Detection of the MDR PGlycoprotein Expression and Function

Eugene Mechetner Summary

Acquired and intrinsic multidrug resistance is the major reason for the failure of anticancer chemotherapy. The most important component of clinical multidrug resistance is mediated by P-glycoprotein (Pgp), an ABC transporter encoded by the MDR1 gene and expressed on the membrane of tumor and normal cells. Sensitive and reproducible detection of Pgp expression and function are critical for the development of new MDR1 drugs and clinical protocols aimed at modulating Pgp-mediated multidrug resistance. The most commonly used methods for detecting Pgp distribution and functional activity have major flaws when used in routine clinical diagnostics. In this chapter, we describe and compare these techniques and introduce a new method for simultaneous detection of Pgp expression and function—the UIC2 Shift assay.

Key Words: MDR1; P-glycoprotein; drug resistance; flow cytometry.

1. Introduction

1.1. MDR1 P-glycoprotein and the Multifactorial Nature of Tumor Drug Resistance

Selection of mammalian cells in vitro and in vivo for resistance to cytotoxic agents frequently results in the development of cross-resistance to many other drugs that share little structural similarity with the primary selective agent and act at different intracellular targets. This phenomenon was termed multidrug resistance (1,2).

Multidrug resistance in tumor cells involves several distinct molecular mechanisms. These mechanisms include ATP-dependent efflux of cytotoxic agents by membrane protein pumps encoded by the MDR or MRP genes,

Lung Resistance Protein (LRP)-related drug sequestration in intracellular vesicles, DNA alterations caused by overexpression of topoisomerase II a, enhanced detoxification of alkylating agents by glutathione-linked enzyme systems, and other yet poorly understood multidrug resistance phenomena. In contrast, individual drug resistance to cancer chemotherapy drugs is caused by elevated levels of enzymes involved in intracellular drug metabolism. This group of drug resistance-associated enzymes includes such enzymes as thymidylate synthase conferring resistance to 5-fluorouracil, and O6-methylguanine DNA methyltransferase contributing to clinical resistance to alkylating agents (3).

The best characterized and understood molecular mechanism of multidrug resistance is membrane efflux mediated by P-glycoprotein (Pgp) encoded by the MDR1 gene. Pgp expression permits tumor cells to evade the cytotoxic effects of several chemotherapy agents, thus contributing to clinical resistance to natural product-based chemotherapeutics, including taxanes, anthracyclines, vInca alkaloids, podophyllotoxins, and camptothecins (1-3). Pgp functions as an ATP-dependent transporter of a structurally diverse group of substances from the cytosol and/or plasma membrane to the extracellular space (4). The MDR1 gene is a member of a large superfamily of integral membrane proteins known as the ATP-binding cassette (ABC) (5). ABC also includes several other multidrug resistance membrane proteins, such as the MRP superfamily, BCRP and SPGP gene products, the product of the cystic fibrosis gene, the TAP peptide transporters encoded by the major histocompatibility complex genes, etc. Overall, more than 100 ABC transporters have been identified in humans, other vertebrates, yeast, insects, plants, and bacteria (6). However, in addition to MDR1, only two ABC genes, MRP and BCRP, contribute to the multidrug hesitance phenotype in clinical situations (6,7). The clinical significance of MDR1 over-expression has been substantiated by studies showing (1) significant increase in Pgp expression at relapse, and (2) presence of functional Pgp on normal cells that constitutively express this efflux pump and give rise to the malignant clones. Thus, both acquired and intrinsic MDR1 expression mechanisms contribute to clinical drug resistance (3,8).

Pgp is constitutively expressed in a variety of human organs and tissues, such as the brush border of renal proximal tubules, the biliary surface of hepatocytes, the apical surface of mucosal intestinal cells, the adrenal cortex, and capillary endothelial cells of the blood-brain and blood-testis barriers (9,10). Pgp expression has also been reported on the membrane of human hematopoietic stem cells and peripheral blood lymphocytes (11,12). Based on the tissue distribution, it has long been proposed that Pgp plays a role in protecting cells and tissues against toxic xenobiotics and metabolites by active extrusion of these compounds into bile, urine, or intestinal lumen, and by preventing accumulation in critical organs, such as the brain (13). The function of Pgp may also include the transmembrane transport of cellular products, such as steroid hormones and growth factors (14-16). Pgp expression in epithelial cells has been shown to be associated with the transmembrane trafficking of ATP, regulation of membrane potentials and cell volume regulation via ATP-dependent, chloride-selective anion channels. Because Pgp is able to transport some natural steroid hormones, a physiological role of this protein in steroid secretion has been suggested (17,18).

Although Pgp-mediated drug efflux appears to be the major component of clinical resistance to chemotherapy, there is accumulating evidence that any of the multidrug resistance mechanisms, or their combinations, can be associated with poor clinical outcomes in different cancer types. It has been also shown that substrate specificities of different multidrug resistance mechanisms may overlap (19,20). Therefore, a sensitive tumor may become resistant to a certain drug via two or more molecular pathways (e.g., doxorubicin in the case of MDR1- and MRP-positive cells). It is because of the multifactorial nature of clinical drug resistance that it is essential to differentially detect individual components of the overall tumor drug response (19-21).

Furthermore, the pharmacokinetics of drugs—MDR1 transport substrates can be significantly altered by the ubiquitous expression of MDR1 and other membrane pumps in normal tissues. For example, because high Pgp expression levels in the blood-brain barrier prevent MDR1 substrates from entering the brain, current chemotherapy protocols for brain tumors do not include several powerful anticancer drugs (such as doxorubicin, taxanes, vinblastine) (22). Additionally, the ubiquitous expression of MDR1 and other ABC transporters results in significant side effects due to drug interference with normal physiological functions of multidrug resistance pumps in normal tissues, such as kidney, liver, gastrointestinal tract, adrenal, brain, hematopoietic stem cells. Although various adverse side effects of several small molecule MDR1 modulators were observed in several Phase I and I/II clinical trials (23), their maximum-tolerated doses yielded serum levels considerably less than the levels needed to effectively inhibit MDR1 function (24). It has been suggested that significant adverse side effects of Cyclosporine A, a transplantation drug and a known Pgp substrate, are associated with its interference with Pgp physiological functions (3,24).

Thus, functional Pgp expressed in tumor cells, as well as normal organs and tissues, can seriously reduce anticancer effects of chemotherapy and alter pharma-cokinetics, thereby negatively affecting the therapeutic window of many commonly used drugs—Pgp efflux substrates. Thus, efficient and selective detection of Pgp expression and function is essential for developing new drugs, conducting clinical trials, and designing new meaningful treatment protocols in cancer patients.

1.2. Existing Methods for Detecting Pgp Expression and Function

It has been shown in several solid tumor and leukemia/lymphoma systems that MDR1 expression levels determined on the protein or mRNA level correlated with in vitro and in vivo drug response and clinical outcomes in cancer patients (19-22). However, a direct evidence of the causative relationship between clinical resistance and any of the known multidrug resistance mechanisms has not been provided. The major reason for this failure is that despite apparent clinical value of Pgp assays for assessing tumor multidrug resistance and extensive efforts to increase their sensitivity and specificity, of existing Pgp tests exhibit significant deficiencies, particularly when used on human-derived specimens in clinical laboratory routine (Table 1).

Although MDR1 testing based on highly sensitive and quantitative molecular biology techniques (reverse transcription-polymerase chain reaction [RT-PCR], Northern blotting) has been widely used in research laboratories, its use for clinical testing was hindered by several essential flaws. First, because every tumor sample contains both tumor and normal (e.g., connective tissue, blood vessels, and infiltrating hematopoietic cells) cells and because total specimen lysates are utilized for RNA isolation, it is impossible to discern specific Pgp expression in tumor vs normal cells. It has been shown that Pgp expression levels in nonmalignant stromal cells and peripheral blood lymphocytes (PBLs) infiltrating the tumor may be comparable to that in the tumor itself (12). Therefore, the assay may generate falsely positive data because of the presence of nontumor Pgp-expressing cells in tissue specimens. Second, molecular biology based techniques are limited by the necessity of using total tumor lysates as the primary source of mRNA, and therefore no morphological analysis of individual Pgp-positive vs Pgpnegative cells can be performed. Finally, the use of this approach leads to ignoring posttranslational modifications of Pgp and does not determine Pgp expression and functional status levels on the protein level. It has been demonstrated that MDR1 mRNA data obtained using RT-PCR-based techniques does not necessarily correlate with the expression levels and functional activity of Pgp on the cell membrane (25).

Pgp detection on the protein level can be carried out by Western blotting, immunohistochemistry (IHC), and flow cytometry using monoclonal antibodies (MAbs) against Pgp. These immunological techniques are potentially more accurate than molecular biology-based testing. For example, IHC allows a pathologist to discern tumor cells from infiltrating PBLs and stromal elements and assess tumor cell heterogeneity in tissue sections. However, Western blotting and conventional immunostaining methods share two common problems: low sensitivity and inability to detect Pgp functional activity.

Table 1

Pgp Detection Methods

Method

End-point

Disadvantages

Advantages

RNA-based (RT-PCR, MDR1 RNA Northern blotting) expression

Highly sensitive

No specific Pgp detection in tumor vs normal cells No cell morphology and biomarker detection No detection of Pgp function

Western blotting Pgp expression

Pgp molecular weight verification

Low sensitivity No specific Pgp detection in tumor vs normal cells No cell morphology and biomarker detection No detection of Pgp function

Immunostaining (flow cytometry, immunohisto-chemistry (IHC))

Pgp expression

Efflux assays Pgp function

(fluorescent dyes in flow cytometry, radioactive Pgp substrates in vitro)

Specific detection of Pgp and biomarkers (flow cytometry) or morphology (IHC) in tumor cells High sensitivity detection of Pgp functional activity

Relatively low sensitivity; no information on Pgp functional activity

No specific detection of Pgp vs other efflux mechanisms; relatively low reproducibility

Efflux assays Pgp function

(fluorescent dyes in flow cytometry, radioactive Pgp substrates in vitro)

In vitro cytoxicity

Pgp function Detection of Pgp

Low sensitivity and

assays

functional activity

reproducibility

Clinically relevant

No specific detection

drug testing

of Pgp vs other

efflux mechnisms

No cell morphology

and biomarker

detection

UIC2 shift assay

Pgp expression,

Higher sensitivity Not applicable

(flow cytometry)

Pgp function

vs conventional flow in IHC

cytometry

Highly specific detection

of Pgp expression and

function can be

combined with efflux and

immunostaining assays

Furthermore, MAbs reactive with internal Pgp epitopes (such as JSB-1 and, particularly, C219) are cross-reactive with other intracellular proteins and, therefore, their specificity to Pgp is limited (26-28). On the other hand, MAbs against Pgp conformational extracellular epitopes are highly specific but sensitive to IHC fixatives and, therefore, can be used almost exclusively on living tumor cells in flow cytometry. Because of relatively low MDR1 gene expression levels in tumor cells (1,21,22), the sensitivity of highly specific monoclonal antibodies against extracellular epitopes of Pgp (such as MRK16, UIC2, 4E3, and MM12.10) has sometimes been too poor to provide accurate assessment of Pgp expression (25). Thus, the sensitivity of both groups of anti-Pgp MAbs is limited, and many cases of low Pgp expressing tumors remain un- or under detected. The problem of low sensitivity is further compounded by cases of membrane expression of nonfunctional Pgp and the inability of MAB-based immunostaining techniques to detect Pgp-mediated transport activity (29,30).

Two groups of tests are currently used to determine Pgp functional activity in tumor cells. While flow cytometry based assays measure direct efflux of fluorescent dyes (e.g., Rh 123, DiOC2, Calcein AM) or drugs (e.g., dauno-rubicin) from MDR1-positive tumor cells under physiological conditions, cytotoxicity-based tests analyze tumor cells exposed in vitro to toxic drugs that are MDR1 efflux substrates. The percentage of cell death or cell growth inhibition is usually used as the end-point in this latter group. The first major deficiency of dye efflux and in vitro cytotoxicity based assays is that they are not suitable for the analysis of cell morphology. Another major limitation is a result of the multifactorial nature of drug resistance, i.e., the expression of other membrane efflux pumps and other mechanisms of drug resistance (25,31). For example, it has been shown that Rho 123, a Pgp-transported fluorescent dye that is commonly used for the detection of Pgp-mediated efflux, is also a substrate for the MRP1 membrane pump (32), thereby preventing accurate discrimination between functional activities of these multi-drug resistance pumps. The overlap in substrate specificities exhibited by various members of the ABC transport system and other extrusion mechanisms (e.g., LRP) is the major reason why it has been so difficult to identify natural compounds that are truly Pgp specific. In addition, these tests are time/cost consuming and highly operator-dependent, making it difficult to standardize functional MDR1 testing between different labs.

Thus, new assays are needed to accurately and specifically detect MDR1 expression and function. In this chapter, we will describe a new method for Pgp detection, the UIC2 Shift assay, which can be utilized to quantitatively assess Pgp expression and function in a range of experimental configurations.

1.3. The UIC2 MAb and the UIC2 Shift Assay

It had been proven difficult to generate MAbs against external epitopes of Pgp because only 7% of the protein is exposed on the cell surface (1). Despite multiple efforts, only a few MAbs reactive with extracellular epitopes have been developed. In 1992, we described a mouse IgG2a MAb, UIC2, that recognized human Pgp on the cell surface of normal and malignant cells (33). We also found that UIC2 inhibited the efflux of Rh123, DiOC2, and other fluorescent dyes and drugs transported by Pgp from MDR cells and reversed the in vitro resistance of MDR1 cells to Pgp-transported drugs in a dose-dependent fashion. Schinkel et al. (12) demonstrated that observed variation in binding specificity between UIC2 vs other anti-Pgp MAbs paralleled the reported functional differences in the ability of these antibodies to inhibit Pgp-mediated drug efflux.

Our epitope mapping studies revealed the conformational nature of the UIC2 epitope and suggested that it consists of (1) several paratopes that belong to the predicted Pgp extracellular domains, and (2) one Pgp domain located inside the cell membrane. Some of the identified UIC2 paratopes completely or partially overlapped with MRK16 paratopes, end these two MAbs strongly competed for Pgp binding in flow cytometry experiments (Mechetner and Roninson, unpublished). However, only UIC2 was capable of efficient inhibition of Pgpmediated efflux function. Therefore, we hypothesized that conformational transitions that are associated with Pgp efflux activity result in "opening up" the Pgp epitope and permit UIC2 binding to the otherwise unavailable intramembrane stretch of amino acids. This hypothesis lead to the prediction that coincubation of MDR1 cells with Pgp transport drugs under physiological conditions should result in altered (increased or decreased) binding of UIC2. In our subsequent flow cytometry experiments, we indeed observed enhanced (up to 10- to 15-fold) binding of UIC2 (but not MRK16) to the cell surface when Pgp-positive cells were incubated at 37°C with multiple Pgp substrates. Both UIC2 affinity and the number of UIC2-binding sites on the cell surface were drastically increased in drug-treated MDR1 cells. This phenomenon, defined as an increase in UIC2 immunoreactivity under physiological conditions with functional vs nonfunctional Pgp, was termed "UIC2 shift" (34,35).

Our subsequent studies on MDR1 cell lines expressing wild-type and mutant forms of Pgp confirmed that UIC2 shift was associated with conformational changes of functioning Pgp and showed that it was related to its ATPase status. This data suggested the existence of at least two possible functional conformations. One of these conformations ("closed") is characterized by lower UIC2 reactivity (revealed by conventional UIC2 staining, i.e., without exposure to a Pgp substrate), binding of two ATP molecules to Pgp, and complete absence of the efflux function. The second ("open") conformation is detectable by UIC2 staining in the presence of a Pgp substrate under physiological conditions, generates UIC2 shift, is linked to the dissociation of ATP from the ATP-binding domains, and involves active Pgp-mediated drug efflux. This model was further supported by data generated in different experimental systems using UIC2 as a probe for Pgp conformation and efflux activities (see ref. 35 for review).

Based on the UIC2 shift phenomenon, we developed and validated a new flow cytometry test, the UIC2 Shift assay, which allows for simultaneous quantitative detection of Pgp expression and function in human MDR1 cells (35). This test is based on the increased reactivity of an anti-Pgp MAb, UIC2, in the presence of Pgp-transported compounds at physiologic conditions (Fig. 1).

1.4. Applications of the UIC2 Shift Assay

0 0

Post a comment