Abstract
Background/Aim: Liposomal Doxorubicin (lipDOX) and free Doxorubicin (DOX) are reported to exhibit similar antitumor efficacy. However, cellular internalization mechanisms of lipDOX are still a subject of controversy. Materials and Methods: Intact and permeabilized cells were exposed for short time to lipDOX and free DOX and drug intracellular content was evaluated by flow cytometry. Then, the antiproliferative capacities of lipDOX and free DOX were compared by the leukocyte nadir test in mice in vivo. Results: The fluorescence increase was 11.2-fold higher in intact cells and 19.7-fold higher in permeabilized cells after exposure to free DOX as compared to lipDOX. Mice injected with DOX showed pronounced antiproliferative activity with a leukocyte count decrease to 2.8±0.65 k/μl (p<0.01) – an effect significantly stronger than that in the lipDOX group. Conclusion: Intact and permeabilized cells internalize free DOX manifold faster than lipDOX. The LipDOX formulation does not induce a remarkable leukocyte nadir effect in vivo.
Anthracyclines (ACs), such as doxorubicin (DOX) and liposomal doxorubicin (lipDOX) represent one of the most important categories of drugs used for the treatment of a wide variety of malignancies. Despite years of research in developing better anticancer drugs or their formulations, ACs remain among the most effective cytotoxic drugs for the treatment of solid tumors. ACs are thought to kill cancer cells through multiple mechanisms that include intercalating between DNA base pairs (1), inhibition of topoisomerase II (2), reactive oxygen species generation (3), DNA fragility (4), and disruption of chromatin dynamics (5). Due to their inherent fluorescent properties and small molecular size, representative of many cytotoxic drugs, ACs have been used as suitable substances to study drug cellular uptake. Intracellular DOX, rather than its concentration in the media, has been demonstrated to be the key factor in cell proliferation inhibition (6). Many attempts have been made to achieve preferential selectivity of ACs for tumor cells by preparing specific carrier agents for anticancer therapy. Thus far, liposomes are the most studied colloidal particles used in antitumor therapy enhancement. Unexpectedly, the increased liposomal DOX uptake demonstrated in certain tumours (7) did not lead to expected improvements in early clinical outcomes (8, 9), indicating that the drug delivered in the liposomal form might not be always active when injected (10).
The mechanism of action of liposomal cytotoxic drugs is not entirely understood. Some studies have indicated that liposomes are degraded in the interstitial matrix accompanied by drug leakage into the extracellular space (11, 12). According to this hypothesis, cellular uptake of released free DOX is a key route involved in anti-tumor activity. An alternative suggestion focuses on the endocytic pathway of liposomal drug internalization. For instance, Seynhaeve et al. have shown that liposomal uptake involving an endocytic pathway might precede intracellular distribution of DOX (13). Finally, macrophages are hypothesized to be part of the mechanism by transferring the drug to the tumor cell after phagocytosis of liposomes (14). The latter mechanism is potentially less feasible in pegylated liposomal formulations, which should not be easily phagocytosed and therefore, macrophage-mediated trafficking of a drug may not be a significant factor.
Our working hypothesis was that lipDOX, namely Caelyx® (CLX), cannot elicit an antitumor effect due to its insufficient accumulation in cells following injection. Its clinically modest antitumor effect might be due to other mechanisms that delay the local recycling of leaked DOX.
In this study, we compared the intracellular content of DOX and CLX after short term exposure to SL-2 cells in vitro. The cellular uptake and DNA staining by free DOX was significantly higher as compared to liposomal CLX in viable and permeabilized cells. The absence of early cellular uptake of CLX in vivo was hypothesized since antiproliferative nadir was marginally detectable in CLX-treated mice.
Materials and Methods
Materials. Doxorubicin hydrochloride 2 mg/ml (Ebewe, Unterach, Austria), Caelyx® – pegylated liposomal doxorubicin hydrochloride 2 mg/ml (Janssen Pharmaceutica NV, Beerse, Belgium), Daunorubicin hydrochloride 2 mg/ml (Neon Laboratories Ltd., Mumbai, India) were obtained from a hospital pharmacy. Acetonitrile, orthophosphoric acid and hydrochloric acid were purchased from Sigma Chemical Company (St Louis, MO, USA).
Cell culture and drug uptake analysis. The SL2 lymphoma line was a kind gift of Prof. Den Otter (University Medical Center Utrecht, Utrecht, the Netherlands). SL2 cells were maintained in RPMI-1640 medium (Thermo Fisher scientific, Waltham, MA, USA) containing 10% heat-inactivated fetal bovine serum (Thermo Fisher Scientific), L-Glutamine and antibiotics (100 U/ml penicillin and 100 μg/ml streptomycin) (Lonza, Portsmouth, NH, USA). For uptake analysis, DOX and CLX were added to a culture of proliferating SL 2 cells at a final concentration of 10 μg/ml. The duration of exposure was 30 min at 37°C for all experiments. At least 10,000 cells were acquired on a FACS-LSRII flow cytometer (FCM) (Becton-Dickinson, San Jose, CA, USA). Analysis of FCM data was performed using FlowJo v10.7 software (Tree Star, Ashland, OR, USA). Apoptosis was verified by staining the cells with Annexin V and propidium Iodide (PI) solution (15). Following exposure to DOX or CLX, SL2 cells suspended in PBS were allowed to adhere on glass coverslips placed at the bottom of a 12-well culture plate (TPP AG. Trasadingen, Switzerland) via 30 min gravity sedimentation as described by Tsang et al. (16). Adhered cells were fixed in 4% paraformaldehyde and then analyzed. The localization of DOX and CLX in SL2 cells was analyzed under a fluorescence microscope Leica DM LS (Leica Microsystems GmbH, Wetzlar, Germany). The fluorescence of DOX and CLX was detected using an excitation 450/490 nm and 515/30 nm emission bandpass filters.
Mice, blood sampling and hematological analysis. C57Bl/6 female mice, aged 8 to 12 weeks, were obtained from the local breeding facility SRICIM, Vilnius, Lithuania. Mice were used and cared for in accordance with the ARRIVE guidelines 2.0 for the Care and Use of Laboratory Animals. All research protocols were approved by the Institutional Animal Care Committee. Animals had ad libitum access to pelleted feed, food supplements, and water. All animals were determined to be specific pathogen-free. DOX or CLX were injected intravenously via retroorbital plexus as a single dose of 15 mg/kg. Light inhalation anesthesia with isoflurane was used for i.v. injections and blood sampling procedures. Blood sampling was performed 48 h after injection. Complete blood counts (CBC) were analyzed using ABX Micros ESV 60 within an hour of sampling.
Permeabilization and flow cytometry. Permeabilization and AC staining of cultured SL-2 cells were performed as previously described for murine thymocytes (17). The concentration of the drugs used was 10 μg/ml and Triton-X100 or RNAse was not added as suggested for anthracycline DNA staining (17). Cells, at a density of 106 cells per sample, were washed once in PBS, chilled to 4°C and rapidly resuspended in 70% ethanol at 4°C. Samples were stored in the dark at 4°C before testing. For staining/exposure to each drug, cells were centrifuged and resuspended in RPMI-1640 media containing 10 μg/ml of DOX or CLX for 30 min at 37°C. FCM evaluations were performed using FACS-LSRII flow cytometer (Beckton Dickinson).
Electron microscopy. For transmission electron microscopy (TEM) analysis, samples were diluted 50x with distilled water, and 6.5 μl of the sample were directly applied on the Holey carbon-coated copper grid (Agar Scientific Ltd, Stansted, UK) the excess liquid was drained with filter paper before staining with two successive drops of 2% uranyl acetate (pH 4.5), dried, and examined in Tecnai G2 F20 X-TWIN transmission electron microscope, operating at 200 kV.
HPLC Chromatographic conditions. High performance liquid chromatography (HPLC) analysis was performed on a Perkin Elmer system (Shelton, CT, USA) that consisted of a Flexar binary LC pump, Flexar 3 CHNL VAC degasser, a Flexar LC autosampler and Flexar fluorescence detector (Xenon lamp). The chromatography data were acquired by Chromera software from Perkin Elmer. The chromatography separation was performed on a Brownlee Bio C18 column (4.6×150 mm, 5 μm particle size, Perkin Elmer). The optimum mobile phase consisting of acetonitrile and water (32:68, v/v), was pH adjusted to 2.6 with 85% orthophosphoric acid. Samples were delivered via isocratic flow at a rate of 0.25 ml/min. The column temperature was maintained at 37°C and excitation and emission wavelengths were set at 475 and 555 nm respectively. The injection volume was 50 μl.
Size analysis by dynamic light scattering (DLS). The size and size distribution of Caelyx was analyzed in a Zetasizer Nano-ZS device (Malvern Instruments Inc., Worcestershire, UK) operating a 4 mW He–Ne laser at a wavelength of 633 nm. Measurement was performed at 25°C, with a count rate of 361.0 kcps, for 50 s.
Statistical analysis. The statistical tests were performed using STATISTICA 12.0 (TIBCO Software Inc Palo Alto, CA, USA). All the results are presented as means and standard error (mean±S.E.). Significance was considered at values of p<0.05.
Results
Identical DOX mobility of free DOX and CLX DOX (provided by different manufacturers) used in in vitro and in vivo experiments was confirmed by HPLC analysis. Specific peaks for both DOX and CLX exhibited retention times at 11.1 min. The retention time of internal standard daunorubicin was found to be 17.2 min in all assays. To verify the liposomal integrity of our samples, we performed TEM and DLS evaluation of specimens, and subsequently used them in the in vitro and in vivo studies. CLX anomalies reported in earlier studies (18) were not found in our samples used for short term drug exposure or in vivo assays. The electron micrographs taken from CLX show homogeneous oval shaped CLX particles or occasionally horseshoe-shaped vesicles (Figure 1). Homogeneity was also recorded by DLS analysis of CLX samples revealing a single peak. Size distribution by volume was 93.74 nm (width 21.74 nm), by intensity 105.00 nm (width 22.22 nm), and by numbers 84.28 nm (width 17.55 nm). The DLS and TEM were also used to rule out the potential decomposition of drug samples deposited in a body temperature environment. Our working hypothesis assumed that the major clinical effect of CLX must be related to the long-term presence of liposomal DOX in tissues. Partial aggregation of particles forming a second peak at >300 nm was observed by DLS only in samples exposed to 37°C for more than 90 days (data not shown). An example of cellular uptake revealed by FCM after exposure of intact and permeabilized cells to the drugs is shown in Figure 2. The mean fluorescence increase, the difference in mean channel fluorescence (dMCF) number, was found to be on average 4,940 for DOX and 233 for CLX in viable intact cells. The difference in fluorescence between drug exposed cells and autofluorescence estimated by the mean channel number (dMCF) was 5,168±11.0 determined for free DOX and 461±10.1 for CLX. Thus, the increase in fluorescence, which can be roughly estimated as uptake levels, was 11.2-fold higher after exposure to free DOX compared to liposomal CLX. Similarly, the dMCF of permeabilized cells was found to be 7,248±25.4 for DOX and 367±17.7 for CLX. The fluorescence increase in permeabilized cells was 19.7-fold higher after exposure to free DOX compared to liposomal CLX. However, this increase would be even higher in permeabilized cells if staining would not be affected by the degraded DNA Exposure of permeabilized cells to DOX results in DNA staining that can be used for cell cycle analysis identical to PI or other stoichiometric stain. The subdiploid population staining was detectable only after exposure of permeabilized cells to free DOX (Figure 2). The presence of the subdiploid population could be attributed to apoptotic cells (30). It was confirmed in our study by Annexin V incorporation. Up to 62% of SL 2 cells could be demonstrated to exhibit subdiploid staining after exposure of permeabilized cells to free DOX (Figure 2). The low staining of subdiploid peak in apoptotic SL2 cell samples was possibly due to insufficiency of free (leaked from CLX) DOX in liposomal DOX exposure. This is an additional evidence that even permeabilized cells exhibit limited internalization of liposomal drug. The white blood cell (WBC) count decrease in the nadir test following 15 mg/ml injection of CLX was significantly lower compared to the free DOX effect (Table I). In fact, the nadir difference between control mice and CLX-injected mice was minimal although significant (p<005).
Discussion
Cellular uptake of lipDOX in our studies was found to be significantly lower compared to free DOX. The antiproliferative effect observed at 48 h by the WBC nadir assay was significantly more pronounced in mice injected with free DOX compared to CLX. The intracellular distribution characteristics of CLX were also inferior to free DOX. Apoptotic bodies are known to exhibit strongly increased permeability, which is a characteristic feature of cells undergoing apoptosis (19). No increase in intracellular internalization of CLX was observed in apoptotic cells. In fact, permeabilized cells exposed to CLX showed very weak, if any, binding to the degraded chromatin fraction (subdiploid peak). This high scale difference between liposomal and free DOX after short term exposure was somewhat unexpected. It might be due to the fact that the levels of leaked DOX available in culture following CLX exposure were not sufficient to bind degraded DNA and therefore, preferentially bound to non-degraded chromatin. Our working concentration of 10 μg/ml of DOX and CLX exposure in vitro for 30 min was based on pharmacokinetic findings in humans (20). The decision to use this concentration was based on evidence that 10 μg/ml is roughly two times higher compared to maximal plasma concentrations found in humans shortly after i.v. administration of free DOX (20), and also, it is in the range of ~8 μg/ml of CLX (22) plasma concentration available following 20 mg/kg i.v. infusion. However, the total exposure area under the concentration-time (AUC) of free DOX was slightly lower in our in vitro system (since the duration was only 30 min) because traces of DOX might be still detectable in the plasma >48 h after i.v. administration (21). It seems unlikely that the endocytic pathway is significantly involved in liposomal CLX internalization. This is due to the fact that tissue and cellular exposure to high DOX concentration was rather short in our in vivo and in vitro studies. The plasma concentration kinetics of i.v. administrated free DOX exhibit a rapid initial decline (half-life 8 min) followed by a slow decline (half-life 30 h), which has been attributed to the ability of tissues to rapidly accumulate the drug, followed by a slow release of the drug from tissues as plasma levels decline due to active drug extrusion from intact cells (21). Thus, the plasma half-life clearance in the initial decline is estimated to be in the order of minutes for free DOX (21), whereas a significantly longer clearance was demonstrated for liposomal CLX (22, 23). Longer presence of encapsulated drug in the circulation displaying slow DOX leakage from liposomes does not automatically imply better tumor control. This can be seen by assessing the volume of distribution - a pharmacokinetic parameter representing a drug’s tendency to either remain in the plasma or to be rapidly absorbed by body tissues. The larger the value is, the more “tissue absorbable” the drug is. The difference in volume of distribution between free DOX and CLX were reported to be 60-300-fold, in favor of free DOX (22, 24), indicating that the initial tissue uptake might be almost absent in CLX as compared to free DOX. The free DOX is entering the cells in significant amounts just after 5 s of exposure (25), a phenomenon that is unrealistic for liposomal CLX. The “leaky vasculature” feature of the tumor tissue frequently reported to be a factor of therapeutic advantage might be playing a negligible role (26). Actually, the easiness of uptake of free DOX by the tumor cells might not compare to liposomal CLX following exposure to the same concentration, 10 μg/ml. Theoretically, only exclusively selective tumor-targeting particles might accumulate in tumors at quantities sufficient enough to eventually elicit uptake.
Our decision to focus on uptake events during short term exposure was based on several assumptions. The in vitro experiments exploring long term exposure, e.g. 96 h and more (27), might not be entirely relevant to clinical situations, since early studies have indicated that long term in vitro exposure should not include in vitro experiments with a free DOX dose exceeding 0.05 μg/ml (21). The only way liposomal CLX can enter the cell via endocytosis is to adhere to every single cell membrane in vivo shortly after injection. Alternatively, liposomes can stay in the vicinity of target cells and continue to elicit antitumor activity by passively extruding DOX into the tumor microenvironment. We assume this is the prevailing mechanism in the activity of liposomal DOX (Figure 3). It is quite possible that DOX leaked from liposomes provides high drug concentration outside the cell for a prolonged period. This slows-down the active drug efflux due to high extracellular content. The active extrusion of DOX continues in viable tumor cells after cessation of therapy. Some of the extrudedµµ DOX is inevitably re-entering the cell by passive diffusion during this period. The role of liposomes located in the vicinity of tumor cells during prolonged efflux via p-glycoprotein is to delay cellular efflux. The most prevailing theory ascribing the antitumor effect of cytotoxic drug assumes that the extent of cellular internalization governs the cytotoxicity of antitumor agents.
The clinical activity of liposomal DOX does not seem to be inferior to free DOX at least in some settings (28). The involvement of host-mediated effects of CLX is also not considered very feasible, since this formulation of liposomal DOX is pegylated. We assume that one of the mechanisms involved in the tumor control elicited by liposomal DOX might be governed by the combination of two factors: 1) long tissue persistence of leaking liposomes and delay of efflux, 2) absence of drug “distraction” into adjacent apoptotic bodies. Insufficient binding to degraded chromatin as seen in our study, might be an advantage when extracellular retention of DOX is required for longer periods. The latter factor might be of significant importance in determining chemotherapy resistance. In fact, any cytotoxic drug tends to preferentially penetrate non-viable, apoptotic cells, or just DNA debris available in excess in the tumor microenvironment. Chemotherapy and radiotherapy from previous treatment cycles might additionally generate these “DNA waste” particles. This DNA waste is overloading phagocyte waste handling function. Thus, free DOX after i.v. injection might be distributed into non-viable, apoptotic deposits or any other DNA waste residues, limiting its activity. Conversely, liposomal drug deposited in the vicinity of tumor tissue does not extrude the whole incapsulated DOX at once and therefore, allows waste handling phagocytes to elicit their function. The hypothesis of waste handling phagocyte failure in the presence of excess DNA is currently used to explain the pathogenesis of lupus (29). We assume that DOX encapsulation into liposomes plays a beneficial role in preserving the functional integrity of phagocytes. Liposomal formulation of CLX does not result in rapid delivery of bulk DOX into permeable cellular debris and therefore, does not deprive it from the target viable cell. The subdiploid peak can represent, in addition to apoptotic cells, nuclear fragments, clumps of chromosomes, micronuclei, and some other structures (30). However, in all scenarios CLX did not rapidly bind to these components in our experiments. Not bound DOX present in the tissue for prolonged period of time might continue to suppress the rapid re-growth of the tumour. Interestingly, free DOX has been shown to persist in human tissues for months and years following cessation of chemotherapy (31). While CLX cannot penetrate permeabilized cells, its presence in liposomal formulation assures leakage of free DOX for an exceptionally long time. The extremely slow availability of the drug passively diffusing from CLX liposomes has been previously demonstrated (32). The accidental extravasation of CLX has indicated that tissue damage by the liposomal formulation is significantly less compared to free DOX extravasation (33, 34). It would be of interest to determine if DOX encapsulated into liposomes retains its functional activity in human tissues for a few years or even for a longer period of time. This implies the possibility that CLX might have additional clinical value in therapies where prolonged local deposition of cytotoxic drug is desirable. An example of this therapy option might be intraarterial chemoembolization, urine bladder instillations or implantable drug delivery systems (35) exploring lipDOX.
Since its FDA-approval in 1995, CLX has not become a standard anticancer drug for multiple indications as originally expected. In vivo and in vitro data have not clarified why DOX and CLX are active against solid tumors to about the same extent. We assume that slow release of free DOX by tissue deposited liposomes might be an important factor indirectly inhibiting active DOX efflux from the tumor. Yet, the antitumor activity and toxicity of CLX is usually attributed exclusively to its pharmacokinetics in the blood (36, 37). Conversely, slow release of free DOX from circulating liposomes might not be the most important factor in tumor control, since miniscule blood-leaked drug fraction is usually eliminated or metabolized too soon.
Conclusion
Viable and permeabilized cells internalize free DOX significantly faster that liposomal CLX. Liposomal CLX binds to degraded DNA weaker that free DOX and does not induce substantial leukocyte nadir in vivo. Slow release CLX liposomes are likely to elicit their antitumor activity during their post-chemotherapy persistence in tissues rather than from their prolonged presence in circulation.
Acknowledgements
The Authors would like to thank Mrs. Nijole Matuseviciene for her assistance with cell culturing and cryopreservation.
Footnotes
Authors’ Contributions
GZ and SG participated in the study design, carried out all experiments and drafted the manuscript, BP performed DLS and HPLC analysis. JAK, MT and LZ assisted in cell handling, flow cytometry and drug exposure experiments, MS carried out electron microscopy experiments. AD and VP performed experiments on mice, sample collection and blood CBC analysis. JAK and VP performed the statistical analyses. All Authors read and approved the final manuscript.
Conflicts of Interest
The Authors declare that they have no competing interests in relation to this study.
- Received March 28, 2021.
- Revision received April 10, 2021.
- Accepted April 13, 2021.
- Copyright © 2021 International Institute of Anticancer Research (Dr. George J. Delinasios), All rights reserved.