Simultaneous release of two drugs from polymer nano-implant inhibits recurrence in glioblastoma spheroids
Greeshma Devassy1, Ranjith Ramachandran1, Kottarapat Jeena1, Vijayabhaskar R. Junnuthula1, Vindya K. Gopinatha2, Cheripelil A. Manju1, Maneesh Manohar1, Shantikumar V. Nair1, Sathees C Raghavan2 and Manzoor Koyakutty1
*These authors contributed equally to this work
1Amrita Centre for Nanosciences and Molecular Medicine, Amrita Vishwa Vidyapeetham, Amrita Institute of Medical Sciences, Kochi 682 041, India.
2Department of Biochemistry, Indian Institute of Science Bangalore, 560012, India.
Local implant-based delivery of rationally selected combination of chemotherapeutics has some major advantages for the treatment of glioblastoma such as: (a) 100 % bio-availability locally in brain can be achieved at the tumor site (b) avoid systemic leakage and associated toxicity, and (c) simultaneous inhibition of multiple, mutually exclusive cancer mechanisms is possible. Here, we report a polymeric brain implant capable of delivering two different drugs in recurrent glioma cells. We have selected a combination of clinically used DNA alkylating agent, Temozolomide, and a DNA mismatch repair protein (Ligase IV) inhibitor, SCR-7, and delivered simultaneously into tumor spheroids formed by rat glioma cells, C6. The dual-drug loaded polymeric wafer, prepared by lyophilization method, could deliver both the drugs in a controlled fashion. To test the efficacy of this system, we have optimized an in vitro recurrent model of glioma spheroids wherein, the implant released both the drugs in a sustained fashion, thereby continuously exposing the cells to DNA methylation while inhibiting the DNA repair pathways. This leads to synergistic toxicity and inhibition of tumor recurrence for extended duration compared to free drug combination.
SCR7, temozolomide, neurospheres, Polymer implants, glioblastoma
- TMZ: temozolomide
Rationale and Purpose
Local recurrence is one of the major challenges in high grade glioma. Prolonged delivery of multiple drugs using nano polymer implants that can enhance the anti-tumoral activity locally in brain will have significant impact in anti-glioma therapy.
Glioma is one of the most devastating primary brain tumors with median survival of 12–15 months.1 Even with recent advancements in surgery, radiation therapy, and chemotherapy, the five-year survival is less than 5%.2 Current clinical drugs are temozolomide (TMZ), given orally, and BCNU given as intracranial implant (GliadelTM). Systemic TMZ therapy faces the challenge of penetrating the blood brain barrier (BBB) which can lead to the lack of an effective concentration of the drug reaching the target brain tissue. Drawbacks of the GliadelTM wafer are a short sustained release of one week and limited tissue penetration. Tumor recurrence is a prime factor deciding patient prognosis, which is almost inevitable for glioma and has been attributed to drug resistances arising from genomic mutations such as EGFR vIII, IDH, MGMT promoter methylation, etc.3-8 In addition, cancer stem cells are another important factor responsible for recurrence due to its capability to repopulate glioma cells even after treatment and also contributes to drug resistance from altered DNA repair mechanisms. Multiple molecular mechanisms play a role in maintaining cancer stem cells, which makes it difficult to target with a single chemotherapeutic agent.9-14 It is also reported that in almost 90% of glioma patients recurrence is at the original tumor location within a depth of 2 to 3 cm from the resection margin.15
To address these multifactorial challenges, multiple chemotherapeutic molecules may need to be delivered together; for example, one for tumor kill, and another to address resistance mechanisms. However, attaining therapeutic concentration of various drugs simultaneously at the tumor vicinity, by crossing BBB, is a major challenge when given systemically. One interesting option for delivering a suitable combination of drugs targeted to different cancer mechanisms is the use of biodegradable, drug-loaded brain implants that can direct 100% delivery in the tumor resected cavity in a controlled fashion.16–19 Polymer drug delivery systems which include drug eluting rods, gels, wafers, nanoparticles, and microcapsules have drawn significant attention over the last few decades.20–23 However, GliadelTM is the only FDA-approved intracranial implant available in clinic which delivers cytotoxic drug BCNU [1,3-bis (2-chloroethyl)-1-nitrosourea]. Although median survival was improved by ̴ 2.3 months as a result of 7 days of BCNU release from GliadelTM, tumor recurrence was reported in almost all patients. The short half-life of BCNU, its limited tissue penetration capacity, and short duration of release (7 days) are proposed as possible reasons for the limited success of GliadelTM.24–26 To combat these challenges we recently developed an electrospun polymeric wafer eluting TMZ from 7 days up to one month in brain resulting in prolonged survival for 85% of treated tumor-bearing animals.27 However, recurrence was seen in 15% of animals indicating that the TMZ-resistant population needs to be addressed judiciously by combining with other novel molecules which can target the DNA repair pathways in the TMZ-resistant population.
In this study, we developed a dual-drug releasing polymer implant loaded with TMZ together with a novel DNA mismatch repair protein (Ligase IV) inhibitor; SCR-7.28–31 TMZ induces DNA damage by adding methyl groups on the O6 position of guanine residues. This will cause futile cycles of DNA repair (mismatch repair) leading to the accumulation of double strand breaks (DSB) triggering apoptosis-mediated cell death.28-30 SCR-7 is an inhibitor of non-homologous end joining (NHEJ), one of the primary mechanisms of DSB repair.31-33 It blocks NHEJ by inhibiting Ligase IV, a mismatch repair protein, leading to the accumulation of DSBs and activates apoptotic pathways.34-35 We hypothesize that SCR7 will enhance the cytotoxicity caused by TMZ, by arresting the ligase IV mediated DSB repair, when delivered together. Thus, by inhibiting the DNA repair mechanism, we propose to improve the sensitivity of cells towards alkylating agents. To deliver TMZ and SCR 7 together, we developed a polymeric brain implant using simple lyophilization method where both the drugs are co-loaded in appropriate concentrations. Moreover, we have developed an in vitro tumor recurrent assay using glioma spheroids wherein the cells survived after the treatment with the test samples were further cultured for 3–12 days to check its ability to recur and form tumor spheroids.
Materials and methods
All the chemicals and reagents used in this study were obtained from Sigma (St. Louis, U.S.A.) unless specified otherwise. Temozolomide was purchased from AK Scientific, USA, SCR-7 was synthesized as described before,31 PLGA (75:25) molecular weight 17,000-20,000 was purchased from Wako Pure Chemical Industries, Ltd., Japan. PLA molecular weight 1,50,000 was purchased from Goodfellow Cambridge Ltd (England).
In vitro cytotoxicity assessment.
C6 rat glioma cell line and T98G MGMT over-expressing human glioma cell line were procured from National Centre for Cell Science, Pune, India. The cells were maintained in Dulbecco's Modified Eagle's Medium (DMEM; Invitrogen, CA, USA) supplemented with 10% Fetal Bovine Serum (FBS; Invitrogen, CA, USA) and with 50 IU/mL penicillin and 50 µg/ml streptomycin. Cells were incubated in a humidified atmosphere of 5% CO2 at 37˚C. Cell proliferation was measured using MTT colorimetric assay. For cytotoxicity assessment of free drug combinations, C6 cells were seeded in 96 well plates at a seeding density of 1×103 cells per well. All concentrations were used in triplicates. After treatment period of 72 hours, supernatant was removed and cells were washed with PBS. 10% Next, MTT reagent in medium was added and plates were further incubated for 4 hours at 37˚C in humidified incubator. After incubation solubilizing buffer is added to dissolve the formazan crystal and mixed well. Absorbance was read at 570 and 660 nm using microplate reader (PowerWave XS, BioTek, Vermont, U.S.A.). In vitro cytotoxicity assessment of TMZ and SCR7 polymer wafer combination were carried out in comparison with two sets of control groups (i) wafer group (namely bare wafer, TMZ wafer, SCR7 wafer and TMZ+SCR7 wafer) and (ii) free drug group (namely media alone, TMZ, SCR7 and TMZ+SCR7). C6 cells were seeded in 24 well plate at a seeding density of 5×103 cells per well. MTT assay was performed as mentioned above.
Colony / neurosphere formation assay.
C6 rat glioma cells were grown in serum free DMEM supplemented with EGF (20 ng/mL, R&D Systems Germany), FGF (20 ng/mL, R&D Systems Germany) and B27 (20 µg/mL, Gibco USA). Cells were seeded at seeding density of 5×103 cells per well in 24 well plate. At 24 hours following the seeding, cells were treated with drug/wafer containing media and maintained at 37˚C in humidified incubator for 12 days without media change. In the wafer group, wafers were weighted and added so that the final doses would be 550 µM TMZ and 100µM SCR7. Photographs were taken at regular intervals using Nikon Eclipse TE2000U at 10× magnification. MTT standard curve was plotted for quantitative estimation of cells in the neurosphere. C6 cells were seeded from a range of 0.1×103 to 1.4×104. MTT reagent was added and incubated for 4 hours at 37˚C in a humidifier. Formazan crystal solubilization was carried out and absorbance was read at specific wavelengths and a standard curve was plotted. For plotting cell growth curve, neurospheres from each time point were treated with MTT and a cell number was obtained from corresponding absorbance value in the standard curve.
To determine apoptosis, treated C6 cells were stained with Annexin V and propidium iodide (PI) using Alexa Fluor Annexin V/Dead Cell Apoptosis Kit (Thermofisher, USA) 72 hours following incubation. Camptothecin was used as positive control and untreated cells as negative control. Assay performed according to manufacturer’s instruction. Stained cells were analyzed by flow cytometry (FACS Aria, BD Biosciences, US)
Characterization of spheres
For studying cell viability within the spheroids, live dead assay was performed using live/dead cell viability assay kit (Thermofisher, USA) following manufacturer’s instruction and viewed under fluorescent microscope (Leica DMI 3000). H & E staining was done for neurospheres. C6 spheroids were grown on acid-etched coverslips. Spheroids were then formalin fixed, dehydrated, and stained with modifications in the conventional H&E staining protocol. For immunostaining for nestin, spheres were grown on cover slips, fixed with 4% PFA (paraformaldehyde), permeabilized using 1% triton, and blocked using 1% FBS. Cover slips were then incubated with specific antibody overnight at 4°C, washed and counter stained with PI. Spheres were imaged using confocal microscope.
Preparation of TMZ and SCR7 loaded wafer.
PLGA:PLA wafer was synthesized by lyophilization method. PLGA: PLA in ratio 25:75 were dissolved in Dichloro Methane (DCM) by stirring at 600 RPM for approximately 5 hours. Once dissolved, ethanol was added to the polymer solution and mixed well by stirring for 15 minutes. The blended polymer solution is then poured in to a glass plate, snap frozen using liquid nitrogen and kept for lyophilization (LABCONCO, Care Biosystems India Ltd). Polymer film thus obtained was stored at 4°C. For the preparation of TMZ-loaded wafers, 10% TMZ was dissolved in DCM along with PLGA: PLA. The wafer was prepared in the way mentioned for bare wafers. Similarly, for SCR7 loaded wafer, 5% SCR7 was dissolved in DCM and wafers synthesized in the same method as mentioned above.
Characterization of lyophilized wafer
Scanning electron microscope analysis was carried out to study the micro morphology, porosity and surface texture of the wafer. A piece of wafer was mounted on a metallic stub using double sided adhesive tape, coated with gold using Auto Fine coater (JFC-1600 JEOL, Japan) and finally imaged by SEM (JSM-6490LA JEOL, Japan). The elemental composition of the wafer was studied by Energy Dispersive Spectral Analysis using JEOL JSM-6490LA system and the spectral images were overlaid with the SEM image at 3000× magnification. The analysis was performed at a counting rate of 1898 cps under accelerating voltage of 15 kV for a sweep count rate of 26 after mounting samples on aluminum stub. FTIR spectra over the wavelength range 4000 – 400cm-1 was recorded for drug-loaded wafers and also free drug using FTIR spectrometer IRAffinity 1S (Shimadzu, Scientiﬁc Instruments, Princeton, NJ, USA)
Estimation of drug loading
HPLC was performed to determine the amount of drug loaded in the wafer. A known amount of drug-loaded wafer was dissolved in 0.5 mL of chloroform. After dissolving completely, an equal volume of methanol was added and mixed well. To this methanol chloroform mix, 0.5 mL of deionized water was added, and centrifuged at 6000 rpm for 10 minutes to get a phase separation. The methanol-water upper phase containing the drug was carefully separated and HPLC analysis was performed. The HPLC system consisted of a Shimadzu LC-9A pump and a Model 3200 PDA detector (Shimadzu, Scientiﬁc Instruments, Princeton, NJ, USA) set at 330 nm. Separation was accomplished on a Qualisil GOLD, ODS, 5 µm, 150×4.6 mm C18 column. For TMZ, mobile phase consisted of 0.5% glacial acetic acid–methanol (90:10, v:v) and was delivered at 1.0 mL/min. The retention time was found to be 7.8 minutes and λmax 330 nm. For SCR7, the mobile phase consisted of a mix of acetonitrile and water (1:1) delivered at a rate of 1.0 mL/min. Retention time was 4.3 minutes and λmax 220 nm.
In vitro drug release study.
Drug release was carried out in artificial cerebrospinal fluid (CSF) (NaCl 148 mM, KCl 3mM, MgCl2 0.8 mM, CaCl2 1.4 mM, Na2HPO4 1.5 mM, NaH2PO4 0.2 mM, BSA 0.1 mg/mL) with pH maintained in the physiological range (7.1–7.3). Drug-loaded wafers were submerged in 5 mL of simulated CSF taken in plastic vials and kept in a shaking incubator at 37°C with stirring of 50 rpm. Next, 0.5 mL of the release media was collected at specific time intervals and replenished with an equal volume of fresh media. HPLC was performed for each sample and the cumulative drug concentration was estimated from the standard curve equation.
Graphs were generated and statistical analysis was done using OriginPro 8 software (Version 8.0724, Origin Lab Corporation, Northampton, USA). P<0.05 was considered as statistically significant.
Results and discussions
SCR7 enhanced TMZ induced cytotoxicity and apoptosis
Preliminary assessment of TMZ+SCR7 combination therapy was studied using MTT assay. First, we tested the cytotoxicity of free TMZ, SCR7 and TMZ+SCR7 combination on two different glioma cell lines, rat C6 and human MGMT over-expressing T98G. Untreated cells served as control. For TMZ, IC 50 was 550 µM for C6 cells (Figure 1A) and ̴ 150µM for SCR7 (Figure 1B). However, when 550µM TMZ was used in combination with 100 µM of SCR7 (IC-90), the response was almost 80% cell toxicity for C6 cells. Similar effect was observed for MGMT over-expressing T98G cells too. Further, we confirmed the cell toxicity using apoptosis assay (Annexin V-PI) in flow cytometry. Untreated cells served as negative control and camptothecin-treated cells as the positive control. After 72 hours of drug treatment, 76.7% cells were live in the SCR7 treatment group. On the other hand, TMZ induced a better cytotoxicity and only 22.4% were alive. Combination therapy (TMZ+SCR7) significantly improved the toxicity, leaving only 7.2% cells alive (Figure 2D), after the treatment.
Figure 1. TMZ and TMZ+SCR7 dose response curve on C6 cells (A) and T98 cells (B) shows enhanced toxicity with TMZ+SCR7 combination (C) SCR7 dose response curve.*P<0.05
Combinatorial effect of free drugs on neurospheres
Compared to typical glioma neurosphere assays, our aim was to create an in vitro tumor recurrent assay using 3D culture model. For this, first, we treated C6 cell lines with chemotherapy drugs of interest, TMZ (550uM), SCR-7 (100 uM) or their combination and cells remaining in the live quadrant were further cultured for their ability to grow back and forms colonies. These residual cells were cultured in serum free DMEM F12 media, supplemented with neural growth factors for a period of 12 days. As seen in Figure 2E, in untreated control, aggressive growth of colonies with cell to cell contact was seen within 3–12 days of culture. The spheroids were stained with LIVE/DEAD (CalceinAM/EthD-1) assay to study the viability of cells within the colony. Calcein is a nonflourescent dye, but after binding to cells gets converted into fluorescent calcein AM whereas EthD-1 binds to necrotic cell-DNA giving red fluorescence. Cells in the neurospheres were stained with calcein indicating live cells throughout the colony (Figure 2A i and ii). H&E staining shows spheroid cells with filopodial extensions and stained uniformly throughout (Figure 2A iii). We used nestin staining to examine the stemness of the cells. Nestin is a marker for identifying neural progenitor cells.36 Neurospheres were stained positive for nestin indicating the presence of neural precursor cells (Figure 2B).
Effect of TMZ (550uM) alone in spheroid recurrence is shown in Figure 2E. The cells cultured from the live quadrant could easily repopulate and form active spheroids by days 9–12. Many active small colonies where seen around the large colony indicating a clear case of recurrence. In SCR7 treated group, by day 3 itself, the spheres started forming. By days 6–12, large interconnected tumor spheres could be seen. This indicates that Ligase IV inhibition alone has no specific impact on the tumor cells. However, when SCR7 was combined with TMZ, the colony formation was found again delayed to days 9 and 12. The study was terminated on day 12 as both the control and SCR7 treated spheroids were over confluent and media depleted. MTT assay performed at each time point quantitatively measured the cell growth (Figure 2C) and showed that, compared to the untreated control, SCR7 and TMZ group, the rate of cell proliferation was relatively restricted in TMZ-SCR7 combination. However, from the TMZ-SCR7 cultures it is evident that although the metabolic activity was reduced, the free drug combination could not effectively inhibit spheroid re-growth which is the prime reason for tumor recurrence in vivo. We believe that, the free drug combination could not provide sustained cytotoxic stress to the spheroid cells and some of the refractory cells could recover over a period of few days. Next, we tested whether prolonged and sustained release of both the drugs using a slow-releasing polymer implant can overcome this challenge.
Preparation and characterization of polymeric wafer for combinatorial sustained drug release
A biodegradable polymeric (PLGA-PLA) wafer loaded with TMZ, SCR7 and TMZ-SCR7 combination was prepared using a lyophilization method (Figure 3A). Different PLGA: PLA concentrations were tried and the combination with desired release kinetics and physical properties such as flexibility, porosity, strength etc., was selected for the experiments (25:75). SEM analysis revealed the formation of highly uniform porous interconnected morphology of the wafers. The pore sizes ranged from 10 to 50µm (Figure 3B). Surface porosity play a key role in drug release and polymer degradation, especially in the case of bulk eroding polymers like PLGA. The presence of porous structure can help in enhanced penetration of physiological fluids leading to better drug release and polymer degradation by hydrolysis.37 The elemental mapping of TMZ and SCR7 wafers (Figure 3C & D respectively) showed uniform distribution of drugs throughout the wafer. Energy Dispersive Spectral (EDS) analysis enabled the qualitative mapping of the element over the substrate. The nitrogen atoms present in the TMZ molecules were mapped on the TMZ-loaded wafer as shown in Figure 3E. Similarly, the nitrogen and sulphur atoms present in the SCR7 were mapped on SCR-7 loaded wafer as shown in Figure 3F. Uniform distribution of the drugs in carrier matrix was important for obtaining reproducible controlled release. FTIR spectra of free drug as well as drug-loaded wafers, depicted in Figure 3G, showed the characteristic vibration of TMZ molecule at 1610 cm−1 due to C=C or C=N stretching vibration. The N-H stretching vibration at 3400 cm-1 was not so evident in the TMZ wafer. FTIR spectra of SCR7 and SCR7 wafer showed C–S stretching vibrations at 3271 cm-1, C–S bending at 1082 cm-1 whereas N-H bending vibrations at 924 cm-1.