It is well known that multicellular spheroids are more resistant to chemotherapeutic drugs than the same cell cultures as monolayers 1112. With the aim to evaluate differences in P-gp activity in multicellular spheroids with respect to monolayers, two P-gp substrates were used (DXR and Rho-123). Both compounds were chosen, because they are good P-gp substrates with an autofluorescence capacity. Our results showed that in monolayers the amount of DXR retained in the cells was in direct proportion to the drug added to the medium (Figure 1). In contrast, multicellular spheroids showed a lower capacity for DXR retention (3-fold lower) than monolayers. Interestingly, this poor retention was not observed when Rho-123 was used as P-gp substrate because the Rho-123 levels retained were equal in both types of cultures (box insert in Figure 1a).

Since the drug-retention is a dynamic process that involves drug uptake (simple diffusion) as well as drug linkage (diffusion vs. expulsion mediated for an active mechanism), we tried to determine whether a more efficient P-gp-dependent drug removal was responsible for allowing lower intracellular DXR concentrations. Thus, drug-removal in monolayers previously loaded with 30 μM of DXR during 30 min showed a first order decay reaction with a half-time of drug concentration at the first 5 minutes interval that indicated the presence of an active transport mechanism (Figure 1b). However, for multicellular spheroids the presence of active transport was not evident since its was not possible to load the cells with sufficient amounts of DXR to determine the efflux values. A common observation in these experiments was that a constant amount of DXR was retained for a longer period of time (45 minutes) in cells grown as monolayers. A possible explanation could be from the presence of positively charged-DXR, which stores in acidic vesicles as chromaffin granules and lysosomes 2021.

With the aim to determine how much the P-gp influences the levels of DXR retained, the modulator agent Cs-A was employed. As shown in the Figure 2, the incubation of monolayers with 5 μM of Cs-A efficiently enhances (2-fold) the intracellular DXR fluorescence in direct relation to reversal agent concentration. Surprisingly, the P-gp-modulation activity of Cs-A was not evident in multicellular spheroids because no effect in the intracellular DXR retention could be seen. Also, other reversal agents, like SDZ PSC 833 and verapamil were not able to increase DXR retention, neither in monolayers nor in multicellular spheroids (data not shown).

P-gp is a protein that belongs to the ABC binding cassette protein, for which efficient drug efflux needs ATP hydrolysis. Thus, with the propose to achieve more information about the P-gp function, we decided to inhibit the P-gp activity through depletion of ATP-pools. Three metabolic poisons were used to be sure of complete ATP-depletion in multicellular spheroids. Figure 3 shows that ATP-depletion did induce a 1.5-fold increase of DXR retention in cells maintained as monolayers but none effect was again seen when in ATP-depleted multicellular spheroids.

With no possibility of obtaining irrefutable evidence of P-gp modulation from multicellular spheroids, we decided to eliminate the P-gp expression from the lung cancer line. Therefore, through the co-selection of the parental cell line INER-51 with 5 μM of DXR and 5 μM of SDZ PSC 833, we were able to obtain one mutant cell clone that did not express P-gp, which was named PSC-1 line (Figure 4a). In vitro RT-PCR analysis of other MDR-related genes did not show qualitative differences between parent and mutant cells (Figure 4b). The amplification experiments showed that both cells lines express with approximately with the same intensity transcripts for topoisomerase I, topoisomerase IIα and topoisomerase IIβ but neither of them expressed the multidrug resistance-associated protein (MRP1) or glutathione-S transferase-μ.

The similarities between INER-51 cells (P-gp^pos) and PSC-1 cells (P-gp^neg), allowed us the possibility to evaluate whether or not the lower DXR-retention levels were mediated by a positively modulated P-gp-mechanism. The incubation of PSC-1 spheroids with increasing DXR concentrations showed a significative increase in the drug retention (1.8-fold) in comparison to INER-51 spheroids and was only 1.6-fold lower than PSC-1 growth as monolayers (Figure 5). When Rho-123 was assayed, an increase in the retention of the dye relative to the parental line INER-51 as monolayers was evident. In PSC-1 spheroids, the levels of Rho-123 retention were similar in both INER-51 monolayers and multicellular spheroids (box insert in Figure 5).

Several techniques have been used to evaluate the MCR in multicellular spheroids. Since in previous experiments we were unable to successfully disaggregate multicellular spheroids through trypsin-based protocols, we decided to evaluate the MCR using the MTT assay (which as been previously evaluated by Furukawa

et al

Doxorubicin (DXR) was provided by Farmitalia-Carlo Erba, whereas Cyclosporin-A was provided by Sandoz Farma, whereas SDZ PSC 833 was gift by Novartis. Rhodamine-123 (Rho-123) and 3-(4,5-dimethylthiazol-2-yl)-2,5 diphenyl tetrazolium bromide (MTT) were purchased from Sigma (St. Luis, MO, USA). All of the working solutions were initially dissolved in dimethyl sulfoxide (DMSO) and posterior dilutions were put in culture medium.

The lung cancer cell line INER-51 and its clone PSC-1 (non-expressing P-gp or P-gp^neg) were grown as monolayer cultures in RPMI-1640 medium at 37°C in 5% CO₂. INER-51 is a NSCLC cell line established in our laboratory from pleural effusion of patient diagnosed with primary lung cancer without previous chemotherapy treatment. The kidney cell line A498 (expressing P-gp) and the lung adenocarcinoma cell line INER-37 (expressing MRP1) were used as controls of P-gp function and MRP1 expression, respectively. The culture medium was supplemented with 10% FCS (Sigma, Co. St. Luis, MO, USA), 1% non-essential amino acids, 1 mM sodium pyruvate, 2 mM L-glutamine, 100 units/ml of penicillin and 100 μg/ml of streptomycin.

Monolayer cells were passed a week by trypsin-EDTA solution (Invitrogen™, USA). To obtain multicellular spheroids, 3.5 × 10⁵ exponentially tumour cells were seeded in 1% agarose-coated (24-well/plate) in RPMI-1640 complete medium 14. Cultures were routinely grown for 72 hours to acquire multicellular spheroids of approximately of 500 μm of diameter.

In order to obtain one mutant clone with the capacity to not express P-gp, INER-51 cells were treated as described by Beketic-Oreskovic,

et al

The level of resistance to drugs was determined with the use of MTT assay as previously described by S. Cole 16 and on multicellular systems as evaluated by Furukawa,

et al

Exponentially growing cells (3.5 × 10⁵) were seeded 72 hours prior to treatment with drugs in 1% agar-coated 24-well tissue culture plates as described above. Either multicellular spheroids or monolayer cells were incubated with increased concentrations of DXR or Rho-123 for 30 min and then washed with fresh drug-free medium twice and PBS once. Cell suspensions from monolayers were obtained by tripsinization whereas multicellular spheroids were recovered using micropipette tips. To exclude artefacts arising from DXR quenching by binding to DNA and accumulating in acidic organelles, DXR was extracted from spheroids with the method described by Wartenberg

et al

The efflux assays were based on those previously described 19. Briefly, monolayers or multicellular spheroids were drug loaded with DXR (30 μM) throughout incubation at 37°C in 5% CO₂ for 30 min. The loading cells were then washed with a complete drug-free ice-cold medium and either placed on ice or incubated at 37°C in 5% CO₂ for different intervals of time. This incubation allowed drug efflux to occur, and during experimental time the remaining drug was quantitated by spectrofluormetry as described above.

For evaluation of P-gp activity, monolayers or multicellular spheroids were pre-incubated for 1 hour with increasing drug concentrations of cyclosporin-A (Cs-A). After pre-incubation, multicellular spheroids were loaded for 30 min with 30 μM of DXR in presence of the reversal agent and afterwards the DXR uptake was assessed again. For ATP-depletion, cells were pre-incubated for 20 min at 4°C in PBS/BSA (1 mg/ml) containing 1 mM sodium-cyanide, 10 mM sodium-fluoride and 10 mM sodium azide prior to the addition of DXR. After centrifugation at 1000 × g for 15 min, both monolayers and multicellular spheroids were loaded with different DXR concentrations in the presence of ATP synthesis inhibitors and the fluorescence was measured as mentioned above.

Total RNA was extracted from the cell lines with Trizol reagent (Invitrogen™, USA) according to manufacturer's instructions. Single stranded cDNA was synthesized by reverse transcription from 5 μg of total RNA using Superscript™ RNAse Reverse Transcriptase (Invitrogen™, USA) and oligo-dT_16–18. The amplification was performed in a final volume of 25 μl, containing 0.5 μl cDNA, 50 pM of each oligonucleotide primer, 30 μM of each dNTPs, 2.5 units of Taq DNA polymerase, 1.5 mM MgCl₂, 20 mM Tris-HCl (pH 8.4) and 50 mM KCl. Amplification was carried out in a Thermal Cycler (Programmed Thermal Controller, model PTC-100, MJ Research Inc., USA) for 35 cycles of denaturalisation at 94°C for 1 min, annealing at 55–60°C for 2 min, and polymerisation at 72°C for 3 min. The PCR primers and expected product size were as follows: For MDR-1, forward: 5'-cccatcattgcaatagcagg-3' and reverse: 5'-gttcaaacttctgctcctga-3 [150 bp]; MRP1, forward: 5'-tctctcccgacatgaccgagg-3' and reverse: 5'-ccaggaatatgatgccccgacttc-3' [140 bp]; topoisomerase IIα, forward: 5'-tttaaggcccaagtccagttaaac-3' and reverse: 5'-gtataacaatatcatcaagattgt [343 pb]; topoisomerase IIβ, forward: 5'-gaagtgttcactagtaaaatacagt-3' and reverse: 5'-cataatctttccatagcgtaaggtt-3' [336 bp]; topoisomerase I, forward: 5'-aagcagaggaagtagctacg-3' and reverse: 5'-gctcatctgtttccgagctt-3' [206 bp]; GST-μ, forward: 5'-gaactccctgaaaagctaaag-3' and reverse: 5'-gttgggctcaaatatacggtgg-3' [250 bp]; G3PDH, forward: 5'-tggggaaggtgaaggtcgga-3' and reverse: 5'-gaaggggtcattgatggcaa-3' [110 bp].