Stem Cells, Vol. 18, No. 6, 422-427,
November 2000
© 2000 AlphaMed Press
Sensitivity of c-erbB Positive Cells to a Ligand Toxin and Its Utility in Purging Breast Cancer Cells from Peripheral Blood Stem Cell (PBSC) Collections
Michelle Keira,
Rod Fiddesb,
James C. Biggsa,
Phillip P. Kearneya
a Haematology Research Laboratory, St. Vincent's Hospital, Sydney, NSW, Australia;
b Cooperative Research Center for BioPharmaceuticals, Sydney, NSW, Australia
Key Words. Purging • PBSC • Breast cancer • Ligand toxin
P. Kearney, MBA, Ph.D., Tumour Biology Division, ActiveBiotech Research AB, P.O. Box 724, SE-22007 Lund, Sweden. Telephone: 46-46-191268; Fax: 46-46-191134; e-mail: phil.kearney{at}activebiotech.com
 |
ABSTRACT
|
|---|
Autografting following high-dose conditioning is being increasingly offered to breast cancer sufferers, without due regard to the reinfusion of malignant cells. We sought to determine if a breast cancer cell line could be successfully purged from peripheral blood stem cell (PBSC) harvests using a ligand-toxin molecule directed to heregulin-activated erbB receptors. Initial experiments demonstrated no reduction in hemopoietic colony-forming ability in the presence of ligand toxin (2 nM). Breast cancer cell lines which demonstrated differing sensitivities to the ligand toxin were subsequently seeded into stem cell collections and incubated with 2 nM ligand-toxin. One cell line, ZR-75-1, was completely sensitive to the ligand toxin in this mixture; a second, MDB-MA-361, was more profoundly sensitive to the ligand toxin in the presence of the PBSC, whereas a third was unaffected by the toxin. These results suggest purging may indeed be possible in the PBSC of breast cancer patients, but the parameters that define sensitivity are as yet unknown.
 |
INTRODUCTION
|
|---|
Breast cancer accounts for almost one-quarter of all tumors in women [1]. It is thought that tumor initiation may precede clinical presentation by two to three decades [2], by which stage the disease is considered to be systemic. The cancer may be graded on diagnosis according to the localization in the breast, extension to adjacent groups of lymph nodes, and metastases to distant sites. There is a clear relationship in breast cancer between the relapse rate and the number of lymph nodes involved at diagnosis. Those women with greater than 10 lymph nodes involved have a 60%-80% relapse rate at five years [3]. Although there is considerable variation in the survival following diagnosis of metastatic breast cancer, the median is two years [3]. Such a poor prognosis and the accepted view that smaller tumors are easier to treat than larger ones, led Peters et al. [4] to the application of high-dose chemotherapy to stage II/III breast cancers. The patient follow-up suggested the disease-free survival at three years was over 70%, supporting the notion that earlier and systemic treatment was of considerable benefit to this patient group.
Irrespective of the therapeutic option pursued, it is noteworthy that the age-adjusted cancer death rate from female breast cancer in the United States has remained stable at 26 deaths per 100,000 females over the past 50 years [5], emphasizing that both new therapeutics and diagnostics are needed in the treatment of this disease. Recent diagnostic approaches include the detection of epithelial cells in the bone marrow or peripheral blood [6, 7]. The degree of overexpression of c-erbB2 has been shown to correlate with the disease stage [8], suggesting it to be a useful marker of advanced disease. A further study has suggested that overexpression of this marker is also associated with resistance of the tumor to treatment with taxol [9] and that such tumors may therefore require more aggressive regimes.
One of the major therapeutic innovations of the last decade is the use of high-dose therapy coupled with hemopoietic rescue by autologous graft [10]. Such an approach is often accompanied by a purge of the graft of malignant cells to reduce the tumor load reintroduced to the patient. Purging has been associated with extended time to relapse in many malignant conditions [11, 12]. The use of cancer-specific or tissue-specific markers to identify breast cancer cells in the autologous graft also allows those antibodies to be used to specifically target the tumor cells with immunotoxin molecules [13]. ErbB-2 linked to Pseudomonas exotoxin or ricin A chain has been shown to be effective against human non-small-cell lung [14], human breast [15, 16], ovarian [17], and squamous carcinomas [18]. In this article, we describe the sensitivity of breast cancer cell lines MDA-MB-361, ZR-75-1, and MCF-7 to a heregulin (HRG)-pseudomonas exotoxin (PE)40 toxin, its ability to reduce tumor burden in an in vitro setting, and the effect of the ligand toxin on the hemopoietic compartment.
 |
MATERIALS AND METHODS
|
|---|
All recombinant DNA manipulations were performed according to National Health and Medical Research Council guidelines. Experiments involving human peripheral blood stem cells (PBSC) were performed in accordance with St. Vincent's Hospital Institute Bioethics requirements.
Ligand Toxin
Recombinant HRG comprised amino acids 177-237 of the ß subunit of HRG [19], while the ligand toxin (HRG-PE40) incorporated the Pseudomonas exotoxin PE40 domain [20]. Proteins were expressed in Escherichia coli and purified as previously described [21].
Tissue Culture
Breast cancer cell lines ZR-75-1, MCF-7, and MDB-MA-361 (ATCC; Manassas, VA; http://www.atcc.org) were cultured in RPMI 1640 (Trace Biosciences; Sydney, NSW, Australia), 10% fetal bovine serum (FBS)/100 U/ml human insulin (Actrapid; Nova Nordisk Pharmaceuticals; Sydney, NSW, Australia; http://www.novo.dk/index.asp), under 5% CO2 at 37°C. Adherent cells were collected by centrifugation following trypsin treatment (Boehringer Mannheim; Sydney, NSW, Australia), and dispensed in growth media into individual wells of a 96-well plate at an initial concentration of 103 cells per well in a volume of 50 µl. After 3 d, recombinant HRG alone, PE40 alone, HRG-PE40, or vehicle control was added in a volume of 50 µl of growth medium. Triplicate plates were prepared to allow cell growth to be determined at three pointsdays 3, 5, and 7 after drug addition. Cell proliferation was determined on six wells per treatment time point as previously described [21].
PBSC Toxin Sensitivity
PBSC collections were obtained from the Cell Separator Unit at St. Vincent's Hospital in accordance with Institute Bioethics Committee regulations. Samples were diluted in Hank's balanced salt solution ([HBSS]; Sigma-Aldrich; Syndey NSW, Australia; http://www.sigma-aldrich.com), 1% acid citrate dextrose solution B.P. ([ACD]-solution A; Domedica; Syndey, NSW, Australia), 10% inactivated (FBS; Cytosystems; Syndey, NSW, Australia), to a cell concentration of 1 x 106 cells/ml. One milliliter of this cell suspension was added to 9 ml 0.3% agar (Bacto-Agar; Difco Laboratories; Detroit, MI), 25% FBS, Iscove's modified Dulbecco's medium ([IMDM], with L-glutamine and 25 mM HEPES buffer, without NaHCO3), 0.075M NaHCO3, 1.65 mM L-asparagine (Sigma-Aldrich), and 0.025 mM 2-mercaptoethanol (Merck; Kilsyth, Victoria, Australia). One milliliter of this cell/agar mixture was plated into 35-mm petri dishes with 0.1 ml human placental conditioned media ([HPCM]; Institute of Medical and Veterinary Science; Adelaide, Australia), and the appropriate ligand-toxin concentration of 0.1 nM, 0.5 nM, or 2.0 nM, or 2.0 nM PE40. Alternatively, methocult GF+ H4535 (StemCell Technologies; Vancouver, BC; http://www.stemcell.com) was used, as this contained all necessary human growth factors to support primitive hemopoiesis (recombinant human stem cell factor [rhSCF], rh interleukin 3 [rhIL-3], rhIL-6, rhGM-CSF, and rhG-CSF). These cultures were incubated in a humidified incubator at 37°C, 5% CO2 for 10 days. Following this time period, the numbers of granulocytic/macrophage colonies (CFU-G/GM) were ascertained under an inverted microscope.
Breast cancer cell lines were cultured overnight in the presence of phosphate buffered saline, toxin alone, or HRG-PE40 as described above. Cells were harvested and 103 cells plated out in triplicate in the same agar mix as described above. The ability of the cell line to form colonies was determined following a seven-day incubation.
In Vitro Cell Line: PBSC Mixing Studies
Breast cancer cell line cells (2 x 105/ml unless otherwise stated) were resuspended alone or in the presence of PBSC (1 x 106/ml) in a flask containing Myelocult H5100 (StemCell Technologies). An aliquot (10 or 100 µl) of this was placed into HPCM/agar to ascertain their colony-forming ability (see above).
The concentrations of each cell type placed in the overnight culture varied and are indicated for each experiment. Initial experiments used ZR-75-1 cell line (2 x 105/ml) and 1 x 106/ml PBSC. In the remaining experiments, cell line concentrations were either 3 x 105 or 3 x 104 total for each cell type. Statistical analysis of the resultant data was performed using the paired t-test.
 |
RESULTS
|
|---|
Effects of HRG on Human Breast Cancer Cell Proliferation and Cytotoxicity of HRG-PE40
Human breast cancer cell lines were used for growth assays as previously described [21]. These cell lines express detectable levels of each of erbB2, erbB3, and erbB4 [22] and A. de Fazio, P.W. Janes, and R.J. Fiddes (unpublished observations). Because, in this study, we utilized MCF-7, MDA-MB-361, and ZR-75-1 cells in the "purging" studies, we wished to determine the in vitro cytotoxicity of HRG-PE40 in these cell lines. The sensitivity of the cell lines to the mitogenic activity of HRG was also tested to demonstrate the expression of active HRG receptors in these cells. We have shown previously that MCF-7 cells responded to HRG by an increase in the rate of proliferation, and the HRG-PE40 ligand toxin exhibited an IC50 of 300-400 pM (Fig. 1
). Each of the cell lines tested showed a significant increase in cell number (Fig. 1
). In response to HRG, MCF-7 and ZR-75-1 cell numbers doubled relative to controls after five days, whereas for MDA-MB-361 cells, which exhibited a slower overall growth rate, cell numbers increased by 40%-50% over controls after five days (Fig. 1
). An increase in growth rate continued to day 7 for the MDA-MB-361 cells, however, the MCF-7 and ZR-75-1 cells treated with HRG reach confluence before day 7 and the growth rate consequently decreases. Each of the breast cancer cell lines treated with HRG-PE40 showed a dramatic decrease in cell number, both relative to untreated control cultures and the initial plating density (Fig. 1
). Ligand-toxin concentrations of greater than 2 pM generally resulted in cytotoxicity, with maximum cell death seen at the highest concentration used (5 nM). ZR-75-1 cells were the most sensitive to HRG-PE40 over four experiments, whereas MDA-MB-361 cells were slightly less sensitive (IC50 of 10-20 pM). MCF-7 cells were the least sensitive of the cell lines tested (IC50 of 300-400 pM). At only the highest concentration of PE40 used (5 nM) was proliferation slightly retarded, as indicated by the IC50 values for PE40, which were more than 2,000-fold higher than for HRG-PE40 in the breast cancer cell lines studied (Fig. 1
).

View larger version (13K):
[in this window]
[in a new window]
|
Figure 1. Effects of HRG or HRG-PE40 on the rate of proliferation of human breast cancer cells. Cells were dispensed into individual wells of 96-well culture plates. HRG (1 nM), HRG-PE40 (5 nM), or PE40 (5 nM) was added at day 0, and cell numbers were evaluated at days 3, 5, and 7 using an indirect MTT assay. Each experiment was performed three times with similar results.
|
|
PBSC Sensitivity to HRG-PE40
PBSC collections were incubated with ligand toxin or toxin alone overnight at 37°C in hemopoietic long-term culture medium. From this mixture, 1 ml was plated into a semisolid colony-forming assay (CFU assay). Three sequential, independent PBSC harvests were used and as a consequence there was considerable variation between the colony-forming ability of the harvests (Table 1
). PE40 toxin alone had no effect on the colony-forming ability of the harvests, nor did the presence of ligand-toxin (Table 1
).
We next sought to determine if breast cancer cells seeded into PBSC would be sensitive to the ligand toxin, and within such a timeframe that would still allow the PBSC to be clinically useful. A number of breast cancer cell lines positive for expression of c-erbB3 and c-erbB4 [22] were mixed with unmanipulated PBSC and incubated with HRG toxin overnight at 37°C (Material and Methods). Both ZR-75-1 and MCF-7 grow in suspension, whereas MDB-MA-361 grows as an attached cell line. Because the assay was a colony-forming one, it was necessary to determine the cloning efficiency of this cell line in dilution experiments. This was determined to be 27 ± 3 per 100 cells plated (data not shown).
Liquid culture toxicity studies had suggested ZR-75-1 to be the cell line most sensitive to HRG-PE40 (Fig. 1
). This was also noted in the PBSC seeding experiments. Cell lines MCF-7 and MDB-MA-361 showed a decrease in their colony-forming ability in the presence of PBSC alone (data not shown), compared with the colony-forming ability in their standard growth medium. Those MCF-7 colonies capable of forming in the PBSC culture medium, Myelocult H5100, were completely insensitive to ligand toxin (Table 2
), mirroring the finding with this cell line in the presence of ligand toxin alone. Cell line MDA-MB-361 was also tested in this assay, at two cell concentrations. At the lower concentration, 104 cells per plate, all cells were sensitive to ligand toxin. However, in the presence of PBSC, colony formation was noted. When 10-fold higher cell numbers were used, colony formation was noted in the presence of ligand toxin and PBSC plus ligand-toxin. Interestingly, the number of colonies that formed in the presence of ligand toxin and PBSC were not significantly higher than at the lower cell number assayed.
 |
DISCUSSION
|
|---|
Recombinant toxins are hybrid molecules made by recombinant DNA methods linking a cell selection moiety to a cytotoxic agent. Such agents may have roles eventually as front-line therapeutics, but in the short term can provide valuable support as adjuvant therapies. In the bone marrow transplant setting, they may be used to reduce the tumor load present in those cells reinfused into the patient during an autologous graft. Ideal targets for these initial studies are those cells expressing HRG receptors, inasmuch as a number of epithelial lineage tumors overexpress this marker whereas hemopoietic lineages do not. Cancer cell lines have been shown to overexpress the HRG receptor protein at levels up to 100 times those of normal cells [23]. This is also seen in patient samples where 20%-30% of invasive breast cancers exhibit amplification of this receptor gene, and the presence of such overexpression is associated with distant metastases [24].
Autologous bone marrow transplantation has been offered to patients with breast cancer for the last decade, and studies suggest that median survival from transplantation was only 16 months with a five-year survival of 19 ± 4% [10]. However, the high rate of relapse remains a universally disturbing problem in this patient population, and thus the means of reducing the level of cancer cells in the marrow or peripheral blood reinfused have been sought.
The activity of ligand toxins is essentially determined by their ability to kill cells that express the target antigen [13]. All cell lines tested responded to the HRG mitogen with a rapid growth shown in liquid culture (Fig. 1
). Equally, all cell lines responded to the ligand toxin, although ZR-75-1 was the most sensitive, and MCF-7 the least. These results are in keeping with the levels of erbB receptor expression in these cells, with ZR-75-1 cells expressing elevated levels of erbB2, 3, and 4, whereas MCF-7 expresses moderate levels of erbB2 and 3 but has undetectable levels of erbB4 (A. de Fazio, P.W. Janes, and R.J. Fiddes, unpublished observations). The third cell line used, MDA-MB-361, expresses high levels of erbB2 but low levels of erb3 and 4 (A. de Fazio, P.W. Janes, and R.J. Fiddes, unpublished observations).
In the setting proposed here, with the recombinant molecule to be used as a purging agent, it was essential to test the activity of the molecule against the hemopoietic stem cells harvested for repopulation of the marrow. Three separate harvests of PBSCs were incubated with up to 2 nM ligand toxin or toxin moiety alone with no impact on the ability of the hemopoietic progenitors to form colonies (Table 1
). The above suggested that the HRG toxin may be useful as an agent to remove HRG receptor-positive cells from PBSC in breast cancer patients prior to autologous stem cell graft. We sought to emulate this process by the incubation of such cell lines in PBSC with and without ligand toxin. The results (Table 2
) demonstrate that HRG receptor-positive cells are indeed sensitive to the ligand-toxin in the stem cell harvest, whereas the toxin alone did not have the same effect. Interestingly, the milieu of the PBSC was associated with reduced efficacy of the ligand-toxin molecule against the tumor cell (Table 2
), which suggests other elements within the collection may sequester or in some other way inactivate the toxin.
Overexpression of erbB2 is thought to occur in 25%-30% of patients, and this correlates with a poor prognosis [24-26]. It is thought that adjuvant immunotherapy may assist those patients found to be erbB-positive. There is, however, considerable heterogeneity in the level of erbB2 expression within tumors, and this may extend to other members of this receptor family. In a study of breast cancer bone marrow metastases, Braun et al. noted that the level of pancytokeratin positive-erbB2 positive cells varied from 0% to 92%, but averaged only 42% [27]. Their conclusion that expression of tumor-associated cell-surface antigens on metastatic cancer cells in the bone marrow was heterogeneous and may limit the efficacy of monovalent immunotherapeutic strategies suggests such methodologies as described here may ultimately be applied as bivalent molecules or as adjuvants.
We have described the first use of a ligand toxin for the selective purging of autografts for breast cancer patients, and shown it to be effective in killing c-erbB-positive cells in a stem cell harvest environment. This suggests the removal of malignant clone from the autograft may be achieved by this approach.
 |
References
|
|---|
-
Baum M. Breast cancer. Clin Oncol 1982;1:647-655.
-
Weatherall DJ. Oxford Textbook of Medicine. Oxford: Oxford University Press, 1983:1203.
-
Myers SE, Williams SF. Cyclophosphamide mobilization of peripheral blood stem cells for use in autologous transplantation after high-dose chemotherapy: clinical results in patients with contaminated or hypocellular bone marrow. J Hematother 1992;1:27-33.[Medline]
-
Peters WP, Eder JP, Henner WD et al. High-dose combination alkylating agents with autologous bone marrow support: a phase 1 trial. J Clin Oncol 1986;4:646-654.[Abstract/Free Full Text]
-
Boring CC, Squires TS, Tong T. Cancer statistics, 1991. CA Cancer J Clin 1991;41:19-36.[Medline]
-
Harbeck N, Untch M, Pache L et al. Tumor cell detection in the bone marrow of breast cancer patients at primary therapy: results of a 3-year median follow-up. Br J Cancer 1994;69:566-571.[Medline]
-
Cote RJ, Rosen PP, Lesser ML et al. Prediction of early relapse in patients with operable breast cancer by detection of occult bone marrow micrometastases. J Clin Oncol 1991;9:1749-1756.[Abstract]
-
Slamon DJ, Clark GM, Wong SG et al. Human breast cancer: correlation of relapse and survival with amplification of the HER-2/neu oncogene. Science 1989;244:707-712.[Abstract/Free Full Text]
-
Yu D, Liu B, Tan M et al. Overexpression of c-erbB-2/neu in breast cancer cells confers increased resistance to Taxol via mdr-1-independent mechanisms. Oncogene 1996;13:1359-1365.[Medline]
-
Laport GF, Grad G, Grinblatt DL et al. High dose chemotherapy consolidation with autologous stem cell rescue in metastatic breast cancer: a 10 year experience. Bone Marrow Transplant 1998;21:127-132.[CrossRef][Medline]
-
Hammert LC, Ball ED. Purging marrow or peripheral blood stem cells for autografting. Curr Opin Hematol 1997;4:423-428.[Medline]
-
Berenson RJ, Shpall EJ, Auditore-Hargreaves K et al. Transplantation of CD34+ hematopoietic progenitor cells. Cancer Invest 1996;14:589-596.[Medline]
-
Pastan I, FitzGerald D. Recombinant toxins for cancer treatment. Science 1991;254:1173-1177.[Abstract/Free Full Text]
-
Skrepnik N, Araya JC, Qian Z et al. Effects of anti-erbB-2 (HER-2/neu) recombinant oncotoxin AR209 on human non-small cell lung carcinoma grown orthotopically in athymic nude mice. Clin Cancer Res 1996;2:1851-1857.[Abstract]
-
Zhu ZP, Lewis GD, Carter P. Engineering high affinity humanized anti-p185 (HER2)/anti-CD3 bispecific F(ab')(2) for efficient lysis of p185 (HER2) overexpressing tumour cells. Int J Cancer 1995;62:319-324.[Medline]
-
Xu FJ, Boyer CM, Bae DS et al. The tyrosine kinase activity of the c-erbB-2 gene product (p185) is required for growth inhibition by anti-p185 antibodies but not for the cytotoxicity of an anti-p185-ricin-A chain immunotoxin. Int J Cancer 1994;59:242-247.[Medline]
-
Deshane J, Siegal GP, Alvarez RD et al. Targeted tumour killing via an intracellular antibody against erbB-2. J Clin Invest 1995;96:2980-2989.
-
Wels W, Beerli R, Hellman P et al. EGF receptor and p185erbB-2 specific single chain antibody toxins differ in their cell killing activity on tumour cells expressing both receptor proteins. Int J Cancer 1995;60:137-144.[Medline]
-
Holmes WE, Sliwkowski MX, Akita RW et al. Identification of heregulin, a specific activator of p185erbB2. Science 1992;256:1205-1210.[Abstract/Free Full Text]
-
Chaudhary VK, FitzGerald DJ, Adhya S et al. Activity of a recombinant fusion protein between transforming growth factor type alpha and Pseudomonas toxin. Proc Natl Acad Sci USA 1987;84:4538-4542.[Abstract/Free Full Text]
-
Fiddes RJ, Janes PW, Sanderson GM et al. Heregulin (HRG)-induced mitogenic signaling and cytotoxic activity of a HRG-PE40 ligand toxin in human breast cancer cells. Cell Growth Differ 1995;6:1567-1577.[Abstract]
-
Plowman GD, Green JM, Culouscou JM et al. Heregulin induces tyrosine phosphorylation of HER4/p180erbB4. Nature 1993;366:473-475.[CrossRef][Medline]
-
King CR, Kasprzyk PG, Fischer PH et al. Preclinical testing of an anti-erbB2 recombinant toxin. Breast Cancer Res Treat 1996;38:19-25.[CrossRef][Medline]
-
Brandt B, Roetger A, Heidl S et al. Isolation of blood-borne epithelium-derived c-erbB-2 oncoprotein-positive clustered cells from the peripheral blood of breast cancer patients. Int J Cancer 1998;76:824-828.[CrossRef][Medline]
-
Schmidt M, Hynes NE, Groner B et al. A bivalent single chain antibody-toxin specific for ErbB-2 and the EGF receptor. Int J Cancer 1996;65:538-546.[CrossRef][Medline]
-
Newman SP, Bates NP, Vernimmen D et al. Cofactor competition between ligand bound oestrogen receptor and an intron 1 enhancer leads to oestrogen repression of ERBB2 expression in breast cancer. Oncogene 2000;19:490-497.[CrossRef][Medline]
-
Braun S, Hepp F, Sommer HL et al. Tumor-antigen heterogeneity of disseminated breast cancer cells: implications for immunotherapy of minimal residual disease. Int J Cancer 1999;84:1-5.[CrossRef][Medline]
Received June 23, 2000;
accepted for publication August 7, 2000.
This article has been cited by other articles:

|
 |

|
 |
 
C. M. Thirukkumaran, J. M. Luider, D. A. Stewart, T. Cheng, S. M. Lupichuk, M. J. Nodwell, J. A. Russell, I. A. Auer, and D. G. Morris
Reovirus oncolysis as a novel purging strategy for autologous stem cell transplantation
Blood,
July 1, 2003;
102(1):
377 - 387.
[Abstract]
[Full Text]
[PDF]
|
 |
|
