Freund's vaccine adjuvant promotes Her2/Neu breast cancer

All animal study protocols were approved by the University of Wisconsin Institutional Animal Care and Use Committee (IACUC). Virgin Wistar-Furth (WF) rats were obtained from Harlan Sprague-Dawley (Indianapolis, IN), housed in our facility on a 12 hour light/dark cycle and given free access to standard lab chow. For all studies, euthanasia was carried out using CO₂ inhalation. Rats were injected subcutaneously in the dorsal region with Freund's complete adjuvant (CFA) or Freund's incomplete adjuvant (IFA) (Sigma-Aldrich) suspended in sterile 0.9% PBS. Three treatment protocols were used: early schedule, late schedule and single dose. Rats on the early schedule protocol received CFA (0.5 ml/kg body weight) at 5 weeks of age and IFA (0.5 ml/kg body weight) at 7, 9 and 13 weeks of age. Rats on the late schedule were injected with CFA at 10 weeks of age and IFA at 12 and 14 weeks. All control groups received an equivalent volume of saline. In the single dose model, CFA or saline was injected into 10 week old rats at a concentration of 0.5 ml/kg body weight, and rats were euthanized 1 (n = 5/group) or 15 days (n = 6/group) after injection.

Virgin WF female rats were injected with adjuvant using the early (n = 15) and late (n = 15) treatment protocols, as described above. Controls (n = 14) received saline using the early schedule. Mammary carcinomas were induced by intraductal infusion of the pJRneu vector into the mammary epithelium (12 glands/rat) of 8 week old rats, using a viral titer of 1 × 10⁵ CFU/ml. The generation and mammary intraductal infusion of the pJRneu retroviral vector has been previously described [9]. Rats were palpated weekly for the presence of mammary carcinomas starting at 9 weeks of age and removed from the study at 17 weeks of age (9 weeks post-infusion). Mammary carcinoma multiplicity was recorded for each rat by counting the mammary carcinomas greater than 3 mm in diameter.

To characterize hyperplastic mammary lesions in FA-treated rats, the fourth mammary glands of rats on the early treatment protocol (n = 20–21/group) were infused with the pJRneu vector at a viral titer (1 × 10⁷ CFU/ml). This higher titer was previously determined to provide sufficient statistical power, while minimizing the number of required animals [10]. Rats were removed from the study 15 days post-infusion, and one of the infused mammary glands per rat was whole mounted and stained with aluminum carmine as previously described [11]. Hyperplastic mammary lesions were identified from the whole mounts by their size and counted using 20× light microscopy to determine lesion multiplicity. Each stained mammary gland was photographed using a SONY Cybershot digital camera using 1.3× optical zoom. The area of each hyperplastic lesion, mammary gland and lymph nodes was determined from digital photographs using ImageJ (Scion Corp.), available from the NIH website http://​rsb.​info.​nih.​gov/​ij/​. Two methods were used for statistical comparison of area measurements from the hyperplastic mammary lesions. First, the average lesion size was determined for each treatment (n = 179 lesions from controls and 258 from adjuvant-treated rats). Secondly, the total lesion area was determined for each mammary gland (n = 20–21/group) and normalized to its respective mammary gland area. The normalized values were then combined for each treatment group for statistical analysis. The contralateral mammary glands (n = 10/group) were fixed in 10% neutral buffered formalin and stained with hematoxylin and eosin for histopathological analysis.

Acute phase serum haptoglobin levels were measured in rats injected with CFA or saline using the single dose protocol, as described above. Following euthanasia, blood was collected from the aorta using a syringe and an 18 gauge needle and allowed to clot. Serum was collected after centrifugation, aliquotted and stored at -80°C. Haptoglobin levels were determined using the Tri-Delta Diagnostics phase haptoglobin colorimetric assay, according to the manufacturers' instructions. All samples, including standards, were run in triplicate.

Mammary epithelial cell proliferation was evaluated by immunohistochemistry for Ki67 in non-infused mammary glands from rats on the early adjuvant schedule, and from hyperplastic mammary lesions in virally infused rats given early treatment of either adjuvant or saline. All rats were sacrificed at 10 weeks of age. All reagents for Ki67 staining were obtained from Vector Laboratories unless otherwise stated. Sections were immersed in antigen retrieval solution at 95°C for 20 minutes following deparaffinization in xylene and rehydration in decreasing concentrations of ethanol. Endogenous peroxide activity was inhibited by incubation with 3% H₂0₂, and nonspecific antibody binding was prevented by treating with 10% normal goat serum and avidin/biotin blocking solutions. Sections were incubated at 4°C overnight with a 1/500 dilution of mouse monoclonal anti-Ki67 antibody. After washing with PBS, sections were then incubated with a 1/200 dilution of biotinylated anti-mouse IgG. Detection of Ki67 antibody binding was done using a Vectastain ABC kit with DAB chromogen (BD Biosciences Pharmingen).

Apoptosis was assessed in hyperplastic mammary lesions by labeling of 5' ends of fragmented DNA with terminal deoxynucleotidyl transferase (TdT) enzyme using a TdT-FragEL kit, following the manufacturers' instructions (Oncogene/Invitrogen). The fraction of apoptotic epithelial cells in hyperplastic mammary lesions was determined as described for Ki67.

All statistical analyses were done using StatView (SAS, Cary, NC). Mammary carcinoma multiplicity was tested using One-Way ANOVA with Bonferroni-Dunn multiple comparison. Serum haptoglobin, cell proliferation/apoptosis immunohistochemistry and hyperplastic mammary lesion multiplicity were analyzed by unpaired t test and nonparametric (Mann Whitney) testing, where appropriate. Statistical analysis of serum haptoglobin values was done separately for the two time points (1 day and 15 days), since each had its own control group. Area measurements (mammary gland, lymph node and hyperplastic mammary lesions) were analyzed by unpaired t test both as raw data and following log transformation to more closely approximate a normal distribution. Since the conclusions were the same for raw and log transformed data, we present the results for the raw data.

Direct transfer of the neu oncogene into the mammary ductal cells results in clonal development of mammary carcinomas (Figure 1A and 1B) [12, 13]. This model thus provides a unique system to assess the effects of chronic inflammation on breast cancer development. Treatment of rats with adjuvant using either the early or late schedule resulted in significantly increased mammary tumor multiplicity following infusion with neu compared with saline controls (Figure 1C). At 9 weeks after ductal infusion, adjuvant-treated rats had approximately twice the number of carcinomas than did the controls (P < 0.05). There were no histopathological differences between mammary carcinomas arising in the various treatment groups.

To assess for evidence of chronic systemic inflammation, we measured serum haptoglobin concentrations [8]. Haptoglobin was elevated (~4-fold) 24 hours after a single injection of CFA and was approximately 2-fold higher than controls 15 days after injection (P < 0.01) (Figure 2).

Since a systemic inflammatory response was present, we hypothesized that the products of inflammation, such as cytokines and growth factors, may affect proliferation in the mammary epithelium. In adjuvant-treated rats on the early schedule protocol without retroviral infusion, staining for Ki67 showed a modest but statistically significant increase in proliferation in the mammary gland epithelium compared to controls (P < 0.05) (Table 1). This was true for both epithelial cells in terminal end buds and for ductal/alveolar structures. Interestingly, we did not observe an increase in inflammatory cells within the mammary parenchyma of adjuvant-only treated rats.

We characterized the proliferative effects of Freund's adjuvant (early schedule treatment) on hyperplastic mammary lesions by analyzing mammary glands 15 days after neu infusion. This time point precedes the formation of palpable tumors, which occurs 3–4 weeks post-infusion. A histological section of a typical hyperplastic mammary lesion is shown in Figure 3A. Lesion multiplicity was higher in rats given adjuvant than in saline controls (12.3 versus 8.9 respectively) but did not differ statistically (P = 0.13, unpaired t test).

The average size of the hyperplastic mammary lesions determined from area measurements was significantly higher (1.6-fold) in adjuvant-treated rats when compared to controls (P < 0.001) (Figure 3B). In addition, the total lesion area per mammary gland (normalized to mammary gland area) was higher (~2-fold) in adjuvant-treated rats versus controls (P < 0.01) (Figure 3C). The larger lesion size led us to evaluate cell proliferation and apoptosis using immunohistochemistry. The epithelial cells within the hyperplastic mammary lesions were significantly more proliferative in adjuvant-treated rats, as determined by the percentage of Ki67+ cells, than in controls (P < 0.001) (Table 1). The apoptotic fraction of cells in the hyperplastic mammary lesions was low, and did not differ significantly between saline (1.2%) and adjuvant (1.9%) groups (P = 0.31, Mann-Whitney Rank test).

In addition to adjuvant-induced proliferation, we also observed localized lymphadenopathy 15 days after neu infusion in adjuvant-treated rats. The average area of lymph nodes residing in the fourth mammary gland was significantly higher (1.6-fold) in adjuvant-treated rats, as compared to controls (P < 0.001) (Figure 3D).