Chemotherapy is used in management of many veterinary cancer patients. For most types of hematopoietic neoplasia, such as canine lymphoma, it is the mainstay of therapy. Lymphoma is the most common hematopoietic canine cancer with an estimated annual incidence of 30 cases/100,000 dogs. Combination chemotherapy is the current standard of care and provides a typical survival of 10 to 12 months. Development of clinical MDR is conferred in part by increased expression and activity of MRPs and is a main reason for treatment failure.1–4
Multidrug resistance–associated proteins are a group of ATP-dependent plasma membrane transporters responsible for cellular drug efflux. The cellular membrane location of these pumps in normal cells suggests an evolutionary role of xenobiotic resistance; however, overexpression of these proteins by neoplastic cells may convey a phenotype of MDR.5 In the classic sense, MDR refers to neoplastic cells that are resistant to structurally and functionally unrelated chemotherapeutic drugs that are hydrophobic, amphipathic, natural products (anthracyclines, vinca alkaloids, taxanes, and epipodophyllotoxins).5
The most widely known and studied efflux pump is P-gp, a 170-kd transmembrane protein on plasma membranes. It is expressed in high concentrations on epithelial cells of the liver, renal tubules, and most secretory organs; the capillary endothelium of the bloodbrain and blood-testis barriers; and nodal lymphocytes. P-glycoprotein extrudes large, hydrophobic, uncharged or positively charged molecules. Known P-gp substrates, in addition to aforementioned chemotherapeutics, are digoxin, opiates, polycyclic aromatic hydrocarbons, technetium, sestamibi, and rhodamine 123.6–11 Other more recently studied members of this transporter family are the group of MRP1, 2, and 3. Multidrug resistance–associated protein 1 is a glutathione-dependent transporter expressed at low basal concentrations in epithelial, endocrine, and muscle tissue in humans as well as in peripheral blood. It transports negatively charged compounds and glutathione or glucuronate conjugates and sequesters some substrates in intracellular vesicles. Multidrug resistance–associated protein 1 is overexpressed in some human cancers, including carcinomas of lung, breast, and bladder, and in acute myeloid leukemia.6,7,11 Canine MRP1 has been characterized via cell transfection studies that reveal 92% homology to human MRP1 and has been identified in a canine osteosarcoma cell line.10,12
Limited veterinary clinical investigations into MDR exist. Immunohistochemical analysis did not detect P-gp expression in normal canine nodal lymphocytes,9 whereas other work that used similar methods revealed P-gp expression in 15% to 20% of the malignant lymphocyte population in untreated canine lymphomas.13 Increases in P-gp expression detected via IHC analysis of nodal lymphocytes obtained via nodal biopsies during combination chemotherapy were associated with a poorer outcome in 22 dogs with lymphoma.14 Another study2 that used western blotting found that only 1 of 30 naïve lymphoma samples expressed P-gp but found detectable concentrations of P-gp in 3 of 8 biopsy samples taken from dogs that had become resistant to chemotherapy.
The exact role the MRP family (MRP1, 2, or 3) plays in development of MDR is also undefined. One study found no relationship between MRP1 expression and response to single-agent doxorubicin therapy in 11 dogs with canine lymphoma.15 Multidrug resistance– associated protein 1 may play a role in conveying drug resistance to other groups of chemotherapeutics, such as alkylating agents; however, few studies of this have been performed, and the association is controversial because of MRP1's ability to remove inactive metabolites and active drug.10,12
The objective of the study reported here was to determine efficacy of flow cytometry for evaluation of expression and activity of P-gp and MRP efflux pumps and to characterize and correlate expression and activity of these membrane proteins in grossly normal canine nodal lymphocytes.
Materials and Methods
Isolation of lymphocytes—Popliteal lymph nodes were collected from 21 clinically normal dogs immediately following euthanasia for population control. Lymph nodes were placed in cold RPMI 1640 medium (minus cations) plus 10% fetal bovine seruma (ie, cell media). Lymph nodes were stripped of subcutaneous tissues, minced, and passed through a cell isolation strainer with 70-μm nylon mesh. Cells were resuspended in FACS buffer (Dulbecco PBS solutionb plus 1% bovine serum albuminc) and washed twice via cold centrifugation at 200 × g for 10 minutes. Cell quantification and viability were performed by use of a hemocytometer and trypan blue exclusion, respectively.
Functional assays—Aliquots of 0.5 mL of media containing 0.5 × 106 cells in pump substrates with or without inhibitor were incubated for 30 minutes at 37°C and 5% CO2. Functional activity of P-gp was detected by use of rhodamine 123d (0.2 μg/mL) with and without inhibitors PSC833e (2 μg/mL) and verapamilf (0.1mM); these inhibitors were used individually and in combination. Multidrug resistance–associated protein activity was determined by use of CFDAg (1MM) with and without inhibitors probenecidh (2.5mM) and MK-571i (2mM); these inhibitors were used individually and in combination. Following the first incubation, samples were centrifuged and then underwent 2 wash steps with FACS buffer and centrifugation. Cell pellets were resuspended in cell media, and inhibitors were added again to the samples. A second 2-hour incubation was performed followed by 2 wash steps. Samples were suspended in FACS buffer for analysis.11,16–21
Protein expression assays—Lymphocytes (150,000 cells in 100 ML of FACS buffer) were incubated with 1.0 ML of rat anti-mouse FcG III/II receptor monoclonal antibodyj for 15 minutes at 4°C in a 96-well plate in the dark to prevent nonspecific antibody binding to cell surface Fc receptors. All samples underwent 2 wash procedures with FACS buffer and 4-minute cold centrifugation at 200 × g. Following the wash procedure, all samples were incubated with 200 ML of a fixative and permeation solutionk for 20 minutes at 4°C in the dark, resulting in cellular permeabilization necessary for staining intracellular proteins. Following 2 wash procedures with a provided wash buffer,k samples were incubated at 4°C in the dark for 1 hour with either 4 μg of mouse anti-human MRP1 antibodyl/million cells, 5 μg of mouse anti-human P-gp antibodym/million cells or 0.5 μg of antibody isotype control/million cells. Finally, incubation at 4°C in the dark for 20 minutes with fluorescein isothyocyanate conjugated rat anti-mouse IgG1 or IgG2a monoclonal antibodiesn was performed.6,9,12,13,20
Flow cytometric analysis—A flow cytometero with excitation wavelengths of 488 and 647 nm was used for analysis of all samples. Lymphocytes were selected by electronically gating on a dual parameter dot plot of forward and side scatter. Ten thousand events were measured per sample. Raw data of fluorescence caused by substrate dyes were measured in channel 1, analyzed by use of flow cytometric software,p and reported as a histogram with logarithmic scale.
Data analysis—Results for functional and expression assays were reported as medians of the activity ratio of fluorescence, as follows18,20:
Activity = Median channel fluorescence of cells and 1 or both inhibitors/Median channel fluorescence of cells without inhibitors
Expression = Median channel fluorescence of cells + pump specific antibody/Median channel fluorescence of cells + isotype control antibody
Expression of P-gp and MRP1 proteins was correlated to function with Pearson correlation for normally distributed data, with significance defined as a value of P ≥ 0.8.
Results
Functional assays—Histograms depicting cellular fluorescence on a logarithmic scale were obtained for each sample. When lymphocytes were incubated with fluorescent substrate and no inhibitor, a lower overall fluorescence was observed because of cellular pump activity and subsequent dye efflux (Figures 1 and 2). In samples containing both substrate and inhibitor, dye efflux was prevented, resulting in a greater overall fluorescence. In comparison to MK-571, probenecid appeared to be a much more potent inhibitor of MRP function, as indicated by a higher mean activity of 20.4 versus an activity of 3.35 when MK-571 was used (Figure 3). A slight synergistic effect appeared to be present with an activity of 21.8 when both inhibitors of MRP were used. The relative potency of P-gp inhibitors verapamil and PSC833 appeared more similar with activities of 1.22 and 1.27, respectively. These inhibitors did not appear to have a synergistic effect; activity was 1.27 when both P-gp inhibitors were used.
Representative flow cytometric histograms (events or cells counted vs fluorescence intensity [FL1-H]) of lymphocytes after incubation with CFDA (white area), compared with lymphocytes incubated without a fluorescent dye (autofluorescent control; gray area). Multidrug resistance–associated protein–mediated efflux of CFDA without inhibition results in less fluorescent cells that shift to the left (A), compared with inhibited efflux of CFDA in B (CFDA plus MRP inhibitor probenecid), C (CFDA plus MRP inhibitor MK-571), and D (CFDA plus probenecid and MK-571). M1 and M2 = Gates. Cells within gate M2 were evaluated to determine MRP activity.
Citation: American Journal of Veterinary Research 69, 10; 10.2460/ajvr.69.10.1310
Representative flow cytometric histograms (events or cells counted vs FL1-H) of lymphocytes after incubation with rhodamine 123 (white area), compared with lymphocytes incubated without a fluorescent dye (autofluorescent control; gray area). (A)—P-glycoprotein–mediated efflux of rhodamine without inhibition results in less fluorescent cells that shift to the left, compared with inhibited efflux of rhodamine in B (rhodamine plus P-gp inhibitor verapamil), C (rhodamine plus P-gp inhibitor PSC833), and D (rhodamine plus verapamil and PSC833). Cells within gate M2 were evaluated to determine the P-gp activity. See Figure 1 for remainder of key.
Citation: American Journal of Veterinary Research 69, 10; 10.2460/ajvr.69.10.1310
Activity of MRP (black bars) and P-gp (gray bars) in canine nodal lymphocytes determined by use of a fluorescent dye, CFDA, or rhodamine 123 and 1 or 2 pump-specific inhibitors, probenecid (P), MK-571 (M), verapamil (V), and PSC833 (PSC). Each bar represents mean ± SEM activity (n = 21).
Citation: American Journal of Veterinary Research 69, 10; 10.2460/ajvr.69.10.1310
Protein expression assays—Two distinct fluorescent peaks were present for each expression assay, which indicated that no nonspecific protein binding was present (Figures 4 and 5). P-glycoprotein expression was evaluated in all 21 samples and was detected on 99.5% of gated cells. Multidrug resistance–associated protein 1 expression was evaluated in 13 samples and was detected on 96.5% of gated cells. The antibodies used in the protein expression assays were specific for MRP1 only and not for MRP2 or 3. Multidrug resistance–associated protein 1 expression on nodal lymphocytes appeared to be lower, compared with P-gp (Figure 6). No significant correlation between expression and activity was found.
Representative flow cytometric histogram (events or cells counted vs FL1-H) of canine lymphocytes after indirect staining with anti-human MRP1 antibodies (white area, black line), compared with lymphocytes incubated with an antibody isotype control (IgG1 [gray area]). See Figure 1 for key.
Citation: American Journal of Veterinary Research 69, 10; 10.2460/ajvr.69.10.1310
Representative flow cytometric histograms (events or cells counted vs FL1-H) of canine lymphocytes after indirect staining with anti-human P-gp antibodies (white area, black line), compared with lymphocytes incubated with an antibody isotype control (IgG2a [gray area]). See Figure 1 for key.
Citation: American Journal of Veterinary Research 69, 10; 10.2460/ajvr.69.10.1310
Expression of MRP1 (black bar) and P-gp (gray bar) in canine nodal lymphocytes determined by use of pump-specific antibodies MRPm6 and C494, respectively. Each bar represents mean ± SEM activity (n = 13 for MRP1; n = 21 for P-gp).
Citation: American Journal of Veterinary Research 69, 10; 10.2460/ajvr.69.10.1310
Discussion
Objectives of this study were to determine clinical efficacy of flow cytometry for expression and activity of P-gp and MRP1 efflux pumps and characterize the correlation, if any, between expression and function of these membrane proteins on normal canine nodal lymphocytes. Results indicated that flow cytometry was a valid test for determining the activity of P-gp and MRP efflux pumps. No strong correlation was observed between protein activity and expression. Even if P-gp, MRP1, or other efflux pumps are expressed on canine lymphocyte surfaces, this does not automatically dictate their activity. This could be caused by individual animal–related variability, which supports the need for a larger study population. In the present study, antibodies specific only to MRP1 were used; a stronger correlation between MRP expression and function may have been detected if antibodies for MRP2 and 3 had also been used. Furthermore, this study evaluated nodal lymphocytes as a whole population and did not take into account possible differences between pump expression of T versus B cells. Because whole lymph nodes were used, including germinal centers, a possibility arises that different levels of expression and function may exist between mature and immature lymphocytes, which may also have contributed to the lack of correlation.
Two previous studies17,20 evaluating P-gp and MRP1 expression and function in human peripheral blood mononuclear cells and leukemic cell lines had differing results; 1 revealed a positive correlation, and the other revealed no correlation. The differing results of those studies and the present study indicated that the relationship between protein expression and function and the best diagnostic approach to evaluate these factors remains undefined. The correlation between expression and activity may have more meaning when evaluated in the context of dogs affected with lymphoproliferative diseases; 1 or both variables may predict remission duration or survival in dogs treated with combination chemotherapy.
The effectiveness of flow cytometry at evaluating pump function and the role of P-gp in drug resistance of clinical canine lymphoma cases have been studied previously. Two studiesq,r evaluated use of flow cytometry for evaluation of P-gp function in canine lymphoma cells. Rhodamine 123 and cyclosporine were used as the pump substrate and inhibitor, respectively, in both studies to evaluate P-gp activity. One studyq evaluated P-gp activity in canine lymphoma cells, canine acute lymphoblastic leukemia cells, and 2 canine lymphoid tumor cell lines after treatment with vincristine, cyclophosphamide, doxorubicin, L-asparaginase, and prednisolone. Another studyr evaluated P-gp activity in cases of canine lymphoma before and after use of combination chemotherapy. Dogs with demonstrable P-gp function had a poorer outcome with chemotherapy, with overall response rates in the P-gp–positive groups of 13% and 27%, compared with 66% and 71% in the P-gp–negative groups.
The present study accounted for the influence of P-gp and MRP on the MDR phenotype in normal canine nodal lymphocytes. Normal canine nodal lymphocytes were evaluated prior to abnormal lymphocytes for assay optimization and characterization of normal canine lymphoid cells, a necessary step before exploring these variables in lymphoid disorders. Two inhibitors for each pump were used to improve assay sensitivity and specificity and more fully define the contribution of each membrane protein, as suggested in a previous report.17
The inhibitors used in this study to assess P-gp and MRP1, 2, and 3 activity were verapamil, PSC833, probenecid, and MK-571. Verapamil is a P-gp modulator that works via competitive inhibition. PSC833, a well-studied second-generation P-gp modulator, is a nonimmunosuppressive derivative of cyclosporine D with 10 to 20 times greater activity than cyclosporine A. Second-generation P-gp modulators are competitive inhibitors as well, but with greater potency and less toxicity than original modulators.5 Probenecid is an inhibitor of MRP activity through blockade of organic ion transport.22 The leukotriene D4 receptor antagonist MK-571 inhibits MRP by binding tightly to the putative binding site.23 In the present study, probenecid, verapamil, and PSC833 performed well in the functional analysis; however, the MK-571 performed less optimally than the others. Reasons for this may include the need for a higher concentration of inhibitor because modulation of drug resistance by MK-571 is dose dependent.24 A combined or synergistic effect on MRP was minimally evident for probenecid and MK-571 and not evident on P-gp for verapamil and PSC833. The lack of significant synergism between probenecid and MK-571 may be attributable to the poor performance of MK-571 in the present study. Unlike probenecid and MK-571, which have different mechanisms of action for pump inhibition, verapamil and PSC833 are both competitive inhibitors of P-gp, with PSC833 having 20 times the activity level; this may not allow for a synergistic effect to be detected when both inhibitors are used simultaneously. However, this depends on the respective affinity each agent has for the receptor and whether they both bind to the exact same point on the membrane protein.
Immunohistochemical analysis,9,13,14 western blotting,2 and real-time reverse transcriptase quantitative PCR8 have been used to evaluate P-gp protein or gene expression in canine tissues. Use of these methodologies can be expensive, time-consuming, and require invasive sample acquisition. Furthermore, only protein or gene expression can be evaluated with these techniques, not the functional activity of the protein pump. The flow cytometric techniques used in the present study evaluated protein expression and functional activity of the protein pumps in approximately 6 hours. Although samples in this study were obtained from whole lymph nodes, the authors have found multiple fine-needle aspirates from enlarged peripheral lymph nodes sufficient to obtain adequate cells to perform these flow cytometric assays, negating the need for anesthesia.
Although a previous study9 evaluating normal canine lymph nodes was unable to detect expression of P-gp via IHC methods, in the present study, flow cytometry was successfully used to evaluate expression of both P-gp and MRP1 in normal canine nodal lymphocytes as present on 99.95% and 96.5% of gated cells. Flow cytometry may be a more sensitive means of detecting membrane protein expression than IHC. The discrepancy detected between the fluorescence value of P-gp antibody and MRP1 antibody in this study may best be explained simply by lack of antibody binding sites; that is, less MRP1 present on lymphocyte membranes. No evidence of cross-reactivity was detected in these assays. Antibodies chosen for this experiment have been validated in species other than dogs; however, these same antibodies have been used in previous immunohistochemical work investigating canine lymphoma.6,9,12,13,20
Flow cytometry may be useful for future evaluations of MRPs. More studies on MDR are necessary to determine its prognostic value for canine lymphoma, understand if it plays a role in poor durability of remission associated with T-cell lymphoma in dogs, aid in development of novel chemotherapeutic protocols, elucidate the role of MDR in other veterinary tumors and nonneoplastic lymphoid disorders, and evaluate MDR inhibitors as therapeutic options.25,26
Abbreviations
| CFDA | 5(6)-carboxyfluorescein diacetate |
| FACS | Fluorescent-activated cell sorting |
| IHC | Immunohistochemistry |
| MDR | Multidrug resistance |
| MRP | Multidrug resistance–associated protein |
| P-gp | P-glycoprotein |
BioWhittaker, Walkersville, Md.
BioWhittaker, Walkersville, Md.
Sigma-Aldrich, Atlanta, Ga.
Sigma-Aldrich, Atlanta, Ga.
Novartis/Pharma Inc, Bazel, Switzerland.
Sigma-Aldrich, Atlanta, Ga.
Sigma-Aldrich, Atlanta, Ga.
Sigma-Aldrich, Atlanta, Ga.
Caymen, Ann Arbor, Mich.
BD Pharmingen, San Jose, Calif.
BD Cytofix/Cytoperm kit, BD Pharmingen, San Jose, Calif.
MRPm6, Signet (Covance), Princeton, NJ.
C494, Signet (Covance), Princeton, NJ.
Fluorescein isothyocyanate–conjugated rat anti-mouse IgG1 and IgG2a monoclonal antibodies, BD Pharmingen, San Jose, Calif.
FACS Vantage SE flow cytometer, Becton-Dickinson, San Jose, Calif.
Cell Quest software, Becton-Dickinson, San Jose, Calif.
Yamanaka I, Nakamura N, Yamada Y, et al. Functional detection of multidrug resistance in canine lymphoma and leukemia cells by flow-cytometry (abstr), in Proceedings. 20th Am Coll Vet Intern Med Forum 2002;161.
Kishida Y. Evaluation of P-glycoprotein function in tumor cells and its relation to drug resistance in canine lymphoma (abstr), in Proceedings. 26th Annu Meet Vet Cancer Soc 2006;49.
References
- 1.
MacEwen EG. Spontaneous tumors in dogs and cats: models for the study of cancer biology and treatment. Cancer Metastasis Rev 1990;9:125–136.
- 2.↑
Moore A, Leveille C, Reimann K, et al. The expression of P-glycoprotein in canine lymphoma and its association with multidrug resistance. Cancer Invest 1995;13:475–479.
- 3.
Page RL, Hughes C, Huyan S, et al. Modulation of P-glycoprotein-mediated doxorubicin resistance in canine cell lines. Anticancer Res 2000;20:3533–3538.
- 4.
Vail DM. Hematopoietic tumors. In: Ettinger SJ, Feldman EC, eds. Textbook of veterinary internal medicine. 5th ed. Philadelphia: WB Saunders Co, 2000;507–523.
- 5.↑
Thomas H, Coley H. Overcoming multidrug resistance in cancer: an update on the clinical strategy of inhibiting P-glycoprotein. Cancer Causes Control 2003;10:159–165.
- 6.
Conrad S, Vietelhaus A, Orzechowski A, et al. Sequencing and tissue distribution of the canine MRP2 gene compared with MRP1 and MDR1. Toxicology 2000;156:81–91.
- 7.
Cui Y, Konig J, Buchholz J, et al. Drug resistance and ATP-dependent conjugate transport mediated by the apical multidrug resistance protein, MRP2, permanently expressed in human and canine cells. Mol Pharmacol 1999;55:929–937.
- 8.↑
Culmsee K, Gruber A, Von Samson-Himmerlstjerna G, et al. Quantification of MDR-1 gene expression in canine tissues by real-time reverse transcriptase quantitative polymerase chain reaction. Res Vet Sci 2004;77:223–229.
- 9.↑
Ginn PE. Immunohistochemical detection of P-glycoprotein in formalin-fixed and paraffin-embedded normal and neoplastic canine tissues. Vet Pathol 1996;33:533–541.
- 10.
Mealey KL, Bentjen S, Gay J, et al. Dexamethasone treatment of a canine, but not human, tumor cell line increases chemoresistance independent of P-glycoprotein and multidrug resistance-related protein expression. Vet Comp Oncol 2003;1:67–75.
- 11.
van der Kolk D, de Vries E, van Putten W, et al. P-glycoprotein and multidrug resistance protein activities in relation to treatment outcome in acute myeloid leukemia. Clin Cancer Res 2000;6:3205–3214.
- 12.
Ma L, Pratt S, Cao J, et al. Identification and characterization of the canine multidrug resistance-associated protein. Mol Cancer Ther 2002;1:1335–1342.
- 13.↑
Lee J, Hughes C, Fine R, et al. P-glycoprotein expression in canine lymphoma: a relevant, intermediate model of multidrug resistance. Cancer 1996;77:1892–1898.
- 14.↑
Bergman P, Ogilvie G, Powers B. Monoclonal antibody C219. Immunohistochemistry against P-glycoprotein: sequential analysis and predictive ability in dogs with lymphoma. J Vet Intern Med 1996;10:354–359.
- 15.↑
Steingold SF, Sharp N, McGahan M, et al. Characterization of canine MDR1 mRNA: its abundance in drug resistance cell lines and in vivo. Anticancer Res 1998;18:393–400.
- 16.
Brugger D, Herbart H, Gekeler V, et al. Functional analysis of P-glycoprotein and multidrug resistance associated protein related multidrug resistance in AML-blasts. Leuk Res 1999;23:467–475.
- 17.↑
Dogan A, Legrand O, Faussat A-M, et al. Evaluation and comparison of MRP1 activity with three fluorescent dyes and three modulators in leukemia cell lines. Leuk Res 2004;28:619–622.
- 18.
Egashira M, Kawamata N, Sugimoto K, et al. P-glycoprotein expression on normal and abnormal expanded natural killer cells and inhibition of p-glycoprotein function with cyclosporine A and its analogue, PSC833. Blood 1999;93:559–606.
- 19.
Laupeze B, Amiot L, Courtois A, et al. Use of the anionic dye carboxy-2', 7'-dichlorofluorescein for sensitive flow cytometric detection of multidrug resistance-associated protein activity. Int J Oncol 1999;15:571–576.
- 20.
Meaden E, Hoggard P, Khoo S, et al. Determination of P-gp and MRP1 expression and function in peripheral blood mononuclear cells in vivo. J Immunol Methods 2002;262:159–165.
- 21.
Rhaaijimakers H, van den Bosch G, Boezeman J, et al. Single-cell image analysis to assess ABC-transporter-mediated efflux in highly purified hematopoietic progenitors. Cytometry 2002;49:135–142.
- 22.↑
Barrand MA, Bagrij T, Neo SY. Multidrug resistance-associated protein: a protein distinct from P-glycoprotein involved in cytotoxic drug expulsion. Gen Pharmacol 1997;28:639–645.
- 23.↑
Williamson G, Aeberli I, Miguet L, et al. Interaction of positional isomers of quercetin glucuronides with the transporter ABCC2 (cMOAT, MRP2). Drug Metab Dispos 2007;35:1262–1268.
- 24.↑
Gekeler V, Ise W, Sanders KH, et al. The leukotriene LTD4 receptor antagonist MK571 specifically modulates MRP associated multidrug resistance. Biochem Biophys Res Commun 1995;208:345–352.
- 25.
Fan TM. Lymphoma updates. Vet Clin North Am Small Anim Pract 2003;33:455–471.
- 26.
Ruslander D, Gebhard D, Tompkins M, et al. Immunophenotypic characterization of canine lymphoproliferative disorders. In Vivo 1997;11:169–172.
