte infiltration of the stroma surrounding the TDLU and induction of anti-p185 neu antibodies. By contrast, no significant CTL response was found, despite all the in vitro restimulation attempts. This provocative finding fits in well with the absence of cytotoxicity we have found following vaccination of BALB/c and BALB-neuT mice against plasmids encoding both the ECD and TM portions of the p185 neu . 19 Even the presence of IL1␤p signal is not enough to activate T killer cells against p185 neu .
The distribution of anti-p185 neu antibodies in the treatment groups also raises some puzzling issues. Only ECD-IL1␤p immunized mice with evident inhibition of Her-2/neu carcinogenesis display a high anti-p185 neu antibody titer. While this may suggest a direct correlation between antibody titer and inhibition, the low titer found in mice with large tumors may be due to antibody absorption by the p185 neu + tumor cells or immunosuppression by such large tumors.
An issue not directly addressed in this paper but raised by the significant evidence in the literature is whether anti-p185 neu antibodies induce a functional block of p185 neu receptor function, 20 down-regulate its expression on the cell surface, 20,21 impede the formation of p185 neu homo-or heterodimers that spontaneously transduce proliferative signals to the cells, 21,22 and block its ability to bind ligands. 23 These antibodies also significantly suppress the growth of transplantable p185 neu + tumors, 24,25 the natural onset of mammary carcinomas in Her-2/neu transgenic mice, 21 and delay tumor growth in patients with Her-2/neu positive tumors. 26 A reduced r-p185 neu expression could be sufficient to reverse their transformed phenotype. 20,21 The marked increase in serum IgG2a and IgA in ECD-IL1␤p-immunized mice suggests a more finely divided reaction scenario wherein leukocytes infiltrating the tumor site may play an important effector role. [27][28][29] These antibody isotypes activate PMN and other cells to mediate antibody dependent cellular cytotoxicity (ADCC) 25,30 and complement-dependent cytotoxicity (IgG2a), and inhibit the growth of p185 neu + tumor in vivo. 25 IgG and IgA may synergistically promote ADCC by PMN and other leukocytes 25,31 that massively infiltrate hyperplastic lesions. It is also possible that the concentration of secretory IgA may reach particularly high levels in the mammary gland, where their inhibitory activity is required.

Material and methods
Mice Inbred BALB-neuT mice overexpressing the transforming rat Her-2/neu oncogene (neuT + /neuT − ) driven by the mouse mammary tumor virus promoter were produced and screened for the presence of the transgene as previously described in detail. 15 Groups of individually tagged virgin BALB-neuT females bred under specific pathogen-free conditions by Charles River, Calco, Italy were treated in accordance with European Union and institutional guidelines. Since all 10 mammary glands of BALB-neuT females naturally undergo carcinogenic transformation with a definite progression, 15 these were inspected weekly, and tumor masses were measured with calipers in the two perpendicular diameters. Progressively growing masses of Ͼ3 mm in mean diameter were regarded as tumors. Growth was monitored until all mammary glands displayed a palpable tumor or until a tumor exceeded an average diameter of 10 mm, when mice were killed for humane reasons. Except where otherwise specified, surviving BALB-neuT mice were killed at 33 weeks. 15 As some immunized mice do not display carcinomas in all mammary glands, the mean number of palpable mammary carcinomas per mouse was calculated as cumulative number of incident tumors per total number of BALB-neuT mice.

DNA expression vectors and vaccination
The pCMV vector was derived from the pcDNA3 plasmid (Invitrogen, San Diego, CA, USA) by deleting the SV40 promoter, neomycin resistance gene and SV40 polyA. The sequence for the extracellular domain of transforming mutated rat p185 neu was generated from the PCR product using the primers 3′-CGCAAGCTTCAT-CATGGAGCTGGC-5′ and 3′-CGGAATTCGGGCTGG CTCTCTGCTC-5′ and the primers 3′-CGCAAGCTTCAT GGAGCTGGC-5′ and 3′-ATGAATTCTTTCCGCATCG TGTACTTCTTCCGG-5′, respectively, as previously described. 17 PCR products of the expected size were isolated by agarose gel electrophoresis, digested with Hin-dIII and EcoRI and cloned into the multiple cloning site of the pCMV plasmid in order to obtain the ECD plasmid used in this work. The immunomodulatory IL1␤p nonapeptide WQGEESNDK corresponding to amino acids 163-171 and the control non-active IL1␤na nonapeptide EGTEKDQVS corresponding to amino acids 189-197 of the human IL-1␤ were cloned in-frame into EcoRI and XbaI sites obtained by incorporating two complementary and overlapping phosphorylated oligonucleotides encoding each of the two peptides with EcoRI and XbaI sites: 5′-AATTCGGTTCAGGGTGAAGAAAGTAACGA TAAATAAT-3′ (IL1␤p-Forw) and 5′CTAGATTATTT ATCGTTACTTTC TTCACCCTGAACCG-3′ (IL1␤p-Rew); 5′-AATTCGGAAGGTACCGAAAAAGATCAGGTTAGTT AAT-3′ (IL1␤na-Forw) and 5′-CTAGATTAACTA ACCTGATCTTTTTCGGTACCTTCCG-3′ (IL1␤na-Rew). Escherichia coli strain DH5 was transformed with ECD, ECD-IL1␤p and ECD-IL1␤na plasmids and then grown in Luria-Bertani medium (Sigma, St Louis, MO, USA) as previously described. 17 Large-scale preparation of the plasmids was carried out by alkaline lysis using Endofree Qiagen Plasmid-Giga kits (Qiagen, Chatsworth, CA, USA). DNA was then precipitated, suspended in sterile saline at the concentration of 1 mg/ml and stored in aliquots at −20°C for subsequent use in immunization protocols. Plasmids were injected (100 g per injection) into the quadriceps muscle through a 28-gauge needle syringe. BALB-neuT mice were immunized at the 6th, 12th, 18th and 24th week.

Morphologic analysis
Groups of three BALB-neuT mice were killed at the indicated times. For histologic evaluation, tissue samples were fixed in 10% neutral buffered formalin, embedded in paraffin, sectioned at 4 m and stained with hematoxylin and eosin or Giemsa.
Cell lines TUBO and N202.1 cells are cloned cell lines established in vitro from lobular carcinomas that arose spontaneously in a BALB-neuT and in an FVB-neuN transgenic mouse, respectively. 19 N202.1A and N202.1E are two clones randomly derived from N202.1. 32 Cytofluorimetric analysis indicates that N202.1A expresses high levels of p185 neu , whereas this expression is not detectable in N202.1E. 32 Both clones were cultured in DMEM (BioWhittaker Europe, Verviers, Belgium) supplemented with 20% FBS (Life Technologies, San Giuliano Milanese, Italy).

Cytotoxicity assays
The cytotoxicity of lymphocytes from the mice in each group was tested immediately or after in vitro restimulation. Lymphocytes (1 × 10 7 ) were stimulated for 6 days with 5 × 10 5 irradiated TUBO cells as previously described. 33 To get better stimulation, this basic design was variously changed in the several repeats of the test. Other rat Her-2/neu expressing BALB/c target cells were also used as stimulator and target cells. Moreover, the suppressor activity of stimulator rat Her-2/neu BALB/c cells was ruled out by adding progressive numbers of third-party TUBO cells in mixed lymphocyte and allogeneic target cell interactions as previously described. 34 Cytotoxicity of fresh and restimulated lymphocytes was assayed in 4 and 18 h 51 Cr release assays, 48 and 72 h [ 3 H]TdR release assays as previously described in detail. 33,34 In all these tests both TUBO cells and other rat Her-2/neu expressing BALB/c target cells were highly lysable by allogeneic cytotoxic T lymphocytes.
Cytofluorimetric evaluation of anti r-p185 neu antibodies Sera of six BALB-neuT mice immunized with ECD, ECD-IL1␤p and ECD-IL1␤na were collected at 33 weeks when they all display 10 palpable tumors and pooled. The sera of ECD-IL1␤p immunized mice displaying only 1-4 palpable tumors at 33 weeks was separately collected. The control sera were a pool from six mice injected with saline only. The ability of sera to bind r-p185 neu was evaluated by flow cytometry. 2 × 10 5 N02.1A or N202.1E cells from in vitro cultures, washed twice with cold PBS supplemented with 2% BSA and 0.05% sodium azide, were stained in a standard indirect immunofluorescence procedure with 50 l of 1:10 dilution in PBS-azide-BSA of control or immune sera. A fluorescein-conjugated rabbit anti-mouse Ig (Dako, Glostrup, Denmark) was used as second-step Ab. The cells were resuspended in PBSazide-BSA containing 1 mg/ml of propidium iodide to gate out dead cells, and evaluated in a FACScan (Becton Dickinson). The specific N202.1A binding potential (Sbp) of the sera was calculated as follows: [(% positive cells with test serum) (fluorescence mean)] − [(% positive cells with control serum) (fluorescence mean)] X serum dilution, as previoulsy described in detail. 34 5 × 10 3 viable cells were analyzed in each evaluation.

Serum concentration of Ig isotypes
The concentration of IgA, IgM, IgG1, IgG2a, IgG2b and IgG3a isotypes in pool of mice injected four times with saline only or immunized with ECD, ECD-IL1␤p or ECD-IL1␤na was determined by the radial immunodiffusion test (The Binding Site, Birmingham, UK).

Statistical analysis
Differences in tumor incidence were evaluated by the Mantel-Haenszel log-rank test, those in tumor/mouse numbers by Wilcoxon's rank sum test and those in the number of tumor infiltrating cells by Student's t test.