Reversal of drug resistance of multidrug-resistant human lung cancer cells by an MDRl ribozyme GAO Zhenqiang, GAO Zhiping and LIU Xifu Department of Patholcgy, Beijing Medical University, Beijing 100083. China

Keywords:

rctmvhl vector, MDRl gem, r i m , moltidrog rdshnce, humno luug ancer.

&.IGNANT tumor is very harmful to human health, and the mortality is very high. About 50 % of the patients with malignancy can be operated on, and the other 50 % patients have to be treated with chemotherapy. Because tumors are mostly chemoresistant, .chemotherapy for these patients often has no effect. The overexpression of MDRl gene is very common in human malignant tumors, about 50% in previously untreated patients and more than 50% in previously treated patients for whom the tumor is resistant to the previous sensitive anticancer drugs. This phenomenon is called rnultidrug resistance'']. Ribozymes, one of ivhich has the character of hammerhead structure, are one kind of catalytic RNA molecules, and can recognize a GUX (X is one of the bases of A, C, U ) sequence in RNA and cleave it most efficiently. According to this principle, the artificially designed ribozyme can cleave RNA at special site and block the protein synthesis[21. In a previous study, MDRl antisense RNA mediated by retrovirus could obviously but not completely inhibit the expression.of pglywprotein (Pgp) which was encoded by MDRl . In the present study, we used the MDRl.ribozyme mediated by retrovirus to inhibit the Pgp expression in multidrug-resistant human lung cancer cells and explored the possibility of reversal of multidrug resistance.

1

Materials and methods

1.1 Cell lines The parent GAO human lung adenocarcinoma cell line and multidrug-resistant subline

GAOK'~]induced by doxorubicin (DUX) were maintained in RPMI 1640 medium containing

Chinese Science Bulletin

VQI.42 No. 1

January 1997

10 % ( V / V ) fetal bovine serum (FBS) .

1 . 2 Ribozyme synthesis and construction of retroviral vector The GUX sequences of MDRl mRNA were analyzed with computer software. According to Symons' theoryL2]and a previous reportL5],we designed a hammerhead ribozyme of MDRl mRNA for the GUC sequence at codon 196. T h e 196 MDRl ribozyme gene was as follows: sense 5'-GATCCGTTGCCATTCTGATGAGTCCGTGAGGACGAAACTGAAACA-', antisense 5 '-AATTCCTTTCAGTTTCGTCCTCACGGACTCATCAGAATGCG ' ( Gibco DNA synthesis center 1. These two singlestranded oligodeoxynucleotides contained flankRibozymxDNA ing BamH I and EcoR I restriction sites on both ends and were 5' phosphorylated by T4 polynucleotide kinase ( New England Biolabs) . The pDORneo retroviral vectorL3'was digested with BamH I and EcoK I and linked with the ribozyme. The ribozyme retrovirus was constructed and named pMR. pDORneo contained Mo-MuLV ( Moloney murine leukemia Amp virus) long terminal repeats ( LTRs), SV40 early promoter, neomycin resistance gene, pBR322 origin of replication and ampicillin resistance gene ( fig. 1 ) . pMR was introduced pBR322 ori into E . coli and selected with ampicillin in an agar plate. The plasmid DNA was extracted Fig. 1. h4ap of retroviral vector of MDRl ribozyme. and identified by restriction enzymes.

1 . 3 Cell transfection The PA317 package cells (NIH, USA) were transfected with 5 pg pMR by using Lipofectin method16]. After G418 (Sigma) selection, the resistant cell clones were cultured again and the supernatants from these cells were then used to infect the GAOK cells. The infected GAOK cells were selected with G418 ( 0 . 5 mg/mL) for two weeks, and the G418-resistant cell clones were analyzed for the quantitative expressions of MDRl genk and Pgp by using flow cytometry (FCM) . The clones with the lowest level of Pgp were isolated and the resistances to DOX, vinblastine (VBL) and colchicine (COL) were examined. 1.4

Northern blot of the ribozyme expression 10 pg total RNA from each transfectant cell clone was hybridized using the ribozyme cDNA as a probe according to the previously reported methodL6].

1.5 Quantitative analysis of MDRl expression RT-PCR (reverse transcription polymerase chain reaction) method was used[41.

FCM quantitative analysis of pgpL3' To quantitate the amount of Pgp on the surface of cells, the cells were analyzed by fluorescence-activated cell sorting. The cells were washed once with 1 X PBS, and stained with Pgp monoclonal antibody MRK16 (Kamiya) for 20 min on ice. The cells washed with 1X PBS contained 3 % FBS and were stained with the second antibody conjugated to fluorescein isothiocyanate (FITC) for 20 min on ice. The cells were again washed with 1X PBS /3 % FBS and resuspended in 1 x PBS containing 3 % FBS. Analysis was performed on FCM (Becton Dickinson) . 1.6

Chinese Science Bulletin

Vol .42 No. 1

January 1997

1. 7

Resistance to cytotoxic drugsi3' T o detect the drug resistance, the transfected cells ( 1 x lo4) were added to the wells of 96well microplates in 100 pI1 medium. The medium containing different concentrations of DOX or VBL or COL was added for about 3 d until cells in control wells were confluent, after which viability was assessed by the reduction of MTT (3-[4, 5-dimethylthiazol-2-yl]-2,5-diphenyltetrazolium bromide). MTT ( 0 . 5 mg/mL) was added to each well of the microtiter plate, and the plates were incubated for another 4 h in the dark at 3 7 C , after which the MTT solution was aspirated from the wells and 0 . 1 mL acid isopropanol was added. The plates were allowed to stand for 10 min, after which they were read in an Anthos microplate reader at 550 nm. The relative drug resistance was determined by comparing the ICSoof transfected and untransfected cells. was defined as the concentration of cytotoxic drug that caused 50% inhibition of cell growth as compared with untreated control. 1. 8

Growth inhibition studies in vivo Male BALB/c-nu/nu 6-week-old nude mice were purchased from Department of Experimental Animal, Heijing Medical University. The transfected cells with the lowest level of Pgp were injected s. c. in the flank of mice as previously describedL7]. When the diameter of the tumor xenografts was 6-8 mm, DOX (6 mg/kg per day) in 0 . 2 mL or 5% glucose solution (GS) was i. p. injected into mice once a day for 22 d. Each group consisted of 5 mi&. Tumor volume ( T V ) was measured every 2 d, and T V = a 2 x 6/2, where a is the tumor width and b is its length in mm. The effect of the drug was expressed as relative tumor volume fRTV), RTV=TV,/TVo, where TV, is the T V at any given time and TVo is the T V at the onset of treatment.

Results

2

2.1

Ribozyme synthesis

Cleavage site codon 196

4 5-UCUUUCAGUC AAUGGCAAC-3 3-AGAAAGUCA UUACCGUUGJ A C A U G A G C G GA A U U GC GC A G GU

Fa. 2 .

There were 37 GUC sequences in the coding region of MDRl mRNA, 6 GUC sequences in the 5'-untranslated region and 3 GUC sequences in the 3'-untranslated region. There were 2 GUC sequences near codon 185 which was the critical site for the activity of MDRl mRNA, one at codon 179 and the other at d o n 196. The 60-base residues on two flanks of these two codons could form hairpin-like secondary structure by computer analysis. Because the cleavage site at codon 196 was crucial"], we selected this site to design the MDRl ribozyme. The secondary structure of the designed ribozyrne was analyzed and could form the essential catalytic center (figure 2)

.

Structure of MDRl mKuA ribozyme.

2 . 2 Ribozyme expression in transfected cells Figure 2 was the map of the ribozyme retroviral vector pMR. The sequence and orientatioh of the ribozyme in pMR were confirmed by cleavage of enzymes of BamH I and Hind or Pst I and EcoR I . The retroviruses produced in PA317 packaging cells were introduced into GAOK cells. After G418 selection, 4 resistant cell colonies (pR1, pR2, pR3 and pR4) were isolated. Total RNAs were extracted from the 4 clones and hybridized with the ribozyme probe. There were expressions of riboz~memRNA in the 4 clones and the order of the ribozyme mRNA levels was pR2 > pR3 > pR1> pR4. No riboz~meexpression existed in the untransfected GAOK cells

m

66

Chinese Science Bulletin

Vol .42 No. 1 January 1997

(figure 3 ) I 2 3 4 5 Quantitative analyses of MDRl gene and Pgp Total RNA was isolated from the individual clones, and the expression of MDRl gene was Ribozyme RNA demonstrated by the quantitative RT-PCR assay. 18s The decreases of MDRl mRNA levels in the 4 clones were obvious, for the MDRl mRNA levels in pR1, pR2 and pR3, as well as in GAO cells were 94.83-fold lower than in GAOK cells; the level in pR4 also was very low and decreased 93.88-fold (table 1) . By FCM analysis, the Pgp contents in pR1, pR2, pR3, as well as in GAO Fig. 3. Expression of ribozyrne in transfected cells. 1, cells decreased 11-fold in GAOK cells; the content pR1; 2, pR2; 3, pR3; 4, pR4; 5, GAOK. in pR4 decreased 10.53-fold in GAOK cells. These results showed that the MDRl ribozyme could cleave MDRl mRNA efficiently and block the synthesis of Pgp.

2.3

Table 1 The contents of MDRl mRNA and PUD -- in -DMR transfected cell line Cell line MDRl content MDRl relative content') Pgp content Pgp relative content') GAO 0.84k0.02 1.00 0.25f0.01 1.00 GAOK 80.50k2.45 95.83 3 . 0 0 f 0.84 12.00 PR~ 0.84k0.03 1.00 0 . 2 5 f 0.04 1.OO pR2 0.84*0.01 1.00 0 . 2 5 k 0.03 1.00 P R ~ 0.84k0.02 1.00 0.25k0.02 1.00 pR4 0.85k0.01 1.01 0.26k0.02 1.04 al Relative content was the content of GAOK or transfected cells relative to the content of GAO cells.

2.4

I n vitro drug sensitivity of transfected cells Drug sensitivity of these four clones was analyzed by M T T assay (table 2 ) . In 4 clones, the pR2 DOX sensitivity was the highest and the drug resistance decreased 90.8 % ; the pR4 was the lowest and the drug resistance decreased 8 9 . 6 % ; the pR1 and pR3 were between pR2 and pR4 and the drug resistances decreased 90.5 % and 90.2 % respectively. The resistances to VBI, and COL of these 4 clones also decreased obviously (table 2 ) . These results showed that the expression of MDRl/Pgp was closely correlative with the drug.resistance. But it did not completely reverse drug resistance to inhibit the expression of MDRl and block Pgp synthesis in GAOK cells, suggesting that there were other resistant mechanisms in GAOK cells. Table 2 Cell line

GAO GAOK PR~ P R ~ pR3 pR4

mx

( 1cso) b,

1.00 12.80 1.21 1.18 1.25 1.43

(28) (358) (34) (33) (35) (40)

a) Relative drug resistance

2.5

Drug sensitivity of pMR transfected cell line i n vitro

= C$~O~/C$:~; b)

Relative drug resistance') VBL ( ICso ) b' 1.0 (0.025) 8.5 (0.210) 1.3 (0.033) 1.1 (0.028) 1.4 (0.035) 1.5 (0.038)

COL 1.0 5.8 1.4 1.2 1.5 1.5

( 1 ~ 5 0b'

(1.04) (6.03) (1.46) (1.25) (1.56) (1.56)

the values in brackets were practical values of ICSo(rrnol).

Growth inhibition in vivo T o examine whether these pMR transfected cells were also sensitive to DOX i n vivo, tu-

Chinese Science Bulletin

Vol.42 No. 1

January 1997

67

12

,-

lo -

4

3

:

a+-a 5

'

tld-, Fig. 4 . In vim doxorubicin effects on pMR transfected cells. RTV, relative tumor volume; 1. CAO + T; 2, GAO; 3, CAOK+ T; 4, GAOK; 5 , p~ + T; 6, p~ ( + T indicat= treatment with doxombicin) .

3

mor cells were inoculated s. c. at the flank of nude mice and treated with DOX (fig. 4 ) . GAOK tumor treated with DOX grew as well as GAOK untreated tumor did; GAO tumor treated with GS grew as well as GAOK untreated tumor did, but that treated with W X grew very slowly ; the tumor volume gradually decreased and disappeared after 13 d; pR2 untreated tumor grew as well as GAOK untreated tumor did; the pR2 treated tumor grew slowly and the volume gradually decreased; however, after 2 weeks, the volume stopped decreasing and began to increase slightly (fig. indicated that the MDR1 ri4, . These bozyme also could but not completely reverse the drug resistance in vim.

Discussion

The main obstacle in the treatment of malignant tumor is the resistance of tumor cells to chemotherapy. Lung cancer, especially non-small lung cancer ( NSCLC) , is mostly chemoresistant. A recent study reported that Pgp overexpression was found in 45 % of 94 previously untreated NSCI-C samples, almost 100 % in previously treated NSCLC samples, and was obviously correlated with multidrug resistance of tumor cells[']. This indicated that MDRl/Pgp played an important role in drug resistance of lung cancers. Since calcium ion was important for the pump function of Pgp, calcium channel blockers were used to restore the drug sensitivity of multidrug-resistant cells. Some of these substances, for example, verapamil, were shown to reverse drug resistance in varying degrees; however, the reverse effect disappeared as verapamil was eliminated, and the dose of verapamil was not suitable for clinical application[']. Antisense RNA could form an RNA duplex with its target mRNA and inhibit protein synthesis. In the previous study, MDRl antisense RNA mediated by retrovirus was introduced into multidrug-resistant GAOK cells; the level of MDRl mRNA decreased and the Pgp synthesis was inhibited; the resistance to DOX of one clone with the lowest level of MDRI mRNA decreased 55 % , but the Pgp content of this clone was 4-5 times higher than the drug sensitive parent cells, indicating that MDRl antisense could not completely inhibit Pgp ~ynthesis'~]. Ribozyme was found and offered a novel method to inhibit gene expression. HIV ribozyme was successfully used to inhibit the synthesis of p24 antigen protein; ras ribozyme reversed the ma~ o b a ~ a s h iused [ ~ ] MDRl ribozyme to relignant phenotype of human bladder ~arcinoma'~]. verse the drug resistance of human acute lymphoblastic leukemia cells; the drug sensitivity increased 700-fold. So ribozyme is a good molecule against virus and tumor. A retrovirus system has potential advantages for gene delivery, such as easiness of producing high titer virus, high infection efficiency, infectivity for many types of cells, and harmlessness to host cells. So retrovirus was often used as expression vector of antisense RNA. In this study, retroviral expression vector of MDRl ribozyme was constructed and expressed in resistant GAOK cells. The MDRl ribozyrne could cleave MDRl mRNA and block Pgp synthesis in vitro and in vi-

68

Chinese Science Bulletin

Vol .42 No. 1 January 1997

BULLETIN vo, but the drug resistance of GAOK cells was not completely reversed. This indicated that

there were other reasons in the resistance of GAOK cells than MDRl gene. We are making efforts to reverse the resistance of GAOK cells completely. (Received August 1 , 1996)

References 1

2 3 4

5 6

7

8 9

.

Gao, Z. Q., Gao, %. P . , Advance in nlultidrug resistance gene (MDRl), W i n . 1. Parhol . 1996, 3: 190. Symons, R . , Self-cleavage of RNA in the replicative of small pathogens of plants and animals, TIBS, 1989, 14: 445. Gao, Z. Q . , Gao. Z. P . , I.iu, X. F . . Reversal of drug resistance in multidmg-resistant human lung cancer cells by retrovirus-mediated MDRl antisense RNA, Chin. Sci . Bull., 1996, 41 (20) : 2166. Gao, Z. Q . , Gao, Z. P . . Quantitative PCR analysis of MDRl gene in human lung cancers, in Basic Medical Sciences (ed. Gc, L . H. ) , Beijing: China Personnel Press, 1995, 574-578. Kobayashi. H.,Dorai. T . . Holland, J . et a l . , Reversal of drug sensitivity in multidrug-resistant cells by an MDRl ribozyme, Cancer R e s . , 1994, 54: 1271. Gao, 2. Q., Gao, Z. P., Zhang, X . H . , Growth inhibition of human lung cancer cells by p53-wt. Chin. J . Preu. and Cantrol of Chronic Non-communi. D i s . . 1996, 2: 66. Cao, Z. Q. Zheng. J . , Wu, B. Q. et a l . , The biol&ical study of xenografts of human lung cancer induced by be-(*)pyrene in nude mice, Chin. J . Pathol., 1996, 2: 106. Safa, A. R . Stern. R . K . Choi, K . et al . , Molecular basis of resistance to colchicine in multiresistant human cells confers by Gly-185 to Bal-185 substitution in Pgp, Proc. Nutl. Acad. Sci. U S A , 1990, 87: 7225. Kashami-sabet, M.. Funato. T . , Tonc. T. et a1 . Reversal of the malignant phenotype by an anti-ras ribozyme, Antise. Res. Dev..1992. 2: 3.

.

.

Chinese Science Bulletin

.

Vol .42 No. 1

January 1997