Journal of Immunology Research

Journal of Immunology Research / 2014 / Article
Special Issue

Cancer Immunology and Cancer Immunodiagnosis

View this Special Issue

Research Article | Open Access

Volume 2014 |Article ID 131494 |

Zhe Bao Wu, Chao Qiu, An Li Zhang, Lin Cai, Shao Jian Lin, Yu Yao, Qi Sheng Tang, Ming Xu, Wei Hua, Yi Wei Chu, Ying Mao, Jian Hong Zhu, Jianqing Xu, Liang Fu Zhou, "Glioma-Associated Antigen HEATR1 Induces Functional Cytotoxic T Lymphocytes in Patients with Glioma", Journal of Immunology Research, vol. 2014, Article ID 131494, 12 pages, 2014.

Glioma-Associated Antigen HEATR1 Induces Functional Cytotoxic T Lymphocytes in Patients with Glioma

Academic Editor: Bin Zhang
Received08 Mar 2014
Revised17 May 2014
Accepted16 Jun 2014
Published09 Jul 2014


A2B5+ glioblastoma (GBM) cells have glioma stem-like cell (GSC) properties that are crucial to chemotherapy resistance and GBM relapse. T-cell-based antigens derived from A2B5+ GBM cells provide important information for immunotherapy. Here, we show that HEAT repeat containing 1 (HEATR1) expression in GBM tissues was significantly higher than that in control brain tissues. Furthermore, HEATR1 expression in A2B5+ U87 cells was higher than that in A2B5−U87 cells (). Six peptides of HEATR1 presented by HLA-A02 were selected for testing of their ability to induce T-cell responses in patients with GBM. When peripheral blood mononuclear cells from healthy donors () and patients with glioma () were stimulated with the peptide mixture, eight patients with malignant gliomas had positive reactivity with a significantly increased number of responding T-cells. The peptides HEATR1682–690, HEATR11126–1134, and HEATR1757–765 had high affinity for binding to HLA-A02:01 and a strong capacity to induce CTL response. CTLs against HEATR1 peptides were capable of recognizing and lysing GBM cells and GSCs. These data are the first to demonstrate that HEATR1 could induce specific CTL responses targeting both GBM cells and GSCs, implicating that HEATR1 peptide-based immunotherapy could be a novel promising strategy for treating patients with GBM.

1. Introduction

Human glioblastoma (GBM) accounts for approximately 60–70% of malignant gliomas, the most common and deadly brain tumors [1]. Despite improvements in standard therapies including surgery, radiation, and chemotherapy, the poor prognosis of patients with GBM has not been obviously improved. Immunotherapy represents a promising treatment designed to reshape the immune system to specifically eradicate malignant cells. The effort of T-cell-mediated immunotherapy to selectively kill remnant glioma cells that could not be completely removed using microsurgery has been highlighted [24].

Glioma stem-like cells (GSCs) may be capable of initiating tumor growth [5, 6] and are likely to be responsible for the malignant behavior of tumors because of their acquired resistance to chemotherapy, radiotherapy, and immunotherapy induced by glioma-associated antigens, which results in the ineffectiveness of existing conventional therapies [79]. Thus, GSCs could be a novel target for cancer therapy, including immunotherapy. Our recent study findings indicated that glioma stem-like cell-associated antigens (SAAs) from CD133+ GSCs bear highly immunogenic antigens and induce significant responses from cytotoxic T lymphocytes (CTLs) [10]. Several other studies have tried immunotherapy targeting GSCs [1116].

A2B5 is considered a marker for immature glial-committed progenitors that are permanently generated in the subventricular zone. Glial progenitor cells are defined as cells that give rise to glial cell types such as astrocytes and oligodendrocytes. In GBM tissues, A2B5+ cells include A2B5+/CD133+ and A2B5+/CD133− cells. Furthermore, A2B5+ cells from human GBM have cancer stem-like cell properties that are crucial for the initiation and maintenance of GBM [17, 18]. Thus, A2B5+ GBM cells could be an ideal target for GBM immunotherapy. Our recent study found that vaccination with A2B5+ GL261 cell lysate-pulsed dendritic cells had a preventive effect for mouse glioma [19]. However, T-cell epitopes derived from A2B5+ GBM progenitor cells for immunotherapy have not been reported.

To identify novel genes selectively overexpressed in A2B5+ GBM as the target for T-cell mediated immunotherapy, we sequenced the mRNA profile of A2B5+ GBM cells from U87 cell lines using fluorescence-activated cell sorting (FACS) by Solexa sequencing (data not shown) and identified that the HEAT repeat containing 1 (HEATR1) gene (gene ID: 55127) was overexpressed in A2B5+ GBM cells. Recently, Bleakley et al. reported that HEATR1 was highly expressed in testis and ovary than in other tissues including liver, colon, small intestine, lung, brain, and heart [20]. Identification of epitope derived from HEATR1 is likely to provide alternative candidates for the design of antitumor vaccine with high efficacy in the future.

In the present study, we confirm the selective HEATR1 overexpression in A2B5+ GBM cells and in the vast majority of GBM. In addition, we identify several HEATR1-derived T-cell epitopes in tumor carrier patients. Our results emphasize the suitability of this protein for T-cell-based immunotherapy in patients with GBM.

2. Materials and Methods

2.1. Ethics Statement

The study protocol was approved by the Local Independent Ethics Committee at Huashan Hospital, Fudan University. Some samples in this study were used in our previous reports [21, 22]. Written informed consent was obtained from each donor of the samples used in our research.

2.2. Cell Lines

Human GBM cell lines U87, A172, and SHG66 were used in this study. SHG66 came from a 47-year-old man with a right parietal glioblastoma (World Health Organization grade IV) [10]. U87 and A172 cells were purchased from the cell bank of the Shanghai Branch of Chinese Academy of Sciences. A172 cells did not express HLA-A02:01 [23, 24], while the other two GBM cell lines expressed HLA-A02:01 according to flow cytometry [25, 26]. The HLA-A02:01, expressing human tumor cells T2 (deficient in TAP1 and TAP2 transporters), and BB7.2 hybridoma, producing anti-HLA-A02 monoclonal antibody (mAb), were purchased from American Type Culture Collection (USA). All cell lines were cultured in Dulbecco’s modified eagle medium (DMEM; Gibco, Grand Island, NY, USA) supplemented with 10% fetal bovine serum (FBS; Hyclone, Logan, UT, USA) and 100 U/mL penicillin/streptomycin (Gibco) and maintained in a humidified atmosphere with 5% CO2 at 37°C.

The GSC lines (U87, A172, and SHG66) were established and characterized as described previously [10, 13]. Short-term tumor spheres of the GBM cell lines were cultured in serum-free medium (SFM) consisting of DMEM/F12 (Invitrogen) supplemented with 20 ng/mL recombinant human basic fibroblast growth factor (bFGF; Chemicon), 20 ng/mL recombinant human epidermal growth factor (EGF; Chemicon), and B27 (Invitrogen). The GSC tumor spheres exhibited stem cell-like characteristics [10, 15].

2.3. Patients

A total of 22 frozen GBM tumor tissues were obtained from the Department of Neurosurgery, Huashan Hospital, to analyze the expression level of HEATR1 mRNA. Additionally, eight control brain tissue samples were obtained from adjacent brain tissues of patients with traumatic brain injury who suffered contusion and laceration. In addition, 10 GBM formalin-fixed, paraffin-embedded (FFPE) tissue sections and 10 normal brain tissues were analyzed by IHC.

Peripheral blood mononuclear cells (PBMCs) were isolated by Ficoll/Paque (Biochrom, Berlin, Germany) density gradient centrifugation of heparinized blood obtained from healthy donors () and patients (benign tumors, 5; grade 2 astrocytoma, 7; grade 3 anaplastic glioma, 10; glioblastoma, 16). The patients’ clinical characteristics are listed in Table 1.

NumberGenderYearsTumor locationPathologyGradeELISpot responseHLA-A2


215F48Right frontalMeningioma0NegativeYes
255F59Left frontalMeningioma0NegativeYes
226F49Sellar regionPituitary adenoma0NegativeYes
261M21Sellar regionPituitary adenoma0NegativeNo

122M45Right temporal Astrocytoma2NegativeNo
135F37Right temporal Astrocytoma2NegativeYes
217M48Right frontal Astrocytoma2NegativeNo
246F30Left parietalAstrocytoma2NegativeYes
252M50Right temporalOligodendroglioma2NegativeNo
264M45Right frontal-callosumAstrocytoma2NegativeYes
262M40Left temporalOligodendroglioma2NegativeYes

218M58Right temporalAA3NegativeYes
238F56Right callosal convolution AO3NegativeNo
254F33Right frontalAOA3NegativeYes
256F46Right temporal-basal gangliaAA3NegativeNo
259F58Right parietalAO3NegativeNo
265M56Right frontal-parietal AOA3NegativeNo
127F42Left frontalAA3PositiveNo
140M45Left temporalAA3PositiveA0201/A1101
156M48Left occipitalAE3PositiveA0203/A3001
239M16Left temporalAA3PositiveA0207/A1102

129F38Left occipitalGBM4NegativeYes
133F48Left temporalGBM4NegativeNo
147F61Left temporalGBM4NegativeYes
150M48Right frontal-temporalGBM4NegativeYes
151M30Right temporalGBM4NegativeYes
214M66Right parietal-occipitalGBM4NegativeNo
223M37Left frontal-temporal GBM4NegativeYes
224M59Left temporal-occipitalGBM4NegativeYes
225M31Left frontal GBM4NegativeNo
231M36Left frontalGBM4NegativeNo
241M58Right frontal GBM4NegativeYes
253M54Left frontalGBM4NegativeNo
132M68Right temporalGBM4PositiveA0201/A0203
141F12Right parietal-occipital GBM4PositiveNo
220M47Left temporalGBM4PositiveA0201/A3303
221F39Right temporalGBM4PositiveNo

: recurrence; F: female; M: male; GBM: glioblastoma multiforme; AA: anaplastic astrocytoma; AO: anaplastic oligodendroglioma; AOA: anaplastic oligoastrocytoma; AE: anaplastic ependymoma.

2.4. FACS with A2B5

The U87 cells were resuspended at a density of 1 × 105 cells/mL in SFM consisting of DMEM/F12 (Invitrogen) supplemented with 20 ng/mL recombinant human bFGF, 20 ng/mL recombinant human EGF, and B27. U87 cells were cultured for 2 weeks. A2B5-PE antibody (Miltenyi Biotec) was used in this study for FACS. Cell sorting was performed on a BD FACSVantage Cell Sorter (BD Biosciences) according to the manufacturer’s instructions.

2.5. Real-Time Reverse Transcription-Polymerase Chain Reaction (RT-PCR) of HEATR1 Expression

Total RNA was extracted from GBM and control brain tissues or from the GBM cell lines using Trizol reagent (Invitrogen) according to the manufacturer’s instructions. First-strand cDNAs were synthesized using a High-Capacity cDNA Archive Kit. Each cDNA (2 μL) was amplified in a SYBR Green Real-time PCR Master Mix (final volume, 20 μL) and loaded on an Applied Biosystems 7900 Real-time PCR Detection System (Applied Biosystems, Foster City, CA, USA). Thermal cycling conditions for quantitative RT-PCR (qRT-PCR) were as follows: the first step, 95°C for 10 min and the ensuing 40 cycles, 95°C for 15 s, 60°C for 60 s, and 72°C for 30 s. The qRT-PCR primers used were as follows: HEATR1 (forward) 5′-TCCTTTTTGATACCCAGCATTTTAT-3′ and HEATR1 (reverse) 5′-TGATCCACCAGAGGCATCATC-3′; actin (forward) 5′-CCCTGGCACCCAGCAC-3′ and actin (reverse) 5′-GCCGATCCACACGGAGTAC-3′. All samples were analyzed in triplicate. To validate that the efficiencies of the target gene amplification and -ACTIN amplification were approximately equal, we plotted standard curves of log input amount versus ΔCT (CTtarget–CTcontrol) for every gene and all the slopes of the plot <0.1. The ΔΔCT method recommended by the manufacturer was used to compare the relative expression levels between samples.

2.6. IHC Analysis

Human GBM FFPE tissue sections were provided and IHC stained with HEATR1-specific antibody made against COOH-terminal peptide of human HEATR1 (Sigma-Aldrich) using a DakoCytomation EnVision+ System-HRP (DAB) detection kit. Briefly, 5 μm tissue sections were dehydrated and subjected to peroxidase blocking. HEATR1 antibody was added at a dilution of 1 : 20 and incubated at room temperature for 30 min on the Dako Autostainer using the DakoCytomation EnVision+ System-HRP (DAB) detection kit. The slides were counterstained with hematoxylin. The stained slides were observed under a microscope and images were acquired. Cytoplasm staining was considered positive. To evaluate HEATR1 expression, 10 high-power fields (400x) within the tumor showing cytoplasm staining were selected. IHC signals were visually quantified by L.F. Sempere using a quick score system combining staining intensity and positive cell percentage (staining intensity: 0 = negative, 1 = weak, 2 = intermediate, and 3 = strong; percentage: 0 = 0%, 1 = <25%, 2 = ≥25%, and 3 = ≥50%). All of the IHC stained sections were evaluated by two senior neuropathologists blinded to the clinical parameters.

2.7. Peptide HLA-A02:01 Binding Affinity

The binding activity of selected peptides to the HLA-A02 molecule was determined semiquantitatively by measuring peptide-induced expression of HLA-A02:01 on T2 cells using flow cytometry. The T2 cells were incubated for 4 h with the candidate peptides, respectively, at a concentration of 20 μg/mL in SFM. After being washed with phosphate buffered saline-fetal calf serum (PBS-FCS), the T2 cells were incubated with supernatant containing murine mAb against HLA-A02:01 derived from BB7.2 cells for 30 min at 4°C. The T2 cells were washed twice with PBS-FCS and stained with 5 μg/mL diluted fluorescein isothiocyanate-conjugated immunoglobulin G which reacts to mouse immunoglobulin for 30 min. The cells were then rinsed three times with PBS-FCS and analyzed using a FACSAria flow cytometer. The percent mean fluorescence index (% MFI) increase of HLA-A02:01 molecules was calculated as follows: % MFI increase = [(MFI with peptide − MFI without peptide)]/(MFI without peptide) × 100 [27].

2.8. Interferon-γ- (IFN-γ-) Based Enzyme-Linked Immunosorbent Spot (ELISpot) Assay

A human IFN-γ ELISpot kit (552138; BD Pharmingen, CA) was used to quantify the CTL response in PBMCs. Several 96-well plates were coated with purified anti-human IFN-γ monoclonal antibodies at the concentration of 5 μg/mL at 100 μL/well and incubated at 4°C overnight and then washed once with 200 μL/well of RPMI-1640 containing 10% FBS and 1% penicillin-streptomycin-L-glutamine (R10) and blocked with 200 μL/well R10 for 2 h at room temperature. PBMCs were then washed twice with R10 and resuspended in R10 complete culture medium. After being counted, the cells were then adjusted to the concentration of 1 × 106 cells/mL and plated onto a 96-well ELISpot plate at 50 μL/well (5 × 104 cells/well) with the addition of 50 μL of the peptide. The final concentration of each peptide was 5 μg/mL. The 96-well ELISpot plates were incubated for about 20 h at 37°C in 5% CO2. After incubation, the ELISpot plates were developed according to the kit instructions. Finally, the plates were air-dried and the resulting spots were counted with ChampSpot IV Bioreader (Beijing SAGE Creation Science, Beijing, China). Peptide-specific IFN-γ ELISpot responses were considered positive only when the number of spots was twofold greater than the control peptide stimulation and there were >50 spots per 1 × 106 PBMCs [28, 29].

2.9. Cytotoxicity Assay by Measuring Lactate Dehydrogenase (LDH) Activity

CytoTox 96 Nonradioactive Cytotoxicity Assay (Cat. number G1780, Promega) was used to determine the cell-mediated cytotoxicity [27, 30]. U87, SHG66, and A172 cells serving as target cells (1 × 105) were loaded with 4 μg/mL peptide for 2 h at 37°C and 5% CO2. Effector PBMCs (1 × 106) were added to peptide-loaded or blank target cells and cultured for additional 4 h at 37°C and 5% CO2. To measure the LDH activity, 50 μL of the reconstituted substrate mix was added to 50 μL of the culture supernatant and incubated at room temperature protected from light for 30 min. A total of 50 μL of the stop solution was added to each well of the plate. The concentrations of the colorimetric product were recorded as absorbance at 490 nm by a spectrometer [27].

2.10. Statistical Analysis

All statistical analyses were carried out using the SPSS 16.0 statistical software package. Continuous variables are expressed as mean ± SEM. Statistical differences between the two groups were evaluated using the unpaired Student’s -test. The correlation between ELISpot response and glioma grades was evaluated using the test. values < 0.05 were considered statistically significant (two-tailed test).

3. Results

3.1. HEATR1 Overexpression in GBM and A2B5+ GBM Cells

First, we investigated whether HEATR1 was overexpressed in GBM cells. We investigated the expression profile of HEATR1 mRNA in 22 primary GBM tissues and eight control brain tissues using quantitative RT-PCR. As shown in Figure 1(a), the expression of HEATR1 mRNA in GBM tissues was higher than that in control brain tissues (). In addition, IHC was initially performed in FFPE tissue sections of primary GBM () and normal brain tissues (). As shown in Figure 1(b), HEATR1 protein was mainly localized in the tumor cell cytoplasm and nuclei. The average IHC score of HEATR1 expression in GBM and normal brain tissues was 4.4 ± 0.7 and 2.1 ± 0.4, respectively. GBM tissues had higher expressions of HEATR1 protein than normal brain tissues (Figure 1(c), ). However, the expression level of HEATR1 proteins did not appear to be correlated with glioma grade (data not shown).

Next, we investigated whether HEATR1 expression in A2B5+GBM cells was higher than that in A2B5−GBM cells. Our previous study showed that U87 cells cultured in SFM for 2 weeks had stem-like features [10]. Furthermore, those A2B5+ U87 cells were double-positive for CD133 and nestin or vimentin (Supplementary Figures 1, 2, and 3, resp., in the Supplementary Material available online at Prior to sorting, the percentage of A2B5+ cells accounted for 6.5%. HEATR1 mRNA in sorted A2B5+ U87 cells was significantly higher than that in A2B5−U87 cells quantified by qRT-PCR (, Figure 1(d)).

3.2. Prediction of Candidate HLA-A02-Binding Peptides Derived from HEATR1

Due to HEATR1 overexpression in GBM, we sought to determine whether HEATR1-derived epitopes that could be presented by antigen process machinery and induce the CTL response in patients with GBM. Since HLA-A02:01 is expressed by 30–40% of Asians as the most common subtype of HLA-I class [31, 32], epitopes potentially binding to HLA-A02:01 were generated using the HLA Peptide Binding Predictions Program ( of the Bioinformatics and Molecular Analysis Section [12]. Six peptides with binding scores >1000 were selected as the candidate epitope peptides (Table 2). Peptides including HEATR12003–2011 (2003–2011, FLFDTQHFI), HEATR11126–1134 (1126–1134, KLLRMLFDL), HEATR12102–2110 (2102–2110, LLPESIPFL), HEATR11411–1419 (1411–1419, FLWILLILL), HEATR1682–690 (682–670, KMVEDLISV), and HEATR1757–765 (757–765, LMLDRGIPV) were synthesized by GL Biochem (Shanghai) Ltd. with >95% purity as indicated by analytic high-performance liquid chromatography and mass spectrometric analysis. The negative control peptides (CFLPVFLAQPPSGQR) were also synthesized.

PeptideHLA moleculeAmino acid positionSubsequence residue listingScore (estimate of half time of disassociation of a molecule containing this subsequence)


was also predicted to bind to HLA-03 and HLA-B08.
was also predicted to bind to HLA-B08, HLA-B40, and HLA-B3801.

3.3. Affinity of Candidate Epitope Peptides for HLA-A02 Molecule

The T2-cell-peptide binding test was used to evaluate the binding affinity of these candidate epitope peptides for HLA-A02 with flow cytometry in vitro (Figure 2(a)). As shown in Figure 2(b), HEATR1682–690 had the highest affinity for HLA-A02:01 and the percentage of MFI increase was 308.5 ± 4.8%. The percentages of MFI increase of HEATR12102–2110, HEATR11126–1134, and HEATR1757–765 were 285.2 ± 49.2%, 287.2 ± 7.7%, and 228.7 ± 5.4%, respectively. HEATR12003–2011 was a lower affinity peptide, while HEATR11411–1419 had the lowest affinity for binding to HLA-A02.

3.4. HEATR1-Derived Peptides Induced CTL Responses

In the next set of experiments, we tested whether those candidate peptides are epitopes that can be recognized by the host immune system in vivo. PBMCs from glioma carriers were incubated with those six mixed peptides and the IFN-γ secretion was tested by the ELISpot. As shown in Table 1, we found that eight patients (anaplastic astrocytomas/ependymoma in four and glioblastoma in four) had positive reactivity with a significant increase of ELISpot-detected spots (Figure 3(a)). The frequency of positive reactivity in malignant gliomas accounts for about 31%. In this study, those positive responses were only observed in the malignant glioma, indicating that those epitopes could be considered specific for malignant gliomas and significantly higher than healthy donors and low-grade glioma carriers (Figure 3(b), ). In addition, three of eight patients with positive reactivity were non-HLA-A02 (Supplementary Table 1), indicating that these peptides might not be exclusively presented by HLA-A02.

Furthermore, we investigated which individual HEATR1-derived peptide could induce the CTL responses. PBMCs from HLA-A02+ patients, five patients with GBM and one control patient with a benign tumor, were stimulated with individual peptide. As shown in Figure 4, HEATR1757–765 had the highest ELISpot response, indicating that it is the most immunogenic in vivo. In addition, the ELISpot responses induced by HEATR1682–690 and HEATR11126–1134 were higher than the others (Figure 5). These data indicate that these three peptides possess the ability to induce CTLs in vivo.

3.5. HEATR1-Specific CTLs Lyse GBM Cells and GSCs

Finally, we evaluated the ability of HEATR1-specific CTLs to lyse GBM cell lines endogenously expressing HEATR1 in vitro; all three GBM cell lines (U87, SHG66, and A172) are capable of expressing endogenous HEATR1 with the highest expression in U87 cell lines (Figure 5(a)). The cytotoxic activity of patients’ PBMCs (effector cells) was evaluated using an LDH-release assay. PBMCs of patient 323 (positive ELISpot response with HLA-A02+; Table 1) were incubated with three GBM cell lines (U87, SHG66, and A172) as target cells, respectively. The results showed that peptide-stimulated PBMCs could lyse 37.4% of U87 and 23.1% of SHG66 target cells expressing both HEATR1 and HLA-A02 at an E : T ratio of 10 : 1 but not A172 cells that are HLA-A02-negative (Figure 5(b)). We further evaluated whether CTLs recognizing the HEATR1 peptides could kill A2B5+ GSCs. PBMCs from patient 323 demonstrated the ability to kill 76.8% of A2B5+ U87 GSCs and 20.4% of A2B5+ SHG66 GSCs at an E : T ratio of 10 : 1 (Figure 5(c)). These data suggest that HEATR1-specific CTLs are effective to lyse target cells endogenously expressing HEATR1; the cytotoxicity is associated with the expression level of endogenous HEATR1.

4. Discussion

To our knowledge, we reported first here that HEATR1 was especially overexpressed in GBM cells and A2B5+GBM cells. T-cell epitopes derived from HEATR1 could significantly induce the CTL response in vivo and these CTLs were able to lyse both GBM cells and GSCs. These results indicate that HEATR1 has great potential for the development of glioma immunotherapy.

The HEATR1 gene is a multiple spliced 7-kb gene that encodes bap28, a protein involved in nucleolar processing of pre-18S ribosomal RNA and ribosome biosynthesis. In the zebrafish central nervous system, bap28 is required for cell survival through its role in rRNA synthesis and processing, and its mutation leads to abnormalities in the brain starting at mid-somitogenesis stages [33]. A recent study indicated that HEATR1 is an ideal minor histocompatibility antigen that is expressed by leukemia stem cells [20, 34]. Moreover, HEATR1 expression detected using TaqMan PCR was higher in testicular and ovarian tissues than in liver, colon, small intestine, lung, brain, and heart tissues [20]. In the meantime, the novel polymorphic minor histocompatibility antigen encoded by the HEATR1 gene could be recognized by one of the CTL clones. In GBM, we first confirmed that HEATR1 expression was significantly higher in most of the GBM samples than in control brain tissues. Although HEATR1 overexpression was not detected in a few cases of GBM, it might contribute to the vast genetic aberrations and their heterogeneity of GBM or GBM samples from the tumor-surrounding tissues. Furthermore, HEATR1 was overexpressed in A2B5+ GSC cells compared to A2B5−tumor cells.

To date, T-cell epitopes derived from several glioma-associated antigens have been shown to elicit T-cell responses against gliomas of several genes, including SART-1 and -3, interleukin-13 receptor a2 chain, ARF4L, GALT3, AIM-2, EphA2, EGFRvIII, HER-2, gp100, MAGE-1, glioma big potassium (gBK), TRP-2, SOX2, SOX11, SOX6, and 3′  β-hydroxysteroid dehydrogenase type 7 gene [12, 24, 3550]. Dutoit et al. recently reported that the peptidomes from ex vivo GBM samples, which consisted of 10 glioblastoma-associated antigen epitopes, induced specific tumor cell lysis by patients’ CD8+ T-cells in vitro and in vivo [51]. Geet al. confirmed that gBK channel-specific peptides could induce HLA-A02-restricted human CD8+ CTLs that killed gBK+ tumor cells [50]. In our study, we confirmed that peptide epitopes derived from HEATR1 could significantly induce the CTL response of killing both GBM cells and A2B5+ GBM progenitor cells.

The CTL response in this study occurred in a non-HLA-A02-dependent manner. We found that HEATR11126–1134 and HEATR11411–1419 were also predicted to bind in the HLA-A03, HLA-B08, HLA-B38:01, and HLA-B40 regions using the epitope prediction system of SYFPEITHI analysis database ( Furthermore, positions 2 and 9 anchor peptides in the HLA-A02-peptide-binding groove are critical for optimal binding to HLA-A02. Positions 2 and 9 anchor peptides of those six peptides derived from HEATR1 were LL, LI, and MV, respectively (Table 2). More than 120 predicted peptides in non-HLA-A02 MHC class I (especially in HLA-B08) were found, where the 2nd and 9th positions were LL, LI, and MV. In addition, the HEATR1 region also was predicted to bind at least 1000 different 15-mers to the HLA-DR regions in the SYFPEITHI analysis database that could stimulate various CD4+ T-cells. Thus, six HEATR1 peptides in this study could cross-bind to the MHC class I or MHC class II region and potentially can be used to treat patients with GBM.

Several studies have used brain tumor stem-like initiating cells or cancer stem-like cells as sources of antigens for DC vaccination against human GBM with the achievement of CSC targeting and enhancing antitumor immunity [1114]. GBM-associated tumor antigens including EGFR, HER2, TRP2, MRP3, AIM2, and SOX2 were twofold to >200-fold higher in CSCs than those in adherent cells [11]. Brown et al. reported that IL13-zetakine+ CTLs were capable of efficient recognition and killing of both GSCs and differentiated cells in vitro and in vivo [15]. Sampson et al. reported that EGFRvIII is expressed in GSC lines and EGFRvIII chimeric antigen receptors-engineered T-cells effectively target these lines [52]. However, the number of GSC-associated proteins’ peptide epitopes known to elicit T-cell responses is rather limited, and sox6 is the first protein expressed in glioma stem cells whose peptides are potentially immunogenic in patients with HLA-A24 or -A02 positive glioma [12]. A2B5 is considered a marker for glioma progenitor cells and A2B5+ cells from human GBM have cancer stem cell properties that are crucial to GBM initiation and maintenance [17, 18]. In our study, we confirmed that HEATR1-derived peptide epitopes could significantly induce the CTL response and then lyse cells from the GBMs and the GSCs, which should be considered a promising strategy for effective T-cell-based immunotherapy for patients with GBM.

HEATR1 expression in normal brain tissues was very low, unlike ARF4L and GALT3, which were markedly expressed in various normal tissues [43, 44]. Interestingly, HEATR1-specific CTLs are only detectable in PBMC derived from patients with malignant gliomas but not in PBMC from healthy donors. Two reasons might account for this discrepancy. First, the induction of HEATR1-specific CTLs may require higher level of HEATR1 expression. As shown in Figure 1, HEATR1 expressions are significantly higher in tumors than in normal tissues. In other words, it is possible that the epitopes expressed in normal tissues are below the threshold level to stimulate T-cell responses [53]. Second, tumor induced microinflammation may result in the increase of permeability of blood-brain barrier and thereby help CTL to access and recognize the presented HEATR1-derived peptide on tumor cells. Furthermore, although the MAGE-1, MAGE-3, Melan-A, gp100, tyrosinase, HER-2, and NY-ESO-1 are expressed in normal testicular, retinal, and/or brain tissues, no autoimmune responses have been elicited in the clinical trials or animal experiments of cancer vaccines [5458]. Of course, our results require further in vivo experiments to confirm the safety and effectiveness of those HEATR1-derived epitope peptides as future immunotherapy for patients with GBM.

5. Conclusion

In this study, we demonstrated the selective overexpression of HEATR1 in A2B5+ GBM cells, whose epitopes could induce specific CTL responses targeting GBM cells and GSCs, suggesting that immunotherapy selectively targeting GSCs could be a novel effective strategy to treat patients with malignant glioma. Combined with other therapeutic avenues, epitope-based GSC-targeting immunotherapy may represent a new promising paradigm for the treatment of patients with GBM [59, 60]. Moreover, those novel CTL epitopes may serve as an attractive component of personalized peptide-based vaccines in the treatment of GBM.


GSCs:Gliomas stem-like cells
CSCs:Cancer stem-like cells
FACS:Fluorescence-activated cell sorting
DC:Dendritic cell
HEATR1:HEAT repeat containing 1
PBMCs:Peripheral blood mononuclear cells
CTL:Cytotoxic T lymphocyte
Q-RT-PCR:Quantitative reverse transcription-polymerase chain reaction
HLA:Human leukocyte antigen

Conflict of Interests

The authors declare that there is no conflict of interests regarding the publication of this paper.

Authors’ Contribution

Zhe Bao Wu, Chao Qiu, and An Li Zhang contributed equally to this paper.


This project was based upon work funded by Grants from Shanghai Science and Technology Research Program (13JC1408000 to Liang Fu Zhou), from the National Natural Science Foundation of China (81271523 to Zhe Bao Wu), and from Zhejiang Provincial Program for the Cultivation of High-level Innovative Health talents (Zhe Bao Wu).

Supplementary Materials

Supplementary Figure 1: Double immunofluorescence staining of A2B5 and CD133.

Supplementary Figure 2: Double immunofluorescence staining of A2B5 and nestin.

Supplementary Figure 3: Double immunofluorescence staining of A2B5 and vimentin.

Supplementary Table 1: HLA subtype of patients with positive ELISpot response.

  1. Supplementary Material


  1. P. Y. Wen and S. Kesari, “Malignant gliomas in adults,” The New England Journal of Medicine, vol. 359, no. 5, pp. 492–507, 2008. View at: Publisher Site | Google Scholar
  2. J. S. Yu, C. J. Wheeler, P. M. Zeltzer et al., “Vaccination of malignant glioma patients with peptide-pulsed dendritic cells elicits systemic cytotoxicity and intracranial T-cell infiltration,” Cancer Research, vol. 61, no. 3, pp. 842–847, 2001. View at: Google Scholar
  3. A. B. Heimberger and J. H. Sampson, “Immunotherapy coming of age: what will it take to make it standard of care for glioblastoma?” Neuro-Oncology, vol. 13, no. 1, pp. 3–13, 2011. View at: Publisher Site | Google Scholar
  4. Z. Li, J. Lee, D. Mukherjee et al., “Immunotherapy targeting glioma stem cells—insights and perspectives,” Expert Opinion on Biological Therapy, vol. 12, no. 2, pp. 165–178, 2012. View at: Publisher Site | Google Scholar
  5. S. K. Singh, C. Hawkins, I. D. Clarke et al., “Identification of human brain tumour initiating cells,” Nature, vol. 432, no. 7015, pp. 396–401, 2004. View at: Publisher Site | Google Scholar
  6. S. K. Singh, I. D. Clarke, M. Terasaki et al., “Identification of a cancer stem cell in human brain tumors,” Cancer Research, vol. 63, no. 18, pp. 5821–5828, 2003. View at: Google Scholar
  7. S. Bao, Q. Wu, R. E. McLendon et al., “Glioma stem cells promote radioresistance by preferential activation of the DNA damage response,” Nature, vol. 444, no. 7120, pp. 756–760, 2006. View at: Publisher Site | Google Scholar
  8. D. Hambardzumyan, M. Squartro, and E. C. Holland, “Radiation resistance and stem-like cells in brain tumors,” Cancer Cell, vol. 10, no. 6, pp. 454–456, 2006. View at: Publisher Site | Google Scholar
  9. A. Murat, E. Migliavacca, T. Gorlia et al., “Stem cell-related “self-renewal” signature and high epidermal growth factor receptor expression associated with resistance to concomitant chemoradiotherapy in glioblastoma,” Journal of Clinical Oncology, vol. 26, no. 18, pp. 3015–3024, 2008. View at: Publisher Site | Google Scholar
  10. W. Hua, Y. Yao, Y. Chu et al., “The CD133+ tumor stem-like cell-associated antigen may elicit highly intense immune responses against human malignant glioma,” Journal of Neuro-Oncology, vol. 105, no. 2, pp. 149–157, 2011. View at: Publisher Site | Google Scholar
  11. Q. Xu, G. Liu, X. Yuan et al., “Antigen-specific T-cell response from dendritic cell vaccination using cancer stem-like cell-associated antigens,” Stem Cells, vol. 27, no. 8, pp. 1734–1740, 2009. View at: Publisher Site | Google Scholar
  12. R. Ueda, K. Ohkusu-Tsukada, N. Fusaki et al., “Identification of HLA-A2- And A24-restricted T-cell epitopes derived from SOX6 expressed in glioma stem cells for immunotherapy,” International Journal of Cancer, vol. 126, no. 4, pp. 919–929, 2010. View at: Publisher Site | Google Scholar
  13. C. E. Brown, R. Starr, C. Martinez et al., “Recognition and killing of brain tumor stem-like initiating cells by CD8+ cytolytic T cells,” Cancer Research, vol. 69, no. 23, pp. 8886–8893, 2009. View at: Publisher Site | Google Scholar
  14. S. Pellegatta, P. L. Poliani, D. Corno et al., “Neurospheres enriched in cancer stem-like cells are highly effective in eliciting a dendritic cell-mediated immune response against malignant gliomas,” Cancer Research, vol. 66, no. 21, pp. 10247–10252, 2006. View at: Publisher Site | Google Scholar
  15. C. E. Brown, R. Starr, B. Aguilar et al., “Stem-like tumor-initiating cells isolated from IL13Rα2 expressing gliomas are targeted and killed by IL13-zetakine-redirected T cells,” Clinical Cancer Research, vol. 18, no. 8, pp. 2199–2209, 2012. View at: Publisher Site | Google Scholar
  16. R. A. Morgan, L. A. Johnson, J. L. Davis et al., “Recognition of glioma stem cells by genetically modified T cells targeting EGFRvIII and development of adoptive cell therapy for glioma.,” Human gene therapy, vol. 23, no. 10, pp. 1043–1053, 2012. View at: Publisher Site | Google Scholar
  17. A. Tchoghandjian, N. Baeza, C. Colin et al., “A2B5 cells from human glioblastoma have cancer stem cell properties,” Brain Pathology, vol. 20, no. 1, pp. 211–221, 2010. View at: Publisher Site | Google Scholar
  18. A. T. Ogden, A. E. Waziri, R. A. Lochhead et al., “Identification of A2B5+CD133- tumor-initiating cells in adult human gliomas,” Neurosurgery, vol. 62, no. 2, pp. 505–515, 2008. View at: Publisher Site | Google Scholar
  19. M. Xu, Y. Yao, W. Hua et al., “Mouse glioma immunotherapy mediated by A2B5+GL261 cell lysate-pulsed dendritic cells,” Journal of Neuro-Oncology, vol. 116, pp. 497–504, 2014. View at: Google Scholar
  20. M. Bleakley, B. E. Otterud, J. L. Richardt et al., “Leukemia-associated minor histocompatibility antigen discovery using T-cell clones isolated by in vitro stimulation of naive CD8+ T cells,” Blood, vol. 115, no. 23, pp. 4923–4933, 2010. View at: Publisher Site | Google Scholar
  21. Y. Yao, A. K. Chan, Z. Y. Qin et al., “Mutation analysis of IDH1 in paired gliomas revealed IDH1 mutation was not associated with malignant progression but predicted longer survival,” PLoS ONE, vol. 8, no. 6, Article ID e67421, 2013. View at: Publisher Site | Google Scholar
  22. Z. B. Wu, L. Cai, S. J. Lin et al., “High-mobility group box 2 is associated with prognosis of glioblastoma by promoting cell viability, invasion, and chemotherapeutic resistance,” Neuro-Oncology, vol. 15, no. 9, pp. 1264–1275, 2013. View at: Publisher Site | Google Scholar
  23. F. Okano, W. J. Storkus, W. H. Chambers, I. F. Pollack, and H. Okada, “Identification of a novel HLA-A*0201-restricted, cytotoxic T lymphocyte epitope in a human glioma-associated antigen, interleukin 13 receptor α2 chain,” Clinical Cancer Research, vol. 8, pp. 2851–2855, 2002. View at: Google Scholar
  24. M. Hatano, J. Eguchi, T. Tatsumi et al., “EphA2 as a glioma-associated antigen: a novel target for glioma vaccines,” Neoplasia, vol. 7, no. 8, pp. 717–722, 2005. View at: Publisher Site | Google Scholar
  25. R. Somasundaram, L. Caputo, D. P. Guerry, and D. Herlyn, “CD8+, HLA-unrestricted, cytotoxic T-lymphocyte line against malignant melanoma,” Journal of Translational Medicine, vol. 3, article 41, 2005. View at: Publisher Site | Google Scholar
  26. G. Z. Jian, J. Eguchi, C. A. Kruse et al., “Antigenic profiling of glioma cells to generate allogeneic vaccines or dendritic cell-based therapeutics,” Clinical Cancer Research, vol. 13, no. 2, pp. 566–575, 2007. View at: Publisher Site | Google Scholar
  27. Z. C. Jia, B. Ni, Z. M. Huang et al., “Identification of two novel HLA-A *0201-restricted CTL epitopes derived from MAGE-A4,” Clinical & Developmental Immunology, vol. 2010, Article ID 567594, 2010. View at: Publisher Site | Google Scholar
  28. A. Bredenbeck, F. O. Losch, T. Sharav et al., “Identification of noncanonical melanoma-associated T cell epitopes for cancer immunotherapy,” Journal of Immunology, vol. 174, no. 11, pp. 6716–6724, 2005. View at: Publisher Site | Google Scholar
  29. T. Hakamada, K. Funatsuki, H. Morita et al., “Identification of novel hepatitis C virus-specific cytotoxic T lymphocyte epitopes by ELISpot assay using peptides with human leukocyte antigen-A*2402-binding motifs,” The Journal of General Virology, vol. 85, no. 6, pp. 1521–1531, 2004. View at: Publisher Site | Google Scholar
  30. M. L. Shinohara, M. Jansson, E. S. Hwang, M. B. F. Werneck, L. H. Glimcher, and H. Cantor, “T-bet-dependent expression of osteopontin contributes to T cell polarization,” Proceedings of the National Academy of Sciences of the United States of America, vol. 102, no. 47, pp. 17101–17106, 2005. View at: Publisher Site | Google Scholar
  31. M. T. Duffour, P. Chaux, C. Lurquin, G. Cornelis, T. Boon, and P. van der Bruggen, “A MAGE-A4 peptide presented by HLA-A2 is recognized by cytolytic T lymphocytes,” European Journal of Immunology, vol. 29, pp. 3329–3337, 1999. View at: Google Scholar
  32. B. Liang, L. Zhu, Z. Liang et al., “A simplified PCR-SSP method for HLA-A2 subtype in a population of Wuhan, China,” Cellular & Molecular Immunology, vol. 3, no. 6, pp. 453–458, 2006. View at: Google Scholar
  33. M. Azuma, R. Toyama, E. Laver, and I. B. Dawid, “Perturbation of rRNA synthesis in the bap28 mutation leads to apoptosis mediated by p53 in the zebrafish central nervous system,” Journal of Biological Chemistry, vol. 281, no. 19, pp. 13309–13316, 2006. View at: Publisher Site | Google Scholar
  34. A. J. Barrett and J. J. Melenhorst, “Minor histocompatibility antigen discovery: turning up the HEATR,” Blood, vol. 115, no. 23, pp. 4630–4631, 2010. View at: Publisher Site | Google Scholar
  35. C. E. Myers, P. Hanavan, K. Antwi et al., “CTL recognition of a novel HLA-A*0201-binding peptide derived from glioblastoma multiforme tumor cells,” Cancer Immunology, Immunotherapy, vol. 60, no. 9, pp. 1319–1332, 2011. View at: Publisher Site | Google Scholar
  36. M. Schmitz, A. Temme, V. Senner et al., “Identification of SOX2 as a novel glioma-associated antigen and potential target for T cell-based immunotherapy,” British Journal of Cancer, vol. 96, no. 8, pp. 1293–1301, 2007. View at: Publisher Site | Google Scholar
  37. G. Liu, H. Ying, G. Zeng, C. J. Wheeler, K. L. Black, and J. S. Yu, “HER-2, gp100, and MAGE-1 are expressed in human glioblastoma and recognized by cytotoxic T cells,” Cancer Research, vol. 64, no. 14, pp. 4980–4986, 2004. View at: Publisher Site | Google Scholar
  38. S. Shimato, A. Natsume, T. Wakabayashi et al., “Identification of a human leukocyte antigen-A24-restricted T-cell epitope derived from interleukin-13 receptor alpha2 chain, a glioma-associated antigen: laboratory investigation,” Journal of Neurosurgery, vol. 109, no. 1, pp. 117–122, 2008. View at: Publisher Site | Google Scholar
  39. K. Kawakami, M. Terabe, M. Kawakami, J. A. Berzofsky, and R. K. Puri, “Characterization of a novel human tumor antigen interleukin-13 receptor α2 chain,” Cancer Research, vol. 66, no. 8, pp. 4434–4442, 2006. View at: Publisher Site | Google Scholar
  40. I. O. Sullivan, M. Blaszczyk-Thurin, C. T. Shen, and H. C. J. Ertl, “A DNA vaccine expressing tyrosinase-related protein-2 induces T-cell-mediated protection against mouse glioblastoma,” Cancer Gene Therapy, vol. 10, no. 9, pp. 678–688, 2003. View at: Publisher Site | Google Scholar
  41. T. Imaizumi, T. Kuramoto, K. Matsunaga et al., “Expression of the tumor-rejection antigen SART1 in brain tumors,” International Journal of Cancer, vol. 83, pp. 760–764, 1999. View at: Google Scholar
  42. K. Murayama, T. Kobayashi, T. Imaizumi et al., “Expression of the SART3 tumor-rejection antigen in brain tumors and induction of cytotoxic T lymphocytes by its peptides,” Journal of Immunotherapy, vol. 23, no. 5, pp. 511–518, 2000. View at: Publisher Site | Google Scholar
  43. Y. Nonaka, N. Tsuda, S. Shichijo et al., “Recognition of ADP-ribosylation factor 4-like by HLA-A2-restricted and tumor-reactive cytotoxic T lymphocytes from patients with brain tumors,” Tissue Antigens, vol. 60, no. 4, pp. 319–327, 2002. View at: Publisher Site | Google Scholar
  44. N. Tsuda, Y. Nonaka, S. Shichijo et al., “UDP-Gal: βGlcNAc β1, 3-galactosyltransferase, polypeptide 3 (GALT3) is a tumour antigen recognised by HLA-A2-restricted cytotoxic T lymphocytes from patients with brain tumour,” British Journal of Cancer, vol. 87, no. 9, pp. 1006–1012, 2002. View at: Publisher Site | Google Scholar
  45. G. Liu, J. S. Yu, G. Zeng et al., “AIM-2: a novel tumor antigen is expressed and presented by human glioma cells,” Journal of Immunotherapy, vol. 27, no. 3, pp. 220–226, 2004. View at: Publisher Site | Google Scholar
  46. A. Wu, J. Xiao, L. Anker et al., “Identification of EGFRvIII-derived CTL epitopes estricted by HLA A0201 for dendritic cell based immunotherapy of gliomas,” Journal of Neuro-Oncology, vol. 76, no. 1, pp. 23–30, 2006. View at: Publisher Site | Google Scholar
  47. M. Schmitz, R. Wehner, S. Stevanovic et al., “Identification of a naturally processed T cell epitope derived from the glioma-associated protein SOX11,” Cancer Letters, vol. 245, no. 1-2, pp. 331–336, 2007. View at: Publisher Site | Google Scholar
  48. H. T. Khong and S. A. Rosenberg, “Pre-existing immunity to tyrosinase-related protein (TRP)-2, a new TRP-2 isoform, and the NY-ESO-1 melanoma antigen in a patient with a dramatic response to immunotherapy,” Journal of Immunology, vol. 168, no. 2, pp. 951–956, 2002. View at: Publisher Site | Google Scholar
  49. G. Liu, H. T. Khong, C. J. Wheeler, J. S. Yu, K. L. Black, and H. Ying, “Molecular and functional analysis of tyrosinase-related protein (TRP)-2 as a cytotoxic T lymphocyte target in patients with malignant glioma,” Journal of Immunotherapy, vol. 26, no. 4, pp. 301–312, 2003. View at: Publisher Site | Google Scholar
  50. L. Ge, N. T. Hoa, A. N. Cornforth et al., “Glioma big potassium channel expression in human cancers and possible T cell epitopes for their immunotherapy,” Journal of Immunology, vol. 189, no. 5, pp. 2625–2634, 2012. View at: Publisher Site | Google Scholar
  51. V. Dutoit, C. Herold-Mende, N. Hilf et al., “Exploiting the glioblastoma peptidome to discover novel tumour-associated antigens for immunotherapy,” Brain, vol. 135, no. 4, pp. 1042–1054, 2012. View at: Publisher Site | Google Scholar
  52. J. H. Sampson, G. E. Archer, D. A. Mitchell, A. B. Heimberger, and D. D. Bigner, “Tumor-specific immunotherapy targeting the EGFRvIII mutation in patients with malignant glioma,” Seminars in Immunology, vol. 20, no. 5, pp. 267–275, 2008. View at: Publisher Site | Google Scholar
  53. N. Renkvist, C. Castelli, P. F. Robbins, and G. Parmiani, “A listing of human tumor antigens recognized by T cells,” Cancer Immunology, Immunotherapy, vol. 50, no. 1, pp. 3–15, 2001. View at: Publisher Site | Google Scholar
  54. X. Hu, N. G. Chakraborty, J. R. Sporn, S. H. Kurtzman, M. T. Ergin, and B. Mukherji, “Enhancement of cytolytic T lymphocyte precursor frequency in melanoma patients following immunization with the MAGE-1 peptide loaded antigen presenting cell-based vaccine,” Cancer Research, vol. 56, no. 11, pp. 2479–2483, 1996. View at: Google Scholar
  55. M. Marchand, N. van Baren, P. Weynants et al., “Tumor regressions observed in patients with metastatic melanoma treated with an antigenic peptide encoded by gene MAGE-3 and presented by HLA-A1,” International Journal of Cancer, vol. 80, no. 2, pp. 219–230, 1999. View at: Google Scholar
  56. B. Thurner, I. Haendle, C. Röder et al., “Vaccination with Mage-3A1 peptide-pulsed nature, monocyte-derived dendritic cells expands specific cytotoxic T cells and induces regression of some metastases in advanced stage IV melanoma,” The Journal of Experimental Medicine, vol. 190, no. 11, pp. 1669–1678, 1999. View at: Publisher Site | Google Scholar
  57. A. Mackensen, B. Herbst, J. Chen et al., “Phase I study in melanoma patients of a vaccine with peptide-pulsed dendritic cells generated in vitro from CD34+ hematopoietic progenitor cells,” International Journal of Cancer, vol. 86, no. 2, pp. 385–392, 2000. View at: Google Scholar
  58. R. M. Prins, S. K. Odesa, and L. M. Liau, “Immunotherapeutic targeting of shared melanoma-associated antigens in a murine glioma model,” Cancer Research, vol. 63, no. 23, pp. 8487–8491, 2003. View at: Google Scholar
  59. S. Phuphanich, C. J. Wheeler, J. D. Rudnick et al., “Phase i trial of a multi-epitope-pulsed dendritic cell vaccine for patients with newly diagnosed glioblastoma,” Cancer Immunology, Immunotherapy: CII, vol. 62, no. 1, pp. 125–135, 2013. View at: Publisher Site | Google Scholar
  60. J. H. Sampson, A. B. Heimberger, G. E. Archer et al., “Immunologic escape after prolonged progression-free survival with epidermal growth factor receptor variant III peptide vaccination in patients with newly diagnosed glioblastoma,” Journal of Clinical Oncology, vol. 28, no. 31, pp. 4722–4729, 2010. View at: Publisher Site | Google Scholar

Copyright © 2014 Zhe Bao Wu et al. This is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

More related articles

 PDF Download Citation Citation
 Download other formatsMore
 Order printed copiesOrder

Related articles

Article of the Year Award: Outstanding research contributions of 2020, as selected by our Chief Editors. Read the winning articles.