Pawłowski et al.

BMC Veterinary Research 2013, 9:119

http://www.biomedcentral.com/1746-6148/9/119

RESEARCH ARTICLE Open Access

Expression and role of PGP, BCRP, MRP1 and

MRP3 in multidrug resistance of canine mammary
cancer cells
Karol M Pawłowski1,2, Joanna Mucha1, Kinga Majchrzak3, Tomasz Motyl1 and Magdalena Król1*

Abstract
Background: In both women and female dogs, the most prevalent type of malignant neoplasm is the
spontaneous mammary tumor. In dogs, half of these are malignant. The treatment of choice for the canine patients
is surgical mastectomy. Unfortunately, it often fails in high-risk, locally invasive mammary tumors as of during the
time of the surgery the micro-metastases are present. Moreover, there are neither large studies conducting to prove
of the benefit from the chemotherapy in dogs nor established chemotherapy treatment protocols available.
Additionally, the effectiveness of each individual chemotherapeutic agent and drug resistance of canine mammary
cancer have not yet been characterized. That has become the aim of our study, to assess the expression of PGP,
BCRP, MRP1 and MRP3 in canine mammary cancer cell lines and to investigate their role in cancer resistance to
vinblastine, cisplatin and cyclophosphamide with using RNAi approach.
Results: The results suggested that in canine mammary cancer, the vinblastine efflux was mediated by PGP and
MRP1 proteins, cisplatin efflux was mediated by all four examined efflux pumps (PGP, BCRP, MRP1 and MRP3),
whereas cyclophosphamide resistance was related to BCRP activity. RNAi silencing of these efflux pumps
significantly decreased IC50 doses of the examined drugs in canine mammary carcinoma cells.
Conclusions: Our results have indicated the treatment of cells involving use of the siRNA targeting efflux pumps
could be a beneficial approach in the future.
Keywords: Multidrug resistance, Chemotherapy, Canine mammary cancer, Vinblastine, Cisplatin, Cyclophosphamide,
PGP, BCRP, MRP1, MRP3

Background unsatisfactory as in case of invasive cancer at the time of

In both women and female dogs, the most prevalent surgery the micro-metastases are present. Moreover,
type of malignant neoplasm is the spontaneous mam- clinical studies reveal that after regional mastectomy
mary tumor. In dogs, half of these are malignant. There 77% of dogs developed a new malignancy [4]. There are
is a threefold incidence in the canine, with middle aged, neither large studies conducting to prove to the benefit
non-spayed female dogs being the most affected [1,2]. of chemotherapy in dogs nor established chemotherapy
The early ovariectomy is thought to reduce the risk of treatment protocols available. Moreover, the effective-
mammary cancer development [3], however, the high ness of each individual chemotherapeutic agent and drug
morbidity and mortality rate due to ineffective treatment resistance of canine mammary cancer have not yet been
strategies makes this problem as being still actual. The characterized. The available data is conflicting, possibly
treatment of choice for the canine patients is surgical due to the small number of studies performed, or the
mastectomy. Unfortunately, it is often recognized - insufficient number of dogs used in these studies. Some
veterinarians adapt human chemotherapy protocols;
* Correspondence: [email protected] however these treatment modalities are often unsatisfac-
1
Department of Physiological Sciences, Faculty of Veterinary Medicine, tory due to similarities observed in recurrence time, time
Warsaw University of Life Sciences - WULS, Nowoursynowska 159, 02-776,
Warsaw, Poland
to metastasis and overall survival [5-7].
Full list of author information is available at the end of the article

© 2013 Pawłowski et al.; licensee BioMed Central Ltd. This is an Open Access article distributed under the terms of the
Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use,
distribution, and reproduction in any medium, provided the original work is properly cited.
Pawłowski et al. BMC Veterinary Research 2013, 9:119 Page 2 of 10
http://www.biomedcentral.com/1746-6148/9/119

The resistance to a specific cytotoxic drug is ensured under optimal conditions: in RPMI-1640 medium
by the activity of efflux pumps that belong to the ABC- enriched with 10% (v/v) heat-inactivated fetal bovine
transporters super family. The most important in breast serum (FBS), penicillin-streptomycin (50 iU mL–1), and
cancer are: PGP (P-glycoprotein), BCRP (Brest Cancer fungizone (2.5 mg mL–1) (reagents obtained from Sigma
Resistance Protein), MRP1 (Multridrug resistance Pro- Aldrich, USA), in an atmosphere of 5% CO2 and 95%
tein 1) and MRP3 (Multidrug Resistance Protein 3). For humidified air at 37°C.
example, in human cancer over-expression of PGP is
linked to resistance to vinca alkaloids, anthracyclines, siRNA transfection
epipodophyllotoxins and also tubulin polymerizing drugs The siRNA transfection procedure used in canine mam-
whereas MRP1 is responsible for resistance to vincristine mary cancer cells was described in details in our pre-
and vinblastine, methothrexate, anthracyclines, etopo- viously published studies [19-21]. The cell density,
side, paclitaxel and irinotecan [8,9]. Unfortunately, ex- transfection reagent toxicity and transfection efficacy
pression of efflux pumps and their specific substrates were optimized according to the procedure described in
have not been assessed in dogs yet. There is just one our previous manuscript: [19]. The canine (Canis lupus
document published showing that BCRP can fully pro- familiaris) pgp, bcrp, mrp1 and mrp3 sequences were
tect canine mammary cancer cells against doxorubicin obtained from Gene Bank with accession numbers:
[10]. NM_001003215, DQ222459.1, NM_001002971, XM_54
That is why the aim of our study was to assess expres- 8204.2, respectively. The siRNA duplexes were designed
sion of PGP, BCRP, MRP1 and MRP3 in five canine by http://www.sigmaaldrich.com/life-science/custom-olig
mammary cancer cell lines and to investigate their role os/sirna-oligos/sirna-design-service.html. The results were
in cancer resistance to vinblastine, cisplatin and cyclo- confirmed using two independent algorithms: Dharmacon
phosphamide which are used in breast cancer treatment (OligoWalk) and Ambion and two duplexes were used for
[11,12]. The examined drugs were selected based on the further experiments (obtained from Sigma Aldrich)
available data in the field of veterinary oncology. They (Table 1). For each gene silencing the mixture of both du-
presented positive results of cisplatin and cyclopho- plexes was used (30 pmol + 30 pmol). All the experiments
sphamide treatment in canine mammary cancer [5,7] with transfected cells were conducted 24–48 hrs after the
whereas vinblastine was selected due to its general toler- transfection.
ance [13]. Moreover, these anticancer agents belong to
various groups, thus we used them to observe efflux Reverse-transcriptase qPCR
pumps activity during treatment with drugs from Total RNA was isolated using a Total RNA kit (A&A
different groups. Biotechnology, Poland) according to the manufacturer‘s
We also treated canine mammary cancer cells with protocol. Isolated RNA samples were dissolved in
siRNAs specific for these efflux pumps to knock-down RNase-free water. The quantity of isolated RNA was
their expression. Treatment of cancer cells by using measured using NanoDrop (NanoDrop Technologies,
classical inhibitors of efflux pumps often fails [8,9], thus USA). The mean concentration of RNA was 173 ng/μl,
we used a novel approach to reduce their activity. and A260/280 ratio was between 1.8 and 2.0. The sam-
Our results have pointed out that treatment of patients ples with adequate amounts of RNA were treated with
with the supportive use of the siRNAs specific for efflux DNaseI to eliminate DNA contamination. The samples
pumps may help with improving results of chemother- were subsequently purified using RNeasy MiniElute
apy. Despite the useful function of RNAi therapeutics Cleanup Kit (Qiagen). Finally RNA samples were ana-
for disease treatment, it, yet - still requires the develop- lyzed on a BioAnalyzer (Agilent, USA) to measure final
ment of clinically suitable, safe and effective drug deliv- RNA quality and integrity. Only RNA with RIN (RNA
ery vehicles, there are some promising data from Integrity Number) > 9 was used for the further analyses.
ongoing clinical trials that give hope for their practical Primers used to detect the expression of pgp, bcrp, mrp1
application in the future [14]. and mrp3 genes were designed using PRIMER3 software
(free on-line access) and checked using Oligo Calculator
Methods (free on-line access) and Primer-Blast (NCBI database).
Cell lines The used sequences are listed in Table 2. rps19 and hprt
The cell lines used for the study have been previously used genes were used as non-regulated references for the
in other published research [14-18]. Two canine mam- normalization of target gene expression [22,23]. Quanti-
mary adenocarcinoma cell lines (CMT-W1, CMT-W2), tative RT-PCR was performed using fluorogenic SYBR
anaplastic cancer cell line (P114), simple carcinoma cell Green and the Sequence Detection System, Fast 7500
line (CMT-U27) and spindle-cell mammary tumor cell (Applied Biosystems). Data analysis was carried out
line (CMT-U309) had been examined. Cells were cultured using the 7500 Fast System SDS Software Version
Pawłowski et al. BMC Veterinary Research 2013, 9:119 Page 3 of 10
http://www.biomedcentral.com/1746-6148/9/119

Table 1 Used siRNA sequences

Gene GenBank ID First strand Second strand
pgp NM_001003215 CGAACUGUUGUUUCUUUGA[dT][dT] UCAAAGAAACAACAGUUCG[dT][dT]
CUUCCGAACUGUUGUUUCU[dT][dT] AGAAACAACAGUUCGGAAG[dT][dT]
bcrp DQ222459.1 GCCCAGGAGUCAAUGUAAC[dT][dT] GUUACAUUGACUCCUGGGC[dT][dT]
CGAAUAAUACCUGUAGCUA[dT][dT] UAGCUACAGGUAUUAUUCG[dT][dT]
mrp1 NM_001002971 GAAAGAGGCUCCCUGGCAA[dT][dT] UUGCCAGGGAGCCUCUUUC[dT][dT]
GGAGUAUUCAGAAACGGAG[dT][dT] CUCCGUUUCUGAAUACUCC[dT][dT]
mrp3 XM_548204.2 GGCUAUGACGGAGAGCCAA[dT][dT] UUGGCUCUCCGUCAUAGCC[dT][dT]
UGUCUACGCUGCCUUGGGA[dT][dT] UCCCAAGGCAGCGUAGACA[dT][dT]
siRNA sequences used to pgp, bcrp, mrp1 and mrp3 genes silencing in canine mammary cancer cell lines. The canine (Canis lupus familiaris) pgp, bcrp, mrp1 and
mrp3 sequences were obtained from Gene Bank with accession numbers: NM_001003215, DQ222459.1, NM_001002971, XM_548204.2, respectively. The siRNA duplexes
were designed by http://www.sigmaaldrich.com/life-science/custom-oligos/sirna-oligos/sirna-design-service.html. The results were confirmed using two independent
algorithms: Dharmacon (OligoWalk) and Ambion.

1.4.0.25 (Applied Biosystems, USA). The results were software (Becton Dickinson, USA) to determine mean
analyzed using comparative Ct method [24]. Relative fluorescence intensity values. This procedure has been
transcript abundance of the gene equals ΔCt values published before [25]. Results represent the average of at
(ΔCt = Ctreference – Cttarget). Relative changes in tran- least three independent experiments.
script were calculated as ΔΔCt values (ΔΔCt = 2-ΔCt). The overlay histogram was created using Flowing
The experiment was conducted three times. Software (Turku University, Finland), available at www.
flowingsoftware.com.

Flow cytometry determination of rhodamine-123

accumulation Cell viability assay (MTT-assay) and IC50 determination
Examined cancer cells (5 × 106 cells per ml) (control cells Cell viability (metabolic activity of viable cells) was
and cells treated with pgp, bcrp, mrp1 and mrp3 –specific quantified by MTT assay. Cells were seeded into 96-well
siRNAs) were incubated for 1 h at 37°C in the 1 mmol of plate (Nunc Inc., Denmark) at the density which ensured
rhodamine-123 (obtained from Sigma Aldrich, Germany). their 50-70% confluence at the day of experiment and: 1)
After the incubation time, cells were washed twice, re- treated with vinblastine (at the following concentrations:
suspended in ice cold PBS and kept at 4°C in the dark 12.3, 36.9, 61.5, 123, 369, 615, 2 460, 4 920, 7 380 nmol),
until analysis in the flow cytometer. At least 50 000 cells cisplatin (at the following concentrations: 46.5, 83, 465,
per sample were counted and analyzed by flow cytometry 830, 4 650, 8 300, 46 500, 83 000 166 000 nmol) or
(BD FACS Aria II, Becton Dickinson, USA). Cells shown cyclophosphamide (at the following concentrations: 425,
in forward scatter and side scatter were gated and 950, 1 900, 3 800, 19 000, 38 000 76 000 nmol) (all drugs
acquired through the fluorescence channel. The amount obtained from Sigma Aldrich, Germany), 2) transfected
of fluorescence was plotted as a histogram within the gate. with pgp, bcrp, mrp1 and mrp3 -specific siRNA and then
Data acquisition was performed using BD FACS Diva treated with the anticancer drugs as given above, 3)

Table 2 Primers used for RT-qPCR

Gene symbol Forward primer Reverse primer Optimum annealing Optimum annealing
temp. (°C) time (sec)
pgp GCTTAACACCCGGCTCACAGAC TAAGAAAGCGGCACCAATAGAAAT 72 10
bcrp GAGCTCCTTGTGGTTGAGAA AGGTGATGGTCATGAGGAGA 72 10
mrp1 TTTACTTTTCCCTCGTGCTG TATTCAGGGACCAAAGGTCA 72 45
mrp3 AGGATGGACCTGATGACAGA AACTTGGGAATCAGGAGACC 72 30
hprt AGCTTGCTGGTGAAAAGGAC TTATAGTCAAGGGCATATCC 59 6
rps19 CCTTCCTCAAAAAGTCTGGG GTTCTCATCGTAGGGAGCAAG 61 10
Primer sequences used in this study and their annealing optimal temperature and time. The mRNA sequences of key genes were obtained from NCBI database.
Primers were designed using PRIMER3 software (free on-line access) and checked using Oligo Calculator (free on-line access) and Primer-Blast (NCBI database).
Primers sequences are listed in Table 1. hprt and rps19 genes were used as non-regulated reference genes for normalization of target gene expression [22,23].
Pawłowski et al. BMC Veterinary Research 2013, 9:119 Page 4 of 10
http://www.biomedcentral.com/1746-6148/9/119

treated with non-coding siRNA (scrambled control) and regression analysis. Determination of IC50 test has been
cyctotoxic drugs as given above. These concentrations of used. The p-value <0.05 was recognized as significant,
anticancer drugs were determined in course of our whereas, p-value <0.01 and p-value <0.001 as highly signifi-
preliminary experiments (data not shown) and chosen cant. The data was expressed as means +/− S.D. For
as appropriate to determine IC50 doses in all of the molarity calculations, the Molarity Calculator (GraphPad,
examined cell lines. USA) on-line platform was used.
Then, cells were incubated in 0.5 mg/ml tetrazolium
salt MTT diluted in phenol red-free RPMI 1640 medium
(Sigma Aldrich) for 4 hrs at 37°C. To complete Results
solubilization of the formazan crystals, 100 μl of DMSO PGP, BCRP, MRP1 and MRP3 are expressed in canine
(Dimethyl sulfoxide, Sigma Aldrich) was added to mammary cancer cells and their expression level changes
each well. Cells viability was quantified by measuring due to anticancer drug in vitro treatment
photometric absorbance at 570 nm in multi well plate RT-qPCR analysis revealed that all of the examined cell
reader Infinite 200 PRO Tecan™ (TECAN, Mannedorf, lines express pgp, bcrp, mrp1 and mrp3 (Table 3). The
Switzerland). All the samples were examined in tripli- pgp expression ranged between 3.50 in CMT-U27 cell
cate, each experiment was conducted seven times line and 17.85 in CMT-W1 cell line. The bcrp expression
(n = 21). Based on these results cytotoxicity was deter- was the lowest in CMT-U309 cell lines (11.86) whereas
mined. It was expressed as a mean percentage decrease the highest in CMT-W2 cell line (36.00). The highest
relative to unexposed control ± S.D. Control values were expression of mrp1 was observed in CMT-W2 cell line
set at 100% viability. Cytotoxicity data were fitted to a (24.02) whereas the lowest expression was noticed in
sigmoidal curve and a four parameter logistic model was CMT-U27 cell line (7.50). CMT-W2 cell line exposed
used to calculate IC50, which is the concentration of the lowest expression of mrp3 (11.35). The highest
agent which reduces cell growth by 50% under the expression of mrp3 was detected in CMT-U309 cell line
experimental conditions (increasing apoptosis, necrosis (23.28).
or causing block in cell cycle). This analysis was Treatment of cancer cells using scrambled siRNA did
performed using GraphPad Prism 5.0 (San Diego, USA). not cause any significant effect on examined gene
This method of IC50 analysis has been previously expression (Table 3), whereas treatment of these cells
published [26]. with pgp, bcrp, mrp1 or mrp3 -specific siRNAs caused
significant decrease in transcript level of targeted genes
Apoptosis assay (the mostly significant effect was noticed in case of pgp
The Annexin V-FITC and propidium iodide (PI) dual and mrp1) (Table 3). The efficacy of silencing reaction
staining was applied for apoptosis analysis. Control cells was very high, as there was observed 88-99% decrease in
(1) and cells treated with (2) anticancer drug at IC50 dose pgp transcript level, 14-100% in bcrp, 94-99% in mrp1
and (3) transfected with pgp, bcrp, mrp1 and mrp3- and 22-100% in mrp3 expression (Table 3).
specific siRNA moreover treated with anticancer drug at Interestingly, we observed that pgp expression increased
IC50 dose were harvested by trypsinization. These cells, as significantly in all the examined cell lines after viblastine
well as the cells floating in medium (RPMI 1640 con- and cisplatin treatment at IC50 doses, whereas decreased
taining 10% FBS) were stained with an Annexin V Kit after cyclophosphamide treatment at IC50 dose (Table 3).
(Becton Dickinson, USA) according to the manufacturer’s Vinblastine treatment increased mrp1 expression in all of
protocol. The cells then were analyzed by flow cytometry the examined cell lines with exception of CMT-W2 cell
(BD FACS Aria II, Becton Dickinson, USA) within 1 hr line where its expression increased also after treatment
after staining. Early apoptotic cells with exposed phos- with cyclophosphamide at IC50 dose (Table 3). In case of
phatidylserine but intact cell membranes bound to bcrp, we observed increased tendency in its expression
Annexin V-FITC however excluded PI. Cells in late apop- after cyclophosphamide treatment and decreased in its
totic stages were labeled with both Annexin V-FITC and expression after vinblastine treatment at IC50 dose
PI, whereas necrotic cells were labeled with PI only. All (Table 3). We are as well able to point out the increase in
samples were assayed in triplicate. The experiment was mrp3 expression in all the examined cell lines after
conducted at least twice. treatment with cisplatin at IC50 dose (Table 3).

Statistical analysis Cancer cell treatment with pgp, bcrp, mrp1 and mrp3 –
The analysis for statistical purposes was conducted using specific siRNA decreases efflux pumps activity enhancing
Prism version 5.00 software (GraphPad Software, USA). rhodamine-123 accumulation
The one-way ANOVA and Tukey HSD (Honestly Signifi- The rhodamine-123 efflux assay was conducted to confirm
cant Difference) post-hoc test were applied as well as our results of gene expression silencing and their functional
Pawłowski et al. BMC Veterinary Research 2013, 9:119 Page 5 of 10
http://www.biomedcentral.com/1746-6148/9/119

Table 3 Expression of pgp, bcrp, mrp1 and mrp3 in various experimental conditions
Efflux Sample Relative expression in cell lines
pump
CMT-U27 CMT-U309 P114 CMT-W1 CMT-W2
PGP ctrl 3.50 9.20 5.00 17.85 16.42
pgp siRNA 0.45*** 0.10*** 0.50*** 0.80*** 0.16***
non-coding siRNA 3.44 9.10 5.10 7.750 16.31
vinblastine 9.50*** 16.20*** 14.70*** 19.56*** 21.52***
cisplatin 8.95*** 35.20*** 31.00*** 36.08*** 26.16***
cyclophosphamide 0.09*** 2.30** 0.22*** 0.08*** 2.87***
BCRP ctrl 19.84 11.86 23.52 23.70 36.00
bcrp siRNA 17.15** 2.77*** 2.77*** 10.46*** 0.00***
non-coding siRNA 19.00 11.20 22.90 23.00 36.18
vinblastine 14.84* 3.80*** 11.80*** 12.30*** 13.24*
cisplatin 30.50*** 22.16*** 40.34*** 36.57*** 37.23*
cyclophosphamide 23.32*** 33.53* 29.23*** 24.50* 41.72**
MRP1 ctrl 7.50 9.40 16.50 20.10 24.02
mrp1 siRNA 0.10*** 0.60*** 0.23*** 0.20*** 0.61***
non-coding siRNA 7.20 6.62 20.12 20.91 24.21
vinblastine 9.70*** 7.00*** 9.76*** 35.01* 30.20***
cisplatin 15.70*** 19.34** 21.84* 30.61*** 26.20**
cyclophosphamide 2.50** 0.09*** 3.32*** ↓0.40*** 6.04**
MRP3 ctrl 14.40 23.28 13.84 17.70 11.35
mrp3 siRNA 1.00*** 0.00*** 4.84*** 13.90** 7.15**
non-coding siRNA 13.87 21.99 13.09 17.10 10.99
vinblastine 14.20 21.97 13.12 17.50 9.78
cisplatin 15.60* 32.20*** 15.80* 21.30*** 14.09*
cyclophosphamide 14.99 24.09 14.25 18.01 11.72
Expression of efflux pumps in canine mammary cell lines in control conditions, after scrambled siRNA treatment and pgp, bcrp, mrp1 and mrp3 -specific siRNA
treatment as well as after treatment with vinblastine, cisplatin and cyclophosphamide at IC50 doses. p < 0.05 was marked as*, p < 0.01 was marked as **, and
p < 0.001 was marked as ***. One-way ANOVA followed by Tukey HSD post-hoc test were applied. Increase in expression was marked as red text, decreased in
expression was marked as green text, whereas no change in transcript level was marked as black text.

inhibition. It showed that after knock-down of pgp, bcrp, Cancer cell treatment with siRNA specific to efflux pumps
mrp1 and mrp3 expression level, the rhodamine-;123 accu- decreases drug-resistance
mulation significantly increased and that shows decreased The MTT assay of cells viability after treatment with
activity of the efflux pumps (Figure 1A, B). Treatment of increasing doses of anticancer drugs showed significant
examined cells with scrambled siRNA did not cause any variations in IC50 doses between the examined cell
significant effect (data not shown). We observed that lines. All IC50 doses are listed in Table 4. Vinblastine
siRNA treatment of CMT-U27 caused significant but slight IC50 dose in CMT-U27 cell line was 1 588 nmol,
effect on rhodamine-123 accumulation (mean fluorescence whereas after PGP and MRP1 knock-down it was only
related to rhodamine-123 accumulation in control cells was 203 nmol and 144 nmol, respectively (Table 4,
97, whereas in siRNA treated cells it was between 108 and Figure 2A). In CMT-W2 cell line vinblastine IC50 was 9
110). In other cell lines the effect was higher (Figure 1), for 434 nmol, whereas after pgp and mrp1 silencing it was
example in P114 cell line the mean fluorescence related to 1264 nmol and 1248 nmol, respectively (Table 4). We
rhodamine-123 accumulation in control cells was only 87 showed that after pgp and mrp1 silencing IC50 was
(the lowest, comparing to other cell lines) however, after significantly lower in all the examined cell lines ranging
the treatment with bcrp-specific siRNA is was 128 203–1391 nmol and 144–1248 nmol, respectively
(Figure 1). The highest effect of rhodamine-123 accumula- (Table 4). Significant variations were also observed in
tion after siRNA treatment was observed in CMT-W2 cell case of IC50 doses of cisplatin – the lowest IC50 dose
line (148 after mrp3-specific siRNA treatment, 111 in was in CMT-U27 cell line (5 669 nmol) (Figure 2B),
control conditions). whereas the highest was in P114 cell line (42 944 nmol)
Pawłowski et al. BMC Veterinary Research 2013, 9:119 Page 6 of 10
http://www.biomedcentral.com/1746-6148/9/119

Figure 1 Activity of PGP, BCRP, MRP1 and MRP3. A.) Graph of mean fluorescence related to rhodamine-123 accumulation inside cancer cells
obtained using FACS Aria II (Becton Dickinson, USA). Error bars refer to S.D. p < 0.05 was marked as *. One-way ANOVA followed by Tukey HSD
post-hoc test was applied. B.) Representative overlay histograms of rhodamine-123 accumulation in CMT-U309 canine mammary cancer cells
(control, markered as green), and treated with pgp -specific siRNA (markerd as red). The overlay histograms were created using Flowing Software
(Turku University, Finland), www.flowingsoftware.com.

(Table 4). Cisplatin IC50 was significantly decreased in Cancer cell treatment with anticancer drug and siRNA
all the examined cell lines after silencing of pgp, bcrp, specific to efflux pumps increases apoptosis
mrp1 and mrp3 (Table 4). Regarding cyclophosphamide, The Annexin V assay was used to examine the influence of
we also noticed significant variations in IC50 values. It pgp, bcrp, mrp1 and mrp3 expression knock-down on apop-
was the lowest in CMT-W1 cell line (8 679 nmol), when tosis induced by vinblastine, cisplatin and cyclophosphamide
on the other hand the highest in CMT-W2 cell line (comparing to anticancer drug given as a single agent at
(25 698 nmol) (Table 4). Treatment of cells with bcrp- IC50 dose). Due to any changes in cells viability based on
specific siRNA decreased cyclophosphamide IC50 in all transfection procedure were observed in MTT assay, this
of the examined cell lines (ranging between 765 and kind of control has been omitted in apoptosis assay. We used
5 705 nmol) (Table 4, Figure 2C). similar approach in our previous study [19].

Table 4 IC50 doses of anticancer drugs in control conditions and after efflux pumps silencing
Anticancer drug Condition IC50 in examined cell lines [nmol]
CMT-U27 CMT-U309 P114 CMT-W1 CMT-W2
vinblastine ctrl 1 588 2 472 2 744 8 072 9 434
pgp siRNA 203*** 464*** 208*** 1 391*** 1 264***
bcrp siRNA 1 497 2 417 2 572 8 002 7 346
mrp1 siRNA 144*** 772** 756** 1 747*** 1 248***
mrp3 siRNA 1 330 2 406 2 561 7 433 8 482
cisplatin ctrl 5 669 12 034 42 944 41 288 9 536
pgp siRNA 2 203** 8 613** 5 000*** 32 663*** 6 707**
bcrp siRNA 3 300** 8 540** 17 699*** 11 228*** 8 177*
mrp1 siRNA 1 339** 10 090* 20 348*** 29 132*** 5 695**
mrp3 siRNA 303*** 7 301** 13 565*** 11 163*** 7 177*
cyclophosphamide ctrl 9 599 12 542 10 340 8 679 25 698
pgp siRNA 8 432 11 755 10 242 7 912 15 443
bcrp siRNA 799*** 2 086*** 5705** 2 463** 765***
mrp1 siRNA 8 762 11 307 9 955 7 247 7 000
mrp3 siRNA 8 802 12 278 10 167 6 932 1 130
IC50 doses [nmol] of anticancer drugs: vinblastine, cisplatin and cyclophosphamide in canine mammary cancer cell lines in control conditions and after silencing
of efflux pumps expression. p < 0.05 was marked as *, p < 0.01 was marked as **, and p < 0.001 was marked as ***. One-way ANOVA followed by Tukey HSD post-
hoc test was applied.
Pawłowski et al. BMC Veterinary Research 2013, 9:119 Page 7 of 10
http://www.biomedcentral.com/1746-6148/9/119

Figure 2 Survival curves of CMT-U27 cells. Survival curves of CMT-U27 cells treated with increasing doses of vinblastine (A), cisplatin (B) and
cyclophosphamide (C). IC50 doses were determined in control conditions and after pgp, bcrp, mrp1 and mrp3 silencing. IC50 was calculated using
GraphPad Prism 5.0 (San Diego, USA).

Annexin V analysis revealed that treatment of cells using transporters. The multidrug resistance is frequently associ-
pgp and mrp1 -specific siRNA together with vinblastine ated with over-expression of two or more membrane
(at IC50 dose) significantly increased number of apoptotic pumps that efflux anticancer drug from the cytoplasm.
cells, compared to apoptosis caused by anticancer drug This protects tumor cells against the drug effects and its
which was given as a single agent (at the IC50 dose as well) correlated molecular processes [8]. Expression of efflux
(Figure 3A). The most significant effect was pointed out in pumps is usually higher in tumors that originate from
CMT-W1 cell line: 25.35% and 17.45% increase in number tissues that normally show their expression and it is always
of apoptotic cells after pgp and mrp1 silencing and vinblast- higher in the tumors than in normal cells [8]. Unfortu-
ine treatment, respectively (Figure 3). Similarly, in nately, the expression and role of proteins that mediate
CMT-W2 cell line 21.78% and 20.88% increase in number drug resistance in canine cancer cells has not yet been
of apoptotic cells was shown after pgp and mrp1 silencing recognized. As increased efflux is such a significant con-
and vinblastine treatment, respectively (Figure 3). tributor to a multidrug resistance in cancer cells, current
In case of cisplatin treatment at IC50 dose, an increase research is aimed at blocking or inhibiting this specific
in number of apoptotic cells was noticed in each cell line mechanism. That is why the aim of our study was to as-
after silencing of all the efflux pumps: pgp, bcrp, mrp1 and sess expression of four the most important efflux pumps:
mrp3. The mostly significant result was observed in P114 PGP, BCRP, MRP1 and MRP3 in canine mammary cancer
cell line, where 40%, 41%, 42% and 40% increase in num- cells as well as to investigate their role in resistance to: cis-
ber of apoptotic cells (p < 0.001) was visible, respectively platin, cyclophosphamide and vinblastine. Because treat-
(Figure 3B, D). In CMT-U309 cell line the highly signifi- ment of cancer cells with the use of classical inhibitors of
cant (p < 0.001) effect on apoptosis was as well as in case efflux pumps often fails [8,9], we used a novel approach,
of cisplatin treatment in cells transfected with siRNA spe- that is: specific RNAi to knock-down their expression.
cific to: pgp, bcrp and mrp3 (26.45%, 26.18% and 21.05% Examination of each efflux pump expression in control
increase in number of apoptotic cells, respectively), conditions and after treatment with anticancer drug was
whereas significant effect (p < 0.05) was observed in cells also quite a novel approach. We observed significant differ-
transfected with mrp1 -specific siRNA (8.25% increase in ences in pgp transcript level in control conditions and after
number of apoptotic cells) (Figure 3B). treatment with anticancer drug. In all of the examined cell
In all the examined cell lines increase in number of lines the pgp expression increased due to vinblastine and
apoptotic cells related to cyclophosphamide treatment at cisplatin treatment whereas decreased due to cyclophos-
IC50 dose was undoubtedly visually recognized after bcrp phamide treatment (Table 3). It ought to be outlined that
knock-down (Figure 3C). Additionally, in CMT-W2 cell these results are in accordance with IC50 doses (Table 4).
line the significant increase in number of apoptotic cells In case of cisplatin treatment, expression of all four efflux
was observed due to cyclophosphamide treatment follow- pumps increased, and thus, any relationship between their
ing transfection with bcrp, mrp1 and mrp3 -specific siRNA transcript level and IC50 doses was very complicated to cal-
(30%, 12.1% and 21%, respectively). culate (Tables 3 and 4). We also observed that cyclophos-
phamide treatment caused increase in bcrp expression in all
Discussion of the examined cell lines. These results are in accordance
Cancer cells retain the important mechanism of self- with previous findings that cytotoxic drugs can initiate or
protection through the activity of multiple drug resistance increase expression of efflux pumps in cancer cells [27].
Pawłowski et al. BMC Veterinary Research 2013, 9:119 Page 8 of 10
http://www.biomedcentral.com/1746-6148/9/119

Figure 3 Apoptosis related to treatment with anticancer drugs. The number of apoptotic cells due to treatment with vinblastine
(A), cisplatin (B) and cyclophosphamide (C) at IC50 doses in control conditions and after efflux pumps silencing in canine mammary cancer cell
lines (CMT-U27, CMT-U309, P114, CMT-W1 and CMT-W2) assessed by the Annexin V/PI test (BD Bioscience, USA). All values were calculated versus
control conditions (spontaneous apoptosis in these cell lines). The numbers of apoptotic cells are represented as a percentages of Annexin
V-positive cells (obtained with FACS Aria II Becton Dickinson). The experiment has been conducted in three replicates. Error bars refer to S.D.
p < 0.05 was marked as *, p < 0.01 was marked as ** and p < 0.001 was marked as ***. One-way ANOVA followed by Tukey HSD post-hoc test was
applied. (D) The representative cytograms of P114 cell line double stained with Annexin V-FITC and propidium iodide (PI). The cytograms show
P114 cells in control conditions, after cisplatin treatment at IC50 dose and after pgp, bcrp, mrp1 and mrp3 -specific siRNA treatment followed by
cisplatin treatment at IC50 dose. On the cytograms are showed normal, early apoptotic, late apoptotic and necrotic cells. Left bottom quadrant
shows normal cells, top left quadrant shows necrotic cells (stained with PI only; damaged cell membrane but no phosphatydilserine exposure),
right bottom quadrant shows the early apoptotic cells (stained with Annexin V only; intact cell membrane) and top right quadrant shows cells in
late stage of apoptosis (stained with Annexin V and PI - phosphatydilserine exposure and damaged cell membrane).

Moreover, the expression level of these efflux pumps corre- cells pointed that it was significantly lower in all of them
lates with response to anticancer therapy and patient follow when compared to control cells (Table 4). In case of
up [28]. Based on our results, we have come up with the cyclophosphamide, bcrp siRNA treatment significantly
conclusion that examination of efflux pumps expression decreased IC50 dose in all of the examined cell lines
before initiation of chemotherapy could help to predict (Table 4).
response to anticancer drugs and could be helpful in assess- Comparison of IC50 doses of anticancer drugs given
ment of their proper doses. However, more studies in this in vitro and their maximum possible plasma concentra-
field are required. tions showed that in case of vinblastine the plasma con-
The RNAi gene silencing was highly successful (compar- centration was much lower than the IC50 doses [29]. It
ing to our previously published studies [19,21]), reaching means that the examined cancer cell lines are vinblastine-
even 100% of gene knock-down in case of bcrp in CMT- resistant and therefore this anticancer drug may be inef-
W2 cell line and mrp3 in case of CMT-U309 cell line. The fective in canine mammary cancer treatment. However,
results of gene knock-down were confirmed by the assay more studies in this field should be conducted. In case of
with rhodamine-123 efflux. To assess whether efflux cisplatin, the IC50 doses in the three examined cell lines
pumps silencing has an influence on cancer cell suscepti- (CMT-U27, CMT-U309 and CMT-W2) were lower than
bility to cytostatic drugs, the IC50 doses were determined the reachable plasma concentrations [30]. However, after
in transfected cells (Table 4, Figure 2). Our results showed bcrp, mrp1 or mrp3 knock-down, the cisplatin IC50 doses
that vinblastine IC50 doses significantly decreased in all were similar or lower than the maximum possible plasma
the examined cell lines after pgp and mrp1 silencing drug concentration in all of the examined cell lines. It
(Table 4). Examination of cisplatin IC50 in transfected means that knock-down of efflux pumps expression
Pawłowski et al. BMC Veterinary Research 2013, 9:119 Page 9 of 10
http://www.biomedcentral.com/1746-6148/9/119

during chemotherapy can reverse cancer drug-resistance. Received: 12 January 2013 Accepted: 13 June 2013
The IC50 doses of cyclophosphamide in all of the exam- Published: 17 June 2013

ined cell lines were significantly lower than its maximum

possible plasma concentration [31]. However, in all the References
1. MacEwen EG: Spontaneous tumors in dogs and cats: models for the study
cases given above, targeting of ABC efflux pumps using of cancer biology and treatment. Cancer Metastasis Rev 1990, 9:125–136.
siRNA significantly decreased IC50 doses of anticancer 2. Midsorp W: Tumors of the mammary gland. In Tumors in domestic animals.
drugs. These results were also confirmed by apoptosis Edited by Meuten DJ. Ames: Iowa State Press; 2002:575–606.
3. Misdorp W: Progestagens and mammary tumours in dogs and cats.
assay (Figure 3). Acta Endocrinol 1991, 125(Suppl. 1):27–31.
In summary: knock-down of efflux pumps expression 4. Stratmann N, Failing K, Richter A, Wehrend A: Mammary tumor recurrence
during chemotherapy could allow decreased doses of anti- in bitches after regional mastectomy. Vet Surg 2008, 37:82–86.
5. Simon D, Schoenrock D, Baumgartner W, Nolte I: Postoperative adjuvant
cancer drugs thereby reducing the risk of their side effects. treatment of invasive malignant mammary gland tumors in dogs with
However, more studies in this field are required as well as doxorubicin and docetaxel. J Vet Intern Med 2006, 20:1184–1190.
large clinical trials to prove this hypothesis. 6. Marconato L, Lorenzo RM, Abramo F, Ratto A, Zini E: Adjuvant gemcitabine
after removal of aggressive malignant mammary tumours in dogs.
Our research is a completely new approach in the field Vet Comp Oncol 2008, 6:90–101.
of veterinary oncology. The results show that targeting of 7. Karayannopoulou M, Kaldrymidou E, Constantinidis TC, Dessiris A: Adjuvant
ABC-efflux pumps using RNAi can be promising strategy post-operative chemotherapy in bitches with mammary cancer.
Transbound Emerg Dis 2001, 48:85–96.
in cancer treatment. However, these in vitro studies 8. Król M, Pawłowski KM, Majchrzak K, Szyszko K, Motyl T: Why chemotherapy
should be confirmed using in vivo animal models. can fail? Pol J Vet Sci 2010, 12:399–406.
9. Szyszko K, Pawłowski KM, Motyl T, Król M: Chemoresistance in cancer: not
just a humans problem. Med Wet 2011, 67(7):453–457.
Conclusions 10. Honscha KU, Schirmer A, Reischauer A, Schoon HA, Einspanier A, Gabel G:
The most important findings of our study are: (1) the ex- Expression of ABC-transport proteins in canine mammary cancer:
consequences for chemotherapy. Reprod Domest Anim 2009, 44:218–223.
pression level of efflux pump reflects the resistance to spe- 11. Roth BJ, Sledge GW, Williams SD, Meyer SC, Ansari R, Fisher WB:
cific anticancer drug thus it should be assessed before the Methotrexate, Vinblastine, Doxorubicin, and Cisplatin in metastatic
initiation of chemotherapy; (2) in canine mammary cancer breast cancer. Cancer 1991, 68:248–252.
12. Briest S, Stearns V: Chemotherapeutic strategies for advanced breast
PGP and MRP1 are responsible for vinblastine resistance, cancer. Oncology 2007, 21(11):1325–1335.
PGP, BCRP, MRP1 and MRP3 are cisplatin transporters 13. Ospovat I, Siegelmann-Danieli N, Grenader T, Hubert A, Hamburger T, Peretz
whereas BCRP is a transporter of cyclophosphamide; (3) T: Mitomycin C and vinblastine: an active regimen in previously treated
breast cancer patients. Tumori 2009, 95(6):683–686.
treatment of cells using siRNA targeting efflux pumps sig- 14. Whitehead KA, Langer R, Anderson DG: Knocking down barriers: advances
nificantly increases cancer susceptibility to anticancer drug in siRNA delivery. Nat Rev Drug Discovery 2009, 8:129–138.
and allows to decrease the effective dose of anticancer drug. 15. Król M, Pawłowski KM, Majchrzak K, Gajewska M, Majewska A, Motyl T: Global
gene expression profiles of canine macrophages and canine mammary
cancer cells grown as a co-culture in vitro. BMC Vet Res 2012, 8:16.
Competing interests
16. Król M, Pawłowski KM, Skierski J, Rao NAS, Hellmen E, Mol JA, Motyl T:
The authors declare that they have no competing interests.
Transcriptomic profile of two canine mammary cancer cell lines with
different proliferative and anti-apoptotic potential. J Physiol Pharmacol
Authors’ contributions 2009, 60:95–106.
KP research design, experimental design, IC50 determination, FACS analyses, 17. Król M, Pawłowski KM, Skierski J, Turowski P, Majewska A, Polańska J, Ugorski
rhodamine-123 accumulation assay, siRNA transfections, JM apoptosis M, Morty RE, Motyl T: Transcriptomic “portraits” of canine mammary
analysis, KM apoptosis analysis, TM manuscript preparation, MK research cancer cell lines with various phenotype. J Appl Genet 2010, 51:169–183.
design, experimental design, FACS analyses, siRNA transfection, RT-QPCR, 18. Król M, Polańska J, Pawłowski KM, Skierski J, Majewska A, Ugorski M, Motyl T:
manuscript preparation. All authors read and approved the final manuscript. Molecular signature of cell lines isolated from mammary
adenocarcinoma metastases to lungs. J Appl Genet 2010, 51:37–50.
Acknowledgements 19. Pawłowski KM, Popielarz D, Szyszko K, Motyl T, Król M: Growth Hormone
This work was supported by grant no N N308 574940 from the Ministry of Receptor RNA interference decreases proliferation and enhances
Sciences and Higher Education. The authors would like to thank Katarzyna apoptosis in canine mammary carcinoma cell line CMT-U27. Vet Comp
Szyszko DVM for her help in optimization of RT-qPCR reaction conditions and Oncol 2012, 10(1):2–15.
Dr. Robert Meyers, M.D. for his contribution in preparation of this manuscript. 20. Majchrzak K, Pawlowski KM, Orzechowska EJ, Dolka I, Mucha J, Motyl T, Król
The authors would like to thank Dr. Eva Hellmen (Swedish University of Life M: A role of ghrelin in canine mam mary carcinoma cells proliferation,
Sciences, Sweden) for her kind donation of CMT-U27 and CMT-U309 cell lines, apoptosis and migration. BMC Vet Res 2012, 8:170.
Dr. Gerard Rutteman (Utrech University, The Netherlands) for his kind donation 21. Król M, Majchrzak K, Mucha J, Homa A, Bulkowska M, Jakubowska A,
of P114 cell line and Prof. Dr. Maciej Ugorski (Wrocław University of Karwicka M, Pawłowski KM, Motyl T: CSF-1R as an inhibitor of apoptosis
Environmental and Life Sciences, Poland) for his kind donation of CMT-W1 and and promoter of proliferation, migration and invasion of canine
CMT-W2 cell lines. mammary cancer cells. BMC Vet Res 2013, 9:65.
22. Brinkhof B, Spee B, Rothuizen J, Penning LC: Development and evaluation
Author details of canine reference genes for accurate quantification of gene
1
Department of Physiological Sciences, Faculty of Veterinary Medicine, expression. Anal Biochem 2006, 356:36–43.
Warsaw University of Life Sciences - WULS, Nowoursynowska 159, 02-776, 23. Etschmann B, Wilcken B, Stoevesand K, von der Schulenburg A, Sterner-Kock
Warsaw, Poland. 2Department of Large Animal Diseases with Clinic, Faculty A: Selection of reference genes for quantitative real-time PCR analysis in
of Veterinary Medicine, Warsaw University of Life Sciences – WULS, canine mammary tumors using the GeNorm algorithm. Vet Pathol 2006,
Nowoursynowska 100, 02-797, Warsaw, Poland. 3Department of Animal 43:934–942.
Environment Biology, Faculty of Animal Sciences, Warsaw University of Life 24. Schmittgen TD, Livak KJ: Analyzing real-time PCR data by the
Sciences - WULS, Ciszewskiego 8, 02-786, Warsaw, Poland. comparative Ct method. Nature Prot 2008, 3:1101–1108.
Pawłowski et al. BMC Veterinary Research 2013, 9:119 Page 10 of 10
http://www.biomedcentral.com/1746-6148/9/119

25. Palmeira A, Vasconcelos MH, Paiva A, Fernandes MX, Pinto M, Sousa E:

Dual inhibitors of P-glycoprotein and tumor cell growth: (Re)discovering
thioxanthones. Biochem Pharmacol 2012, 83:57–68.
26. Nalbantsoy A, Karabay-Yavasoglu NU, Sayim F, Deliloglu-Gurhan I, Gocmen
B, Arikan H, Yildiz MZ: Determination of in vivo toxicity and in vitro
cytotoxicity of venom from the Cypriot blunt-nosed viper Macrovipera
lebetina lebetina and antivenom production. JVATiTD 2012, 18:208–216.
27. Grube S, Langguth P: Modulation of drug-carrier activity by excipients.
In Drug delivery research advances. Edited by Mashkevich BO. New York:
Nova Science Publishers, Inc; 2007:81–104. ISBN 978-1-60021-732-6.
28. Sun J, He ZG, Cheng G, Wang SJ, Hao XH, Zou MJ: Multidrug resistance P-
glycoprotein: crucial significance in drug disposition and interaction.
Med Sci Monit 2004, 10:RA5–RA14.
29. Creasey W, Scott AI, Wei CC, Kutcher J, Schwartz A, Marsh JC:
Pharmacological studies with Vinblastine in the dog. Cancer Res 1975,
35:1116–1120.
30. Schroeder A, Honen R, Turjeman K, Gabizon A, Kost J, Barenholz Y:
Ultrasound triggered release of cisplatin from liposomes in murine
tumors. J Control Release 2009, 137(1):63–68.
31. Struck RF, Alberts DS, Horne K, Phillips JG, Peng YM, Roe DJ: Plasma
pharmacokinetics of cyclophosphamide and its cytotoxic metabolites
after intravenous versus oral administration in a randomized, crossover
trial. Cancer Res 1987, 47:2723–2726.

doi:10.1186/1746-6148-9-119
Cite this article as: Pawłowski et al.: Expression and role of PGP, BCRP,
MRP1 and MRP3 in multidrug resistance of canine mammary cancer
cells. BMC Veterinary Research 2013 9:119.

Submit your next manuscript to BioMed Central

and take full advantage of:

• Convenient online submission

• Thorough peer review
• No space constraints or color figure charges
• Immediate publication on acceptance
• Inclusion in PubMed, CAS, Scopus and Google Scholar
• Research which is freely available for redistribution

Submit your manuscript at

www.biomedcentral.com/submit