| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Experimental Therapeutics |
Departments of Surgery [M. A. C.], Pediatrics [S. S.], and Obstetrics and Gynecology [W. H.], Women and Infants Hospital, Brown University, Providence, Rhode Island 02905; Department of Urology, University of Iowa, Iowa City, IA 52242 [Y. L., M. O.]; and Department of Surgery, University of California Los Angeles School of Medicine, Los Angeles, California 90095 [H. R. C.]
 |
ABSTRACT |
|---|
 Top ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
|
INTRODUCTION |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
There have been several attempts at using MUC1 as a cancer vaccine. Most of the work has focused on the use of a synthetic peptide containing five or seven of the tandem repeats, either by itself or conjugated to a carrier protein (11, 12, 13) . These MUC1 vaccines have been able to stimulate a modest humoral MUC1 response. Importantly, reactivation of patient-derived memory T cells has also proved to be efficacious against breast cancer cells in a non-obese diabetic/SCID3 mouse model (14) . As a matter of fact, MUC1-specific cytotoxic T cells have been isolated from MUC1 transgenic mice that, when adoptively transferred in vivo, eradicate tumors (15) . However, despite the encouraging results obtained from these approaches, it would be preferable to use a live MUC1-cytokine-coexpressing molecular construct as a vaccine. Attempts at using viral vectors have met with limited success (16) . The MUC1 protein expressed by host cells infected with MUC1-viral vectors is predominantly glycosylated, mimicking the benign form of MUC1 protein. Additionally, the MUC1 proteins expressed in other settings are heterogenous, suggesting a varying pattern of glycosylation or some instability in expression of the recombinant protein (17) .
The present study reports on the use of the attenuated form of Mycobacterium bovis, BCG, as the vehicle for delivering MUC1 protein. BCG, in its attenuated form, has been shown to be safe in humans. We have engineered BCG to express a truncated form of MUC1 protein containing 22 tandem repeats and to simultaneously secrete hIL-2. The ability of BCG-hIL2MUC1 to inhibit the growth of MUC1-positive tumor cells in a xenograft model of human breast cancer was evaluated in hu-PBL-SCID mice (SCID mice reconstituted with human PBLs).
|
MATERIALS AND METHODS |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
. The first MCS is located after the first promoter, a start codon, the BCG 
secretion signal (Fig. 1A
, SS), and a marker epitope, the influenza virus HA epitope tag (Fig. 1A
, TAG). The second MCS is located after the HSP70 promoter. A selectable antibiotic marker for kanamycin (Fig. 1A
, KAN) is also encoded by the plasmid. The plasmid PMOD12 was genetically modified to create pIL2MUC1, a plasmid capable of secreting IL-2 and expressing a truncated form of MUC1 protein. A schematic diagram of pIL2MUC1 is illustrated in Fig. 1B
. Plasmid pIL2MUC1 was constructed in a stepwise manner by inserting the cDNA for IL-2 into PMOD12 to create P12IL2 (intermediate plasmid), followed by the cDNA for the truncated form of MUC1 containing 22 tandem repeats to create pIL2MUC1.
|
Construction of pIL2MUC1 Plasmid.
The plasmid containing the human IL-2 cDNA was obtained from ATCC (Manassas, VA), and the cDNA was amplified by PCR using IL-2 sense and antisense primers encompassing the restriction enzyme sequences for BamHI and EcoRI, respectively. The purified IL-2 cDNA fragment was ligated into PMOD12 to create the intermediate plasmid, P12IL2. Using P12IL2, pIL2MUC1 was constructed by inserting a cDNA containing a truncated form of MUC1 with 22 tandem repeats through standard subcloning techniques. Briefly, a 1.7-kb fragment containing 22 tandem repeats of MUC1 was derived from pDKOF (kindly provided by Dr. O. J. Finn; Ref. 19
). The MUC1 1.7-kb fragment was subcloned into P12IL2 using cohesive and blunt-end ligations. P12IL2 was digested with NcoI and treated with calf intestinal phosphatase to prevent self-ligation. The two DNA fragments were then ligated at the NcoI site, and the free ends were blunt-ended by filling in the missing nucleotides with Klenow polymerase. The newly created blunt ends were religated with T4 ligase to create the new plasmid, phIL2MUC1. The new plasmid phIL2MUC1 was used to transform competent bacteria and colonies grown on selectable media. All restriction endonucleases and enzymes were obtained from New England Biolabs (Beverly, MA), and the reactions were performed according to the manufacturer’s guidelines.
The presence of the correct DNA inserts in phIL2MUC1 was confirmed using restriction enzyme mapping. The plasmid DNA was digested with BamHI, EcoRI, HindIII, and NcoI. The hIL-2 insert was released from the plasmid by codigestion with BamHI and EcoRI, and the truncated version of MUC1 was released by codigestion with HindIII and NcoI. The site at which the majority of the DNA manipulations for inserting the truncated MUC1 fragment had occurred involved the first series of ATG after the TATAA box. To ensure that the correct ORF had been maintained after the HSP70 promoter (the second promoter), phIL2MUC1 was sequenced from the second promoter using primer 3 and fluorescent end-labeled nucleotides (Brown Sequencing Facility; Ref. 20 ). Of 14 clones sequenced, only 4 maintained the correct reading frame. After confirmation of the correct DNA sequence of phIL2MUC1 by restriction enzyme analysis and DNA sequencing, phIL2MUC1 was used to transform competent BCG (Pasteur strain; Laval, Canada) by electroporation. BCG-conditioned media were tested for the presence of the recombinant proteins, hIL-2 and truncated MUC1.
Western Blotting.
Production of IL-2 and MUC1 by phIL-2MUC1 plasmid was confirmed by immunoblotting as described below. Protein extracts from equivalent optical densities of cultured BCG were prepared in the following manner for one-dimensional protein electrophoresis. One ml of cultured media of BCG-hIL2MUC1 was centrifuged, and the pellet was resuspended in SDS sample buffer and boiled for 10 min. The supernatant was loaded on a 15% bis-acrylamide resolving gel, and the protein was separated by electrophoresis. Molecular mass determinations were made by calibration of the gels with protein standards. At the completion of electrophoresis, the proteins were transferred to a nitrocellulose membrane (Bio-Rad Laboratories, Hercules, CA), and nonspecific sites were blocked with 10% nonfat powdered milk in PBS. The presence of hIL-2 was determined by immunoblotting with an anti-hIL-2 antibody (Amersham Pharmacia Biotech, Piscataway, NJ). After completion of the primary incubation, the membranes were incubated with goat antimouse peroxidase-labeled secondary antibody. The immunoblot was developed by the enhanced chemiluminescence method (Amersham Pharmacia Biotech) as directed by the manufacturer. To confirm that the IL-2 detected arose from BCG-hIL2MUC1, a parallel Western blot was performed with a primary antibody against HA, the epitope present on our recombinant IL-2 (Amersham Pharmacia Biotech). Cultured media of BCG-261 (BCG containing vector plasmid pMV261) were used as a negative control for these experiments. Commercially available IL-2 (Boehringer Mannheim, Indianapolis, IN) was used as a positive control. MUC1 was detected essentially as described above, except that the protein mass was separated on a 10% gel. Anti-MUC1 antibody was obtained from Sigma (St. Louis, MO).
Human Breast Cancer Cells.
Three human breast cancer cell lines, ZR75-1, MCF-7, and MDA-MB-175, were obtained from ATCC and evaluated for MUC1 expression. ZR75-1 (ATCC CRL 1500) is a human breast carcinoma cell line derived from the malignant ascites of a postmenopausal Caucasian female with infiltrating ductal carcinoma (21)
. MCF-7 (ATCC HTB 22) is a breast carcinoma cell line established from the pleural effusion of a postmenopausal Caucasian female (22)
. MDA-MB-175-VII (ATCC HTB 25) is also a breast carcinoma cell line derived from a pleural effusion, but in this instance, the source was a postmenopausal African-American female (23)
. The cell lines were maintained in RPMI 1640 supplemented with 10% fetal bovine serum using standard cell culture technique. Total MUC1 expression was determined by immunocytochemistry. Trypsinized cells were incubated with BC3 (anti-MUC1 antibody; a gift from Dr. O. J. Finn) in a 1:20 dilution. The antigen-antibody complex was then detected by a goat antimouse antibody and visualized using the streptavidin peroxidase method. Cell surface expression of MUC1 expression was determined by flow cytometry. Equivalent aliquots of exponentially growing breast cancer cells were stained with BC3 followed by incubation with antimouse fluorescence-labeled antibodies. Presence of the antigen-antibody complex was quantified in a Becton Dickinson FASCalibur machine. MUC1 staining was localized to the cell surface, and >95% of the cells stained strongly for this protein in the three cell lines tested. ZR75-1 cells were chosen as the source of human breast cancer cells for the animal model because of abundant MUC1 expression on the cell surface and because of their easy and rapid growth in culture. PANC1 (ATCC CRL-1469), a pancreatic carcinoma cell line known to express MUC1, was used as a positive control for these experiments. The negative control included incubation of the ZR75-1 breast cells without any primary antibody.
Reconstitution of SCID Mice (hu-PBL-SCID Mice).
Female SCID mice (CB17 scid/scid; 3–4 weeks old; Taconic Farms, Inc., Germantown, NY) were reconstituted with 50 x 106 human PBLs to create a xenograft of human lymphocytes in SCID mice (hu-PBL-SCID mice). PBLs were procured as described from human buffy coats obtained from the Rhode Island Blood Center (24)
. The buffy coats were resuspended in HBSS (pH 7.4) and layered under a high-density solution of Ficoll-Paque plus (Amersham Pharmacia Biotech, AB, Sweden) and centrifuged for 30 min at 1500 rpm. The interface containing the PBLs was then harvested and washed twice in HBSS. The washed cell pellet was resuspended in PBS, and 50 x 106 PBLs were injected i.p. into the SCID mice.
Vaccination of hu-PBL-SCID Mice.
The vaccines used in this experiment included BCG-hIL2MUC1 (0.5 cfu; experimental vaccine), MUC1 peptide (100 µg; MUC1 control), BCG-261 (0.5 cfu; BCG control vaccine), and PBS (sham vaccine) diluted in a total volume of 200 µL of PBS. BCG-hIL2MUC1 is our recombinant vaccine consisting of BCG that expresses a truncated form of MUC1 protein with 22 tandem repeats while simultaneously secreting human IL-2. MUC1 synthetic peptide was a custom peptide consisting of five tandem repeats of MUC1 (GVTSAPDTRPAPGSTAPPAHGVTSAPDTRPAPGSTAPPAHGVTSAPDTRPAPGSTAPPAHGVTSAPDTRPAPGSTAPPAHGVTSAPDTRPAPGSTAPPAH; Boston Biomolecules, Warhem, MA). BCG-261 vaccine consisted of the starting plasmid used to construct pIL2MUC1 and served as the control for BCG-stimulated antineoplastic activity. PBS was a sham vaccine and served as an internal control. hu-PBL-SCID mice were reconstituted as described above with 50 x 106 PBLs on day 0. One day after lymphocyte reconstitution, hu-PBL-SCID mice received three i.p. vaccine injections at biweekly intervals. Each group of animals had eight mice. Two weeks after the third vaccination, 4 x 106 ZR75-1 breast cancer cells were injected s.c. into the right flank. The mice were observed until impending death or 150 days after tumor inoculation and then sacrificed. Outcomes of interest were time to tumor detection, size of primary tumor, and rate of tumor growth. Median time to tumor onset was determined for each experimental group. Mean tumor size was plotted over time after tumor engraftment for each group, and polynomial regression analysis was performed to determine whether the rate of tumor growth differed between the groups of mice.
To further verify that the tumor growth inhibition observed in animals immunized with BCG-hIL2MUC1 was solely due to a MUC1-specific response and was not due to the enhanced T-cell function from BCG-mediated IL-2 expression, hu-PBL-SCID mice were immunized with BCG-IL2 with and without exogenous MUC1 peptide and xenografted with 4 x 106 ZR75-1 human breast cancer cells. The primary tumors obtained from these animals were evaluated for MUC1 expression and infiltration of CD8-positive human lymphocytes.
When the animals became gravely ill, as manifested by inability to move or groom, they were sacrificed. All surviving mice were sacrificed 24 weeks (168 days) after tumor engraftment. Three mice receiving BCG-hIL2MUC1 and one mouse receiving MUC1 peptide were sacrificed at 10 weeks after receiving the tumor xenograft to allow for completion of the experiments in a timely fashion. The mice were anesthetized by 100% CO2 insufflation, and necropsy was performed on all mice. All tumor deposits were carefully measured and fixed in a 10% formalin solution. All enlarged masses or abnormalities within the abdominal cavity or detected on the liver and spleen were harvested, measured, and fixed in formalin. Liver, spleen, and lung samples were also obtained. The harvested tissue was fixed in a 10% formalin solution and embedded in paraffin wax. The tissue blocks were then sectioned for H&E staining and immunohistochemistry.
Histological Analysis of Tissue Samples.
Immunohistochemistry of tissue sections was performed using the LSAB+ peroxidase kit (DAKO Corp., Carpinteria, CA). Briefly, slides containing the tissue sections were deparaffinized and rehydrated with alcohol and xylene. Endogenous peroxidase activity was suppressed with a 3% hydrogen peroxide solution, and nonspecific antigen binding was blocked by incubation with swine serum. The tissue was then incubated with the appropriate primary antibodies. The following antibodies were used: anti-episialin antibody (Sigma) was used for detection of MUC1; anti-cytokeratin clone AE1/AE2 antibody was used for detection of cytokeratins, antihuman CD8 was used for detection of CD8-positive human lymphocytes; anti-mycobacterial antibody was used to detect the presence of BCG; and anti-CD45 antibody was used to detect human lymphocytes (DAKO Corp.). Incubation of tissue sections with murine-derived control IgG served as the negative control for these experiments (DAKO Corp.). After incubation with the primary antibody, the secondary antibody consisting of a biotinylated antirabbit, antimouse, and antigoat immunoglobulin antibody was used. Streptavidin peroxidase was then added. Color development followed incubation with a substrate chromogen solution (3',3'-diaminobenzidine chromogen solution). Slides were then counterstained with hematoxylin (Mayer’s hematoxylin; Lillie’s modification; DAKO Corp.). Slides were finally dehydrated with alcohol and xylene and mounted for histological evaluation.
Statistical Analysis.
The Kaplan-Meier method was used to estimate survival curves. Polynomial regression analysis of tumor size over time and within groups showed a P < 0.01.
|
RESULTS |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Production of IL-2 and MUC-1 in BCG-hIL2MUC1.
Cultured BCG-hIL2MUC1 mycobacteria samples were evaluated for the expression of hIL-2 and MUC1 (Fig. 2)
by immunoblotting as described in "Materials and Methods." Fig. 2A
shows data for detection of IL-2. Lane A represents commercial IL-2 protein that served as our positive control. The protein mass extracted from BCG-261 served as a negative control (Lane B). Proteins extracted from BCG-hIL2MUC1 were found to contain IL-2 as shown in Lanes C and D. Immunoblotting with an anti-IL-2 antibody resulted in the detection of a unique protein in Lanes A and C. The protein detected in Lane C represented IL-2 and was approximately 20 kDa in size. The size of the BCG-expressed IL-2 was slightly larger than that obtained for its commercially available counterpart (Lane A) and is due to the additional sequences for the BCG secretion signal and the HA marker epitope tag. To confirm that the IL-2 detected by immunoelectrophoresis was from BCG-IL2MUC1 and of the same size, a parallel immunoblot was performed, but the blot probed with an antibody directed against HA, the epitope tag (Lane D).
|
. Recombinant MUC1 protein with a molecular mass of 87 kDa was expressed by BCG-hIL2MUC1 (Lane F). Lanes A and C represent our positive controls (ZR-75-1 and E. coli-phIL2MUC1, respectively). Lanes B, D, and E are the negative controls consisting of extracts from E. coli-PMOD12, BCG-261, and BCG-hIL2, respectively. MUC1 expressed by BCG-hIL2MUC1 was a single band, indicating that no heterogeneity occurred due to glycosylation as has been described with recombinant viral vectors expressing MUC1 (25)
.
Tumor Kinetics of ZR75-1 Cells in hu-PBL-SCID Mice.
Before embarking on our vaccine experiments, it was important to determine the growth kinetics of ZR75-1 cancer cells in our animal model. As described earlier, SCID mice were reconstituted with 50 x 106 human PBLs on day 0 to create a xenograft of a human immune system (hu-PBL-SCID). One day after lymphocyte reconstitution of the hu-PBL-SCID mice, varying concentrations of exponentially growing ZR-75-1 cells were injected s.c. into the right flanks of these mice. Sham inoculation of the contralateral flank with PBS was performed. There were three groups with 4 mice/group. Group A received 1 x 106 ZR75-1 cells, group B received 2 x 106 ZR75-1 cells, and group C received 4 x 106 ZR75-1 cells. The mice were observed until death or 150 days, whichever was later. Outcomes of interest were time to tumor detection, size of primary tumor, and the rate and pattern of metastatic disease. At the time of necropsy, primary tissue, liver, lung, and abnormal masses were harvested for histological analysis. All animals developed a gross primary tumor with "the mean time to gross tumor detection" inversely proportional to tumor inoculum. Fifty percent of mice developed metastatic breast cancer irrespective of tumor inoculum. One animal (in group C) developed a lymphoma and died on day 133. The results indicated that 4 x 106 ZR75-1 cells was the suitable concentration for the aggressive tumor model described in this study.
Overall Survival of Vaccinated Animals.
The overall survival of the animals was evaluated to determine the toxicity of the various vaccines as well as death from breast cancer. All of the vaccines (see details in "Materials and Methods") were well tolerated by the animals with the exception of BCG-261. Nine weeks after receiving their first immunization, and 3 weeks after tumor engraftment, 75% of the hu-PBL-SCID mice that had been vaccinated with BCG-261 had died. The cause of death was not breast cancer because it was observed in animals who had received the vaccine but were never xenografted with breast cancer. The two remaining mice who had received BCG-261 vaccine subsequently died at days 52 and 57 after tumor engraftment. These results indicated that BCG-261 alone, at the concentration used, was lethal to the hu-PBL-SCID mice. Necropsy of these animals did not reveal an obvious cause of death. However, it is important to point out that coexpression of hIL-2 or the truncated form of MUC1 had protected the hu-PBL-SCID mice from the toxic effects of the same dose of BCG.
All of the animals who received BCG-hIL2MUC1 vaccine survived until termination of the experiment, with the exception of three mice in the group sacrificed at the 10-week interval. All of the animals in the MUC1 peptide vaccine group had died by 16 weeks after tumor engraftment; one was sacrificed at the 10-week interval. Animals with sham inoculation (PBS) had died by week 9 after tumor engraftment. The cause of death in the MUC1 peptide vaccine- and sham-inoculated animals was similar. These animals developed large intra-abdominal masses and were lethargic and preterminal within a week of detection of these masses. The etiology of these masses will be discussed later.
Time to Tumor Detection and Tumor Growth After Immunization.
hu-PBL-SCID xenografted with human breast cancer cells were examined for the development of primary tumors. Mice with the MUC1 synthetic peptide-based or sham immunization developed a primary tumor with a median time to tumor detection of 34 and 32 days, respectively. Most of the animals receiving BCG-hIL2MUC1 vaccine developed a gross tumor at the site of injection of the breast cancer cells. The median time for gross tumor detection in this group of mice was 43 days. As mentioned earlier, only 2 mice in the BCG-261 group survived beyond week 6 after tumor engraftment. In all of the surviving BCG-261-immunized mice, a primary tumor was detected with a median time to tumor detection of 33 days. The delay in median time to tumor detection in the group of mice immunized with BCG-hIL2MUC1 was different from that of the other groups and suggested that the experimental vaccine, BCG-hIL2MUC1, was capable of suppressing the growth of breast cancer cells.
The primary tumors detected in the animals were measured and plotted over time (Fig. 3)
. The size of the primary tumors in the control animals (PBS, MUC1 synthetic peptide, and BCG-261) showed a continual increase in size with time. The curves for these three groups of animals were essentially identical. However, the rate of tumor growth in the animals immunized with BCG-hIL2MUC1 was much slower than that observed in the control groups. In these animals, the mean tumor size, stratified by treatment group and time after tumor inoculation, was significantly smaller (polynomial regression analysis, P < 0.001). The rate of tumor growth in BCG-hIL2MUC1-immunized animals was also significantly slower than that observed in the control animals (P < 0.01). These results suggest that the control vaccines, i.e., MUC1 peptide or BCG-261, exhibited no inhibitory effect on the growth of the primary tumor. However, BCG-hIL2MUC1 vaccination inhibited the rate of tumor growth.
|
. The breast cancer cells formed ductal-like structures and had a Bloom-Richardson appearance with a nuclear grade of 2. The breast cancer cells were moderately differentiated. The primary breast tumors from mice that had received BCG-hIL2MUC1 were of a similar morphology as that detected in the control groups (Fig. 4E)
.
|
. The MUC1 protein was present on the cell surface of virtually all breast cancer cells. There was no MUC1 staining observed on the surrounding stromal cells or in the negative control (Fig. 4F)
. In addition, a satellite deposit detected in a MUC1 peptide-immunized animal was also strongly positive for MUC1 protein (data not shown). The MUC1 staining pattern detected in mice receiving BCG-hIL2MUC1 vaccine was different from that detected in the control animals. Although many of the breast cancer cells maintained their cell surface expression of MUC1 protein, a significant proportion of the breast cancer cells were negative for specific MUC1 staining (Fig. 4E)
. These cells maintained the histological characteristics of breast cancer cells. This observation suggested that our vaccine, BCG-hIL2MUC1, inhibited the growth of MUC1-positive breast cancer cells, resulting in the maintenance of probably less aggressive MUC1-negative breast cancer cells.
Induction of Cell-mediated Immunity by BCG-hIL2MUC1.
It was not possible to obtain viable human lymphocytes for evaluation of anti-MUC1 cell-mediated cytotoxicity in hu-PBL-SCID mice. Alternatively, immunohistochemistry of primary tumors for CD8-positive human lymphocytes was performed. As illustrated in Fig. 5D
, only tumors obtained from animals immunized with BCG-hIL2MUC1 had any detectable infiltration of CD8-positive cells. There were no detectable CD8-positive cells in tumors obtained from mice immunized with BCG-hIL2, with or without MUC1 peptide, or sham-immunized animals (Fig. 5)
.
|
, spleens from BCG-hIL2- and BCG-hIL2MUC1-treated animals were strongly positive for mycobacteria. However, only spleens from BCG-hIL2MUC1-immunized animals expressed MUC1 protein 168 days after inoculation (Fig. 6D)
. This result suggested that BCG retained phIL2MUC1 for a significant time in vivo without any selectable pressure.
|
. The exact primary source of the masses was not clear because of their large size but involved the mesenteric lymph nodes, peripancreatic region, and liver. In one mouse (in the PBS group), the mass arose in the retroperitoneum and impinged on the nerves of the left hind leg. The masses contained large areas of necrosis on gross inspection. Histological examination of these masses showed a diffuse infiltration of small cells that lacked an organoid pattern (Fig. 7B)
. This lack of histological arrangement was different from the primary breast tumors. The histological pattern observed in these masses was that of a lymphoma or poorly differentiated carcinoma. In addition to the cells noted, large areas of necrosis were documented by histology. Immunohistochemistry with anti-CD45 antibody (a marker of human lymphocytes), a pan-cytokeratin antibody (a marker of epithelial cells), and MUC1 antibody was performed. As shown in Fig. 7C
, the malignant cells were strongly positive for CD45. Only the cells within the masses were CD45 positive (Fig. 7C
,solid arrow). There was no appreciable CD45 staining identified in the adjacent organs such as the pancreas or liver (Fig. 7C
, open arrow). The cells within these abdominal masses were also negative for both cytokeratin (Fig. 7D)
and MUC1 staining (Fig. 7E)
. Specificity of the staining was confirmed by a negative control consisting of mouse-derived sera (Fig. 7F)
. Based on these results, the masses were compatible with a lymphoma.
|
|
DISCUSSION |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Analysis of the ability of BCG-hIL2MUC1 to prevent and/or inhibit the growth of xenografted human breast cancer cells in hu-PBL-SCID mouse model reveals that human breast cancer cells were capable of growing a primary tumor. There was a high rate of tumor engraftment observed in all animals, as observed by others (26) . The tumors in BCG-hIL2MUC1-immunized mice grew significantly more slowly than the tumors observed in animals immunized with PBS, BCG-261, or MUC1 peptide. The size of the tumor detected in the current protocol was approximately 1.0 cm, which was much smaller than that reported from other studies involving SCID mice (27 , 28) . This difference in maximal tumor size may have been due to the presence of the reconstituted human immune system in hu-PBL-SCID mice because a smaller tumor size has been reported in SCID mice reconstituted with human lymphocytes and grafted with human tumors (29) .
It has been observed that human cancers, after transplantation into SCID mice, maintain the histological features of the donor (30) . Thus, the primary tumors were stained for MUC1 expression to determine whether our vaccine had selected for MUC1 negative cells. Although some heterogeneity in MUC1 expression has been reported in SCID mice (31 , 32) , virtually all of the breast cancer cells in the primary neoplasm stained positive for MUC1 protein in the hu-PBL-SCID mice receiving MUC1 peptide, BCG-261, or sham vaccine. The staining pattern in the hu-PBL-SCID mice vaccinated with BCG-hIL2MUC1 was different from that of the control animals. A substantial proportion of the cancer cells in these primary neoplasms did not stain for MUC1 protein. Furthermore, the presence of CD8-positive cells in the primary tumor was detected only in animals immunized with BCG-hIL2MUC1. These observations suggested that BCG-hIL2MUC1 had stimulated a cell-mediated immune response resulting in the selective deletion of MUC1-positive cells and are supported by recent findings. It has been shown that a single transfer of restimulated bone marrow cells from breast cancer patients caused regression of xenografted autologous tumor in NOD/SCID mice (14) . Such a phenomenon has also been observed in MUC1 transgenic mice, in which MUC1-specific cytotoxic T cells, when adoptively transferred, were effective in controlling mammary gland and melanoma tumors (15) .
When the survival trends were compared in the vaccine groups, it appeared that control BCG-261 was lethal to SCID mice. These mice died within 9 weeks after lymphocyte reconstitution, with some dying before engraftment of the human cancer cells. This fatal outcome with BCG-261 was not expected from our prior studies; however, those experiments were completed within 6 weeks. The cause of death remains unclear but was not due to disseminated breast cancer. Even though our vaccine was BCG-based, a similar fatal outcome was not observed in animals that had received BCG-hIL2MUC1. A more rapid intrinsic growth of BCG-261 in animals is unlikely because both BCG-261 and BCG-hIL2MUC1 have similar doubling rates in vitro (data not shown). It is feasible that the presence of either IL-2 or MUC1 protein protected the animals from the toxic effects of BCG (33) . If BCG were lethal, then the length of time that BCG retained phIL2MUC1 is important. Spleens obtained from BCG-hIL2MUC1-immunized animals and stained for MUC1 protein were positive for MUC1 protein. Therefore, despite no in vivo selective pressure, our data suggest that the plasmid hIL2MUC1 is retained by the BCG for at least 168 days. Similar results reported by Luo et al. (34) indicate that the recombinant BCG is stable in vivo for up to 16 weeks. An explanation for the lethality of BCG-261 observed in our model is the use of SCID mice reconstituted with a limited supply of human peripheral lymphocytes. In the absence of IL-2, this small compartment of T cells cannot control the BCG infection. It would be expected that in a competent animal, the lethality of BCG-261 would not be observed. Competent mice have been vaccinated with BCG-261 with no morbidity. Duda et al. (35) reported no morbidity or lethality in C57BL/6 mice immunized with BCG-261 after a total dose of 3 x 106 cfu administered intratumorally. Taken together, these results suggest that BCG-hIL2MUC1 is a safe vaccine.
The majority of hu-PBL-SCID mice in the sham-inoculated and MUC1 peptide-vaccinated groups developed large intra-abdominal or retroperitoneal masses. Only one mouse in the BCG-hIL2MUC1-immunized group developed a similar mass. Few mice immunized with BCG-hIL2 or BCG-hIL2MUC1 developed these masses. Histological examination of these masses indicated that they had different histological characteristics than those observed in the primary tumors, appeared to arise from nodal tissue, and were consistent with a lymphoma by immunostaining. The development of lymphoma in hu-PBL-SCID mice has been reported previously and is not a unique phenomena (36) . Lymphomas that arise in hu-PBL-SCID mice are derived from B cells, and virtually all are EBV positive (37) . Most adult humans are EBV seropositive, and therefore the source of the lymphomas observed in our animal model was most likely secondary to the use of EBV-positive lymphocytes for reconstitution. However, the incidence of lymphoma development and the rate of growth were delayed in the BCG-hIL2MUC1-immunized animals. It has been reported that in hu-PBL-SCID mice, T cells are critical for control of EBV infection (38) . In hu-PBL-SCID mice, the administration of low doses of human IL-2 may abolish the development of lymphoproliferative disease in these animals (33) . Therefore, it is feasible that animals vaccinated with BCG-hIL2MUC1 were protected from lymphoma development by IL-2 secreted by the recombinant vaccine. Our data thus warrant the use of EBV-negative human PBLs in SCID mice reconstitution experiments.
In summary, we have constructed a novel BCG-based MUC1 vaccine for the treatment and prevention of breast cancer. This is the first report of a BCG-based vaccine engineered to express a tumor antigen (in this case, MUC1). Our vaccine, BCG-hIL2MUC1, inhibited the growth of breast cancer in hu-PBL-SCID mice and, in a minority of animals, even prevented primary tumor development. Future experiments are planned to evaluate the efficacy of this vaccine in the adjuvant treatment of breast cancer and the immune mechanisms involved in tumor immunity.
|
FOOTNOTES |
|---|
1 Supported in part by grants from Association of Academic Surgery (to M. A. C.), Expeditions Inspirations Outstanding Young Investigator Award (to M. A. C.), and HD41701-01 (to S. S.). 
2 To whom requests for reprints should be addressed, at Department of Pediatrics, Women and Infants’ Hospital-Brown University, 101 Dudley Street, Providence, RI 02905. Phone: (401) 274-1122, ext. 1289; Fax: (401) 453-7571; E-mail: ssharma{at}wihri.org
3 The abbreviations used are: SCID, severe combined immunodeficient; BCG, Bacillus Calmette-Guérin; hIL, human interleukin; IL, interleukin; cfu, colony-forming unit(s); HSP, heat shock protein; MCS, multiple cloning site; HA, hemagglutinin; ATCC, American Type Culture Collection; PBL, peripheral blood lymphocyte; ORF, open reading frame.
Received 4/22/02. Accepted 1/15/03.
|
REFERENCES |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
This article has been cited by other articles:
|
|
 |
|
  L. M. Herbert, J. F. Grosso, M. Dorsey Jr., T. Fu, I. Keydar, M. A. Cejas, D. H. Wreschner, N. Smorodinski, and D. M. Lopez A Unique Mucin Immunoenhancing Peptide with Antitumor Properties Cancer Res., November 1, 2004; 64(21): 8077 - 8084. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. L. Sabichi, M.-F. Demierre, E. T. Hawk, C. E. Lerman, and S. M. Lippman Frontiers in Cancer Prevention Research Cancer Res., September 15, 2003; 63(18): 5649 - 5655. [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Cancer Research | Clinical Cancer Research |
| Cancer Epidemiology Biomarkers & Prevention | Molecular Cancer Therapeutics |
| Molecular Cancer Research | Cancer Prevention Research |
| Cancer Prevention Journals Portal | Cancer Reviews Online |
| Annual Meeting Education Book | Meeting Abstracts Online |
