Intracellular particle dissolution in alveolar macrophages.

Aerosol particles deposited in the lungs that are not readily soluble in the epithelial lining fluid will be phagocytized by alveolar macrophages (AM). Inside the phagolysosomal vacuole, the constituents of the plasma allow dissolution of a variety of compounds at a higher rate than dissolution in extracellular lung fluids. Chelator concentration and a pH value of about 5 were found to control intracellular particle dissolution (IPD). Hence, IPD is the initial step of translocation of dissolved material to blood, which is an important lung clearance mechanism for particles retained long term. IPD rates of uniform test particles determined in human, baboon, and canine AM cultures were similar to initial translocation rates determined in lung clearance studies of the same species after inhalation of the same test particles. IPD rate in cultured AM proved to be a sensitive functional parameter of AM, which was used to identify changes in the clearance mechanism of translocation during different exposure conditions.

Multidrug resistance (MDR) of tumour cells is considered to be one of the major obstacles to effective cancer chemotherapy (Curt et al., 1984;Bradley et al., 1988;van der Bliek & Borst, 1989). Such resistance can be an intrinsic property of a tumour or may be acquired following courses of chemotherapy. MDR has been shown to be associated with reduced intracellular drug content due to the over-expression of the P-glycoprotein (P-170), an energy dependent efflux pump that prevents intracellular accumulation of antineoplastic drugs (Juliano & Ling, 1976).
Since many mechanisms for the development of MDR are associated with energy metabolism pathways, the high aerobic glycolysis rate showed by resistant cells (Lyon et al., 1988;Kaplan et al., 1990) could make them targets for drugs inhibiting the energy metabolism.
Lonidamine (LND, Angelini ACRAF, Pomezia, Rome, Italy), a dichlorinated derivative of indazole-3-carboxylic acid, has been proven to strongly influence the energy metabolism of neoplastic cells inhibiting their aerobic lactate production (Floridi & Lehninger, 1983). LND appears also to be able to induce wide changes in plasma membrane due to its high affinity for the inner leaflet of the lipid bilayer (Malorni et al., 1988). Both the main LND effects, the reduction of ATP production and the membrane damage, could impair the metabolic adaptations associated with the development of drug resistance.
Previous data obtained in our laboratory demonstrated that LND differently affected the cell survival of melanoma lines and when used in combination with ADR determined a synergistic effect, according to the sequence employed (Zupi et al., 1986). Recently, it was reported that LND also enhanced the cytotoxic effect of cis-platinum on a human squamous cell carcinoma (Raaphorst et al., 1990). Moreover, when combined with radiation LND potentiates the lethal effect of ionising radiation on fibrosarcoma tumour cells (Kim et al., 1984).
In the present study we have investigated the efficacy of LND given in combination with Adriamycin (ADR, Adriblastina, Farmitalia Carlo Erba, Milano, Italy) on a human breast cancer cell line, MCF-7 wild type, and its derivative ADR-resistant line MCF-7 ADRR.
Both tumour lines (MCF-7 WT and MCF-7 ADRR, kindly provided by Dr K. Cowan from NCI, Bethesda, Maryland, USA) were maintained as monolayer cultures in 25 cm2 Corning flasks in supplemented RPMI 1640 medium (Gibco). ADRR cells were grown in medium containing 10 JiM ADR and passaged for at least 2-4 weeks in medium lacking the drug prior to their use in experiments. WT and ADRR cells were exposed to ADR (from 0.01 to 50 JAM) alone and in combination with a non-cytotoxic dose of LND (50 fig ml-'), tested in preliminary experiments. 1 x 106 ADRR and WT cells were plated in 25 cm2 Corning flasks in supplemented RPMI 1640 medium. The next day, medium containing the varying concentration of drug was added to the cells. After 7 days, the cells were harvested, assayed for cell viability (Trypan Blue exclusion test) and counted (Coulter Counter, Kontron, model, ZM). Drug sensitivity was evaluated by calculating the ADR dose that caused 50% of growth inhibition (ICy, value). Samples from cells exposed for different times to ADR or to ADR + LND were twice washed in PBS and kept frozen (106 cells ml-') for intracellular ADR determination.
Purified specific anti-ADR IgGs from polyclonal anti-ADR immune serum were employed to determine the intracellular ADR content as previously described (Citro et al., 1988). Briefly, the assay was performed by a competitive ELISA using a suspension from treated cells in cold PBS (106 cells ml-' sonicated at 100 W for 1 min). Experimental data were compared using Student's t-test, and the results were considered statistically significant when P <0.05.
The sensitivity of both cell lines to ADR and to ADR + LND exposures is reported in Table I. WT and ADRR MCF-7 cells showed a marked difference in the ICm value following exposure to ADR: 0.03 and 9 JAM, respectively. The simultaneous exposure to both drugs (ADR + LND) enhanced the ADR lethal effect, as shown by the decrease in the IC50 values in both tumour lines. However, the enhancement of ADR cytotoxicity induced by LND is more significant for ADRR cells than for their ADRsensitive counterparts. The ICm value of sensitive cells treated with the combination ARD + LND fell to 0.008 gM (about a 4-fold decrease of that observed in ADR alone treated cells) while the ADR dose able to kill 50% of the ADRR cells exposed to both agents fell to 0.007 JM (about a 1,300-fold decrease). The selectivity elicited by LND on the ADRresistant cell line appeared also by the significant decrease in the resistance index (0.875 for ADR + LND exposure vs 300 for ADR exposure). Tables II and III show the intracellular ADR content in both tumour lines upon ADR and ADR + LND treatments. In order to compare cellular ADR content and drug activity, the corresponding cell surviving fractions were also reported. The results are expressed as the mean ± s.e. of three separate determinations. Remarkably, differences in ADR intracellular content were observed between the two MCF-7 cell lines.
In the sensitive cell line treated with ADR as a single agent, an increase of the ADR intracellular content as a function of both dose and exposure time was observed (Table  II). The statistical analysis performed comparing the ADR content for each dose at the same exposure time demonstrated that there was a significant difference between each dose, with a P value ranging from 0.0002 to 0.04. A statistical significance was also obtained comparing the ADR content for each exposure time at the same dose (P value ranging from 0.0002 to 0.03).
Otherwise, in the resistant line treated with ADR alone the intracellular ADR content was not significantly modified by increasing both the dose and the exposure time (Table III). In fact, the intracellular ADR contents analysed either in function of doses or times did not significantly differ ( Figure 1 shows the intracellular ADR content of the two cell lines treated with the highest ADR dose as a function of exposure time. The trend of the two curves illustrates the different behaviour showed by the sensitive and the resistant lines. ADR concentration in WT cells increased steadily reaching the value of 78 ng ADR 10-6 cells within 72 h. On the contrary, in ADRR cells no marked increase in the drug content was observed during the treatment. These results demonstrate that the accumulation rate is consistently reduced in ADRR cells, indicating that the increase of drug doses and exposure times could be uneffective on drugresistant tumour cells. The ADR + LND combination differentially influenced the intracellular ADR accumulation of the two cell lines employed. Sensitive cells treated with the combination (Al, B1, Cl, D1; Table II) showed values of drug concentration similar to those obtained after their exposure to ADR as a single agent (A, B, C, D; Table II). In fact as also demonstrated by the statistical analysis, the differences in ADR content between the two treatments were not statistically significant.
These data correlate with the cell response to the combination; ADR + LND gave rise to a moderate enhancement of ADR cytotoxicity, as shown by the values of cell survival reported in Table II. On the contrary, the ADR + LND association strongly affected the intracellular ADR accumulation of the resistant cells. In fact, the presence of LND nearly allowed to double the intracellular ADR content for all doses and exposure times employed. The differences of intracellular ADR level detected between the two treatments came out highly significant from the statistical analysis (A, B, C, D vs Al, B1, Cl, D1; Table III).
Comparing the data reported in Tables II and III, it appears that the combination allowed ADR to achieve in the resistant cells an intracellular amount similar to that obtained for the wild type cells, indicating that LND had the ability to restore the in vitro sensitivity of ADR-resistant cells. In particular, LND allows ADR to reach the same intracellular content otherwise achievable with a 1,000-fold higher ADR dose given as single agent (Table III). The values of cell survival demonstrate the ability 'of LND to overcome ADR resistance (Table III). This effect is particularly relevant since at the lowest ADR dose, uneffective on the resistant cells, the combination determines a remarkable increase in cell lethality. In conclusion, these results indicate that the combination ADR + LND could be useful to improve the therapeutic index of ADR treatment.
The enhancement of the ADR cytotoxicity observed on the MCF-7 ADRR cells could be the result of two simultaneous selective LND effects: (1) the LND-lipid bilayer interactions give rise to clustering of intrinsic intramembrane proteins like P-170 glycoprotein, which is present on the MCF-7 ADRR cell membranes (Fairchild et al., 1987), impairing its biological functions and thus reducing drug efflux and detoxification; (2) Lonidamine specifically inhibits the activity of mitochondria bound hexokinase (Floridi et al., 1989) affecting the aerobic glycolysis of tumour cells. Since ADRR cells show an enhanced rate of glycolysis (3-fold), as compared to drug sensitive cells (Kaplan et al., 1990), LND could selectively impair cellular energy-dependent mechanisms like drug efflux, drug-conjugation, enzyme synthesis, and lethal damage recovery. Table II Intracellular  of LND treated cells was 95 ± 0.6. Not statistically different: A vs A1 P < 0.8; B vs B, P < 0.9; C vs C, P <0.8; D vs Di P <0.5.  The relationship between the high rate of aerobic glycolysis and drug resistance of MCF-7 ADRR cells was recently demonstrated by Kaplan et al. using 2-deoxyglucose, a specific glycolysis inhibitor. This drug was found to be extremely toxic for the cells which had acquired resistance to Adriamycin (Kaplan et al., 1990). These data are in agreement with our results, demonstrating that a specific inhibitor of aerobic glycolysis like Lonidamine can play a remarkable role to reduce or overcome multidrug resistance.
In conclusion, the present study indicates the possibility to use Lonidamine as potentiating agent to interfere with MDR properties. Considering its good tolerance reported in preliminary clinical trials (Ozols et al., 1983;Evans et al., 1984;Pronzato et al., 1989) Lonidamine, in combination with antitumour drugs, could improve cancer therapy when tumours initially responsive to chemotherapy develop resistance to the drugs following treatment with one of them.