Human glioma cell culture: Two FCS-free media could be recommended for clinical use in immunotherapy

Human glioma cell culture: two FCS-free media

could be recommended for clinical use in immunotherapy

Anne Clavreul_&Isabelle Jean_&Laurence Preisser_&

Agnès Chassevent&Anne Sapin&Sophie Michalak&

Philippe Menei

Received: 7 November 2008 / Accepted: 29 April 2009 / Published online: 16 June 2009 / Editor: J. Denry Sato

#The Society for In Vitro Biology 2009

Abstract Immunotherapy, particularly active vaccination, may be developed as an effective and safe treatment modality for malignant gliomas, which continue to have a poor prognosis, despite advances in surgical techniques and adjuvant chemotherapy and radiotherapy. Since no glioma- specific tumor-associated antigens (TAAs) have been discov- ered, autologous tumor cells or well-established glioma cell lines could be used in future vaccination protocols to induce antitumour immunity against unknown TAAs. One obstacle for clinical use of these tumour cell vaccines is related to foetal

calf serum (FCS). Efforts are currently being directed toward developing FCS-free media and serum-free alternatives to culture these cell vaccines. In this study, a medium containing human serum and one serum-free medium (UltraCulture), supplemented or not with epidermal growth factor, were tested on morphology, survival, DNA content and TAA expression of human glioma cell lines and glioma biopsy primary cultures.

Their effects were compared on FCS-containing medium.

Results show that, whatever the medium used, no significant variations in morphology and survival were observed. Fur- thermore, human serum-containing medium or UltraCulture preserved at early passage cultures the cell population of interest present in the biopsies before culture. In addition, the expression profile of eight TAAs was similar between these media. These data indicate that human serum-containing medium and UltraCulture serum-free medium could be promising candidates to produce tumour-cell vaccines.

Keywords Cell culture . Glioma . Immunotherapy . Tumour-associated antigens

Introduction

Malignant glioma has a poor prognosis despite aggressive treatment using surgery, radiotherapy and chemotherapy (Westphal et al. 2003; Stupp et al. 2005). One therapy emerging over the last few years is active immunotherapy to initiate T-cell-mediated antitumour immunity (Delhaye et al.

2003; Carpentier and Meng 2006; Das et al. 2008;

Yamanaka 2008). This strategy requires the administration of an antigenic target. Since no glioma-specific, immunologic- relevant tumour-associated antigens (TAAs) have been identified, several approaches were considered to provide these antigens. One method uses whole-tumour homogenates DOI 10.1007/s11626-009-9215-4

A. Clavreul

:

:

A. Clavreul (*)

:

:

:

10, rue André Boquel, 49100 Angers, France

e-mail: anne.clavreul@univ-angers.fr L. Preisser

INSERM, U564,

Bâtiment Montéclair, 4, rue Larrey, 49933 Angers, France

L. Preisser

Service Commun de Cytométrie et d’Analyse Nucléotidique (SCCAN),

Bâtiment Montéclair, 4, rue Larrey, 49933 Angers, France

A. Chassevent

Centre Paul Papin, CRLCC, 2, rue Moll,

49933 Angers, France S. Michalak

Laboratoire Pathologie Cellulaire et Tissulaire, CHU, 49933 Angers, France

made from surgically resected tumour (Rutkowski et al.2004;

Yu et al.2004; Yamanaka et al. 2005; Walker et al.2008;

Wheeler et al.2008). Although this method is quite rapid to obtain TAAs, the contamination of these tumour homoge- nates by normal cells such as normal glial cells, fibroblasts, endothelial cells, microglial cells and macrophages, could induce autoimmunity after their injection. Besides, the induction of lethal experimental allergic encephalomyelitis has been described in primates and guinea pigs after vaccination with human glioblastoma (GBM) tissue (Bigner et al. 1981b). Other methods implicate the isolation of a patient’s own tumour cells by culture or the use of well- established glioma cell lines as a source of TAAs. In this regard, these whole tumour cell vaccines, either irradiated (Mahaley et al. 1983), co-injected with adjuvants such as granulocyte monocyte colony stimulating factor (GM-CSF) (Borrello and Pardoll 2002; Menei et al. 2002; Jean et al.

2004; Chen et al.2006), fused with dendritic cells (Kikuchi et al. 2001; Kikuchi et al. 2004), genetically modified in vitro to increase their immunogenicity (Parney et al.1997;

Okada et al.2001; Smith et al.2007; Ma et al.2008), or used in dendritic cell-based immunotherapy (Liau et al. 2005;

Zhang et al.2007), showed promising data in several animal experiments and clinical trials. The use of autologous tumour cells (ATC) or well-established glioma cell lines as tumour cell vaccines needs to cultivate these cells in vitro. To this date, major studies have expanded glioma cells in a culture medium supplemented with foetal calf serum (FCS). FCS, which has antioxidant properties, supplies cells with essential nutrients allows growth to take place. On the other hand, the use of FCS could lead to different problems, such as possible viral or bacterial infections and prions (bovine spongiforme encephalopathy) which can cause variant Creutzfeldt–Jackob disease. Furthermore, several studies show that FCS repre- sents a powerful allergen able to induce strong immune responses (Mackensen et al. 2000; Warncke et al. 2006;

Kadri et al. 2007). Therefore, for clinical purposes, it is of major importance to find media allowing an optimal expansion of glioma cells without risk of infection and that preserve their TAA expression. One alternative is the use of human serum which could be obtained directly from the patient. However, this approach can present some problems for clinical application such as the amount of autologous serum necessary to expand glioma cells. Nevertheless, as for FCS, a lot of secured human AB serum (HABS) can be produced and its biological activity can be tested. Another alternative is the use of serum-free media which, in comparison to serum-containing media, have a simplified and better-defined composition, a reduced source of infec- tious agents and a lower cost.

In this study, we compared the effect of a medium containing either FCS or HABS and a serum-free medium (UltraCulture) supplemented or not with human recombinant

epidermal growth factor (hrEGF) on established glioma cell lines and on early passage cultures from human glioma biopsies. The characteristics of these cells cultured in these four media are described in terms of morphology, survival, DNA content and TAA expression.

Materials and Methods

Culture media tested. Four culture media were tested on human glioma cell lines and glioma primary cultures: medium

#1: RPMI 1640 or DMEM (Lonza, Verviers, Belgium) complemented with 10% FCS (Hyclone, PerbioScience, Bredières, France) and 1% antibiotics (Sigma, Saint Quentin Fallavier, France); medium #2: RPMI 1640 or DMEM supplemented with 10% HABS (EFS, Lyon, France) and 1%

antibiotics; medium #3: UltraCulture (Lonza), complemented with L-glutamine 100 mM (Lonza) and 1% antibiotics;

medium #4: corresponds to medium#3 supplemented with 10 ng/ml of hrEGF (R&D Systems, Lille, France). Ultra- Culture medium consists of a DMEM:F-12 base supple- mented with recombinant human insulin, bovine transferrin and a purified mixture of bovine serum proteins, including albumin. This medium is manufactured according to current GMPs and is listed with the FDA in a product masterfile.

UltraCulture medium is used in various applications such as production of microcapsules for cellular gene therapy (Shen et al.2009), chondrocyte and fibroblast cell culture (Lonergan et al. 2003; Kita et al. 2006) and development of a bioartificial pancreas (Delaunay et al.1997).

Human glioma cell culture and cell lines. Twelve fresh tumour biopsies (seven de novo GBM, three grade-III oligodendrogliomas and two grade-III oligoastrocytomas) were obtained from patients treated at the Department of Neurosurgery (University Hospital, Angers, France). Samples were minced into small (1–2 mm³ diameter) pieces and mechanically dissociated in DMEM. This tumour slurry was passed through filters with decreasing pore sizes (BD Falcon, Dominique Dutsher, Brumath, France) and aliquots of cell suspensions were conserved at−80°C for further analysis by real-time RT-PCR. For seven tumour biopsies (Tumour 1–7;

Table1), early passage cultures (<3) were established. Cells, after filtration through tissue culture sieves, were seeded into T80 flasks (Nunc, Dominique Dutsher) in the four culture media described above. Cells were grown at 37°C, 5% CO2, and the medium was changed twice a week. The time required for cultures to be passaged once varied from 2 to 3 wk. Six human glioma cell lines (A172, HS683, U138MG, U87MG) and (Lab1, Lab2) were provided by ATCC (LGC Promochem, Molsheim, France) and our laboratory, respec- tively (Table1). These cells were grown at 37°C, 5% CO2, in the four culture conditions.

Morphological study.The morphology of cells cultured in the four media was observed according to several criteria such as density, shape, cellular process and the presence of in-culture spheroids. Observations were made in six separated fields, with an inversed optic microscope (phase contrast, Axiovert, Carl Zeiss SAS, Le Pecq, France), at

×10 magnification in culture flasks.

Survival assay. After the first passage, human glioma cell lines and primary glioma cultures maintained in the four media were harvested with Hanks’BSS containing 3 mM EDTA (Lonza). Cells were seeded in 24-well plates at a concentration of 1×10⁵cells/ml for primary glioma cultures and 2×10⁴cells/ml for glioma cell lines. After 72 h at 37°C, 5% CO2, cell survival was determined using the CellTiter 96® AQueous non-radioactive cell proliferation assay kit

(Promega, Charbonnieres, France). As recommended by the manufacturer, 100μl/well of combined MTS/PMS solution was added. After 3 h at 37°C, the absorbance at 490 nm was recorded using an ELISA plate reader (Labsystem, Multiskan Ascent).

Flow cytometry DNA analysis. Three tumour fragments (Tumour 5, Tumour 6 and Tumour 7) and their corresponding cultures in either HABS-containing medium or UltraCulture were tested. Vindelov’s protocol was used to stain the DNA content (Vindelov et al. 1983) and flow cytometry was performed with a FACScan flow cytometer (BD Biosciences, Le Pont de Claix, France). DNA index (DI) and cell population percentages discriminated by different DNA contents were calculated with Modfit, version 5.2, software (Verity Software House, Topsham, ME). The DNA diploid peak was located on DNA histogrammes according to an external standardisation procedure using normal human lymphocytes and an internal standardisation procedure using trout blood cells. The DI was calculated as the ratio of mean DNA content of tumour cells on the mean DNA content of diploid cells. Cases with a DI value of 1 were classified as diploid, and cases with DI values <1 or >1 were classified as aneuploid. Multiploidy characterised cell suspensions with more than two different peaks on DNA histogrammes.

EGF receptor expression analysis. To detect the expression of EGF receptors (EGFR), cells were analysed by flow cytometry. Briefly, glioma cells, resuspended in PBS containing 5% FCS and 0.02% NaN3, were incubated on ice for 1 h with anti-human EGFR (R&D Systems). An isotype-matched mouse antibody was used as a control.

Following incubation, the cells were washed and further Table 1. Cell line and tumour characteristics

Cell lines/tumours Histology

A172 GBM (ATCC description)

HS683 Glioma (ATCC description)

U138MG GBM (ATCC description)

U87MG GBM, astrocytoma (ATCC description)

Lab1 De novo GBM

Lab2 De novo GBM

Tumour 1 Oligoastrocytoma grade III Tumour 2 Oligodendroglioma grade III

Tumour 3 De novo GBM

Tumour 4 De novo GBM

Tumour 5 De novo GBM

Tumour 6 Oligodendroglioma grade III Tumour 7 Oligodendroglioma grade III

Gene Primer sequences forward and reverse Length (bp) Tm(°C) Accession number

IL-13Rα2 ACCTTTGGGACCTATTCC 196 79.5 NM_000640

CCACTCACTCCAAATTCC

SART-1 AGACCACAGTGCAGAAGG 171 86 NM_005146

TAACGTCGGGTTTGTAGC

SART-3 CTACAACTGCCATGTGGAC 180 86 NM_014706

TCATACACGTGCTCTCTGTC

HER-2 GCCTGTCCCTACAACTACC 189 87 NM_004448

ATTGGCACTGGTAACTGC

gp100 CACCATTACTGACCAGGTG 187 86 NM_006928

CAGGGTTCCACTACTGTCTC

p97 CATTGACACCCTGAAAGG 196 88.5 NM_005929

GAGGGACTCAGAGTAACTGG

MAGE-1 GAAAAGTACCTGGAGTACCG 184 85 NM_004988

CTTCCTCCTCTCTCAAAGC

TRP-2 GGACCTGCATTTGTTACC 187 84.5 AJ000503

GACTAATCAGAGTCGGATCG Table 2. Primers used in the

experimental studies

stained with fluorescein isothiocyanate (FITC)-conjugated goat F(ab′)2 anti-mouse immunoglobulin (Dako, Trappes, France) for 30 min on ice. After washing, the cells were fixed in 2% formaldehyde. The stained cells were analysed using a FACScan flow cytometer with CellQuest Software (BD Biosciences).

Total RNA extraction and real-time RT-PCR. The analysis of TAA expression by glioma cell lines and glioma primary cultures was done by real-time RT-PCR:

Total RNA extraction. Total tumour cell RNA was isolated using a total RNA isolation kit (Macherey-Nagel Inc., Hoerdt, Figure 1. Photomicrographs of glioma cell lines cultured in serum-containing media and UltraCulture serum-free medium, supplemented or not with hrEGF. Thebarindicates 50μm.

France) according to the manufacturer’s instructions. RNA concentrations were determined by optical density at 260 nm, and RNA was stored at−80°C until use. Three RNA samples from normal human brain were obtained from Ambion (Huntingdon, UK), Clinisciences (Montrouge, France) and Ozyme (St Quentin-en-Yvelines, France).

Reverse transcription. cDNA was prepared from 0.3 to 1μg purified RNA (first-strand cDNA synthesis kit, Amersham Biosciences, Orsay, France). Then, cDNA was purified using a Qiaquick PCR purification Kit (Qiagen, Courtaboeuf, France). Purified cDNA was aliquoted and stored at−20°C.

qPCR primer design.Most of the qPCR primer sequences used in this study were designed using primer3 freeware (http://web.unmassmed.edu/bioapps/primer3_www.cgi) and obtained desalted from MWG Biotech (Roissy CDG, France). The length of the amplicons was kept as close as possible to 100–200 bp and the melting temperature of the primers was set at 55°C. Details of the primers and

Figure 2. Photomicrographs of early passage cultures from human glioma biopsies in serum-containing media and UltraCulture serum-free medium supplemented or not with hrEGF. Thebarindicates 50μm.

Figure 3. Abar graphshowing survival analysis of glioma cell lines and early passage glioma cells cultured in serum-containing media and UltraCulture serum-free medium supplemented or not with hrEGF.

Columns represent mean±SEM of results obtained from six glioma cell lines (A172, U87MG, HS683, U138MG, Lab1 and Lab2) or four glioma primary cultures (Tumour 1, Tumour 2, Tumour 3 and Tumour 4). *P< 0.05 (paired t test), significantly different from FCS- containing medium and UltraCulture.

the GenBank accession numbers are given in Table2. The specificity of the primers was confirmed by BLAST searching. Each pair of primers had to present a good amplification curve, a single melting peak showing no artefact amplification and a single band with the expected size after gel electrophoresis of each PCR product. Primers for GFAP and SSX4 genes were purchased from Qiagen.

Six housekeeping genes were selected (ACTB, B2M, GAPDH, HPRT1, RPS18 and HSPCB) and were provided from the ‘Service commun de cytométrie et d’analyse nucléotidique’(SCCAN, Angers, France). The efficacy for each pair of primers was determined by the reamplification of the PCR product and must be >80%.

Real-time RT-PCR.Transcript levels were derived from the accumulation of SYBR green fluorescence in a Chrom 4

(Bio-Rad, Marnes-la-Coquette, France). Reactions were carried out in duplicate in 96-well plates in a final volume of 15 μl. For housekeeping genes and target genes, the reaction mix contained 7.5μl of iQ SYBR Green Supermix 2× (Bio-Rad), 1.9μl RNAse-free water, 0.6μl forward and reverse primer mix (5 μM) and 5μl of cDNA diluted 10- fold in RNAse-free water. The PCR reaction consisted of an initial enzyme activation step at 95°C for 3 min, followed by an amplification and quantification programme repeated for 40 cycles (95°C for 10 s, 55°C for 15 s and 72°C for 15 s, with a single fluorescence measurement of SYBR green I at each end of cycle) and a melting curve programme (55–98°C with a heating rate of 0.5°C and continuous fluorescence measurement). A cycle threshold value (Ct) was obtained for each sample and duplicate sample values were averaged. In a given cohort, the Ct

Figure 4. Flow cytometry DNA analysis was performed on three glioma biopsies and their corresponding cultures (Tumour 5—a,b,c;

Tumour 6—d,e,f; Tumour 7—g,h,i). DNA histogrammes obtained before culture are shown in (a), (d), (g). For each biopsy, the DI of the different cell groups is indicated on their respective peak. The proportion of these populations is then calculated. The graphs (b),

(c), (e), (f), (h), (i) represent the evolution of the proportion of different cell groups present in glioma biopsies before culture and during their culture in HABS-containing medium (b, e, h) or UltraCulture serum-free medium (c,f,i). (DNA–PIDNA–propidium iodide).

values were transformed to relative quantities by using the delta-Ct method (Q=2^{(minCt−sampleCt)}) normalised by the geometric mean of the three most stable housekeeping genes selected by geNorm-assisted analysis (Vandesompele et al.2002).

Statistical analysis. Results are given as mean±standard error of mean (SEM). To describe statistical differences, the Mann–Whitney test and pairedttest were used. APvalue of <0.05 was considered statistically significant.

Results and Discussion

ATC or well-established glioma cell lines are promising candidates for active vaccination against glioma. One major obstacle for their clinical use is the biosafety of FCS, which is currently used for the culture of glioma cells. In this study, the effect of a medium containing FCS or HABS and a serum-free medium, UltraCulture, supplemented or not with hrEGF, has been compared for the morphology, survival, DNA content and TAA expression of human glioma cell lines and glioma primary cultures. HrEGF was used because this factor has been described to enhance in vitro proliferation of malignant glioma cell lines and primary human tumours of the nervous system (Frappaz et al. 1988; Pollack et al. 1990; Engebraaten et al. 1993;

Pedersen et al.1994).

The morphology of six glioma cell lines (A172, HS683, U138MG, U87MG, Lab1 and Lab2) and four early passage

cultures of human glioma biopsies (Tumour 1, Tumour 2, Tumour 3 and Tumour 4) cultured in the four culture conditions are presented in Figs.1and2, respectively. The results obtained with UltraCulture plus hrEGF were not presented since they were similar to those obtained with UltraCulture alone. The different culture conditions tested did not seem to affect the morphology of human glioma cell lines. These cell lines were composed of a homogeneous population of either ramified cells (A172 and Lab2) or short processes cells (HS683, U138MG, U87MG and Lab1). In comparison with glioma cell lines, primary glioma cultures presented some morphology differences among the media used. In particular, spheroids, which are clusters of cells, were more often observed in HABS-containing medium and in UltraCulture with or without hrEGF than in FCS- containing medium. These spheroids were generally well individualised and could be stuck on the flask bottom or floated. HABS has already been described as containing active spheroid-inducing signalling factors (Chun 2000).

The presence of spheroids was not observable in glioma cell lines except in Lab1.

In addition to morphology, the capacity of glioma cells to survive in the different media has been studied. For glioma cell lines, survival was not statistically affected by the culture conditions (Fig.3). In contrast, the survival rate of glioma primary cultures seemed lower in HABS- containing medium in comparison with FCS-containing medium (P=0.03) and UltraCulture (P=0.03). Neverthe- less, the existence of large spheroids in the presence of HABS which could not be dissociated mechanically nor with HBSS/EDTA might underestimate our cell count and

Figure 5. Level and frequency of TAA mRNA expression in normal brain samples (n=3), fresh glioma biopsies (n=12) and glioma cell lines cultured in the presence of FCS (n=6). Total RNA was reverse transcribed into cDNA, and real-time RT-PCR was performed using primers specific for IL-13Rα2, SART-1, SART-3, HER-2, gp100, p97, MAGE-1 and TRP-2. The conversion of Ct values in relative

quantities was performed with the delta-Ct method (see ‘Materials and Methods’). Data are presented as the mean±SEM. *P<0.05 (Mann and Whitney test), significantly different from normal brain samples. Values in parentheses represent the number of positive samples (mRNA expression was considered negative when the Ct score was >33 cycles).

might have limited the incorporation of MTS tetrazolium salt. The addition of hrEGF in UltraCulture did not increase the survival rate of glioma cell lines although all these cell lines showed an expression of EGFR included in 5% to 25% of cases (data not shown). On the contrary, for the four primary glioma cultures analysed, a slight increase of the survival rate was observed when the cells were cultured in UltraCulture supplemented with hrEGF in comparison with UltraCulture alone (P=0.11). This survival rate was similar with that observed in the presence of FCS. Nevertheless, we noted that, after the second passage, glioma primary cultures maintained in UltraCulture with or without hrEGF presented slowed growth when compared to those cultured in serum-containing media (data not shown). To determine if HABS-containing medium and UltraCulture had the same influence on biopsy cultures, the stability of DNA content between the surgical sample and each passage of the resulting primary culture in these media was analysed by flow cytometry. DNA ploidy, estimated by measuring

DI, is closely linked to the modal chromosome number of a tumour cell group. Figure4shows the results obtained with three glioma biopsies (Tumour 5, Tumour 6 and Tumour 7).

Before culture, all these tumours included a diploid cell group (DI=1) and several aneuploid cell groups (DI range=

1.05–1.98). The evolution of DNA profiles showed that notably the culture influenced the respective proportions of these different cell groups. Usually, cells with a DI=1 of which the normal cells belong to decreased progressively in the course of in vitro passages, which was particularly obvious in Tumour 5 and Tumour 6. Interestingly, biopsy culture could give rise to an enrichment of certain aneuploid cell groups. Nevertheless, after the second passage (5 to 7 wk of culture), aneuploid groups, which were in small quantities in the initial biopsy, and which did not develop during culture, tended to disappear (data not shown). No significant difference was observed on the evolution of different cell groups depending on whether the culture was made in HABS-containing medium or Ultra-

Figure 6. amRNA expression profile of IL-13Rα2, SART-1, SART- 3, HER-2, gp100, p97, MAGE-1 and TRP-2 in six glioma cell lines cultured in FCS-containing medium (A172, U87MG, HS683, U138MG, Lab1 and Lab2). b The comparison of TAA mRNA

expression levels in glioma cell lines cultured in serum-containing media (FCS or HABS) and UltraCulture serum-free medium with or without hrEGF. The columns represent the mean±SEM of results obtained from the six glioma cell lines.

Culture. During early passage cultures, these two media preserved the cell population of interest present in the biopsies before culture.

Several studies have shown that the in vitro propagation of tumour cells can increase or decrease antigen expression (Colombatti et al.1989; Bilzer et al.1991; Anderson et al.

2002; Vogel et al. 2005). Such changes are particularly important when investigating tumour-cell-based vaccination strategies. In this study, the influence of culture conditions on the expression of several TAAs was analysed by real- time RT-PCR. The use of this technique may not be the most efficient to predict alterations in tumour antigens which may be important for subsequent immunotherapy trials. Nevertheless, the quantity of cells derived from biopsy cultures did not allow us at the same time to carry out the analysis described above and the study of TAA expression at the protein level. We chose real-time RT-PCR, which is a powerful method for quantifying several transcript expressions from small cell samples. Since a universal glioma-specific antigen was not found, nine TAAs were selected (IL-13Rα2, p97, SART-1, SART-3, MAGE-1, SSX4, HER-2, TRP-2 and gp100). These TAAs have been found on glioma cells (Hiesiger et al.1993; Chi et al.1997; Scarcella et al.1999; Sahin et al.2000; Saikali et al.2007; Shimato et al. 2008). Furthermore, several of

these antigens were recognised by CTLs (Imaizumi et al.

1999; Murayama et al. 2000; Okano et al.2002; Liu et al.

2003; Liu et al.2004).

Figure 5 indicates the level of expression and the frequency of TAA transcripts by normal brain samples, fresh glioma biopsies and glioma cell lines cultured in FCS- containing medium. IL-13Rα2, SART-1, SART-3 and HER- 2 mRNA were expressed strongly and in high frequency (>11/

12) in fresh glioma biopsies. On the contrary, gp100, p97, MAGE-1 and TRP-2 mRNA were hardly expressed and were presented in 10/12, 8/12, 5/12 and 11/12 glioma biopsies, respectively (Fig.5). Comparison of the expression levels of these TAAs with normal brain samples indicated that TAA transcripts were overexpressed in glioma biopsies (Fig. 5).

This overexpression has been observed at the protein level in other studies (Imaizumi et al. 1999; Murayama et al.2000;

Koka et al. 2003; Saikali et al. 2007). No expression of SSX4 mRNA was observed in normal brain samples and glioma biopsies (data not shown). On the contrary, all glioma biopsies and normal brain samples expressed GFAP tran- script which showed glial origin (data not shown). Glioma cell lines also expressed these TAA mRNA but a reduction of the expression of SART-1, SART-3, gp100 and TRP-2 mRNA and an increase of MAGE-1 mRNA expression were observed in comparison to fresh glioma biopsies (Fig.5). A

Figure 7. amRNA expression profile of IL-13Rα2, SART-1, SART- 3, HER-2, gp100, p97, MAGE-1 and TRP-2 in four fresh glioma biopsies (Tumour 1, Tumour 2, Tumour 3 and Tumour 4). b The comparison of TAA mRNA expression levels in glioma biopsies

before culture and after culture in serum-containing media (FCS or HABS) and UltraCulture serum-free medium with or without hrEGF.

Thecolumnsrepresent the mean±SEM of results obtained from the four glioma samples.

slight SSX4 mRNA expression was observed on three glioma cell lines (U87MG, U138MG, Lab1) and GFAP mRNA expression was not observed on glioma cell lines except for Lab2 (data not shown). This last result is in accordance with previous studies which showed that a large number of glioma cell lines did not express GFAP at the protein level (Bigner et al. 1981a; Westphal et al. 1990;

Bocchini et al.1991; Kharbanda et al.1993; Farr-Jones et al.

1999).

The TAA expression profile of each glioma cell line cultured in FCS-containing medium and each glioma biopsy (Tumour 1, Tumour 2, Tumour 3 and Tumour 4) before culture is presented in Figs.6aand7a, respectively.

We can observe that glioma cell lines and glioma biopsies present a variable TAA expression profile between them confirming the heterogeneous nature of malignant glioma cell surface receptor expression. This result helps to explain why there has been no one single TAA that is consistently expressed on all glioma cells that could be used as a target for experimental therapeutics. Furthermore, this indicates that if a glioma cell line is used in a vaccination protocol, it must be verified that its TAA expression profile resembles that of the patient’s own tumour cells. Culture of glioma cell lines in HABS-containing medium or UltraCulture supplemented or not with hrEGF did not affect their TAA mRNA expression level (Fig. 6b). For glioma biopsies, comparison of the TAA mRNA expression level before culture and after culture shows that culture in the four culture conditions did not influence the expression of HER- 2, MAGE-1, gp100, IL-13Rα2 and p97 mRNA (Fig. 7b).

On the contrary, culture of these biopsies induced a reduction of SART-1, SART-3 and TRP-2 mRNA expres- sion. This reduction was observed in the four tested media and took place from the first passage (2 to 3 wk of culture).

After the second passage (5 to 7 wk of culture), a stabilisation of their expression was noted (data not shown).

This reduction could be due to the loss of normal glial cells which also express these antigens (Fig. 5). In the same manner, GFAP mRNA expression diminished after culture in three of the four glioma primary cultures and this diminution was observed in the four tested media (data not shown). This decrease has been observed at the protein level in other published studies (Westphal et al. 1990;

Bocchini et al.1991; Kharbanda et al.1993; Farr-Jones et al.1999).

Conclusion

All these data indicate that HABS-containing medium or UltraCulture serum-free medium could be good candi- dates to replace FCS-containing medium. These two media preserved at early passage cultures the cell

population of interest present in the biopsies before culture. Furthermore, expression profiles of TAA tran- scripts of human glioma cell lines and glioma biopsy primary cultures were similar between these media.

Recently, Lee et al. (2006) showed that tumour stem cells derived from GBM cultured in bFGF and EGF closely mirror the phenotype and genotype of primary tumours. Our study was carried out prior to this publica- tion. It will be interesting in future studies to evaluate the expression of TAA in cancer stem cells and their efficiency in vaccination protocols.

Acknowledgments We would like to thank Denise Joyaux and Cécile Henry for the FCM-DNA analyses and the members of the

‘Service Commun de Cytométrie et d’Analyse Nucléotidique’, SCCAN, Angers, for the facilities provided. This work is part of the

‘Grand Ouest Glioma Project’ (Cancéropôle Grand-Ouest) and is supported by the ‘Ligue Contre le Cancer’, a regional PHRC and INCA.

References

Anderson R. C.; Elder J. B.; Brown M. D. et al. Changes in the immunologic phenotype of human malignant glioma cells after passaging in vitro. Clin. Immunol. 102: 84–95; 2002.

doi:10.1006/clim.2001.5152.

Bigner D. D.; Bigner S. H.; Ponten J. et al. Heterogeneity of genotypic and phenotypic characteristics of fifteen permanent cell lines derived from human gliomas. J. Neuropathol. Exp.

Neurol.40: 201–229; 1981a. doi:10.1097/00005072-198105000- 00001.

Bigner D. D.; Pitts O. M.; Wikstrand C. J. Induction of lethal experimental allergic encephalomyelitis in nonhuman primates and guinea pigs with human glioblastoma multiforme tissue.J.

Neurosurg.55: 32–42; 1981b.

Bilzer T.; Stavrou D.; Wechsler W. et al. Antigen variation in a human glioblastoma: from the primary tumor to the second recurrence, permanent cell line and xenotransplantation tumors.Anticancer.

Res.11: 547–553; 1991.

Bocchini V.; Casalone R.; Collini P. et al. Changes in glial fibrillary acidic protein and karyotype during culturing of two cell lines established from human glioblastoma multiforme. Cell Tissue Res.265: 73–81; 1991. doi:10.1007/BF00318141.

Borrello I.; Pardoll D. GM-CSF-based cellular vaccines: a review of the clinical experience.Cytokine Growth Factor Rev.13: 185–

193; 2002. doi:10.1016/S1359-6101(01)00034-X.

Carpentier A. F.; Meng Y. Recent advances in immunotherapy for human glioma. Curr. Opin. Oncol. 18: 631–636; 2006.

doi:10.1097/01.cco.0000245321.34658.f4.

Chen J. C.; Chen Y.; Wu J. M. et al. Effects of irradiated tumor vaccine and infusion of granulocyte-macrophage colony- stimulating factor and interleukin-12 on established gliomas in rats. Cancer Immunol. Immunother. 55: 873–883; 2006.

doi:10.1007/s00262-005-0077-7.

Chi D. D.; Merchant R. E.; Rand R. et al. Molecular detection of tumor-associated antigens shared by human cutaneous melano- mas and gliomas.Am. J. Pathol.150: 2143–2152; 1997.

Chun M. H. Serum signaling factors and spheroids. Crit. Rev.

Oncol. Hematol.36: 89–98; 2000. doi:10.1016/S1040-8428(00) 00079-2.

Colombatti M.; Dipasquale B.; Del-l’Arciprete L. et al. Heterogeneity and modulation of tumor-associated antigens in human glioblas- toma cell lines.J. Neurosurg.71: 388–397; 1989.

Das S.; Raizer J. J.; Muro K. Immunotherapeutic treatment strategies for primary brain tumors.Curr. Treat Options Oncol.9: 32–40;

2008. doi:10.1007/s11864-008-0055-3.

Delaunay C.; Honiger J.; Darquy S. et al. Bioartificial pancreas containing porcine islets of Langerhans implanted in low-dose streptozotocin-induced diabetic mice: effect of encapsulation medium.Diabetes Metab. 23: 219–227; 1997.

Delhaye M.; Clavreul A.; Garcion E. et al. Anti-glioma vaccination.

Neurochirurgie49: 459–463; 2003.

Engebraaten O.; Bjerkvig R.; Pedersen P. H. et al. Effects of EGF, bFGF, NGF and PDGF(bb) on cell proliferative, migratory and invasive capacities of human brain-tumour biopsies in vitro. Int. J. Cancer 53: 209–214; 1993. doi:10.1002/ijc.

2910530206.

Farr-Jones M. A.; Parney I. F.; Petruk K. C. Improved technique for establishing short term human brain tumor cultures.J. Neurooncol.

43: 1–10; 1999. doi:10.1023/A:1006115608103.

Frappaz D.; Singletary S. E.; Spitzer G. et al. Enhancement of growth of primary metastatic fresh human tumors of the nervous system by epidermal growth factor in serum-free short term culture.

Neurosurgery 23: 355–359; 1988. doi:10.1097/00006123- 198809000-00012.

Hiesiger E. M.; Hayes R. L.; Pierz D. M. et al. Prognostic relevance of epidermal growth factor receptor (EGF-R) and c-neu/erbB2 expression in glioblastomas (GBMs). J. Neurooncol. 16: 93–

104; 1993. doi:10.1007/BF01324695.

Imaizumi T.; Kuramoto T.; Matsunaga K. et al. Expression of the tumor-rejection antigen SART1 in brain tumors.Int. J. Cancer 83: 760–764; 1999. doi:10.1002/(SICI)1097-0215(19991210) 83:6<760::AID-IJC11>3.0.CO;2-R.

Jean W. C.; Spellman S. R.; Wallenfriedman M. A. et al. Effects of combined granulocyte-macrophage colony-stimulating factor (GM-CSF), interleukin-2, and interleukin-12 based immunother- apy against intracranial glioma in the rat.J. Neurooncol.66: 39– 49; 2004. doi:10.1023/B:NEON.0000013477.94568.0f.

Kadri N.; Potiron N.; Ouary M. et al. Fetal calf serum-primed dendritic cells induce a strong anti-fetal calf serum immune response and diabetes protection in the non-obese diabetic mouse. Immunol. Lett. 108: 129–136; 2007. doi:10.1016/j.

imlet.2006.11.007.

Kharbanda K.; Sarkar C.; Dinda A. K. et al. Morphological appearance, growth kinetics and glial fibrillary acidic protein (GFAP) expression in primary in vitro explant culture of astrocytic neoplasms. Acta. Oncol. 32: 301–306; 1993.

doi:10.3109/02841869309093599.

Kikuchi T.; Akasaki Y.; Abe T. et al. Vaccination of glioma patients with fusions of dendritic and glioma cells and recombinant human interleukin 12. J. Immunother. 27: 452–459; 2004.

doi:10.1097/00002371-200411000-00005.

Kikuchi T.; Akasaki Y.; Irie M. et al. Results of a phase I clinical trial of vaccination of glioma patients with fusions of dendritic and glioma cells.Cancer Immunol. Immunother.50: 337–344; 2001.

doi:10.1007/s002620100205.

Kita M.; Hanasono M. M.; Mikulec A. A. et al. Growth and growth factor production by human nasal septal chondrocytes in serum- free media. Am. J. Rhinol. 20: 489–495; 2006. doi:10.2500/

ajr.2006.20.2927.

Koka V.; Potti A.; Forseen S. E. et al. Role of Her-2/neu over- expression and clinical determinants of early mortality in glioblastoma multiforme. Am. J. Clin. Oncol. 26: 332–335;

2003. doi:10.1097/00000421-200308000-00004.

Lee J.; Kotliarova S.; Kotliarov Y. et al. Tumor stem cells derived from glioblastomas cultured in bFGF and EGF more closely

mirror the phenotype and genotype of primary tumors than do serum-cultured cell lines.Cancer Cell9: 391–403; 2006.

Liau L. M.; Prins R. M.; Kiertscher S. M. et al. Dendritic cell vaccination in glioblastoma patients induces systemic and intracranial T-cell responses modulated by the local central nervous system tumor microenvironment.Clin. Cancer Res.11:

5515–5525; 2005. doi:10.1158/1078-0432.CCR-05-0464.

Liu G.; Khong H. T.; Wheeler C. J. et al. Molecular and functional analysis of tyrosinase-related protein (TRP)-2 as a cytotoxic T lymphocyte target in patients with malignant glioma. J.

Immunother. 26: 301–312; 2003. doi:10.1097/00002371- 200307000-00002.

Liu G.; Ying H.; Zeng G. et al. HER-2, gp100, and MAGE-1 are expressed in human glioblastoma and recognized by cytotoxic T cells.Cancer Res.64: 4980–4986; 2004. doi:10.1158/0008-5472.

CAN-03-3504.

Lonergan D. M.; Mikulec A. A.; Hanasono M. M. et al. Growth factor profile of irradiated human dermal fibroblasts using a serum-free method. Plast. Reconstr. Surg. 111: 1960–1968; 2003.

doi:10.1097/01.PRS.0000055065.41599.75.

Ma Y. H.; Yu J. B.; Yao H. P. et al. Treatment of intracerebral glioblastomas with G422 tumour cell vaccine in a mouse model.

J. Int. Med. Res.36: 308–313; 2008.

Mackensen A.; Drager R.; Schlesier M. et al. Presence of IgE antibodies to bovine serum albumin in a patient developing anaphylaxis after vaccination with human peptide-pulsed den- dritic cells.Cancer Immunol. Immunother.49: 152–156; 2000.

doi:10.1007/s002620050614.

Mahaley M. S. Jr.; Bigner D. D.; Dudka L. F. et al. Immunobiology of primary intracranial tumors. Part 7: Active immunization of patients with anaplastic human glioma cells: a pilot study. J.

Neurosurg.59: 201–207; 1983.

Menei P.; Clavreul A.; Piard N. et al. Vaccination contre les gliomes malins par inoculation de cellules tumorales autologues et microperfusion locale de GM-CSF (étude de phase I). Proceeding of Société Francophone de Thérapie Cellulaire et Génique; 2002 Murayama K.; Kobayashi T.; Imaizumi T. et al. Expression of the SART3 tumor-rejection antigen in brain tumors and induction of cytotoxic T lymphocytes by its peptides.J. Immunother.23: 511– 518; 2000. doi:10.1097/00002371-200009000-00001.

Okada H.; Villa L.; Attanucci J. et al. Cytokine gene therapy of gliomas: effective induction of therapeutic immunity to intracra- nial tumors by peripheral immunization with interleukin-4 transduced glioma cells. Gene. Ther. 8: 1157–1166; 2001.

doi:10.1038/sj.gt.3301496.

Okano F.; Storkus W. J.; Chambers W. H. et al. Identification of a novel HLA-A*0201-restricted, cytotoxic T lymphocyte epitope in a human glioma-associated antigen, interleukin 13 receptor alpha2 chain.Clin. Cancer Res.8: 2851–2855; 2002.

Parney I. F.; Petruk K. C.; Zhang C. et al. Granulocyte-macrophage colony-stimulating factor and B7-2 combination immunogene therapy in an allogeneic Hu-PBL-SCID/beige mouse–human glioblastoma multiforme model. Hum. Gene Ther. 8: 1073– 1085; 1997. doi:10.1089/hum.1997.8.9-1073.

Pedersen P. H.; Ness G. O.; Engebraaten O. et al. Heterogeneous response to the growth factors [EGF, PDGF (bb), TGF-alpha, bFGF, IL-2] on glioma spheroid growth, migration and invasion.

Int. J. Cancer56: 255–261; 1994. doi:10.1002/ijc.2910560219.

Pollack I. F.; Randall M. S.; Kristofik M. P. et al. Response of malignant glioma cell lines to epidermal growth factor and platelet-derived growth factor in a serum-free medium. J.

Neurosurg.73: 106–112; 1990.

Rutkowski S.; De Vleeschouwer S.; Kaempgen E. et al. Surgery and adjuvant dendritic cell-based tumour vaccination for patients with relapsed malignant glioma, a feasibility study.Br. J. Cancer91:

1656–1662; 2004.

Sahin U.; Koslowski M.; Tureci O. et al. Expression of cancer testis genes in human brain tumors.Clin. Cancer Res.6: 3916–3922; 2000.

Saikali S.; Avril T.; Collet B. et al. Expression of nine tumour antigens in a series of human glioblastoma multiforme: interest of EGFRvIII, IL- 13Ralpha2, gp100 and TRP-2 for immunotherapy.J. Neurooncol.

81: 139–148; 2007. doi:10.1007/s11060-006-9220-3.

Scarcella D. L.; Chow C. W.; Gonzales M. F. et al. Expression of MAGE and GAGE in high-grade brain tumors: a potential target for specific immunotherapy and diagnostic markers.Clin. Cancer Res.5: 335–341; 1999.

Shen F.; Mazumder M. A.; Burke N. A. et al. Mechanically enhanced microcapsules for cellular gene therapy.J. Biomed. Mater. Res. B Appl. Biomater(in press, July 2009).

Shimato S.; Natsume A.; Wakabayashi T. et al. Identification of a human leukocyte antigen-A24-restricted T-cell epitope derived from interleukin-13 receptor alpha2 chain, a glioma-associated antigen. J. Neurosurg. 109: 117–122; 2008. doi:10.3171/JNS/

2008/109/7/0117.

Smith K. E.; Janelidze S.; Visse E. et al. Synergism between GM-CSF and IFNgamma: enhanced immunotherapy in mice with glioma.

Int. J. Cancer120: 75–80; 2007. doi:10.1002/ijc.22286.

Stupp R.; Mason W. P.; van den Bent M. J. et al. Radiotherapy plus concomitant and adjuvant temozolomide for glioblastoma. N.

Engl. J. Med.352: 987–996; 2005. doi:10.1056/NEJMoa043330.

Vandesompele J.; De Preter K.; Pattyn F. et al. Accurate normali- zation of real-time quantitative RT-PCR data by geometric averaging of multiple internal control genes. Genome Biol. 3;

2002. RESEARCH0034

Vindelov L. L.; Christensen I. J.; Nissen N. I. A detergent-trypsin method for the preparation of nuclei for flow cytometric DNA analysis.

Cytometry3: 323–327; 1983. doi:10.1002/cyto.990030503.

Vogel T. W.; Zhuang Z.; Li J. et al. Proteins and protein pattern differences between glioma cell lines and glioblastoma multi- forme. Clin. Cancer Res. 11: 3624–3632; 2005. doi:10.1158/

1078-0432.CCR-04-2115.

Walker D. G.; Laherty R.; Tomlinson F. H. et al. Results of a phase I dendritic cell vaccine trial for malignant astrocytoma: potential interaction with adjuvant chemotherapy.J. Clin. Neurosci. 15:

114–121; 2008. doi:10.1016/j.jocn.2007.08.007.

Warncke M.; Dodero A.; Dierbach H. et al. Murine dendritic cells generated under serum-free conditions have a mature phenotype and efficiently induce primary immune responses.J. Immunol.

Methods310: 1–11; 2006. doi:10.1016/j.jim.2005.09.012.

Westphal M.; Hilt D. C.; Bortey E. et al. A phase 3 trial of local chemotherapy with biodegradable carmustine (BCNU) wafers (Gliadel wafers) in patients with primary malignant glioma.

Neuro-oncol5: 79–88; 2003. doi:10.1215/15228517-5-2-79.

Westphal M.; Nausch H.; Herrmann H. D. Antigenic staining patterns of human glioma cultures: primary cultures, long-term cultures and cell lines.J. Neurocytol. 19: 466–477; 1990. doi:10.1007/

BF01257237.

Wheeler C. J.; Black K. L.; Liu G. et al. Vaccination elicits correlated immune and clinical responses in glioblastoma multiforme patients.Cancer Res. 68: 5955–5964; 2008. doi:10.1158/0008- 5472.CAN-07-5973.

Yamanaka R. Cell- and peptide-based immunotherapeutic approaches for glioma.Trends Mol. Med.14: 228–235; 2008. doi:10.1016/j.

molmed.2008.03.003.

Yamanaka R.; Homma J.; Yajima N. et al. Clinical evaluation of dendritic cell vaccination for patients with recurrent glioma:

results of a clinical phase I/II trial.Clin. Cancer Res.11: 4160– 4167; 2005. doi:10.1158/1078-0432.CCR-05-0120.

Yu J. S.; Liu G.; Ying H. et al. Vaccination with tumor lysate-pulsed dendritic cells elicits antigen-specific, cytotoxic T-cells in patients with malignant glioma. Cancer Res. 64: 4973–4979;

2004. doi:10.1158/0008-5472.CAN-03-3505.

Zhang J. G.; Eguchi J.; Kruse C. A. et al. Antigenic profiling of glioma cells to generate allogeneic vaccines or dendritic cell- based therapeutics. Clin. Cancer Res. 13: 566–575; 2007.

doi:10.1158/1078-0432.CCR-06-1576.