An N-butyl-N-(4-hydroxybutyl)nitrosamine-induced bladder cancer model was developed in human mucin 1 (MUC1) transgenic mice for the purpose of testing MUC1-directed immunotherapy. After administering a MUC1-targeted peptide vaccine, a cytotoxic T lymphocyte response to MUC1 was confirmed by measuring serum cytokine levels and T-cell specific activity.
A preclinical model of invasive bladder cancer was developed in human mucin 1 (MUC1) transgenic (MUC1.Tg) mice for the purpose of evaluating immunotherapy and/or cytotoxic chemotherapy. To induce bladder cancer, C57BL/6 mice (MUC1.Tg and wild type) were treated orally with the carcinogen N-butyl-N-(4-hydroxybutyl)nitrosamine (OH-BBN) at 3.0 mg/day, 5 days/week for 12 weeks. To assess the effects of OH-BBN on serum cytokine profile during tumor development, whole blood was collected via submandibular bleeds prior to treatment and every four weeks. In addition, a MUC1-targeted peptide vaccine and placebo were administered to groups of mice weekly for eight weeks. Multiplex fluorometric microbead immunoanalyses of serum cytokines during tumor development and following vaccination were performed. At termination, interferon gamma (IFN-γ)/interleukin-4 (IL-4) ELISpot analysis for MUC1 specific T-cell immune response and histopathological evaluations of tumor type and grade were performed. The results showed that: (1) the incidence of bladder cancer in both MUC1.Tg and wild type mice was 67%; (2) transitional cell carcinomas (TCC) developed at a 2:1 ratio compared to squamous cell carcinomas (SCC); (3) inflammatory cytokines increased with time during tumor development; and (4) administration of the peptide vaccine induces a Th1-polarized serum cytokine profile and a MUC1 specific T-cell response. All tumors in MUC1.Tg mice were positive for MUC1 expression, and half of all tumors in MUC1.Tg and wild type mice were invasive. In conclusion, using a team approach through the coordination of the efforts of pharmacologists, immunologists, pathologists and molecular biologists, we have developed an immune intact transgenic mouse model of bladder cancer that expresses hMUC1.
Bladder cancer is the fourth most common form of cancer and the eighth leading cause of cancer deaths in American men. In the United States, an estimated 72,500 new cases and 15,000 deaths from bladder cancer are expected among men and women combined in 20131. The incidence of bladder cancer is approximately three times as high in men compared to women. In the United States, transitional cell carcinomas (TCC) account for over 90% of cases, while squamous cell carcinomas (SCC) have an incidence of less than 2%2. The overall relative 5-year survival rate for papillary TCC is 91.5% compared to only 30.9% for SCC2. Although noninvasive papillary TCCs account for approximately 75% of cases at the time of diagnosis, even with treatment more than 50% of patients will experience a recurrence within 5 years, with up to 30% of these patients progressing to muscle invasive disease3,4. Typical treatment regimens for non-muscle invasive disease include transurethral resection (TUR) followed by intravesical chemotherapy. In patients with high-grade Ta or T1 tumors, a repeat TUR may be performed prior to chemotherapy3,4. For those patients with low-grade Ta recurrences or high-grade Ta or T1 lesions, TUR followed by adjuvant chemotherapy or immunotherapy in the form of Bacillus Calmette-Guerin (BCG) may be used3,4. Intravesical BCG has been shown to be superior to intravesical mitomycin C with respect to time to recurrence5. For T2 muscle invasive disease, radical cystectomy with or without neoadjuvant chemotherapy is the recommended course of treatment3. In patients with SCC, radical cystectomy appears to be the most effective treatment6. Given the very high rates of recurrence despite the best treatments available, there is clearly a need for new, more effective therapies for bladder cancer.
Expanding new immunotherapies for bladder cancer is one possible approach that may hold promise for extending disease-free survival. Historically, BCG has been the only effective immunotherapy for bladder cancer. Its mechanism of action is thought to involve the nonspecific induction of a T-helper 1 (Th1) type immune response via increasing levels of interleukin-2 (IL-2) and interferon gamma (IFN-γ)4. Cellular, or Th1 immunity, is critical in cancer immunotherapy as humoral, or Th2, immunity has never been shown to be effective against solid tumors, with the exception of antibodies directed against growth factor receptors7. In an attempt to improve upon the benefits of BCG monotherapy, IFN-α 2B/BCG combination immunotherapy was evaluated in a phase II clinical trial with inconclusive results8. An alternative approach to immunotherapy for bladder cancer may be to target tumor-associated antigens (TAAs), the identification of which has made cancer immunotherapy more specific7.
One such TAA is mucin 1 (MUC1), which is a cell surface glycoprotein overexpressed in many epithelial cell cancers such as bladder, breast, lung, and pancreatic cancer9,10. The expression and modification of MUC1 is also substantially altered during carcinogenesis, such that underglycosylation exposes antigenic sequences of amino acids known as variable number of tandem repeats (VNTR) on the peptide core. While MUC1 is a self-molecule, these immunodominant VNTR regions are not normally exposed due to extensive glycosylation, and thus they are seen by the immune system as foreign11,12. Cytotoxic T-lymphocytes (CTLs) that specifically recognize MUC1 epitopes have been isolated from the tumor-draining lymph nodes of breast cancer patients13, as well as the blood and bone marrow of myeloma patients14,15, making MUC1 a potential target for a cellular immune response. The immunodominant VNTRs of the underglycosylated form of MUC1 are recognized by CTLs, resulting in the destruction of tumor cells16-19. Native cellular and/or humoral immune responses to cancerous MUC1 are, however, not strong enough to eliminate tumors. To augment the already existing weak immune response to MUC1, synthetic immunodominant peptides can be introduced through vaccination to generate a CTL response strong enough to be of clinical benefit18,20. A MUC1 liposomal vaccine has already been shown to increase survival in lung cancer patients21,22, generate CTLs capable of killing MUC1-positive tumor cells, and produce a Th1-polarized cytokine response23,24. With a high level of MUC1 expression9,11,25, bladder cancer is a logical candidate for testing MUC1-directed immunotherapy26,27. Furthermore, MUC1 has potential as a prognostic factor in bladder cancer28, MUC1 expression in TCC is significantly associated with stage and grade, and metastatic TCC has been shown to continue to express MUC129.
In order to evaluate the potential utility of MUC1-directed immunotherapy in bladder cancer, we developed an immune intact human MUC1 (hMUC1)-expressing transgenic (MUC1.Tg) mouse model of bladder cancer congenic on the C57BL/6 background30. Human MUC1 is expressed as a self-protein under the control of its own promoter, resulting in a tissue expression pattern consistent with that observed in humans30,31. The mice were induced with the known bladder carcinogen N-butyl-N-(4-hydroxybutyl)nitrosamine (OH-BBN)32, and then the resulting tumors were evaluated for hMUC1 expression and tumor type and grade. To assess the effect of the carcinogen on Th1/Th2 cytokine levels during tumor development, serum samples were collected periodically for multiplex analysis. Mice were then treated with a MUC1-targeted peptide vaccine, and the serum cytokine and immune responses were evaluated by multiplex fluorometric microbead immunoassay and ELISpot.
All animal studies and experiments were conducted under a protocol approved by the University of California, Davis Institutional Animal Care and Use Administrative Advisory Committee.
1. MUC1.Tg Mouse Breeding and Propagation
2. Study Design
3. Molecular Biology/Western
The following procedures were performed to verify the expression of MUC1 in mouse bladder tumor tissue using standard Western Blot protocol (data not shown).
4. Multiplex Fluorometric Microbead Immunoassay
5. IFN-γ/IL-4 ELISpot Preparation and Analysis
6. Immunohistochemistry (IHC) and Hematoxylin & Eosin (H&E) Staining
7. Statistical Methods
For the Multiplex Fluorometric Microbead Immunoassay, use a two-tailed Student's t-test to compare the average observed serum cytokine concentrations between the treatment and control groups. For ELISpot, use a one-way ANOVA to compare the spot forming colonies between the media control, scrambled peptide and peptide groups. Use Dunnett's Multiple Comparison Test to lessen the likelihood of a false positive result. A p-value of ≤0.05 is considered significantly different for all analyses.
The preclinical assessment of the effects of novel immunotherapies and combinations in bladder cancer requires the development of an appropriate animal model. In our transgenic mouse model, induction with the chemical carcinogen OH-BBN resulted in a high rate of bladder cancer incidence of predominantly TCC with some SCC, which is similar to bladder cancer in humans. To determine tumor histology, MUC1 expression status and the immune response to the peptide vaccine treatment, 21 MUC1.Tg and 18 wild type mice were euthanized for the collection of blood, bladders, and spleens (Figure 1) eight weeks following OH-BBN induction (Week 28). The bladder cancer incidence rate for both MUC1.Tg (14/21) and wild type (12/18) mice was 67%. Hematoxylin and Eosin (H&E) staining confirmed the presence of both TCC and SCC, with TCCs predominating at a 2:1 ratio. Among these, we observed a range of low and high-grade noninvasive to high-grade invasive tumors. All MUC1.Tg bladder cancer specimens were positive for MUC1 expression by IHC (Figure 2). It should be noted that the antibody used for MUC1 IHC recognizes both normal and cancerous human MUC1.
During model development, the serum levels of inflammatory cytokines were monitored serially between Weeks 8-28. We observed that inflammatory cytokine levels increased with time from induction through the end of the study (Figure 3). This cytokine pattern is very similar to what we observed previously in our lung cancer model33, which strongly suggests that increasing inflammatory cytokine levels may correlate with tumor development.
To assess the Th1 serum cytokine response to the peptide vaccine, 15 vaccinated and 14 placebo-treated MUC1.Tg mice were euthanized and blood was collected at the end of the study in Week 28, 24 hr after the last vaccine treatment. Multiplex analysis (Figure 4) shows increased Th1 serum cytokine levels of TNF-α, IFN-γ, IL-2, IL-12 (p70), and IL-17 in the vaccine group compared to the placebo group. Levels of TNF-α, IFN-γ, and IL-17 were significantly higher (p<0.05) in the vaccine-treated mice. These results suggest a Th1 polarized cytokine response to the peptide vaccine.
In order to evaluate the Th1/Th2 immune response to the peptide vaccine, splenocytes were assessed by IFN-γ/IL-4 ELISpot. Twenty-four hours after the last treatment, spleens were collected and processed to isolate lymphocytes for ELISpot analysis. Lymphocytes were counted and assessed for viability by Muse Analyzer (Figure 5). ELISpot plates were seeded with 1 x 106 viable cells per well and developed 48 hr later. Representative results (Figure 6) show a clear and specific IFN-γ response to the peptide, which confirms a Th1 immune response to the peptide vaccine.
Figure 1. Mouse Necropsy. Necropsy was performed at 28 weeks, 8 weeks after the end of OH-BBN induction. Liver, bladder tumor, and spleen are indicated. Asterisk (*) marks puncture point for blood collection. In this example, a high-grade, invasive SCC was observed. Click here to view larger image.
Figure 2. Representative bladder tissue sections stained with H&E (left) and human MUC1 IHC (right) of normal bladder, invasive squamous cell carcinoma, and invasive transitional cell carcinoma. (A) Normal urinary bladder with mucosa lined by transitional epithelium, which shows diffuse MUC1 reactivity. (B) Nests of invasive SCC (arrow) in submucosa. Organized keratin layers (asterisk) line the bladder mucosa. Diffuse MUC1 reactivity is seen in nests of SCC. (C) Mucosa contains TCC projecting into the lumen (left, at right). Transitional cell carcinoma is anaplastic with invasion into submucosa and muscle (arrow and inset). Mucosa and TCC projecting into lumen show diffuse MUC1 reactivity (right, at right), while invasive TCC has less prominent reactivity (right, at left). Bar= 200 μm (main panel) and 50 μm (inset). Click here to view larger image.
Figure 3. Inflammatory serum cytokines at different stages of tumor development. Serial serum specimens were collected by submandibular bleeds at baseline (8 weeks), then every 4 weeks thereafter until study termination. Blood was pooled (n=4), and the serum was isolated and analyzed for the presence of 20 cytokines. Concentrations represent the mean of pooled samples and bars represent the range. Arrows indicate the point at which OH-BBN dosing concluded. Click here to view larger image.
Figure 4. Th1 serum cytokines following peptide vaccine treatment. Serum samples were collected at study termination, 24 hr after the final dose of the vaccine (n=15) or placebo (n=14) and analyzed for the presence of 20 cytokines. Data is shown as mean cytokine concentrations and bars represent positive standard deviation. * p<0.05 Click here to view larger image.
Figure 5. Representative mouse splenocyte histogram. Mouse splenocytes were isolated at study termination and assessed for count and viability using a Muse Analyzer. Left panel, cell viablility based on cell size. Right panel, cell viability based on nucleated cells (live cells in green zone, dead cells in white zone). Click here to view larger image.
Figure 6. Splenocyte LISpot analysis at study termination. (A) Representative wells showing IFN-γ (red spots) and IL-4 (blue spots) production in response to media, scrambled peptide, and peptide. A clear IFN-γ, antigen-specific response was observed with peptide exposure. (B) Graphical representation of typical IFN-γ ELISpot data showing the mean (± standard deviation) spot forming colonies in response to media, scrambled peptide and peptide. Click here to view larger image.
The successful induction of invasive transitional and squamous cell bladder carcinoma in human MUC1.Tg mice offers a preclinical model for Immunotherapy development. Immunotherapeutic studies require the use of a spontaneous, immune intact model in order to evaluate the inflammatory response to tumor progression over time as well as the immune response to immunotherapy. In a spontaneous tumor development model, the tumor microenvironment remains intact and the tumors develop at a more representative growth rate that allows for the assessment of the antitumor effects of treatment. Moreover, the immune system can be measured and monitored through biomarkers, allowing for the evaluation of treatment efficacy.
Other tumor bearing mouse models described in the literature for testing immunotherapy include both xenograft models and transplant models. Although these models are convenient and have been extensively employed in cancer research, there are a number of important limitations to consider when conducting immunotherapeutic studies. Neither xenografts nor transplanted tumors develop spontaneously, and they proliferate in a microenvironment that is not representative of the tissue from which the tumors were originally derived. Furthermore, xenografts and transplanted tumors grow more rapidly than spontaneous tumors, allowing less time to study the immune effects of therapy. Most importantly, these models require hosts with compromised immune systems.
In addition to our model, there are a number of other chemically induced bladder cancer models. For example, N-[4-(5-nitro-2-furyl)-2-thiazolyl] formamide (FANFT) and N-methyl-N-nitrosourea (MNU) have also been shown to induce bladder cancer in both rats and mice. However, these chemicals are slightly different with respect to the histology of the tumors they induce. FANFT primarily induces urothelial cell carcinoma (UCC) with some SCC, while MNU induces papillary carcinoma that eventually results in muscle-invasive tumors with a low incidence of metastasis34. OH-BBN is commonly used for bladder cancer induction in rodent models because the TCCs that develop closely resemble high grade human TCCs34. Urinary bladder cancer has also been induced in canines, rabbits and rats as well as in mice using OH-BBN. Although canines share similar metabolic processes with humans in respect to bioactivation of carcinogens35, the latency period for bladder cancer development in beagles is 37 weeks36, and experimentations with dogs have both financial and ethical considerations. Rabbits have even longer latency periods, and when added to the dosage period, a minimum of 21 months is required for TCC and SCC development37. Similar to mice, rat models develop tumors that are histopathologically similar to humans, with short dosage periods of 8 weeks and latency periods of 5 weeks38. However, a human MUC1 transgenic rat does not currently exist. Therefore, the transgenic mouse model described here is currently the only suitable animal model for studying human MUC1-directed immunotherapy.
Previously we developed two immune intact human MUC1-expressing spontaneous tumor models for both lung33 and breast cancer39. In order to evaluate immunotherapies in bladder cancer, we developed an OH-BBN-induced, spontaneous mouse model of bladder carcinoma. Similar to the previously described models33,39, the tumors that develop in this model express the tumor-associated antigen MUC1 as a self-molecule, which is the target of the peptide vaccine. This model showed a 67% incidence of bladder tumors, all of which were positive for the expression of human MUC1. Histological assessments showed that TCCs predominated, which is consistent with what is observed in human bladder cancer. This model is ideal for studying carcinogenesis, prevention strategies and the treatment of localized and advanced bladder cancer in humans. In the future, we plan to pursue additional studies of immunotherapy in combination with chemotherapeutics and radiation therapy in the described bladder cancer model.
The authors have nothing to disclose.
The authors would like to thank the UC Davis Mouse Biology Program for breeding the mice. This research was supported by a grant from Merck KGaA, Darmstadt, Germany.
Reagent | |||
N-butyl-N-(4-hydroxybutyl)-nitrosamine (OH-BBN) | TCI America | B0938 | |
20 G Gavage Needles | Popper & Sons, Inc. | 7921 | Stainless steel |
Peptide Vaccine | N/A | N/A | investigational agent |
BD Microtainers | BD | 365957 | |
Tissue Cassettes | Simport | M490-12 | |
10% Neutral Buffered Formalin | Fisher Scientific | SF100-4 | |
Lysis Buffer | Pierce | 87787 | |
Halt Protease & Phosphatase inhibitor cocktail | Thermo Scientific | 78444 | |
Pierce BCA Protein Assay Kit | Pierce | 23225 | |
Mouse Cytokine 20plex Kit | Invitrogen | LMC006 | |
Magnetic Microsphere Beads | Luminex | MC100xx-01 | xx is the bead region |
Anti-mouse TNF- Capture Antibody | BD Pharmingen | 551225 | |
Anti-mouse TNF- Detection Antibody | BD Pharmingen | 554415 | |
Anti-mouse IFN- Capture Antibody | Abcam | ab10742 | |
Anti-mouse IFN- Detection Antibody | Abcam | ab83136 | |
PBS, pH 7.4 | Sigma | P3813-10PAK | |
Tween-20 | Fisher | BP337-500 | |
Assay Buffer | Millipore | L-MAB | |
Cytokine Standard | Millipore | MXM8070 | |
Multi-screen HTS 96well filter plates | Millipore | MSBVN1210 | |
SA-PE | Invitrogen | SA10044 | |
100 m Nylon Tissue Sieves | BD | 352360 | |
Splenocyte Separation Media | Lonza | 17-829E | |
TNF- /IL-4 ELISpot plates | R&D Systems | ELD5217 | |
Rabbit Anti-MUC1 monoclonal antibody | Epitomics | 2900-1 | |
Goat Anti-actin monoclonal antibody | Sigma | A1978 | |
Anti-rabbit HRP antibody | Promega | W401B | |
Goat anti-mouse HRP antibody | Santa Cruz Biotechnology, Inc. | SC-2005 | |
PVDF membrane | BioRad | 162-0174 | |
Mini Protean TGX Precast Gels | BioRad | 456-1083 | |
Muse Count & Viability Kit | Millipore | MCH100104 | |
MUC1 Antibody | BD Pharmingen | 550486 | IHC antibody |
Animal Research Peroxidase Kit | Dako | K3954 | IHC staining |
[header] | |||
Equipment and Software | |||
Millipore plate vaccum apparatus | Millipore | MSVMHTS00 | |
Luminex Lx200 | Millipore / Luminex | 40-013 | Manufactured by Luminex, distributed by Millipore |
Luminex Xponent Software | Millipore / Luminex | N/A | Version 3.1; included with Luminex Lx200 |
Milliple Analyst Software | Milliplex / VigeneTech | 40-086 | Version 5.1 |
Muse Cell Analyzer | Millipore | 0500-3115 | |
Muse Software | Millipore | N/A | Version 1.1.0.0; included with Analyzer |
Dissecting Microscope | Unitron | Z730 | |
Graphpad Prism Software | Graphpad Software Inc. | N/A | Version 5.1 |
Mini Protean Tetra Cell Gel apparatus | BioRad | 165-8001 | |
Trans Blot SD Cell and PowerPac | BioRad | 170-3849 |