Increased efficacy of doxorubicin delivered in multifunctional microparticles for mesothelioma therapy



New and effective treatment strategies are desperately needed for malignant mesothelioma (MM), an aggressive cancer with a poor prognosis. We have shown previously that acid-prepared mesoporous microspheres (APMS) are nontoxic after intrapleural or intraperitoneal (IP) administration to rodents. The purpose here was to evaluate the utility of APMS in delivering chemotherapeutic drugs to human MM cells in vitro and in two mouse xenograft models of MM. Uptake and release of doxorubicin (DOX) alone or loaded in APMS (APMS-DOX) were evaluated in MM cells. MM cell death and gene expression linked to DNA damage/repair were also measured in vitro. In two severe combined immunodeficient mouse xenograft models, mice received saline, APMS, DOX or APMS-DOX injected directly into subcutaneous (SC) MM tumors or injected IP after development of human MMs peritoneally. Other mice received DOX intravenously (IV) via tail vein injections. In comparison to DOX alone, APMS-DOX enhanced intracellular uptake of DOX, MM death and expression of GADD34 and TP73. In the SC MM model, 3× weekly SC injections of APMS-DOX or DOX alone significantly inhibited tumor volumes, and systemic DOX administration was lethal. In mice developing IP MMs, significant (p < 0.05) inhibition of mesenteric tumor numbers, weight and volume was achieved using IP administration of APMS-DOX at one-third the DOX concentration required after IP injections of DOX alone. These results suggest APMS are efficacious for the localized delivery of lower effective DOX concentrations in MM and represent a novel means of treating intracavitary tumors.

New treatment strategies are desperately needed for the intracavitary tumor, malignant mesothelioma (MM), which typically originates from mesothelial cells of the pleura, peritoneum or pericardium. Current therapies for MM include systemic chemotherapy, gene therapy, surgery, radiation or palliative procedures.1 The effectiveness of intracavitary chemotherapy is also a topic of intense examination,2, 3 including the administration of intracavitary hyperthermic chemotherapy.4, 5 Single modality therapies have little effect on patient survival, while multiple modality therapies are only slightly more effective.6–11 For example, doxorubicin (DOX) has been used in the treatment of a variety of cancers, including MM and breast cancer but can cause cardiac and systemic toxicity.12 The ineffectiveness of systemic chemotherapy in treating MM may be due to inefficient delivery and uptake of chemotherapeutics. Thus, to increase cell uptake of DOX and to circumvent its systemic toxicity, we explored the use of a novel system to deliver DOX directly to MM cells via local administration.

Vehicles for loading and delivery of chemotherapeutic drugs or molecular constructs have been featured recently in the cancer literature.13–15 Although nanomaterials have been advocated for drug delivery to some tumor types, their limitations include penetration and consequent dysfunction of cellular organelles, and the possibility of systemic toxicity due to entrance into the blood stream and transport to other organs.16, 17 To overcome these limitations and to avoid systemic dissemination, we developed nontoxic, micron-sized particles to allow their injection directly into established tumors, sites of surgical resection of tumors and peritoneal or thoracic cavities.

Acid-prepared mesoporous spheres (APMS) are porous, amorphous silica microspheres (1–2 μm in diameter) with a pore diameter that is tunable between 40 and 100 Å to allow loading of drugs, functional plasmids18 or DNA/siRNA constructs.19, 20 In contrast to crystalline silicas, amorphous silicas are less durable and are nontoxic.21 We have shown recently that APMS functionalized with tetraethylene glycol (TEG) are nontoxic in vitro or after intranasal instillation or intrapleural injection into mice, and that this functionalization promotes phagocytosis and uptake by murine lung epithelial and human MM cells in vitro.18 In studies here, we hypothesized that using APMS at sites of MM tumor development might be advantageous as a novel form of localized chemotherapy delivery to avoid systemic toxicity.

We first demonstrated that APMS-TEG loaded with DOX in comparison to DOX alone led to significantly (p < 0.05) increased delivery and drug uptake by MMs, increased MM cell death in vitro and enhanced expression of genes linked to DNA damage and death (cell cycle arrest and apoptosis) pathways. Next, we performed in vivo studies with fluorescently labeled APMS-TEG to confirm particle uptake by subcutaneous (SC) MMs growing in severe combined immunodeficient (SCID) mice, as well as by pleural cells and cells in peritoneal lavage fluid (PLF) after injection of APMS-TEG into the pleural or intraperitoneal (IP) cavities. In the SC human MM xenograft model, we found that localized injection of APMS-TEG loaded with microgram quantities of DOX was more effective in inhibition of SC tumor growth and nontoxic to mice when compared to systemic DOX administration. Finally, we used an IP human MM xenograft model to demonstrate that threefold lower concentrations of APMS-TEG-loaded DOX (APMS-DOX) are equally effective in inhibiting IP tumor growth as higher concentrations of DOX alone. Our results suggest that APMS are an effective approach for the localized delivery of chemotherapeutic drugs typically associated with systemic toxicity.


AOD: assay on demand; APMS: acid-prepared mesoporous spheres; APMS-DOX: doxorubicin-loaded APMS; DOX: doxorubicin; HPLC: high-pressure liquid chromatography; IP: intraperitoneal; IV: intravenous; LDH: lactate dehydrogenase; ME26: epithelioid MM cells; MM: malignant mesothelioma; MO: sarcomatoid MM cell line; PCR: polymerase chain reaction; PLF: peritoneal lavage fluid; QRT-PCR: quantitative real-time PCR; SC: subcutaneous; SCID: severe combined immunodeficient; TEG: tetraethylene glycol

Material and Methods

Cells and reagents

Human pleural MM cell lines isolated from surgical resection of MMs or at autopsy were kind gifts from Drs. Luciano Mutti (Maugeri Foundation, Pavia, Italy; MO), Maurizio Bocchetta (Loyola University, Mayfair, IL; ME26) and Harvey Pass (NYU School of Medicine, New York, NY; PPM Mill, PPM Rob, PPM Ada, PPM Gar, PPM Gates, PPM Gord). All isolates were confirmed as MM cells by immunohistochemistry using an antibody to calretinin and verified for lack of mycoplasma contamination using a polymerase chain reaction (PCR) assay prior to use in studies described here. Hmeso cells, originally designated H-MESO-1, were initially isolated by Reale et al.,22 and supplied by Drs. Joseph Testa and Deborah Altomare (Fox Chase Cancer Institute, Philadelphia, PA). Cell culture procedures were as described.18

Synthesis of APMS-TEG and preloading with DOX

Synthesis of APMS and surface modification with TEG have been described previously.18 Hereafter, APMS-TEG will be referred to simply as APMS. APMS were loaded with DOX via incubation with a saturated aqueous solution of DOX-HCl (Sigma, St. Louis, MO) for 16 hr and then isolated with centrifuge membrane filters (Nanosep MF 0.45 μm, Pall Corporation, Ann Arbor, MI) and dried under vacuum at room temperature in the dark.

Treatment of MM cells with APMS, APMS-DOX, DOX and cisplatin

For initial studies examining differences in DOX delivery via high-pressure liquid chromatography (HPLC), the sarcomatoid (MO) MM cell line was plated in complete medium at 37°C. When cells reached 70–80% confluence, the complete medium was aspirated and replaced with maintenance medium (containing 0.5% fetal bovine serum) for 24 hr. APMS were then suspended in maintenance medium at a concentration of approximately 6 × 107 APMS/100 μl, and sonicated 5× for 2 sec to achieve an even suspension. Fifty microliters were then added to cells at a final concentration of 7.5 × 106 APMS/cm2 surface area dish (∼185 APMS/cell). DOX (160 nM) was added directly to maintenance medium and APMS-DOX (160 nM DOX equivalent) was prepared and administered as described above. Cell killing effects of DOX alone, APMS alone, DOX added simultaneously with APMS and APMS loaded with DOX (APMS-DOX) were assessed using both the MO MM line and an epithelioid MM line (ME26) cultured and treated as described above. Finally, seven MM cell lines (Hmeso, PPM Mill, PPM Rob, PPM Ada, PPM Gar, PPM Gates and PPM Gord) were evaluated in dose–response studies to determine their relative sensitivity to DOX and cisplatin. Cells were grown as described above to near confluence, and DOX (0–25 μM) and cisplatin (0–75 μM) were added directly to maintenance medium.

Determination of extracellular and intracellular DOX concentration

After incubation of MO cells with either DOX (160 nM) in medium or APMS-DOX (160 nM DOX equivalent), medium was centrifuged at 10,000g at 4°C. The supernatant was then transferred to fresh vials, daunorubicin (a fluorescent internal standard) was added (final concentration of 1 μM), and the vial was vortexed prior to incubation at 37°C for 15 min. Proteins were precipitated by adding 250 μl of acetone and 50 μl of aqueous ZnSO4 solution (400 mg/ml) to the sample and vortexing prior to incubation at 37°C for 15 min. Samples were subsequently centrifuged at 20,000g (10 min) at 4°C and supernatants transferred to new vials and dried at room temperature for 2 hr under vacuum. The residue was solubilized in 500 μl of MeOH, and a volume of 100 μl used for HPLC analysis. To determine the intracellular DOX concentration, MO cells were trypsinized, centrifuged at 1,000 rpm in cold phosphate-buffered saline (PBS), supernatants were removed, and cells were resuspended in 190 μl of PBS and 5 μl lysis solution (Triton X-100, 3%). Then, 5 μl of proteinase K (10 mg/ml stock solution) was added. Samples were incubated for 45 min in a water bath at 65°C, and 56.4 μl of 10 μg/ml daunorubicin was then added to a final volume of 400 μl with PBS. Five microliters of phenylmethylsulfonyl fluoride was added for 10 min prior to the addition of 10 μl MgCl2 (0.4 M) and 20 μl DNase I (1 mg/ml). After centrifugation, the sample was incubated in a water bath at 37°C for 30 min; 450 μl of methanol and 45 μl of ZnSO4 (400 mg/ml) were added to 450 μl of each sample. After mixing, the sample was centrifuged for 5 min, and a volume of 100 μl was used for HPLC analysis.

High-pressure liquid chromatography

A published HPLC method for detection of DOX was modified to allow quantitation of extracellular and intracellular amounts of DOX.23–25 The HPLC system consisted of a 1525 Binary HPLC Pump, a 717 Plus Autosampler and a 2475 Multi-Wavelength Fluorescence Detector (Waters Corporation, Milford, CT). Chromatography was performed on a BDS Hypersil C18 column (4 × 150 mm2, particle size 5 μm). The mobile phase consisted of water:acetronitrile:tetrahydrofuran (75.5:24:0.5, v/v/v), with an apparent pH adjusted to 2.0 with perchloric acid. The flow rate was maintained at 1.25 ml/min with an injection volume of 40 or 100 μl. The effluents were monitored at an excitation wavelength of 480 nm and an emission wavelength of 560 nm at 40°C. Detection and integration of chromatographic peaks was performed using Breeze Chromatography v.3.20 software.

Lactate dehydrogenase assay

Cell lysis was measured by determining levels of lactate dehydrogenase (LDH) in the medium of MO or ME26 cells treated with either APMS (7.5 × 106 APMS/cm2), varying concentrations of DOX (40–800 nM) or APMS-DOX (7.5 × 106 APMS/cm2; 10–80 nM DOX equivalent) using a Cytotox 96® Non-Radioactive Cytotoxicity Assay (Promega Corporation, Madison, WI) as per the manufacturer's recommendations.18 The percentage of LDH release was calculated based on complete lysis induced by a positive control lysis buffer (0.09% Triton-X).

3-(4,5-dimethylthiazol-2-yl)-5-(3'-carboxymethoxyphenyl)-2H-tetrazollium (MTS) assay

MO cell viability after treatment with medium alone, DOX alone (80 nM), APMS (7.5 × 106 APMS/cm2) or APMS-DOX (7.5 × 106 APMS/cm2; 80 nM DOX equivalent) was measured in vitro using the colorimetric MTS assay, CellTiter 96® AQueous One Solution Cell Proliferation Assay (Promega) as per the manufacturer's recommendations.

Preparation of RNA and PCR array analyses

To determine if different patterns of gene expression related to cellular DNA damage/repair were observed after addition of DOX alone or APMS-DOX, RNA from MO cells treated with medium alone, APMS alone (7.5 × 106 APMS/cm2), DOX alone (80 nM) or APMS-DOX (7.5 × 106 APMS/cm2; 80 nM DOX equivalent) for 24 hr was prepared and purified using a RNeasy® Plus Mini Kit (Qiagen, Valencia, CA). After quality assessment, 1 μg of RNA was used for cDNA synthesis using the RT2 First Strand Kit (SABiosciences, Frederick, MD). Quantitative real-time PCR (QRT-PCR) was performed by the Vermont Cancer Center DNA Analysis Facility using RT2 Real-Time™ SYBR Green PCR Master Mix and Human DNA Damage Signaling Pathway RT2 Profiler™ PCR Arrays (SABiosciences; 7900HT Sequence Detection System, Applied Biosystems, Foster City, CA). QRT-PCR (TaqMan) was used to validate selected genes using assay on demand (AOD) primers and probes (Applied Biosystems). For the purposes of these studies, only gene changes ≥threefold are provided.

Confocal scanning laser microscopy for determination of APMS-Alexa-568 and APMS-Alexa-488 localization in vivo

APMS were covalently labeled with the fluorescent dye, Alexa Fluor 568 (APMS-Alexa-568), by standard peptide-bond forming methodology.18 APMS-Alexa-568 were suspended in PBS at a concentration of approximately 3.3 × 107 APMS/100 μl PBS. C57BL/6 mice (n = 3; Jackson Laboratories, Bar Harbor, ME) received intrapleural injections and were euthanized after 72 hr.18 Rib cage and adjacent diaphragm were removed surgically and either placed in Tissue-Tek O.C.T.® compound and snap-frozen or fixed in 4% paraformaldehyde and paraffin embedded. For determination of tumor uptake, 5 × 106 Hmeso cells were injected into four SC sites on the dorsa of 6-week-old male Fox Chase strain SCID mice (Jackson Laboratories, Bar Harbor, ME). Tumors appearing at 21 days were injected on day 28 with 3 × 106 APMS labeled with Alexa Fluor 488 succinimide ester (APMS-Alexa-488) for 24 hr at the distal pole of the tumor and removed for processing and step sectioning. Tissue section slides from frozen tissues were prepared,18 and representative images were acquired using a Zeiss LSM 510 META confocal scanning laser microscope (Carl Zeiss Microimaging, Thornwood, NY). All animal studies were approved by the Institutional Animal Care and Use Committee at the University of Vermont.

Inhibition of human MM growth in SC and IP mouse xenograft models

In the SC xenograft model, Hmeso cells (5 × 106 cells/injection site in 50 μl of 0.9% NaCl, pH 7.4, hereafter referred to simply as saline) were injected into four sites on the dorsa of 6-week-old male Fox Chase strain SCID mice (Jackson Laboratories) under isoflurane anesthesia. This MM line was selected because of its intermediate sensitivity to DOX and cisplatin in vitro (see Results) and its reproducible tumor growth and phenotype (biphasic MM). Tumors were first characterized by a board-certified pathologist using previously described criteria26, 27 to confirm their mesothelial origin. In an initial study (Experiment 1) using the SC model (n = 2 mice or eight MM tumors/group), Hmeso tumors appearing at 19 days postinjection were injected 1× weekly at their distal poles for 3 weeks with 50 μl saline alone, APMS (3 × 108 APMS/tumor, equivalent to 160 mg/kg APMS) in saline, DOX (6.15 μg DOX/tumor injection, equivalent to a total concentration of 1 mg/kg DOX/mouse) in saline or APMS-DOX (3 × 108 APMS; 6.15 μg DOX equivalent/tumor) in saline. Another group of mice received injections of DOX (5 mg/kg) at concentrations known to inhibit MM growth in rats28via intravenous (IV) tail vein injection. All mice were weighed weekly and examined every other day for tumor growth and morbidity. When the largest axes of the tumors in the control mice reached 1 cm, mice were euthanized as described above, necropsied to determine possible metastases, and major organs were removed and stored in 4% paraformaldehyde before processing for histopathology and evaluation by a board-certified pathologist. Tumors were weighed and were measured using digital calipers. Tumor volumes were calculated using the formula (π × long axis × short axis × short axis)/6. As the 1× weekly injection regimen failed to inhibit tumor growth, a second experiment was performed (n = 4 mice or 16 MM tumors/group) in which agents were administered 3× weekly (Experiment 2).

In the IP xenograft model (Experiment 3), Hmeso cells (5 × 106 cells in 50 μl saline) were injected into the lower left quadrant of the peritoneal cavity of 6-week-old male Fox Chase strain SCID mice (n = 5 mice/group). After 2 weeks, a time period in which all mice developed nonadherent spheroid MMs (see below), mice were injected IP 3× over the course of 1 week with 500 μl saline alone, DOX (1 mg/kg) in saline or APMS-DOX (7.8 × 108 APMS/mouse, equivalent to 0.33 mg/kg DOX) in saline. Based on our results in Experiment 2, in which we did not see a difference in tumor growth between APMS-DOX and DOX alone groups when administered SC at a higher concentration (1 mg/kg; see Results), we administered a lower dose of DOX (0.33 mg/kg) in APMS to demonstrate increased efficiency of this delivery method. In addition, this DOX concentration is similar to the dose of DOX delivered during intracavitary MM chemotherapy in humans (15 mg/m2, equivalent to 0.41 mg/kg).3 An APMS alone group was not included in Experiment 3 as separate studies have demonstrated that IP doses of APMS up to 500 mg/kg (equivalent to 3.8 × 109 APMS/mouse) are well tolerated in mice (data not shown). Mice were euthanized as described above, and tumors were classified as either “spheroid” (small, free-floating nodules) or larger “mesenteric” tumors attached to the peritoneal mesentery. The total number of spheroid or mesenteric tumors/mouse was determined along with individual tumor volume and weight. For each mouse, total tumor weight and volume were calculated separately for spheroids and mesenteric tumors by multiplying the average tumor weight and volume by the total number of spheroid and mesenteric tumors present, respectively. A summary of all in vivo xenograft studies and treatment regimens is presented in Table 1.

Table 1. Summary of in vivo treatment regimens following establishment of Hmeso tumors SC or IP
inline image

Collection and preparation of PLF for cytospins

To determine the presence of free APMS and those associated with cells in PLF in Experiment 3, the peritoneal cavity of each mouse was instilled with 5 ml of cold sterile PBS using an 18-gauge needle. Leaving the needle in place, the abdomen was gently massaged and PLF was aspirated back into the syringe and placed on ice. Cytospin slides were generated as described,18 and representative photographs were taken using an Olympus BX50 microscope (Olympus America, Lake Success, NY) and QCapture Pro v.6.0 software.

Statistical analysis

For all in vitro experiments, at least three independent experiments were performed (n = 2–4 samples/experiment). For in vivo tumor experiments, results are representative of two to five mice/group/experiment. Statistical significance was evaluated by ANOVA using the Student–Neuman–Keul's procedure for adjustment of multiple pairwise comparisons between treatment groups or using the nonparametric Kruskal–Wallis, Mann–Whitney or Tukey's honestly significant difference (HSD) tests. Values of p < 0.05 were considered statistically significant.


We first studied the amounts of DOX delivered to MO cells in vitro after administration in medium alone or via APMS-DOX. As DOX becomes rapidly bound to DNA, use of DNase to release DOX from DNA and subsequent addition of MeOH and ZnSO4 were necessary for accurate quantitation by fluorescence HPLC. Similar methodology was used to quantify the amount of extracellular DOX. Whereas the left axis (Figs. 1a and 1b) displays the percentage of the total dose of DOX, the right axis shows the actual dose (ng) of DOX/plate of cells. As the total amount of DOX contained within APMS prior to their addition to MMs was known, the amount of DOX remaining within the APMS over time could be determined. A small amount (10–15%) of DOX was initially detected in medium, presumably through leakage from APMS after sonication (Fig. 1a). Over a 24-hr period, intracellular DOX increased until a maximum of approximately 50% of the initial amount within the APMS was released. More importantly, the balance of DOX remained within the APMS for 48 hr, allowing further release over time. In contrast, the maximum amount of DOX transferred to cells when DOX was added directly to medium was approximately 25% of the initial dose (Fig. 1b). The drop in intracellular DOX concentration observed at 48 hr in APMS-DOX-treated cultures (Fig. 1a) is likely due to decreases in the overall number of cells available for analysis as a result of increased APMS-DOX cell death between 24 and 48 hr of exposure.

Figure 1.

Determination of DOX uptake in vitrovia HPLC after two different methods of administration to sarcomatoid (MO) MM cells for up to 48 hr. Extracellular and intracellular DOX levels in MO cells measured by HPLC show greater cell delivery and retention of DOX when (a) loaded and released from APMS (APMS-DOX; 7.5 × 106 APMS/cm2; 160 nM DOX equivalent) in comparison to (b) direct addition to medium (160 nM DOX). Points represent the mean ± SEM (n = 3 in triplicate experiments).

To comparatively assess the effects of DOX alone, DOX was added simultaneously (but separately) with APMS, and APMS-DOX dose–response studies were first performed on the sarcomatoid MM line. Initially, a range of concentrations of DOX were incubated with MO cells for 24 or 48 hr, and LDH release into the medium was determined as a measure of lytic cell damage. Little LDH release (<10%) was seen with doses of DOX alone ranging from 40 to 800 nM (Fig. 2a; top panel). In contrast, addition of APMS-DOX caused a dose-related release of LDH. At 48 hr, approximately 50% of total cell lysis was achieved at 65 nM DOX (Fig. 2a; middle panel). This potency was not achieved when individual preparations of APMS and DOX were incubated simultaneously with MO cells (Fig. 2a; bottom panel). These data also show that APMS alone at the highest concentrations (7.5 × 106 APMS/cm2 surface area of dish, equivalent to amounts loaded with 65 nM DOX in the APMS-DOX group) were nontoxic to cells over a 48-hr period.

Figure 2.

APMS-DOX delivery increases cytotoxicity in both sarcomatoid (MO) and epithelioid (ME26) MM cells in vitro. Cell lysis and viability were determined using the LDH and MTS assays, respectively. (a) LDH release by MO cells was measured under several treatment conditions including exposure to DOX alone (40–800 nM; top panel), exposure to APMS preloaded with varying concentrations of DOX (7.5 × 106 APMS/cm2, 10–65 nM DOX equivalent; middle panel) and exposure to individual preparations of unloaded APMS (7.5 × 106 APMS/cm2) and DOX (40–800 nM) simultaneously (bottom panel). (b) MTS viability assay confirming increased cytotoxicity of APMS-DOX (7.5 × 106 APMS/cm2; 80 nM DOX equivalent) in MO cells. (c) LDH release by ME26 cells. For LDH, percent release was calculated based on a complete lysis induced by a positive control lysis buffer (0.09% Triton-X; data not shown). For [(a), (b) and (c)], 0 = medium only control; bars represent the mean ± SEM of n = 2–4 treatment groups in duplicate or more experiments; *p < 0.05 in comparison to medium alone at same time point; †p < 0.05 in comparison to DOX alone group. (d) PCR array experiments reveal alteration in several genes related to DNA damage and repair in sarcomatoid (MO) MM cells treated with DOX (80 nM) or APMS-DOX (7.5 × 106 APMS/cm2; 80 nM DOX equivalent) for 24 hr. Upregulation of six genes and downregulation of one gene was observed after treatment with DOX or APMS-DOX compared to cells treated with medium alone. No significant gene expression alterations were observed with APMS alone. Only gene changes ≥threefold are provided. *p < 0.05 in comparison to DOX alone group. †Not significant compared to medium control group. Bars represent the mean ± SEM of n = 3 plates/group.

To further verify the more potent toxic effects of APMS-DOX on MM cells, we assessed cell viability using an MTS assay over 96 hr (Fig. 2b). When MO cells were exposed to either 80 nM DOX or APMS alone (7.5 × 106 APMS/cm2 surface area of dish, equivalent to amounts loaded with 80 nM DOX), there were no significant differences in cell viability in comparison to untreated cells. However, when cells were treated with APMS-DOX, significant (p < 0.05) decreases in cell viability occurred at 48 hr and thereafter. We next verified the increased potency of APMS-DOX in the LDH assay using ME26 cells (Fig. 2c). By 48 hr, the APMS-DOX (80 nM DOX equivalent) group showed a significantly (p < 0.05) higher amount of lytic cell death, i.e., ∼20% of total release, in contrast to 80 nM DOX alone. Because the greatest cytotoxic effects of APMS-DOX were achieved in sarcomatoid MMs, which have the worst prognosis,29 we performed further mechanistic studies on the MO MM line.

To shed light on the mechanisms of DNA damage by DOX and whether these were exacerbated in DOX delivery by APMS, we performed PCR array analysis on genes associated with apoptosis, cell cycle regulation and various forms of DNA repair. Treatment of MO cells with DOX (80 nM) or APMS-DOX (80 nM DOX equivalent) caused at least threefold (p < 0.05) increased expression of six genes (GADD45A, GADD153, GADD34, IHPK3, BTG2, TP73) in comparison to untreated MO cells (Fig. 2d). In contrast, mRNA levels of GSTE1 were significantly (p < 0.05) decreased. Messenger RNA levels of RAD50 were increased (p < 0.05) in response to APMS-DOX (80 nM DOX equivalent) but not DOX alone (80 nM). In addition, GADD34 and TP73 gene expression were significantly (p < 0.05) elevated in response to APMS-DOX when compared to the DOX alone group. No significant gene expression alterations were observed with APMS alone. QRT-PCR was used to confirm significantly (p < 0.05) increased gene expression of GADD45A and GADD34 in both DOX and APMS-DOX-treated cells (data not shown). These data indicate increased responses to DNA damage after delivery of APMS-DOX versus DOX alone to MM cells.

To determine whether APMS could enter and remain in tissues in vivo adjacent to sites of APMS injection, 3.3 × 107 APMS-Alexa-568 were injected into the pleural cavities of C57BL/6 mice (n = 3). Using confocal scanning laser microscopy, APMS-Alexa-568 was located adjacent to sites of injection in the soft tissue surrounding the ribs (Figs. 3a and 3b). As reported previously,18 APMS-Alexa-568 were seen occasionally in the diaphragm, lung and spleen using fluorescence microscopy. We then used SCID mice developing MMs after SC injection of Hmeso cells to determine whether APMS-Alexa-488 injected at the distal pole of the tumor migrated to the interior of the tumor. As shown in Figure 3c, APMS-Alexa-488 (green) were detected in the cytoplasm of MM cells in tumor masses. When cytospins were evaluated after IP administration of APMS-DOX (Experiment 3), APMS were detected in PLF samples, indicating these particles remain within the peritoneal cavity up to 10 days (Fig. 3d). Further, it appears as if most APMS are associated with MMs and macrophages, although additional studies are required to confirm this observation.

Figure 3.

Several in vivo experiments demonstrate that APMS are taken up by a variety of cell types and remain adjacent to sites of injection. (a, b) APMS-Alexa-568 (red) injected into the pleural cavity of C57BL/6 mice were found in cells lining the rib at 72-hr post injection. Cell nuclei are stained blue (TOTO-3). Arrows indicate the approximate site of APMS injections. Approximately 3.3 × 107 APMS-Alexa-568 were injected into individual mice (400×, scale bars = 20 μm). (c) APMS-Alexa-488 (green) injected at the distal pole of Hmeso SC tumors grown in SCID mice were detected within the cytoplasm of individual MM cells toward the center of the tumor mass. Tumor cell nuclei are stained blue (DAPI). Arrows indicate the presence of APMS-Alexa-488 signal. Approximately 3 × 106 APMS-Alexa-488 were injected into individual tumors (400×, scale bar = 20 μm). (d) Representative PLF cytospin showing colocalization of APMS with macrophages (arrow), sloughed mesothelial cells (black arrowhead), and MM cells (white arrowhead) (400×, scale bar = 20 μm).

Our final goal was to demonstrate the efficacy of APMS-DOX in a SCID mouse xenograft model. We first tested a number of human MM lines for their sensitivity to DOX or cisplatin over a range of concentrations in vitro and for reproducible tumor growth in SCID mice. The maximum concentration of DOX used in these dose–response studies (25 μM) was approximately equal to DOX concentrations achieved in peritoneal fluids (18.4 μM or 10 μg/ml) after intracavitary DOX chemotherapy (15 mg/m2 or 0.41 mg/kg) in patients with MMs.3 Dose–response experiments with cisplatin were also performed in each cell line to determine if patterns of drug resistance were the same as DOX in individual MM lines.4, 30 These studies revealed that different human MMs were heterogeneous in their sensitivity to either DOX or cisplatin (Figs. 4a and 4b). Moreover, the patterns of sensitivity to both drugs were similar in individual MMs. As the Hmeso MM line expressed intermediate sensitivity to drugs in comparison to other MM lines and grew reproducibly after injection into SCID mice, this line was selected for in vivo studies.

Figure 4.

Antitumor effect of APMS-DOX in a SC xenograft model of MM. Preliminary in vitro studies demonstrated human pleural MM cell lines are differentially sensitive to (a) DOX or (b) cisplatin as measured by the MTS assay when drugs are added at a range of concentrations to maintenance medium. Because the Hmeso MM line exhibited intermediate sensitivity to DOX or cisplatin and grew reproducibly when injected SC or IP in a SCID mouse xenograft model, it was used for (c) 1× weekly (Experiment 1) and (d) 3× weekly (Experiment 2) injection studies to evaluate the effects of DOX when administered systemically or loaded into APMS (APMS-DOX) on inhibition of MM growth. Systemic administration of 5 mg/kg DOX IV 3× weekly resulted in death of mice after two sequential doses (data not shown) *p < 0.05 in comparison to saline control groups. White arrows indicate time of injection of Hmeso cells (5 × 106), and black arrows indicate days at which mice were euthanized in respective experiments. Black arrowheads indicate times at which saline, APMS alone (160 mg/kg), DOX alone (1 mg/kg) and APMS-DOX (160 mg/kg APMS; 1 mg/kg DOX equivalent) were injected into mice. Each point represents the mean volume of all tumors within a particular treatment group ± SEM.

As shown in Experiment 1 (Fig. 4c), 1× weekly SC injections of APMS alone, DOX (1 mg/kg), APMS-DOX (1 mg/kg) or IV administration of DOX (5 mg/kg) over a 3-week period did not affect Hmeso tumor growth (n = 2 mice/group). Although histopathological analysis of organs revealed diffuse hepatocytic ballooning in the livers of mice administered DOX IV, no other adverse events were observed in organs evaluated from other treatment groups. However, when agents were administered 3× weekly (Experiment 2), tumor volume failed to increase after day 40 after SC injections of DOX (1 mg/kg) or APMS-DOX (1 mg/kg) and was significantly (p < 0.05) lower in these groups in comparison to saline- or APMS-treated groups at the termination of the experiment (n = 4 mice/group) (Fig. 4d). In these experiments, systemic administration of DOX (5 mg/kg) was lethal to all mice after two sequential IV doses (data not shown).

In Experiment 3, using an IP xenograft model (n = 5 mice/group), we reduced the concentration of DOX loaded in APMS (0.33 mg/kg) to determine if we could demonstrate increased or equivalent efficacy as DOX alone at threefold higher concentrations (1 mg/kg). In comparison to saline controls, mesenteric MM tumor numbers (Fig. 5a), volumes (Fig. 5b) and weights (Fig. 5c) were significantly (p < 0.05) less in mice administered with DOX (1 mg/kg) or APMS-DOX (0.33 mg/kg) IP 3× weekly. Although not statistically significant, it appears that administration of DOX and APMS-DOX also decreased growth of spheroid tumors. Hematoxylin and eosin staining of representative spheroid (Fig. 5d) and mesenteric (data not shown) tumors indicate a biphasic MM tumor type.

Figure 5.

Antitumor effect of APMS-DOX in an IP xenograft model of MM. DOX (1 mg/kg) and APMS-DOX (104 mg/kg APMS; 0.33 mg/kg DOX equivalent) administered IP 3× weekly for 1 week were equally effective in decreasing Hmeso IP mesenteric tumor (a) numbers, (b) volumes and (c) weights. Trends suggest that DOX and APMS-DOX were also effective in decreasing IP spheroid tumor numbers, volumes and weights, although statistical significance was not achieved. In (b) and (c), the left axis corresponds to spheroid tumors and the right axis corresponds to mesenteric tumors. *p < 0.05 in comparison to saline control group. For each mouse, total tumor weight and volume were calculated separately for spheroids and mesenteric tumors by multiplying the average tumor weight and volume by the total number of spheroid and mesenteric tumors present, respectively. Bars represent the mean value for each treatment group ± SEM of n = 5 mice/group. (d) Representative hematoxylin and eosin staining of spheroids from IP xenografts indicates a biphasic MM tumor type (40×, scale bar = 200 μm).


Limited effective treatment strategies exist for patients with MM. Many chemotherapeutic drugs, including DOX, have been used in single or combination therapies with most current treatments resulting in a mean survival time of 12–15 months.1, 6–11, 31, 32 Most cancer chemotherapeutic drugs cause adverse systemic side effects because of their lack of tumor specificity and necessary administration at high doses. Thus, we hypothesized that using APMS at sites of MM tumor development might be advantageous as a novel form of localized chemotherapy delivery to avoid systemic toxicity. We show here in preclinical studies that APMS are an attractive vehicle to administer chemotherapeutic drugs and to enhance their delivery, cellular uptake and cytotoxic effects. The APMS surface can be modified to contain a myriad of functional moieties, including fluorescent molecules for detection, gadolinium for magnetic resonance imaging (MRI) (Steinbacher et al., in press; Lathrop et al., unpublished data) and antibodies for targeting specific tumor types (Cheng et al., in press). Functionalizing the exterior surface of APMS with molecules such as TEG enables rapid uptake by target cells by a process evading lysosomal degradation.18 Other important characteristics of mesoporous microparticles are their large surface areas (up to 1,100 m2/g) and pore volumes (>1.0 cm3/g) that are advantageous in loading a large amount of “cargo”.33–37

Cisplatin and DOX have proven to be more effective than a variety of single chemotherapeutic drugs in single modality treatment of MM.1, 6–10 Because DOX is more chemically stable and quantifiable using HPLC, it was selected for our studies. In addition to showing increased drug delivery and toxicity of APMS-DOX to MMs in vitro, we reveal the increased expression of a number of genes influencing DNA damage/repair. DNA damage is generally regarded as the primary cause of DOX-induced cell death in tumor cells. The molecular events linking DOX-induced DNA signaling to cell death have not been defined, but may involve cell signaling cascades, free radical-mediated events and DOX–DNA adduct formation.38, 39 Upregulation of three GADD (growth arrest and DNA damage response) genes was observed in DOX or APMS-DOX-treated MM cells. Although GADD45 is an oxidative stress responsive gene induced by DOX in a number of cell types, expression of GADD153 (DNA damage inducible, alpha or DDIT3), a redox-activated gene that plays a role in cell cycle arrest, apoptosis and activation of signal transduction pathways after DNA damage, is a novel finding that may be important in combination chemotherapy of MMs. Increased mRNA expression of IHPK3 (inositol hexaphosphate kinase 3), a gene linked to the development of apoptosis, and BTG2 (BTG family member 2), a less well-characterized p53-mediated DNA damage response gene associated with DNA repair, were also increased by DOX or APMS-DOX administration in our studies. Moreover, GADD34 and TP73 mRNAs were increased (p < 0.05) more markedly in MM cells after exposure to APMS-DOX in comparison to DOX alone, presumably because of the increased delivery of DOX when loaded into APMS. The increased expression of TP73 (encoding Tumor protein p73) in MMs treated with APMS-DOX was another novel observation in our gene array studies. This gene has been previously implicated in regulating a p53-dependent apoptotic pathway in tumors treated with chemotherapeutic drugs.40 In contrast to the well-studied p53 gene that encodes a tumor suppressor gene mediating cell cycle arrest or apoptosis in response to DNA damage, the TP73 gene encodes an array of isoforms not possessing typical tumor suppressor gene functions (reviewed in Refs.41, 42).

Significant (p < 0.05) decreases in GTSE-1 expression, which encodes a cell cycle-regulated protein (hGSTE-1 or human G(2) and S-phase expressed-1), were also observed in both DOX- and APMS-DOX-treated MM cells. As hGSTE-1 is able to downregulate p53 levels and activity,43 its decreased expression in drug-treated MM cells may represent a repair mechanism whereby p53 function is kept intact.

Overall, studies have demonstrated that APMS are a novel and effective tool for enhanced delivery of DOX. Localized administration of APMS-DOX inhibits tumor cell growth in mouse models of MM via SC or IP injection in the absence of systemic toxicity. Most importantly, threefold lower effective concentrations of DOX were achieved after loading in APMS as opposed to injection of DOX alone in an IP mouse xenograft model of MM. In this model, APMS-DOX or DOX was injected when all of the mice had pre-established MM spheroids as demonstrated in preliminary experiments. Although DOX has been incorporated into RGD-modified44 and polyethylene glycol-derivatized liposomes45 and nanoparticles46 to increase its stability in the intravascular compartment, our studies are unique in that APMS were engineered in a size range (1–2 μm diameter) favoring tumor cell uptake (via TEG functionalization) at sites of local and intracavitary injection and prohibiting entrance into the systemic circulation, a major problem in DOX-associated cardiotoxicity.47 Infusion of APMS-DOX or other chemotherapeutic drugs into the pleural or peritoneal cavity could potentially inhibit the growth of premalignant MM cells in pleural/peritoneal fluids. Additionally, the ability to functionalize drug-loaded APMS with targeting moieties such as antimesothelin antibodies (Cheng et al., in press), and the ability of APMS to release DNA20 and introduce functional plasmids into tumor cells,18 are exciting approaches that could potentially result in more effective treatment regimens for MMs and other cancers.


We thank Jennifer Díaz and Trisha Barrett for technical assistance with this article, and Drs. Sean McCarthy and Albert van der Vliet (Department of Pathology, UVM) for help with HPLC analyses. Dr. Pamela Vacek (Department of Medical Biostatistics, UVM) was valuable in performing statistical analyses. We also appreciate the technical assistance of the Vermont Cancer Center DNA Facility staff with PCR Array and QRT-PCR analyses. This work was supported by the Mesothelioma Applied Research Foundation (MARF to B.T.M.), a Small Business Technology Transfer (STTR) grant from the National Cancer Institute (R41 CA126155 to C.C.L.) and a training grant from the National Institute of Environmental Health Sciences (T32 ES007122 to J.M.H.).