February 25, 2016
Anandamide Prevents Nerve Growth Factor-Dependent Neurite Outgrowth in PC12 cells
Khadijah Hamid '16 is a senior from Niles, Illinois. She is a biology and neuroscience double major and spa...
Charles Alvarado, Khadijah Hamid, & Alexandra Roman
Department of Biology
Lake Forest College
Lake Forest, IL 60045
Download PDF
Abstract
When PC12 cells are treated with nerve growth factor (NGF), they become differentiated and have extensive neurite growth (Greene & Tischler, 1976). The endocannabinoid anandamide (AEA) influences PC12 differentiation both in vitro and in vivo by activating CB1 receptors (Rueda et al., 2002). The dose-dependent effects of AEA on current models of neuronal differentiation are currently unclear. In NGF-differentiated PC12 cells, 5 μM AEA decreases PC12 differentiation, whereas 10 μM AEA induces apoptosis (Rueda et al., 2002; Sarker et al., 2000). However, in neuronal stem cells, 1 μM AEA promotes neurite outgrowth (Compagnucci et al., 2013). In this study, we investigated whether AEA exhibits dose-dependent effects on PC12 cell differentiation. We hypothesized that lower doses (≤1 μM) of AEA would promote NGF-dependent neurite growth and treatment with 5 μM AEA would prevent this differentiation. We found that AEA treatment overall resulted in decreased total branch length; however, no dose-dependent changes of branch length, primary branches, or branch points were observed. This suggests that AEA treatments may be detrimental for NGF-dependent PC12 cell differentiation, and may not serve as an accurate model for general endocannabinoid signaling in neuronal populations.
Introduction
When rat pheochromocytoma (PC12) cells are exposed to nerve growth factor (NGF), they differentiate and extend their neurites (Greene & Tischler, 1976). There are a few distinct signaling cascades that allow for NGF induced differentiation and PC12 cell survival (Klesse et al., 1999). Previous studies have shown that the activation of the Ras/Erk signaling cascade is crucial for Erk phosphorylation (induced by NGF), and this mechanism is also essential for PC12 cell differentiation. Therefore, when the Ras signaling pathway is inhibited, the differentiation of the PC12 cells is also inhibited (Klesse et al., 1999). This particular signaling cascade is comprised of the Erk cascade which involves kinases such as Raf, Mek, and Erk (Klesse et al., 1999). Another proposed mechanism of NGF growth induction occurs through the Rap1/Raf/B-Raf pathway, which is activated when Raf-1 and B-Raf kinases are stimulated by NGF activation in a Ras-dependent manner and differentiating signals are transduced to al- low for differentiation (Wu et al., 2001; Wood et al., 1993). A study by Rueda et al. (2002) demonstrated that anandamide (AEA) binding results in CB1 mediated inhibition of the TrkA-induced Rap1/B-Raf/ERK activation, leading to a decrease of PC12 cell differentiation. Based on the proposed mechanisms above, it is crucial to further investigate the role of AEA and the endocannabinoid system on NGF induced PC12 differentiation.
Cannabinoid type 1 receptor (CB1) activation initiates multiple signaling pathways involved in the regulation of neurotransmitter release and neurogenesis (Compagnucci et al., 2013). Anandamide (AEA), an endogenous cannabinoid receptor ligand (CB1 receptor agonist), has various effects on cell differentiation both in vitro and in vivo (Rueda et al., 2002; Compagnucci et al., 2013; Marzo et al., 2002). Inhibition of the TrkA-in- duced Rap1/B-Raf/ERK cascade in PC12 cells when CB1 receptors were activated by AEA, lead to a decrease of neurite outgrowth (Rueda et al., 2002). AEA reduces neurite outgrowth in PC12 cells, likely by decreasing NGF induced TrkA tyrosine phosphorylation and decreased ERK activation (Rueda et al., 2002; Tahir et al., 1992). Consistently, Sarker et al. (2000) showed that AEA application induces apoptosis in PC12 cells through the accumulation of oxidative stress and even activation of CPP32-like pro- tease activation in a time-dependent manner in PC-12 cells. However, Compagnucci et al. (2013) showed that AEA promoted neuronal stem cell differentiation and maturation. Overall, these studies report different effects of AEA on cell differentiation and survival within PC12 cells and when they were compared to neuronal stem cells, indicating that these mechanisms need to be explored further.
The studies discussed above tested different concentrations of AEA on PC12 cells or neuronal stem cells at various time periods during their trials. The variation across the studies suggests that observations of cell cultures at various time periods (Sarker et al., 2000), AEA concentrations, and cell cultures may account for the observed differences. Undif- ferentiated PC12 cells treated with a range of 10 μM of AEA for 24 hours, experienced apoptosis at 10 μM of AEA (Sarker et al., 2000). Similarly, when 10 μM of AEA were used on NGF differentiated PC12 cells for 24 hours, apoptosis was also observed (Sarker & Maruyama, 2003). Differ- entiated PC12 cells treated with 5 μM of AEA for 48 hours were seen to exhibit inhibition of neurite outgrowth (Rueda et al., 2002). However, in neural stem cells, AEA demonstrated pro-neural differentiation properties at a 1 μM concentration of AEA (Compagnucci et al., 2013). This shows that AEA inhibits differentiation and even induces apoptosis of PC12 cells, but stimulates neural stem cell differentiation.
The different dose-dependent effects of AEA on NGF-induced PC12 cell differentiation have not yet been fully explored and need to be studied further. After examining the previous studies, we hypothesized that lower doses (≤1 μM) of AEA would promote NGF-dependent neurite growth while 5 μM AEA treatment would prevent this differentiation. We found that AEA treatment resulted in decreased total branch length; however, no dose-dependent changes of branch length, primary branches, or branch points were observed. This suggests that AEA treatments may be detrimental for NGF-dependent PC12 cell differentiation, and may not serve as an accurate model for general endocannabinoid signaling in neuronal populations.
Materials and Methods
Cell Maintenance and Differentiation
PC-12 cells were used for all experiments. They were maintained in 25 cm2 flasks with Cell Maintenance Media (CMM). CMM consist- ed of high-glucose DMEM, 5% horse serum, 5% calf serum, and 1% penicillin-streptomycin (Pen-Strep). For experimentation, cells were plated in 60-mm cell culture dishes coated with 0.1 mg/ml rat collagen I (Invitrogen, A1048301) and grown in Cell Differentiation Media (CDM). CDM consisted of high-glucose DMEM, 1% horse serum, and 1% Pen-Strep and NGF (50 ng/mL, Sigma, NO513-.1MG). Cells were passaged once the flask reached ~80% confluency, and were plated at a density of 2 x 104 cells/mL in the flasks and 5 x 104 cells/mL in the plates.
Anandamide Treatment
Anandamide (AEA, 5 mg/mL, Tocris, 1339) was dissolved in an- hydrous ethanol. Drug treatment was administered after cells had differ- entiated (which was defined by neurite outgrowths being twice as long in length than the cell body), which usually took place after a week of NGF treatment. Four different concentrations were administered to evaluate dosage-dependent effects (5 μM, 1 μM, 0.5 μM, 0.1 μM), maintaining an equivalent amount of anhydrous ethanol, including for control.
Live-cell Microscopy Imaging
Images of cells were taken using a Nikon light microscope at a 40X objective connected to a Nikon Elements Software (NES). Twenty random fields were imaged for each condition in a blinded fashion. Images of differentiated cells were analyzed using the Simple Neurite Tracer plugin in Fiji (Schindelin et al., 2012) to measure the number of primary branches, the number of branch points, and the average branch length to evaluate differentiation.
Statistical Analysis
One trial was completed. A total of 157 cells across the five conditions were analyzed 24 hours after drug administration. Standardized values, as well as skewness and kurtosis, were tabulated to determine the data spread. Standardized z-scores with a higher absolute value of three were considered for removal. We removed three outliers for branch length and two outliers for branch points. After calculating standardized z-scores for a second time, we removed four additional outliers for branch points. When final standardized scores were calculated, one outlier still remained, but was not deleted from the data set (z = 3.21). Ultimately, we analyzed a total of 148 cells for statistics purposes. Means and standard deviations calculated in SPSS 21.0 (IBM) are reported parenthetically throughout the manuscript. Numerical data were imported into Prism 5 (Graphpad) for Mann-Whitney U-tests (α = 0.05) and plotting. Box plots were formatted in Illustrator CS6 (Adobe).
Results
Anandamide Reduces NGF-Dependent Neurite Branch Length At 24 hours, PC12 cells treated with 0.1 μM AEA demonstrated significantly shorter (p < 0.0001) total branch length (N = 27, M = 2.285 μm, SD = 1.537) compared to the control condition (N = 30, M = 4.494 μm, SD = 2.294). Likewise, there was a significant difference (p < 0.005) between the 0.5 μM AEA condition (N = 30, M = 2.875, SD = 1.451) and the control. The trend continued with the 5.0 μM AEA condition (N = 30, M = 2.588, SD = 1.905), as a significant difference (p < 0.001) in total branch length was seen in comparison to the control (Fig 1). The 1.0 μM (N = 31, M = 3.525, SD = 2.525) condition did not demonstrate a significant difference (p = 0.08) with control in regards to branch length as we had expected. The effects of AEA on total branch length were not dose-dependent; however AEA did reduce branch length across the four conditions. This may be because the EC50 value for anandamide on CB1 receptors was 31 nM, and the concentrations used in this study were at a much higher concentration, potentially fully activating the receptor (Di Marzo et al., 2006).
Inconclusive Effects of Anandamide on Primary Branches and Branch Points
There were multiple significant differences in the number of primary branches between the conditions (Fig 2). In relation to control (M = 4.400, SD = 2.608), the 0.1 μM AEA condition (M = 2.815, SD = 1.570), as well as the 1.0 μM AEA condition (M = 3.097, SD = 2.315) demonstrated significantly less primary branches (p < 0.05). The 0.1 μM AEA condition exhibited significantly less primary branches compared with the 0.5 μM AEA condition (M = 4.467, SD = 2.097) and the 5.0 μM AEA condition (M = 4.433, SD = 2.373), 0.5 μM: U(55) = 211, Z = -3.142, p = .002; 5.0 μM: U(55) = 239, Z = -2.696, p = .007. Other dosage differences were seen between the 0.5 μM and 1.0 μM AEA conditions, as the 0.5 μM condition had more primary branches in its cells, U(59) = 279, Z = -2.716, p = .007. Similarly, the 5.0 μM AEA condition had significantly more primary branch- es than the 1.0 μM AEA condition, U(59) = 304, Z = -2.358, p = .018. The morphological differences of decreased branch lengths were evident in all of the AEA treatment conditions; however, the expected dose-dependent decline of primary branches was not observed.
There were a few significant differences in branch point num- ber across the data set (Fig 3). In relation to the control (M = 1.300, SD = 1.263), there were less branch points for the 0.5 μM (M = 0.433, SD = 0.728)and 5.0 μM (M = 0.567, SD = 0.971) AEA conditions, but there was no significance with the other two conditions and control, 0.5 μM: U(58) = 270, Z = -2.910, p = .004; 5.0 0.5 μM: U(58) =295.5, Z = -2.481, p = .013. There was a marginal difference between 0.1 μM and 0.5 μM AEA conditions in reference to branch points, although it was not deemed significant (p = .055). Similar to the degree of primary branching, the effects of AEA were not dose-dependent.
Discussion
Main Findings
When anandamide was administered to the PC12 cells, the branch length decreased regardless of the concentration; however, one of the concentrations (1.0 μM AEA) did not show this trend. Perhaps with more cells analyzed within the conditions or additional replications of the experiment, we will be able to see this trend emerge, as both 5.0 μM AEA and 0.5 μM AEA treatments prevent NGF-dependent neurite outgrowth. The number of primary branches and branch points gave a less conclu- sive picture as to anandamide’s effects on differentiation. Although there were some significant differences between concentrations, no clear trend emerged, raising questions about statistical significance. Once again, analyzing more cells within each condition may help in making a trend ap- pear across concentrations, or simply removing the irrelevant significant results between certain concentrations. Taken together, our data suggest that anandamide causes NGF-dependent neurite retraction at the doses administered in our study, with no dose-dependent effects.
Evaluating Current Model Systems of Neuronal Differentiation
Other studies that used NGF-differentiated PC12 cells as their model system observed similar effects of anandamide on neurite outgrowth. Rueda et al. (2002) found that treatment of PC12 cells treated with NGF and 5.0 μM AEA resulted in the inhibition of neurite outgrowth, compared to cells treated with NGF alone. Treatment with higher concen- trations, such as 10 μM AEA, results in PC12 cell death; however, this is independent of traditional cannabinoid receptors (Sarker et al., 2000; Sarker & Maruyama, 2003). Activation of CB1 receptors using other agonists, such as THC, reduces or disrupts microtubules and microfilaments in differentiated PC12 cells in a dose-dependent manner (Tahir et al., 1992). Because our PC12 data support the general findings thus far, it can be con- cluded that anandamide prevents neurite outgrowth within this particular cellular model.
However, studies evaluating the effects of cannabinoids in other cellular models disagree with the PC12 data, bringing into question the validity of using NGF-dependent PC12 neurite outgrowth as a way to provide additional insight into the brain. Compagnucci et al. (2013) found that treatment with anandamide of neural stem cells actually promotes neuronal differentiation and maturation. Neural progenitor cells experience induction of neuronal differentiation after anandamide treatment, and are unaffected in regards to apoptosis induction (Soltys et al., 2010). Glial pathways are important to note, since anandamide also promotes the differentiation of astroglial cells (Aguado et al., 2006), which is required for neurite outgrowth support (Wang et al., 1994). Thus, it is difficult at this point to definitively state the role of endocannabinoid signaling in the brain and its effects on neurotrophic pathways.
Real-World Application and Future Studies
The endocannabinoid signaling pathways ultimately cause a decrease in neurotransmitter release probability. Thus, they may play import- ant roles in neurogenesis, as many brain regions which perform neurogenesis express an abundance of CB1 receptors, and long-term depression (Rueda et al., 2002; Flores et al., 2013; Peterfi et al., 2012). It is still yet to be determined how CB1 activation regulates NGF and its effects on neurons. Conflicting research on whether CB1 receptor stimulation inhibits (Rueda et al., 2002; Compagnucci et al., 2013) or stimulates (Dalton & Howlett, 2011) ERK, and whether stimulation (Rueda et al., 2002; Dalton & Howlett, 2011) or inhibition (Compagnucci et al., 2013) of ERK promotes neurite outgrowth. Thus, future studies might involve using Western blots to evaluate the phosphorylation of ERK in response to dose-dependent treatment of anandamide. Additionally, while more primary branches and branch points have been used as two more measures to evaluate differentiation (Compagnucci et al., 2013), most papers strictly evaluated neurite length. Based on our results, primary branches and branch points did not show the same trend that neurite length did compared to control, which leads us to question how sensitive the two measurements are at defining differentiation. Finally, evaluating β-tubulin through Western blots and microscopy methods can give us more information into the disruption of differentiation by CB1 receptor activation. This is because CB1 acts through the Rap1/Raf/B-Raf pathway in order to disrupt TrkA’s activation by NGF to promote neurite outgrowth (Rueda et al., 2002; Lu et al., 2010), and prevent NGF-dependent TrkA activity from binding microtubules to lipid rafts in the cellular membrane (Pryor et al., 2012). In conclusion, understanding endocannabinoid signaling and its roles in neuronal differentiation offers a new perspective in the mechanisms underlying neurogenesis, and war- rants further research.
Acknowledgements
We would like to thank Professor Pitt for his guidance and eagerness to help us at all hours of the day; as well as Greg Jamieson and John Vinkavich for assisting us in preparation and maintenance of our cells. Lastly, we would also like to thank the students and affiliates of BIOL346 for their company and friendship throughout the semester.