Detection of PD-L1 expression in temozolomide-resistant glioblastoma by using PD-L1 antibodies conjugated with lipid-coated superparamagnetic iron oxide

doi:10.21203/rs.3.rs-269802/v1

Download PDF

Research

Detection of PD-L1 expression in temozolomide-resistant glioblastoma by using PD-L1 antibodies conjugated with lipid-coated superparamagnetic iron oxide

https://doi.org/10.21203/rs.3.rs-269802/v1

This work is licensed under a CC BY 4.0 License

Version 1

posted

You are reading this latest preprint version

Background: Targeted superparamagnetic iron oxide (SPIO) nanoparticles are a promising tool for molecular magnetic resonance imaging (MRI) diagnosis. Lipid-coated SPIO nanoparticles have a nonfouling property that can reduce nonspecific binding to off-target cells and prevent agglomeration, making them suitable contrast agents for molecular MRI diagnosis. PD-L1 is a poor prognostic factor for patients with glioblastoma. Most recurrent glioblastomas are temozolomide-resistant. Diagnostic probes targeting PD-L1 could help early diagnosis and be used to predict the responses of targeted PD-L1 immunotherapy in patients with primary and recurrent glioblastoma. In this study, we conjugated lipid-coated SPIO nanoparticles with PDL1 antibodies to identify PDL1 expression in glioblastoma or temozolomide-resistant glioblastoma by using MRI.

Results: The synthesized PD-L1 antibody–conjugated SPIO (PDL1-SPIO) nanoparticles were characterized using dynamic light scattering, zeta potential assays, transmission electron microscopy images, in vitro cell affinity assay, and in vivo MRI analysis. These data demonstrated that PDL1-SPIO has a specific binding capacity to PD-L1 of the mouse glioblastoma cell line (GL261). The presence and quantity of PDL1-SPIO in temozolomide-resistant glioblastoma cells and tumor tissue were confirmed through Prussian blue staining and in vivo T2* map MRI, respectively.

Conclusions: These findings revealed that PDL1-SPIO can specifically target temozolomide-resistant glioblastoma with PD-L1 expression and can be quantified with MRI analysis, thereby making it suitable for the diagnosis of PD-L1 expression in temozolomide-resistant glioblastoma in vivo.

Nanoscience

PDL1

superparamagnetic iron oxide (SPIO)

magnetic resonance imaging (MRI)

lipid-coated nanoparticle

glioblastoma

Glioblastoma multiforme (GBM) is the most malignant brain tumor with a poor prognosis. Following a standard treatment regimen (surgical resection, radiotherapy, and chemotherapy), the average survival time remains as low as 14 months [1]. Immune therapy targeting the negative regulatory molecules at the immune inhibitory checkpoint axis (i.e., programmed cell death protein 1 (PD-1) and its ligand PD-L1) has achieved breakthroughs in many cancers [2]. PD-L1 is widely expressed in almost all tumor cells and in many normal cells [3]. Its binding to PD-1 is the major inhibitor of the activation and proliferation of cytotoxic T lymphocytes (CTL) [4] in the tumor microenvironment [5]. PD-L1 is mainly upregulated in GBM and is negatively correlated with the survival time of patients [6]; thus, PD-L1 level in GBM is considered a major predictive marker of PD-1/PD-L1 antibody response [7]. Taken together, the precise detection of PD-L1 expression in GBM tissues can help predict PD-1/PD-L1 antibody response.

Because of intratumoral heterogeneity, tumor biopsy may not represent the pathological molecular profile of the whole tumor [8]. The current reference standard is immunohistochemistry, which allows the evaluation of PD-L1 expression on tumor cells or immune cells in a biopsy sample. However, immunohistochemistry is invasive and limited by the temporal and spatial heterogeneity of PD-L1 expression in GBM tissue, thus a small portion of tumor tissue may not accurately represent the PD-L1 expression level in whole tumor. In addition, dynamic changes in immune cell populations and PD-L1 expression levels occur during chemotherapy or immune checkpoint blockade in multiple cancer types [9]. The nature of GBM heterogeneity, limitation of tissue biopsy, and dynamic changes in the PD-L1 expression profile may explain the poor efficacy of PD-L1 antibody therapy in GBM [8, 10]. Thus, efficiently and precisely quantifying the PD-L1 level in the whole tumor may improve the treatment outcome.

Superparamagnetic iron oxide (SPIO) nanoparticles with biodegradable and low toxicity properties are suitable for biomedical imaging applications [11]. SPIO nanoparticles can be manipulated to conjugate with functional groups for cross-linking with antibodies, peptides, or small-molecule drugs for diagnostic imaging or delivery of therapeutic drugs [11]. They possess unique paramagnetic properties, which serve as contrast agents for multiple parametric magnetic resonance imaging (MRI) modalities, including T2-, T2*-, and susceptibility-weighted imaging (SWI) [12]. To identify patients who may benefit from PD-L1 targeted therapy, evaluating the PD-L1 expression level and its distribution is critical. SPIO nanoparticles have many benefits in biomedical imaging applications, including tunable size-dependent magnetic properties, easy modification, and lack of ionizing radiation [12, 13]. Nonfouling lipid-coated surfaces can efficiently reduce nonspecific binding to synthesized ex vivo surfaces [14-16]. In the present study, we manipulated lipid-coated SPIO nanoparticles for conjugation with PD-L1 antibodies (PDL1-SPIO) to detect PD-L1 expression using both in vitro and in vivo MRI. We demonstrated that PDL1-SPIO could specifically target PDL1-positive GBM cells and generate distinguishable molecular MRI images for noninvasive PD-L1 expression in temozolomide (TMZ)-resistant GBM in vivo.

Characterization of PDL1‑SPIO nanoparticles

To generate an MRI contrast agent with targeting specificity for PD-L1, the PD-L1 antibody was conjugated with the 10-nm lipid-coated SPIO nanoparticles (Figure 1). In brief, the amine-functionalized lipid-coated SPIO nanoparticles were reacted with sulfo-SMCC to obtain maleimide-functionalized SPIO nanoparticles. PD-L1 antibodies were treated with iminothiolane to form free thiol groups before spontaneous reaction with the maleimide-functionalized lipid-coated SPIO nanoparticles. This coupling strategy enabled direct robust covalent conjugation of the PD-L1 antibodies to SPIO nanoparticles. We first determined the hydrodiameter of PDL1-SPIO, which peaked at 25.8 ± 1.8 nm on the volume distribution graph, and unconjugated lipid-coated SPIO nanoparticles peaked at 21.9 ± 1.4 nm (Figure 2A). We also determined the diameter of SPIO nanoparticles and PDL1-SPIO by using transmission electron microscopy (TEM), which were observed to be 11.2 ± 0.6 nm and 12.7 ± 0.7 nm, respectively (Figure 2B); furthermore, their zeta potentials peaked at −7.8 ± 0.9 mV and −24.5 ± 3.3 mV, respectively (Figure 2C). These data suggest that the colloidal dispersion of PDL1-SPIO is more stable than that of SPIO nanoparticles and does not aggregate.

Evaluation of the targeting specificity of PDL1-SPIO to PDL1‑expressing GBM cells

To evaluate the PD-L1 targeting specificity of the PD-L1 antibodies after conjugation with SPIO nanoparticles, we used PDL1-SPIO to detect surface PD-L1 expression on a mouse GBM cell line (GL261). Flow cytometry analysis demonstrated that PD-L1 on GL261 cells can be detected using both PDL1-SPIO nanoparticles and PDL1 antibodies (Figure 3A). The mean fluorescence intensity (MFI) of PD-L1 expression was higher in the PDL1-SPIO staining group than in the PD-L1 antibody staining group (Figure 3A). Next, the targeting specificity of PDL1-SPIO was evaluated using an immunofluorescence (IF) labeling technique. The results revealed markedly strong green fluorescence signals specifically on the GL261 cells, indicating the affinity and binding magnitude of the PDL1-SPIO to GL261 cells (Figure 3B). By contrast, no signal was detected in the lipid-coated SPIO nanoparticle controls or isotype control antibodies. Taken together, the findings indicate that the PD-L1 antibodies retained affinity to PD-L1 after conjugation to SPIO nanoparticles.

Detection of PDL1-SPIO in GBM cells

To confirm the presence of targeted PDL1-SPIO in GBM cells, Prussian blue staining was performed to detect the presence of iron oxide nanoparticles in GL261 cells that were incubated with PDL1-SPIO or SPIO nanoparticles for 24 h. Consistent with the IF staining results, blue signals were detected in the GL261 cells after incubation with PDL1-SPIO, indicating the presence of iron oxide nanoparticles in these cells, but not in those incubated with SPIO nanoparticles (Figure 4A). In addition, to validate the cellular localization of PDL1-SPIO and confirm the nanoparticle uptake after treatment, the presence of nanoparticles in the cytoplasmic region of GL261 cells was detected using TEM. The TEM images of SPIO-treated GL261 cells revealed only sparse nanoparticles, whereas those of PDL1-SPIO-treated cells revealed electron-dense clustering of SPIO nanoparticles (>50 particles) (Figure 4B). These data indicate that GBM cells can uptake PDL1-SPIO.

PDL1-SPIO can be used to detect PD-L1 expression in TMZ-resistant GBM tissue

Most recurrent GBMs are resistant to TMZ [17]. TMZ treatment can upregulate PD-L1 expression in human GBM cell lines [18]. PD-L1 in the GBM microenvironment is contributed mainly by tumor-infiltrating monocytic myeloid derived suppressor cells (MO-MDSCs) and GBM tumor cells themselves [18, 19]. Thus, to detect PD-L1 expression in TMZ-resistant GBM tissue, we generated a TMZ-resistant GBM cell line (GL261/TMZ-R) from GL261 cells that were incubated for a long time in a culture medium containing TMZ. The clonogenic assay revealed that GL261/TMZ-R cells had fewer colonies than GL261 cells under normal culture conditions, suggesting that long-term TMZ treatment decreases the clonogenic activity of GBM cells; moreover, the number of GL261/TMZ-R cells colonies was not affected by TMZ treatment (Figure 4C). By contrast, GL261 cells could not form colonies after TMZ treatment (Figure 4C). These results indicate that GL261/TMZ-R cells can be resistant to the cytotoxicity effects of TMZ. Next, the level of PD-L1 expression was slightly increased in GL261/TMZ-R cells compared with GL261 cells (Figure 4D). To test PDL1-SPIO binding specificity to tumor tissue of GL261/TMZ-R-bearing mice, we treated tumor and peritumoral tissue with PDL1-SPIO or SPIO and performed Prussian blue assay. Both peritumoral and tumor regions in brain slices had positive blue signals after treatment with PDL1-SPIO, but no signals were observed after treatment with SPIO control (Figure 4E). We next examined the level of PD-L1 on MO-MDSCs (CD45^hiCD11b^hi Ly6C⁺Ly6G⁻) and nonhematopoietic cells (CD45⁻ cells) in tumor tissues by using flow cytometry and found that both subsets expressed PD-L1; however, the expression was low in MO-MDSCs and nonhematopoietic cells in tumor-bearing hemispheres (Figure 4F). Taken together, these findings indicated that PDL1-SPIO can be used to detect PD-L1 expression in intratumoral and peritumoral regions from TMZ-resistant GBM-bearing mice.

MR fitting curve and T2 relaxation time of PDL1‑SPIO phantoms

To evaluate the feasibility of using PDL1-SPIO for PDL1-positive cell detection through MRI, PDL1-SPIO phantom scans were performed to investigate the MRI signal intensity and response to T2 relaxation time at various concentrations. The signal intensity and T2 relaxation time of PDL1-SPIO suspended in agarose gel with known Fe concentrations (0–15.7 μg/mL) were measured using 7-T MRI. The T2-fitting curve of PDL1-SPIO exhibited a correlation between the signal intensity and nanoparticle concentration at TE values ranging from 8.8 to 352 ms (Figure 5A). Moreover, the PDL1-SPIO indicated a decrease in MR T2 relaxation time with an increase in Fe concentration (Figure 5B). The T2 relaxation time of PDL1-SPIO nanoparticles was 92.4 ms at 0.98 μg/mL, and this decreased to 22.9 ms when the Fe concentration was increased to 15.7 μg/mL. Figure 5C presents the MRI signal intensities of different concentrations of PDL1-SPIO phantoms, indicating that the PDL1-SPIO concentration was positively correlated with the dark signals in T2-weighted in vitro MR images. The positive correlations of the PDL1-SPIO concentration with the T2 relaxation time and signal intensity suggest that the PDL1-SPIO can be quantitatively detected using in vitro MRI.

Identification of PDL1‑positive GBM cells through in vitro MRI

To test whether PDL1-SPIO can be used for PD-L1-positive GBM cell detection, GL261 cells were incubated with PDL1-SPIO or SPIO nanoparticles, and SPIO signals were detected in T2-weighted MRI images. The T2 relaxation time was lower in PDL1-SPIO-treated GL261 cells than in the SPIO control group (Fig. 5D). These findings indicate that PDL1-SPIO can be used for identifying PDL1-positive GBM cells on in vitro MRI images.

Detection of PDL1 in GL261/TMZ-R GBM through in vivo MRI

To detect PD-L1 expression in TMZ-resistant GBM in vivo, PDL1-SPIO was administered intravenously into GL261/TMZ-R GBM-bearing mice. The heads of tumor-bearing mice were scanned using T2-, T2*-, and susceptibility-weighted MRI at indicated postinjection times (0, 1, 4, and 24 h); the dark signals in the tumor were most prominent in SWI at 4 h (Figure 6A). To confirm the presence of targeted PDL1-SPIO in GBM tissue, Prussian blue staining was performed to detect the presence of Fe in tumors from GBM-bearing mice after PDL1-SPIO or SPIO nanoparticle injection. Prussian blue staining revealed blue signals in the tumor region in PDL1-SPIO-treated mice but not in SPIO-treated mice, indicating that iron oxide nanoparticles were present in tumor tissue after PDL1-SPIO injection (Figure 6B). To quantify and further test the specificity of the binding of PDL1-SPIO in brain tumors of TMZ-resistant tumor-bearing mice, SPIO or PDL1-SPIO nanoparticles were injected into tumor-bearing mice, and PD-L1 expression was quantified using a T2* map scan before and after PDL1-SPIO or SPIO injection. Some hypointense areas in the tumor region of GBM-bearing mice were identified at 4 h after PDL1-SPIO injection (Figure 6C). By contrast, we did not detect notable hypointense areas in GBM-bearing mice that received SPIO nanoparticles (Figure 6C). In addition, DR2* analysis quantified PDL1-SPIO expression in the tumor region of tumor-bearing mice through T2* mapping (Figure 6D), whereas no signaling was observed in the GBM-bearing mice that received SPIO. Taken together, the findings imply that PDL1-SPIO can target PD-L1 expression in tumors and can be detected on MRI images.

In this study, we demonstrated a method of conjugating lipid-coated SPIO nanoparticles with PD-L1 antibodies (PDL1-SPIO) for MRI analysis. PDL1-SPIO nanoparticles specifically target PD-L1-expressing GL261 cells and generate distinguishable signals in TMZ-resistant GBM-bearing mice in MRI analysis. In addition, lipid-coated SPIO nanoparticles exhibit a potent nonfouling property that efficiently suppresses the nonspecific sorption of SPIO nanoparticles by GBM cells. This can reduce system noise in MRI, thus preventing false-positive signals. Conjugating PD-L1 antibodies to lipid-coated SPIO nanoparticles enables specific targeting toward PD-L1, and the conjugation process did not alter the antibody affinity, as confirmed using flow cytometry analysis, IF staining, Prussian blue staining, and TEM. Furthermore, PDL1-SPIO nanoparticles were successfully used for in vitro and in vivo MRI analysis. According to multiple-parameter in vivo MR images, including SWI, T2*, and T2* map images, PDL1-SPIO nanoparticles can noninvasively reveal PD-L1 expression levels in brain tumors. According to our review of relevant literature, this study is the first to describe the conjugation of PD-L1 antibodies to lipid-coated SPIO nanoparticles.

The development of novel noninvasive approaches for early detection of PD-L1 is beneficial for PD-L1 diagnosis. Several studies have demonstrated that PD-L1 expression can be detected by radiolabeled PD-L1 antibodies through positron emission tomography imaging analysis [20]. However, these studies have also reported that these radiolabeled PD-L1 antibody–mediated imaging has some limitations. Imaging with radiolabeled antibodies is associated with considerable background signals and requires imaging several days after injection of the mAbs to obtain the best contrast image [20]. Moreover, because radiolabeled antibodies have radioactivity, the radiation burden would be high for highly perfused and radiosensitive organs [20]. The PDL1-SPIO nanoparticles used to detect PDL1 expression in vivo are nonradioactive, thus avoiding this radiation burden.

TMZ treatment causes differential effects on PD-L1 expression levels in primary and recurrent GBM. TMZ treatment leads to a down-regulation of PD-L1 in IFN-g stimulated human primary GBM cells [21]. By contrast, another study demonstrated that TMZ treatment can upregulate PD-L1 expression in U87 and U251 human GBM cell lines [17]. These data suggest that TMZ treatment differentially affects PD-L1 expression levels in different types of human GBM cells under different culture conditions. High expression of PD-L1 in GBM tumor tissues is associated with poor survival of patients [22]. Recent clinical trials have revealed a lower response of targeted PD-L1 immunotherapy in recurrent GBM [23]. A potential explanation for this lower response may be the reduced expression of PD-L1 in recurrent GBM [24]. Limitation of tumor biopsy, heterogeneity of PD-L1 expression in intratumoral GBM tumor tissue, and dynamic changes of PD-L1 expression profile in primary and recurrent GBM are possible reasons for poor responses to targeted PD-L1 immunotherapy. Thus, precisely diagnosing PD-L1 expression levels in whole tumors would be beneficial for patients with GBM who request targeted PD-L1 therapy. In this study, PDL1-SPIO nanoparticles targeted PD-L1 in both the edge and core regions of the tumor, indicating that they were perfused into the TMZ-resistant GBM tissue. Furthermore, we also quantified the PD-L1 expression level in tumors by T2* map MR imaging. Although some hypointensity signals have been observed in the liver and spleen after intravenous administration of PDL1-SPIO nanoparticles, possibly due to them being trapped by reticuloendothelial cells in the liver and spleen (data not shown) [25], they can still target PD-L1 and generate prominent signals in brain tumors. Thus, PDL1-SPIO nanoparticles could serve as an MR diagnostic contrast agent to determine the PD-L1 expression level in GBM.

Lipid-coated SPIO nanoparticles are a suitable diagnosis contrast reagent for application in MRI analysis. A previous study concluded that lipid-coated SPIO nanoparticles conjugated with EGFR antibodies (EGFR-SPIO) can specifically detect EGFR expression in a human GBM cell line [16]. However, animal experiments were not performed to test whether EGFR-SPIO revealed EGFR expression in GBM tumors when conducting in vivo MRI analysis. In this study, we demonstrated that intravenous delivery of PDL1-SPIO nanoparticles into tumor-bearing mice can specifically target PD-L1 in the GBM. Nanoparticles smaller than 50 nm can passively permeate the blood brain barrier (BBB) and accumulate in the brain [26]. Our study demonstrated PDL1-SPIO nanoparticles are sufficiently small (hydrodiameter of 25 nm and diameter of 12.7 nm) to passively permeate the BBB and target PD-L1 molecules in GBM. Chen et al. used a high molar ratio of iminothiolane to antibodies (2000:1) to thiolate the EGFR antibodies, and they conjugated these thiolated EGFR antibodies to lipid-coated SPIO nanoparticles. EGFR-SPIO nanoparticles exhibited lower binding efficiency compared with the control EGFR antibodies [16]. Because Traut’s reagent with iminothiolane can randomly modify antibody molecules with sulfhydryl groups, the antibody binding specificity may be lost when the amino groups are excessively thiolated in antigen-binding regions [27]. Thus, we reduced the molar ratio of iminothiolane to PD-L1 antibodies from 2000:1 to 20:1. The present data demonstrated that thiolated PD-L1 antibodies have normal binding efficacy compared with control PD-L1 antibodies (data not shown), and they can be conjugated to SPIO nanoparticles through the linker SMCC.

We developed a PD-L1 MRI diagnosis probe that can quantify the PD-L1 expression in GBM tumors in vitro and in vivo. Such MRI measurements can facilitate the identification of PD-L1 expression level, which may improve prognosis and clinical outcomes for patients with GBM.

Construction of PDL1-SPIO nanoparticles

The construction of PDL1-SPIO nanoparticles followed the procedures described in previous publications with some modifications [16]. Briefly, 0.25 mL of amine-functionalized lipid-coated SPIO nanoparticles (OceanNanoTech Inc., 1 mg/mL (Fe) = 0.86 nmol/mL nanoparticles, 10 nm in size) were reacted with 37.5 μL/mL sulfo-SMCC (10 mg/mL, 858 nmol) at room temperature for 1 h to obtain maleimide-functionalized SPIO nanoparticles. The maleimide-functionalized SPIO nanoparticles were washed with 10 mL of phosphate buffered saline (PBS) to remove the excess free sulfo-SMCC with an LS column (Miltenyi Biotech, Germany) and then eluted in 800 mL of PBS. Second, 0.43 nmol of PD-L1 antibodies (10F.9G2, BioXCell) was treated with iminothiolane (1.2 μg, 8.7 nmol) (Thermo, USA) dissolved in 200 μL of Traut’s reagent (50 mM NaHCO₃, 150 mM NaCl, and 10 mM EDTA, pH 8.6) and reacted at room temperature for 60 min. After thiolation, 200 mL of 10 mM tris(2-carboxyethyl)phosphine (TCEP, Sigma, USA) was added at room temperature for 30 min. The solution was replaced with a 5 mM EDTA solution in PBS via Vivaspin (10-kDa MWCO polyethersulfone, Sartorius). Finally, the thiolated antibodies and maleimide-functionalized SPIO nanoparticles were mixed and reacted at 4 °C overnight. The unused maleimide-functionalized groups were then blocked by excess cysteine for 15 min at room temperature. The PD-L1 antibody–conjugated SPIO nanoparticles were separated using an LS column and washed with PBS at a volume > 25 times greater than the column bed volume to remove unconjugated antibodies; then, the nanoparticles were eluted in 600 μL of PBS. The number of immobilized PDL1 antibodies per SPIO nanoparticle was estimated to be 2 based on the molarities of the components in the reaction.

Dynamic light scattering and zeta potential measurement

The hydrodynamic size and zeta potential of the SPIO and PDL1-SPIO nanoparticles were determined using Zetasizer Nano ZS (Malvern Instruments, UK) according to the manufacturer’s instructions. The nanoparticles were diluted in a 10 mM NaCl solution.

Generation of the GL261/TMZ-R cell line

The mouse GBM cell line (GL261) was obtained from Deutsche Sammlung von Mikroorganismen und Zellkulturen. Cells were cultured in a DMEM culture medium supplemented with 10% fetal bovine serum (FBS). We established a TMZ-resistant subline (GL261/TMZ-R) by continuous exposure of GL261 cells to TMZ in our lab. These cells were treated with 50 μM TMZ for 1 day before being divided into 96 wells. Cells were incubated with 150 mM TMZ for 21 days, and surviving cells were cultured in the culture medium containing 150 mM TMZ for an additional 39 days.

Colony formation assay

We placed 2000 GL261 cells or GL261/TMZ-R cells into six-well plates and incubated them for 5 days, followed by treatment with 150 mM TMZ (Cayman, USA) for 12 days. The cells were fixed with 4% paraformaldehyde, stained with 0.5% crystal violet, and the colony number was counted.

Brain tumor animal model

For the GL261/TMZ-R GBM animal model, C57BL/6JNarl mice were purchased from the National Laboratory Animal Center. The tumor cell inoculation assay followed that described in a previous study [39]. Briefly, 20000 GL261/TMZ-R cells in 4 μL of PBS solution were slowly injected into the right-brain region at a rate of 0.5 mL/min. The injection site was 2.5 mm lateral, 0.1 mm posterior, and 3.1 mm ventral from the central bregma. The injection hole was sealed with bone wax to prevent leakage of cells, and the skin was closed with sterilized sutures.

Isolation of tumor-infiltrating leukocytes

To isolate brain-infiltrating leukocytes, we anesthetized mice and transcardially perfused them with 5 mL of cold PBS. Tumor-infiltrating leukocytes were dissociated into single-cell suspensions using a Tumor Dissociation kit and Multi Tissue Dissociation Kit 1 (Miltenyi Biotec, Germany), respectively, in combination with gentle MACS™ Octo Dissociator with Heaters (Miltenyi Biotec, Germany).

Flow cytometry

To detect PD-L1 expression on GL161 cells, cells were detached using Accutase® (Thermo Fisher Scientific, USA) and then stained with PDL1-SPIO nanoparticles, SPIO nanoparticles, PDL1 antibodies, or isotype control antibodies, followed by PE-conjugated antirat IgG2b antibodies (MRG2b-85, BioLegend, USA). To detect PD-L1 expression on brain-infiltrating leukocytes, cells were stained with specific antibodies following standard protocols and analyzed on a CytoFLEX™ flow cytometer (Beckman Coulter, USA). The following antibodies conjugated with fluorochrome were used: Ly6G-PB (RB6-8C5), Ly6C-FITC (HK1.4), CD45-APC-Cy7, (30-F11), CD11b-PB (M1/70), and PD-L1-PE (10F.9G2) (BioLegend. USA). Data were analyzed using FlowJo software.

Immunofluorescence assay

To detect PD-L1 expression in GL161 cells, 10⁵ GL261 cells were seeded on a 4-well chamber slide for 24 h. Cells were treated with SPIO nanoparticles, PDL1-SPIO nanoparticles, PDL1 antibodies, or isotype control antibodies for 24 h at 37 °C and then fixed with 4% paraformaldehyde for 30 min. Slides were then stained with FITC antimouse Rat IgG2b antibody and mounted with a mounting medium containing DAPI (Vectashield, USA). IF images of stained cells were acquired through microscopy (Olympus/Bx43, Canada).

Prussian blue staining assay for GBM cells and TEM for SPIO nanoparticles

To identify iron in GBM cells by using Prussian blue staining, 10⁵ GL261 cells were seeded on a 4-well chamber slide (SPL Life Sciences Co., USA) with DMEM culture medium supplemented with 10% FBS for 24 h. Next, cells were treated with SPIO or PDL1-SPIO nanoparticles and incubated at 37 °C for 24 h. The presence of iron was detected using an Abcam iron stain kit (ab150674, Abcam). In brief, hydrochloric acid solution (2%) and potassium ferrocyanide solution were mixed at equal volumes (1:1) and incubated with cells or tumor tissue slides for 20 min; then, the cells or slides were rinsed with distilled water and stained with nuclear fast red. Images of stained cells were acquired using a microscope (Olympus BX43). TEM images were acquired using an HT-7700 microscope (Hitachi, Japan), and the analysis was performed according to the manufacturer’s instructions.

Detection of iron in GBM tissues treated with PDL1-SPIO nanoparticles by using Prussian blue staining assay

To test the binding of PDL1-SPIO nanoparticles for tumor tissue, PDL1-SPIO nanoparticles were used to stain tumor slices, after which Prussian blue staining assay was performed. Briefly, GL261/TMZ-R tumor-bearing mice were sacrificed and transcardially perfused with 5 mL PBS and 5 mL formalin. Whole brains were fixed with formalin for approximately 3 days. Formalin-fixed 2-mm coronal slices were embedded in paraffin and cut into 5-μm-thick sections. The sections were then deparaffinized, rehydrated through a graded series of ethanol solutions, subjected to an antigen retrieval process, and then stained using PDL1-SPIO or SPIO nanoparticles at 4 °C for 24 h. The presence of iron on these sections was detected by an Abcam iron stain kit (Abcam). These sections were observed through microscopy (Olympus/Bx43). In some experiments, after tumor-bearing mice were injected with PDL1-SPIO nanoparticles or PDL1 antibodies for 4 h, the brains were directly subjected to Prussian blue assay to detect the presence of iron caused by PDL1-SPIO nanoparticles in the tumor tissue.

Cellular MRI measurements

MRI images were obtained using a 7T Bruker PharmaScan MRI scanner using a volume coil with an inner diameter of 72 mm (Bruker BioSpin, MA, USA). T2-weighted images were acquired using spin-echo sequences with an echo time (TE) of 8.8 ms, a repetition time (TR) of 3500 ms, 50 echoes, a field of view of 50 × 50 mm, a resolution of 256 × 256, and a slice thickness of 1 mm. The MRI samples were PDL1-SPIO nanoparticle phantoms and GL261 cells treated with various concentrations of PDL1-SPIO nanoparticles and then suspended in 1% agarose gel.

Animal MRI measurements

In vivo MRI images of mouse brains were obtained using a 7T Bruker PharmaScan MRI scanner using a volume coil with an inner diameter of 72 mm (Bruker BioSpin, MA, USA). MRI was performed in mice anesthetized by 2% isoflurane in the coronal plane. The MRI protocol included a T2-weighted image (TR, 2500 ms; TE, 33 ms; flip angle, 45°; FOV, 16 × 16 mm; matrix, 256 × 256; 2D; slice thickness, 0.75 mm; number of excitations, 8; resolution, 0.0625 × 0.0625 × 0.75 mm), a T2*-weighted image (TR, 1000 ms; TE, 12 ms; flip angle; FOV, 16 × 16 mm; matrix, 256 × 256; 2D; slice thickness, 0.75 mm; number of excitations, 2; resolution, 0.0625 × 0.0625 × 0.75 mm), an SWI (TR, 39 ms; TE, 51.8 ms; flip angle, 15°; FOV, 50 × 50 mm; matrix, 128 × 128; 3D; slice thickness, 0.5 mm; number of excitations, 3; resolution 0.39 × 0.39 × 0.5 mm), and a T2* mapping (16-echo gradient echo sequence; TR,1150 ms; minimum TE, 3.3 ms; ∆TE, 3 ms; flip angle, 80°; FOV, 16 × 16 mm; matrix, 256 × 256; 2D; slice thickness, 0.5 mm; number of excitations, 2; resolution, 0.0625 × 0.0625 × 0.5 mm). The initial T2* mapping scan was performed before injection of PDL1-SPIO or SPIO nanoparticles. The second T2* mapping session was started at 4 h after injection. The T2* value of each voxel was calculated through exponential fitting performed using in-house by using MATLAB (version R2020a, MathWorks, Sherborn, MA, USA). The T2* map was first converted to an R2* relaxivity map by taking its reciprocal. The increase in R2* value (∆R2*) after PDL1-SPIO or SPIO injection was calculated using the following formulas: ∆R2* = R2*_PDL1-SPIO– R2*_pre or ∆R2* = R2*_SPIO– R2*_pre, where R2*_pre refers to the R2* value measured in precontrast MR images, and R2*_PDL1-SPIO or R2*_SPIO are the R2* values measure in MR images following PDL1-SPIO or SPIO administration. The ∆R2* value, which indicates the change in relaxivity due to the local aggregation of PDL1-SPIO or SPIO nanoparticles, has a linear relationship with the local PDL1-SPIO or SPIO nanoparticle concentration. The coefficient of the linear transformation function between ∆R2* and the local PDL1-SPIO or SPIO nanoparticle concentration was estimated through linear regression analysis of agarose phantoms containing various concentrations of PDL1-SPIO or SPIO nanoparticles. The ∆R2* map was then converted into a local PDL1-SPIO or SPIO nanoparticle concentration map using the estimated linear transformation function. A disk of nickel-coated neodymium iron boron (Nd2Fe14B) with a diameter of 8 mm, a height of 5 mm, and a 0.43-T N42 grade magnet was placed on the tumor site of tumor-bearing mice. After placing the magnet, 60 mg of PDL1-SPIO or SPIO nanoparticles was administered via tail vein injection. The magnet was maintained on the tumor site for 1 h and then removed. MRI scans were performed at the indicated time. A radiologist evaluated the MRI images to identify dark signals due to PDL1-SPIO nanoparticles on T2-weighted images, T2*-weighted images, SWI, and T2* mapping.

Ethics approval and consent to participate: This study was approved by the IACUC at Taipei Medical University.

Consent for publication: Not applicable.

Availability of data and materials: The datasets used and/or analyzed during the current study are available from the corresponding author on reasonable request.

Competing interest: The authors have no potential conflicts of interest to disclose.

Funding: This work was supported by MOST 108-2314-B-038-023 and MOST 109-2314-B-038-128

Authors’ contributions: GAL and CYC were responsible for the study concept and design, modification of the study design, and review and interpretation of the data. GAL drafted the manuscript. GAL, WLL, and YWC conducted the experiments. YTL and SWH made modifications to the study design. GAL, WLL, DPK, YTL, JBKH, YCC, and CYC contributed to data collection, analysis, interpretation, and manuscript revision.

Acknowledgment: We gratefully acknowledge the services and facilities at the Core Facility Center and Laboratory Animal Center of Taipei Medical University. This manuscript was edited by Wallace Academic Editing.

Stupp R, Hegi ME, Mason WP, van den Bent MJ, Taphoorn MJ, Janzer RC, Ludwin SK, Allgeier A, Fisher B, Belanger K, et al: Effects of radiotherapy with concomitant and adjuvant temozolomide versus radiotherapy alone on survival in glioblastoma in a randomised phase III study: 5-year analysis of the EORTC-NCIC trial.Lancet Oncol 2009, 10:459-466.
Nduom EK, Wei J, Yaghi NK, Huang N, Kong LY, Gabrusiewicz K, Ling X, Zhou S, Ivan C, Chen JQ, et al: PD-L1 expression and prognostic impact in glioblastoma.Neuro Oncol 2016, 18:195-205.
Dong Y, Sun Q, Zhang X: PD-1 and its ligands are important immune checkpoints in cancer.Oncotarget 2017, 8:2171-2186.
Dong H, Strome SE, Salomao DR, Tamura H, Hirano F, Flies DB, Roche PC, Lu J, Zhu G, Tamada K, et al: Tumor-associated B7-H1 promotes T-cell apoptosis: a potential mechanism of immune evasion.Nat Med 2002, 8:793-800.
Alsaab HO, Sau S, Alzhrani R, Tatiparti K, Bhise K, Kashaw SK, Iyer AK: PD-1 and PD-L1 Checkpoint Signaling Inhibition for Cancer Immunotherapy: Mechanism, Combinations, and Clinical Outcome.Front Pharmacol 2017, 8:561.
Chen RQ, Liu F, Qiu XY, Chen XQ: The Prognostic and Therapeutic Value of PD-L1 in Glioma.Front Pharmacol 2018, 9:1503.
Sanmamed MF, Chen L: A Paradigm Shift in Cancer Immunotherapy: From Enhancement to Normalization.Cell 2018, 175:313-326.
Bergmann N, Delbridge C, Gempt J, Feuchtinger A, Walch A, Schirmer L, Bunk W, Aschenbrenner T, Liesche-Starnecker F, Schlegel J: The Intratumoral Heterogeneity Reflects the Intertumoral Subtypes of Glioblastoma Multiforme: A Regional Immunohistochemistry Analysis.Front Oncol 2020, 10:494.
Rochigneux P, Garcia AJ, Chanez B, Madroszyk A, Olive D, Garon EB: Medical Treatment of Lung Cancer: Can Immune Cells Predict the Response? A Systematic Review.Front Immunol 2020, 11:1036.
Wang X, Guo G, Guan H, Yu Y, Lu J, Yu J: Challenges and potential of PD-1/PD-L1 checkpoint blockade immunotherapy for glioblastoma.J Exp Clin Cancer Res 2019, 38:87.
Peng XH, Qian X, Mao H, Wang AY, Chen ZG, Nie S, Shin DM: Targeted magnetic iron oxide nanoparticles for tumor imaging and therapy.Int J Nanomedicine 2008, 3:311-321.
Nduom EK, Bouras A, Kaluzova M, Hadjipanayis CG: Nanotechnology applications for glioblastoma.Neurosurg Clin N Am 2012, 23:439-449.
Maier-Hauff K, Ulrich F, Nestler D, Niehoff H, Wust P, Thiesen B, Orawa H, Budach V, Jordan A: Efficacy and safety of intratumoral thermotherapy using magnetic iron-oxide nanoparticles combined with external beam radiotherapy on patients with recurrent glioblastoma multiforme.J Neurooncol 2011, 103:317-324.
Wu JC, Tseng PY, Tsai WS, Liao MY, Lu SH, Frank CW, Chen JS, Wu HC, Chang YC: Antibody conjugated supported lipid bilayer for capturing and purification of viable tumor cells in blood for subsequent cell culture.Biomaterials 2013, 34:5191-5199.
Lai JM, Shao HJ, Wu JC, Lu SH, Chang YC: Efficient elusion of viable adhesive cells from a microfluidic system by air foam.Biomicrofluidics 2014, 8:052001.
Chen HL, Hsu FT, Kao YJ, Liu HS, Huang WZ, Lu CF, Tsai PH, Ali AAA, Lee GA, Chen RJ, Chen CY: Identification of epidermal growth factor receptor-positive glioblastoma using lipid-encapsulated targeted superparamagnetic iron oxide nanoparticles in vitro.J Nanobiotechnology 2017, 15:86.
Wang S, Yao F, Lu X, Li Q, Su Z, Lee JH, Wang C, Du L: Temozolomide promotes immune escape of GBM cells via upregulating PD-L1.Am J Cancer Res 2019, 9:1161-1171.
Mi Y, Guo N, Luan J, Cheng J, Hu Z, Jiang P, Jin W, Gao X: The Emerging Role of Myeloid-Derived Suppressor Cells in the Glioma Immune Suppressive Microenvironment.Front Immunol 2020, 11:737.
Antonios JP, Soto H, Everson RG, Orpilla J, Moughon D, Shin N, Sedighim S, Yong WH, Li G, Cloughesy TF, et al: PD-1 blockade enhances the vaccination-induced immune response in glioma.JCI Insight 2016, 1.
Broos K, Lecocq Q, Raes G, Devoogdt N, Keyaerts M, Breckpot K: Noninvasive imaging of the PD-1:PD-L1 immune checkpoint: Embracing nuclear medicine for the benefit of personalized immunotherapy.Theranostics 2018, 8:3559-3570.
Heynckes S, Daka K, Franco P, Gaebelein A, Frenking JH, Doria-Medina R, Mader I, Delev D, Schnell O, Heiland DH: Crosslink between Temozolomide and PD-L1 immune-checkpoint inhibition in glioblastoma multiforme.BMC Cancer 2019, 19:117.
Hao C, Chen G, Zhao H, Li Y, Chen J, Zhang H, Li S, Zhao Y, Chen F, Li W, Jiang WG: PD-L1 Expression in Glioblastoma, the Clinical and Prognostic Significance: A Systematic Literature Review and Meta-Analysis.Front Oncol 2020, 10:1015.
Caccese M, Indraccolo S, Zagonel V, Lombardi G: PD-1/PD-L1 immune-checkpoint inhibitors in glioblastoma: A concise review.Crit Rev Oncol Hematol 2019, 135:128-134.
Heynckes S, Gaebelein A, Haaker G, Grauvogel J, Franco P, Mader I, Carro MS, Prinz M, Delev D, Schnell O, Heiland DH: Expression differences of programmed death ligand 1 in de-novo and recurrent glioblastoma multiforme.Oncotarget 2017, 8:74170-74177.
Weissleder R, Elizondo G, Wittenberg J, Rabito CA, Bengele HH, Josephson L: Ultrasmall superparamagnetic iron oxide: characterization of a new class of contrast agents for MR imaging.Radiology 1990, 175:489-493.
Sonavane G, Tomoda K, Makino K: Biodistribution of colloidal gold nanoparticles after intravenous administration: effect of particle size.Colloids Surf B Biointerfaces 2008, 66:274-280.
Ji T, Muenker MC, Papineni RV, Harder JW, Vizard DL, McLaughlin WE: Increased sensitivity in antigen detection with fluorescent latex nanosphere-IgG antibody conjugates.Bioconjug Chem 2010, 21:427-435.

Download PDF

Version 1

posted

You are reading this latest preprint version

Detection of PD-L1 expression in temozolomide-resistant glioblastoma by using PD-L1 antibodies conjugated with lipid-coated superparamagnetic iron oxide

Status:

Version 1

Abstract

Figures

Background

Results

Discussion

Conclusions

Methods

Declarations

References

Status:

Version 1