Closing the phenotypic gap between transformed neuronal cell lines in culture and untransformed neurons

Abstract

Studies of neuronal dysfunction in the central nervous system (CNS) are frequently limited by the failure of primary neurons to propagate in vitro. Neuronal cell lines can be substituted for primary cells but they often misrepresent normal conditions. We hypothesized that a three-dimensional (3D) cell culture system would drive the phenotype of transformed neurons closer to that of untransformed cells, as has been demonstrated in non-neuronal cell lines. In our studies comparing 3D versus two-dimensional (2D) culture, neuronal SH-SY5Y (SY) cells underwent distinct morphological changes combined with a significant drop in their rate of cell division. Expression of the proto-oncogene N-myc and the RNA-binding protein HuD was decreased in 3D culture as compared to standard 2D conditions. We observed a decline in the anti-apoptotic protein Bcl-2 in 3D culture, coupled with increased expression of the pro-apoptotic proteins Bax and Bak. Moreover, thapsigargin (TG)-induced apoptosis was enhanced in the 3D cells. Microarray analysis demonstrated significantly differing mRNA levels for over 700 genes in the cells of the two culture types, and indicated that alterations in the G1/S cell-cycle progression contributed to the diminished doubling rate in the 3D-cultured SY cells. These results demonstrate that a 3D culture approach narrows the phenotypic gap between neuronal cell lines and primary neurons. The resulting cells may readily be used for in vitro research of neuronal pathogenesis.

Figure 1. 3-D culture-induced changes in cell division rates and morphology

(A) After 3 weeks in RWV culture, the doubling rate of SY cells that were transferred back into 2-D culture for 5 days dropped from 1x/40 h to 1x/65 hours, with no change in viability. Data are shown as the mean (n=4) ± SD. * P<0.001 (B) SY cells grown in standard 2-D tissue culture flasks sediment to the bottom surface and have a flattened morphology. Culture in a RWV promotes 3-D assembly of the individual cells into large tissue-like aggregates. SEM: scanning electron micrograph.

Figure 2. Decreased expression of N-myc and HuD in 3-D versus 2-D-cultured SY cells

(A) Western blot analysis reveals a decrease in the expression of N-myc and HuD proteins in cells that were propagated for 4 weeks in 3-D culture that does not occur during growth in 2-D. (B) Confocal images showing expression of the N-myc oncogene and the neuron-specific RNA-binding protein HuD. The 3-D culture was maintained for 4 weeks. The secondary antibody to N-myc and HuD is labeled with Alexa 488 (green). PI (red) was used as the nuclear stain. The scale bar on each image represents 20 μm.

Figure 3. Resistance to apoptosis is diminished in 3-D-cultured SY cells

(A) 1-Confocal microscopy reveals diminished expression of the anti-apoptotic protein Bcl-2 in SY cells cultured for 3 weeks in a RWV. Pro-apoptotic Bax and Bak proteins are up-regulated in 3-D culture. The secondary antibody (AB) to Bcl-2, Bax and Bak is labeled with Alexa 488 (green). PI (red) or To-Pro (blue) was used to stain the nucleus. Scale bars on the images are: Bcl-2 20 μm, Bax 23.81 μm, Bak 40 μm. (A) 2- Western analysis of whole-cell lysates collected from the 2-D and 3-D cultures at 3 weeks confirms that Bcl-2 expression is down-regulated in 3-D cells, and expression of Bax is up. (B)1- The percent of TUNEL-positive SY cells in 3-D culture increased 4 to 7-fold above those in 2-D when treated with TG (10 nM). (B) 2- TUNEL-positive PC12 cells in 3-D increased 3-fold above those in 2-D. For (B) 1 and 2: 3-D pre-tx = 3-D cells from RWV just before transfer to dish. 2-D+0 = 2-D cells, unstimulated, 2-D+TG = 2-D cells stimulated with TG, 3-D+0 = 3-D cells, unstimulated, 3-D+TG = 3-D cells removed from RWV to dish, stimulated with TG, 3-D(RWV) +TG = 3-D cells treated with TG inside of the RWV. Data are shown as the mean (n=3) ± SD. *P<0.01 in SY (except for the 3-DRWV+TG, where n=1), *P<0.035 in PC12. L axis: actual percent apoptosis R axis: arbitrary units of fold change representing the actual apoptosis.

Figure 4. 3-D culture-driven changes in the phenotypic differentiation markers N-myc and Bcl-2 are still apparent after 5 d of return to 2-D growth in tissue culture flasks

After 10 days of re-introduction to 2-D growth, marker expression in the 3-D cultured cells has returned to a level more analogous to those of the 2-D-cultured cell line. The secondary AB to N-myc and Bcl-2 is labeled with Alexa 488 (green). PI (red) was used as the nuclear stain. The scale bars on the 2-D and 3-D images represent 20 μm except for 5 days, where the bar represents 40 μm.

Figure 5. Comparison of gene expression in 2-D and 3-D-cultured SY cells using microarray analysis

(A) Changes in gene expression due to cell culture conditions affect cellular disease-related pathways (top three of 63, shown in order of significance). Threshold = cutoff for P< 0.05. (B) Ten canonical pathways most affected in SY cells grown in 3-D rather than 2-D. Bar graph = ratio of gene expression in 3-D cultured cells as compared to those grown in 2-D. Line graph represents significance as −log(p-value) with P<0.05. (C) Gene expression pathway and results for G1/S cell cycle progression. Values were obtained using IPA software, version 5.0. Minimum fold change ≥1.5. (D) (Table 2) Confirmation of array results using QRT-PCR. Reactions were run in triplicate with GAPDH gene expression being used as the reference. PCR efficiencies, average fold change and statistical analysis was performed using the REST© software program. All genes in this pathway were represented on the chips. Red: increased gene expression. Green: decreased gene expression. mRNA for the QRT-PCR and array analysis was collected at passage 8 (2-D and 3-D cultures) with n = 2 for each culture type.

Figure 5. Comparison of gene expression in 2-D and 3-D-cultured SY cells using microarray analysis

(A) Changes in gene expression due to cell culture conditions affect cellular disease-related pathways (top three of 63, shown in order of significance). Threshold = cutoff for P< 0.05. (B) Ten canonical pathways most affected in SY cells grown in 3-D rather than 2-D. Bar graph = ratio of gene expression in 3-D cultured cells as compared to those grown in 2-D. Line graph represents significance as −log(p-value) with P<0.05. (C) Gene expression pathway and results for G1/S cell cycle progression. Values were obtained using IPA software, version 5.0. Minimum fold change ≥1.5. (D) (Table 2) Confirmation of array results using QRT-PCR. Reactions were run in triplicate with GAPDH gene expression being used as the reference. PCR efficiencies, average fold change and statistical analysis was performed using the REST© software program. All genes in this pathway were represented on the chips. Red: increased gene expression. Green: decreased gene expression. mRNA for the QRT-PCR and array analysis was collected at passage 8 (2-D and 3-D cultures) with n = 2 for each culture type.