Optimising the chick chorioallantoic membrane xenograft model of neuroblastoma for drug delivery

SK-N-BE(2)C (human NB, ECACC No. 95011817) and IMR-32 (human NB, ECACC No. 86041809) were grown in DMEM (Life Technologies), 10% Foetal Bovine Serum (Biosera, East Sussex, UK), 100 U/ml penicillin,100 μg/ml streptomycin (Sigma, P0781) and 1% Non-Essential Amino Acids (Sigma, M7145). They were maintained at 37 °C with 5% CO₂ in humidified incubator. Passaging was carried out using 0.05% Trypsin/EDTA (Sigma Aldrich) as required. Cell lines were transduced with green fluorescent protein (GFP) lentivirus as described previously [7, 15].

1 × 10⁴ of BE(2)C cells and IMR32 cells were plated onto coverslips in a 24 well plate, incubated for 18-24 h. Medium containing either 10 μM RA, 4 μM of MLN8237 or DMSO alone 0.06% or 0.04% final concentration was added and cells were analysed after 72 h of incubation. To assess the morphology of cells, images of cells were obtained using an inverted microscope (Leica DMIRB) prior to fixation. For immunocytochemistry, coverslips were removed from wells and fixed with 4% paraformaldehyde for 10 min, blocked with 1% BSA, 0.1% Triton X100 in 0.12 M phosphate pH 7.4 for 30 min and stained overnight at 4 °C with 1:50 dilution of Ki67 (Abcam ab16667) followed by 1:500 Goat anti rabbit Alexa 594 (Life Technologies) for one hour at room temperature both diluted in blocking buffer. Cell nuclei were stained with DAPI. Proliferating cells were quantified by Ki67 staining and normalised to the total number of nuclei stained by DAPI. At least three fields per coverslip and 3 coverslips per experiment were counted and a minimum of 300 cells per condition.

Fertilised white leghorn chicken eggs were obtained from Lees Lane Poultry, Wirral, or Tom Barron, Preston, UK. Eggs were incubated at 38 °C and 35–40% humidity and windowed at E3 as described previously [15]. GFP-labelled cells were initially seeded onto the CAM as tumourspheres, in matrigel or as a cell suspension. A cell suspension of 2 × 10⁶ in 5 μl of DMEM seeded onto a slightly injured CAM was found to be most efficient [7]. The CAM was injured by laceration with a pipette tip or traumatisation using a strip of sterile lens tissue causing small bleed [16]. Traumatisation was found to be the most reproducible method and was used for all experiment. To further enhance the efficiency of tumour formation 5 μl of 0.05% trypsin 0.5 mM EDTA was added immediately prior to the addition of cells. For confocal analysis, 10% GFP with 90% unlabelled cells were used to facilitate observing any morphological changes inside the tumours. Eggs were resealed and incubated until E11 [17].

Embryos were treated either topically to the CAM or by injection into the allantoic cavity between E11 and E13. ATRA was used at 10 μM and 100 μM for 3 days at E11, E12 and E13 or 40 μM was used at E11 and E13. Concentration was determined based on the volume of an egg of 45 ml. 2.8 μl, 11.25 μl or 28 μl DMSO diluted to 200 μl in PBS was injected into control embryos. Embryos were dissected on E14 and tumours analysed.

In vitro samples: Each cell line was seeded at a density of 2 × 10⁶ per 75cm² flask and after 24 h, medium was replaced with fresh medium containing either ATRA (10 μM) or MLN8237 (4 μM) or DMSO. Every 48 h the medium was replaced with fresh medium containing RA, MLN8237 or DMSO. After 3 or 6 days, RNA was extracted using RNA mini Kit (QIAGEN) according to manufacturer’s instructions. qPCR was carried out on CFX Connect (Biorad) thermocycler using iTaq Universal SYBR green mix (Biorad) 0.5 μM primers and up to 2 μl cDNA for 35 cycles. An annealing temperature of 60 °C was used for all primer pairs and three technical replicates and three biological replicates were carried out for each sample. qPCR data analysis was carried out using Bio-Rad CFX Manager 3.0 software. Normalised relative expression of target genes was calculated using the ΔΔCq analysis mode. A list of the primers used is provided in Table 1.

In-vivo tumours: Tumours were harvested from the CAM, rinsed in phosphate-buffered saline (PBS), then transferred into RNAlater solution (QIAGEN), and stored at initially at 4 °C or −20 °C for longer term storage prior to RNA extraction. Tissue was first removed from the RNAlater and transferred to a clean RNase free falcon tube. Liquid nitrogen was used to freeze the tissue before a pestle and mortar was used to disrupt it. RNA was then extracted using RNA mini Kit (QIAGEN). qPCR was performed as described above.

Tumours which were harvested for paraffin embedding were fixed overnight in 10% neutral buffered formalin and embedded in paraffin using standard protocols. Prior to staining, the slides underwent deparaffination and high temperature antigen retrieval using a DAKO PT link. Following antigen retrieval, the slides were incubated in EnvisionTM FLEX Wash Buffer (1× working solution pH 7.67; DAKO, K8007) for 5 mins prior to loading onto the DAKO Autostainer (K8012). Sections were incubated for 30 min with Ki67 antibody (1:200) (DAKO M7240) in 5% BSA in Tris Buffered Saline followed by goat anti-mouse HRP (Abcam) and staining with 3,3′-diaminobenzidine. Haematoxylin staining was performed on all the slides and some slides were also stained with eosin to assist in distinguishing between tumour and chick nuclei. A total of 12 fields from 3 slides were counted per tumour and at least two tumours per condition were analysed.

Tumours required for confocal imaging were fixed in 4% paraformaldehyde for one hour, trimmed into small pieces <2mm³ and mounted into slides using DAKO mounting medium. The images were observed using the Leica DMIRE2 microscope at X40 objective to assess the morphology of cells within the tumours.

Statistical significance was computed using Student’s t-test or one-way ANOVA followed by a post-hoc tukey test using SPSS. All data are presented as mean + S.E.M. (standard error of the mean).

The effect of ATRA on MNA neuroblastoma cells has been well characterised in terms of morphology and immunofluorescence of differentiation markers [18‐20]. We wished to develop assays that would enable us to quantify the extent of differentiation upon drug treatment and be suitable to compare effects in culture and in CAM tumours. Differentiation usually goes in parallel with a decrease in proliferation, which can be measured by Ki67 staining. The expression of differentiation and stem cell markers allows direct quantification of the differentiation process and can be assessed by qPCR. Two MNA cell lines BE2C and IMR32 were selected as they respond well to RA. In cell culture, 10 μM ATRA treatment for 3 days prompted the expected change in morphology (Fig. 1a) and a 65% decrease in the proliferation rate in both cell lines (Fig. 1b‐d). The expression of two differentiation markers (STMN4 and ROBO2) and one stem cell marker (KLF4) was further analysed by qPCR. These three markers have previously been shown to exhibit significant changes in IMR32 cells in response to ATRA by qPCR [21]. Here, STMN4 increased more than 6 fold (p < 0.001) and KLF4 decreased by 5.5 fold (p < 0.05) in IMR32 treated with ATRA for 3 days (Fig. 2a). A non-significant increase in ROBO2 and small decrease in MYCN expression was also observed. The results with BE2C cells were similar (Fig. 2a). To test whether the changes with ROBO2 and MYCN would become significant we treated both cells lines with ATRA for 6 days in culture. By this stage ROBO2 was significantly up regulated in both cell lines and MYCN was significantly down regulated confirming the trends observed at three days (Fig. 2b).

Aurora A kinase has been shown to stabilise MYCN [22] and early results in preclinical models have proven promising with data suggesting that MLN8237 might reduce MYCN protein levels, decrease cell proliferation and increase differentiation of Neuroblastoma cells [23, 24]. We tested whether this drug was more effective than RA, prior to further validation of drug delivery to the CAM tumours. We tested 1, 4 and 10 μM MLN8237 in both cell lines. Whilst 10 μM showed some toxicity, 4 μM showed some change in morphology after 3 days (Fig. 3a). MLN8237 also reduced cell proliferation by 22% (BE2C) and 24% (IMR32) a much smaller extent than with ATRA (Fig. 3b). The qPCR results showed no significant change for any of the markers after 3 days treatment in culture although similar trends as for ATRA were observed (Fig. 3c). ATRA was therefore used for subsequent in vivo experiments.

Some cell lines efficiently form tumours on the CAM whilst others, including BE2C and IMR32, do so less frequently [25]. This was surprising since MNA cells are thought to be both aggressive and invasive cells and BE2C and Kelly cells readily extravasate into the embryo following intravenous injection [15]. SKNAS cells form tumours efficiently [7] and BE2C cells mixed with SKNAS cells also formed tumours. However BE2C or IMR32 cells alone more typically formed a sheet of dried cells on the surface of the CAM (data not shown), suggesting that the cells lacked invasive potential. This prompted us to test treating the CAM surface with trypsin immediately prior to adding the cells to facilitate invasion and this indeed improved the efficiency of tumour formation for BE2C cells to more than 70% (Fig. 4a). Trypsin treatment also improved the efficiency of tumour formation for IMR32 and Kelly cells although these remained less efficient than BE2C cells. Tumours formed beneath the surface of the CAM became visible under the fluorescent stereomicroscope within 3–4 days and reached 1-5 mm in size by E14 (Fig. 4b-c). Tumours had clearly recruited blood vessels from the surrounding CAM and were highly vascularised although the IMR32 tumours were less vascularised than the BE2C. The histology of the tumours formed reflected the histology of patient tumours suggesting the tumours are a good model for preclinical analysis (Fig. 4d).

Tumours could be reliably observed by E11 so ATRA treatment was initiated at E11 and repeated at E12 and E13. The volume of eggs was approximately 45 ml and ATRA was added to give a final concentration of 100 μM which was equivalent to 30 mg/kg used for treating mouse xenograft tumours [26]. ATRA is insoluble in aqueous solutions however survival of embryos is compromised by more than 100 μl of DMSO and by the addition of 100% DMSO [27]. Hence the ATRA diluted in maximally 28 μl of DMSO was further diluted up to 200 μl with PBS to form a colloidal suspension. Topical addition of the suspension was used successfully in some experiments however in some cases not all the ATRA re-dissolved. Hence, in later experiments the colloidal suspension was injected into the allantoic sac where it reproducibly dissolved within a few hours, aided by the movements of the chick embryo. Tumours treated with and without ATRA were analysed for changes in the differentiation markers and also for MYCN (Fig. 5a). For IMR32 cells, the results were similar to those observed in culture, with KLF4 down regulated 4.2 fold and STMN4 upregulated 4.4 fold (p < 0.05). ROBO2 and MYCN showed an appropriate trend that although it was not statistically significant. For BE2C cells both ROBO2 and STMN4 were significantly upregulated (5.9 fold and 5.0 fold respectively; p < 0.05) and KLF4 and MYCN were down regulated by 3.4 and 2.2 fold respectively although these latter results were not statistically significant. Survival of the embryos was unaffected by the introduction of ATRA compared to either 28 μl of DMSO per injection or no treatment (Fig. 5b).

The effect of ATRA on cell proliferation of BE2C cells within the tumour was also assayed. The percentage of dividing cells was reduced by 43% compared to the DMSO control (Fig. 6a). Treatment of tumours with 100 μM every 24 h was a considerably higher dose than the 10 μM every 48-72 h used in culture. We therefore tested the effect of lower doses of ATRA. As shown in Fig. 6b, 10 μM ATRA (3 doses) reduced proliferation by 22% (not significant) while 40 μM (2 doses at E11 and E13) reduced proliferation by 37% (p < 0.05) (Fig. 6b and c). Thus four times the dose used in culture was sufficient to produce statistically significant results in vivo (Fig. 6c).

Finally we sought to test whether 40 μM ATRA would also change the morphology of cells within the tumour as is typically observed in culture. For this experiment, only 10% of the cells within the tumour expressed GFP with the remaining 90% were unlabelled BE2C cells. Confocal images of tumours treated with ATRA or DMSO are shown in Fig. 6d. ATRA treated cells exhibited a more elongated shape with cells often having small extensions resembling neurites.

Taken together these results demonstrate, using ATRA as a proof of principle, that the chick CAM model can be used successfully for drug treatment thereby providing a platform of choice for further evaluation of drug efficiency in neuroblastoma.