Sage Journals: Discover world-class research

Abstract

One of the primary obstacles in the restoration or repair of damaged tissues is the temporospatial orchestration of biological and physiological events. Cellular transplantation is an important component of tissue repair as grafted cells can serve as replacement cells or as a source of secreted factors. But few, if any, primary cells can perform more than a single tissue repair function. Epithelial cells, derived from the choroid plexus (CP), are an exception to this rule, as transplanted CP is protective and regenerative in animal models as diverse as CNS degeneration and dermal wound repair. They secrete a myriad of proteins with therapeutic potential as well as matrix and adhesion factors, and contain responsive cytoskeletal components potentially capable of precise manipulation of cellular and extracellular niches. Here we isolated CP from neonatal porcine lateral ventricles and cultured the cells under a variety of conditions to specifically modulate tissue morphology (2D vs. 3D) and protein expression. Using qRT-PCR analysis, transmission electron microscopy, and gene microarray studies we demonstrate a fine level of control over CP epithelial cell clusters opening further opportunities for exploration of the therapeutic potential of this unique tissue source.

Keywords

Choroid plexus (CP)Spheroid Growth factors 3D culture Vascular endothelial growth factor (VEGF)Neuroprotection

Introduction

The choroid plexus (CP) is made up of ciliated columnar epithelial cells that secrete a large, but relatively uncharacterized, number of bioactive factors apically into the cerebrospinal fluid and basally into the circulation (19). While the primary function of the CP is to produce and enrich cerebrospinal fluid in the adult (5), it also plays an instructive role during embryologic development (12,13,17). Because of the wide range of activities encompassed within the CP's complex secretome, cultured CP epithelial clusters are being explored as a therapeutic source of cell-based regenerative molecules. An abundance of regenerative factors secreted from CP have been identified in vitro, including vascular endothelial growth factor (VEGF), connective tissue growth factor (CTGF), fibroblast growth factor 2 (FGF2), transthyretin, insulin-like growth factor II, netrin 1, pentraxin 3, pigment epithelium-derived factor (PEDF), cardiotrophin-1, Wnt-7B, Wnt-10A, nerve growth factor (NGF), neurotrophin-4 (NT-4), transforming growth factor-β (TGF-β), growth hormone (GH), a variety of inflammatory factors, and others (5,8,10,11,16,20,22–24).

Encapsulated within alginate-polyornithine-alginate (APA) microcapsules, transplanted CP clusters are neuroprotective in animal models of Huntington's disease (4,7) and stroke (1,2), and can markedly accelerate the rate and quality of healing of full thickness dermal wounds in rats (22). While these benefits could conceivably be related to secretion of individual factors from CP, it is more logical that such a broad range of therapeutic effects can, in part, be explained by factor synergy within the secreted cocktail. Consistent with this notion is the observation that while delivery of VEGF from hydrogels is potently neuroprotective in a rat model of Huntington's disease (6), comparable benefits have been achieved (3) with CP transplants delivering several log-fold reductions in VEGF together with a cocktail of other CP-secreted neuroprotective factors (23). Together, these data support the notion that CP may exert its therapeutic effects through the secretion of multiple synergistic factors, capable of achieving equivalent potency of monofactor therapy at very low doses.

The therapeutic potential of CP, whether in encapsulated form or as a purified mixture of proteins derived from supernatant, would be significantly enhanced by gaining a greater understanding of the mosaicism of the secreted factors as well as being able to specifically modulate the dose of identified molecules delivered from CP. Indeed, previous studies have shown that regenerative factor candidates in the transcriptome of porcine CP are exceptionally stable over time and that they can be modulated in tissue culture (22). Here we extend these data by examining the effects of media type, extracellular matrix, and CP cluster morphology (three-dimensional cultures). Using VEGF as a candidate factor combined with qRT-PCR analysis, transmission electron microscopy, and gene microarray studies we demonstrate a fine level of control over CP epithelial cell clusters opening further opportunities for exploration of the therapeutic potential of this unique tissue source.

Materials and Methods

Choroid Plexus Isolation

All procedures adhered to NIH and Society for Neuroscience guidelines for the use of research animals. Surgical procedures were conducted under aseptic conditions and all efforts were made to minimize animal suffering and to reduce the number of animals used. Neonatal Yorkshire pigs aged 7–10 days, anesthetized initially with ketamine (500 mg/kg) and xylazine (0.15 mg/kg), were maintained on 3% isoflurane during exsanguination. Following a midline incision, the overlying skull was removed with rongeurs, and the brain was hemisected to expose the CP in the lateral ventricles. The CP were acquired and placed in Hanks balanced salt solution (HBSS, 4°C) supplemented with ciproflaxocin (Sigma, USA, 3 μg/ml), 2% human serum albumin and digested as previously described (3,7). After mechanically disrupting the tissue with scissors, CP epithelial clusters were generated with light enzymatic digestion over 20 min (Liberase HI, Roche, NJ). Clusters of CP epithelial cells were washed and maintained in RPMI-1640 (Sigma, St. Louis, MO) with 2% neonatal porcine serum and 3 μg/ml ciproflaxocin (RPMI-CPN) and placed in nonadherent flasks at 37°C with 5% CO₂. Half of the media was removed and replaced with fresh media after 24 h, 48 h, and every other day thereafter. After 0, 2, and 8 days in RPMI, clusters were counted and separated into individual flasks. After 14 days, the CP clusters were divided into groups incubated in RPMI-CPN as a suspension, immobilized in Matrigel-HC (BD Biosciences), or suspended in serum-free endothelial cell culture medium (ENDO-SFM, Gibco, USA).

Tissue Culture Media Screen and Formation of 3D Hollow Shells

After 2 weeks in RPMI-CPN, cultures were divided into groups in RPMI-CPN, ENDO-SFM, Matrigel-HC in RPMI-CPN, or Matrigel-HC in ENDO-SFM. Briefly, clusters were sedimented, counted, and placed into groups of 65,000 clusters each into four nonadherent plates in either ENDO-SFM or RPMI-CPN. 3D culture groups were created in 24-well plates loaded with 3,000 clusters per well suspended in 50 μl Matrigel-HC diluted with 50 μl RPMI-CPN or ENDO-SFM at 4°C. Matrigel was allowed to gel at 37°C for 15 min, and then 1 ml medium was added to each well. All groups were maintained in an incubator at 37°C with 5% CO₂ for either 7 or 14 days with at least three media changes per week, and then samples were assessed for VEGF secretion by incubation in 1 ml fresh medium for 24 h, followed by extraction of RNA for PCR and microarray. Samples cultured in Matrigel were photographed and the diameter of developing epithelial shells was measured using ImageJ image analysis software. At each time point, at least 100 cell clusters were analyzed. In each culture group, a sample cohort was processed for transmission electron microscopy, wherein epithelial shells (capsuloids) or epithelial clusters were immersion fixed in Karnovsky's fixative, embedded in Spurr epoxy, and thin sectioned for viewing at 80 kV on a Hitachi microscope.

Detection of VEGF

VEGF was measured as a marker of potency as previously described (23). Samples containing 3,000 clusters were incubated with 1 ml fresh medium (37°C, 5% CO₂), and the supernatant was removed and frozen for subsequent analysis. Thawed samples were analyzed in triplicate using a human VEGF ELISA (Quantikine®, R&D Systems, USA). Sequence homology between porcine and human VEGF is greater than 90% (18). The value obtained, expressed in pg/ml, was compared to a standard curve of recombinant human VEGF. Background samples including culture media and Matrigel-HC showed negligible signal and were subtracted to obtain final VEGF values. Groups were statistically compared in a one-way ANOVA and post hoc Dunnett's test against the group cultured in RPMI-CPN using Prism software (GraphPad, USA).

Isolation of RNA, Quantitative Real-Time PCR (QRT-PCR), and Microarray Analysis

QRT-PCR was used to confirm the expression of VEGF and transthyretin. RNA was extracted using RNAeasy (Qiagen, USA) per manufacturer's protocol and optimized for real-time PCR. Total RNA (5 μg) was treated with DNase (Invitrogen, Carlsbad, CA) before reverse transcription with iScript Reverse Transcriptase (Bio-Rad, Hercules, CA). A method was optimized for real-time PCR wherein samples were run on Bio-Rad's iCycler iQ thermocycler at 95°C for 4.5 min, then cycled 40 times at 95°C for 10 s and 59.5°C for 45 s, with a final hold at 95°C for 1 min. Samples were run in triplicate in 25-μl reactions with 25 ng/well of sample (based on total RNA) using iQ SYBR green supermix (Bio-Rad). The primer sequences were: VEGF, 5′-gca gaaggagaccagaaacc-3′ and 5′-actcgatctcatcggggtact-3′; transthyretin, 5′-aggagggagctctttgaagg-3′ and 5′-ttgtctct gcccgagtttct-3′; and β-actin (housekeeping gene), 5′-gg gcagtagcatcgctttag-3′ and 5′-ggctgcccactcaaaataaa-3′. Relative expression of VEGF was calculated using the mathematic model previously described (23). The primer efficiency was calculated by the BioRad iCycler program and is based on the slope of Ct values obtained from standard curves of control tissue samples. Primer efficiencies for the genes of interest (VEGF and transthyretin) and the housekeeping gene (β-actin) were used to determine the relative expression of VEGF between samples from Ct values using the following equation: [2.068^(Ct FACTOR control – Ct FACTOR samples)]/ [2.087^(Ct β-actin control – Ct β-actin samples)],where Ct FACTOR (or Ct β-actin) is the cycle number in which the fluorescent signal from VEGF or transthyretin primers (or β-actin) crosses the noise threshold in the “control” tissue (or unknown “samples”). Statistical analysis was performed between groups using a one-way ANOVA and post hoc Dunnett's test against the group cultured in RPMI-CPN with Prism software (GraphPad, USA).

Separately, RNA was isolated from three of the four groups (RPMI-CPN, ENDO-SFM, Matrigel in RPMI-CPN) followed by Affymetrix Microarray Analysis. RNA sample integrity was confirmed on electropherograms acquired from a Bioanalyzer 2100 (Agilent Technologies, USA). Once samples were confirmed to be suitable for array analysis, an Affymetrix Porcine Microarray was used. Sample preparation and hybridization protocols were used as supplied and recommended by Affymetrix. Samples were run in duplicate and compared in Microsoft Excel against RPMI-CPN controls. Data presented meets the criteria of having fold-changes less than or greater than 2, with p < 0.05.

Results

Morphology

Epithelial clusters ranging between 50 and 200 μm were obtained from the collagenase digestion of fresh CP, with a yield of approximately 100,000 clusters per CP after 8 days in RPMI-CPN. Following a transition to Matrigel-HC, epithelial shells formed that increased in size throughout the course of the experiment. As shown in Figure 1, CP clusters deposited in Matrigel-HC in RPMI-CPN remained as solid masses for only 1 day, as the outer layer of epithelium formed projections toward the basement membrane-rich substrate. After 2 or 3 days, a fluid-filled space was formed in the interior of the cluster that increased in size over the next several days. By 5 days, most of the remaining epithelium that spanned the interior migrated toward the outer shell, causing further overall expansion. Eventually, epithelial shells in close proximity joined to create very large cell assemblies or complex tubular networks, as shown on day 14 in Figure 1. This can also be seen in Figure 2, which shows a grouping of capsuloids at low magnification over the course of 2 weeks. Table 1 shows a morphometric comparison of capsuloids formed in Matrigel-HC cultured in RPMI-CPN or ENDO-SFM. Capsuloids increased in diameter by almost twofold in RPMI-CPN and 1.7x in ENDO-SFM, with relatively consistent overall growth kinetics between groups.

Table 1.

Capsuloid Diameter (μm) as Measured by Image Analysis From a Sample Set of 100 Capsuloids

	Matrigel in RPMI-CPN				Matrigel in ENDO-SFM
	Day 1	Day 3	Day 7	Day 14	Day 1	Day 3	Day 7	Day 14
25th percentile	62.4	70.6	74.8	104.6	56.1	58.2	79.0	95.9
Median	74.8	99.8	103.8	139.4	70.6	74.8	103.8	123.2
75th percentile	103.8	120.4	147.5	179.2	91.4	95.6	128.8	151.9

Figure 1.

Light micrographs of choroid plexus (CP) capsuloids formed in Matrigel-HC. Scale bars: 50 μm (top row; day 1–5) and 100 μm (bottom row, day 7–14).

Figure 2.

Light micrographs of CP capsuloids in Matrigel-HC after 4, 7, 11, or 14 days. Scale bars: 100 μm.

At the ultrastructural level, the reorientation of the CP epithelium was evident through the inverted polarity of cilia and microvilli. In Figure 3A, these structures are seen on the outside of the cluster, which in this illustrated case is 4–5 cells deep and bound by tight junctions (Fig. 3B). In contrast (Fig. 3C), epithelial shells formed in RPMI-CPN contain a fluid-filled interior with motile cilia and microvilli, also with apparent tight junctions within intercellular borders (Fig. 3D). Epithelial shells formed in ENDO-SFM also displayed similar morphology, but with a marked increase in the density of cytoplasmic vesicles and secretory granules (Figs. 3E, F).

Figure 3.

Transmission electron micrographs of CP epithelial clusters maintained in RPMI-CPN (RPMI + ciproflaxocin; A, B), CP epithelial capsuloids maintained in RPMI-CPN (C, D), and CP epithelial capsuloids maintained in ENDO-SFM (serum-free endothelial cell culture media; E, F). Scale bars: 100 μm (A, C, D, E), 1 μm (B), 5 μm (F).

VEGF Expression

Secretion of VEGF from cultured epithelial clusters ranged significantly over the 2-week period. Wells containing 3,000 CP clusters were analyzed in each group, followed by RNA extraction for PCR or microarray. Successful RNA extraction prohibited the destructive measurement of DNA for normalization by cell number, and thus the results are reported as gross secretion per 3,000 cell clusters. Overall, as shown in Figure 4, secreted VEGF was increased in response to capsuloid formation in Matrigel, and was further enhanced with the addition of ENDO-SFM, as previously reported (23). In suspension culture, ENDO-SFM was responsible for an increase in VEGF detected in supernatant by 194% after 7 days and 38% after 14 days, both in comparison to RPMI-CPN. In contrast, Matrigel in RPMI-CPN was associated with a 337% increase in VEGF production after 7 days and 240% after 14 days. The addition of ENDO-SFM to the Matrigel system resulted in a further improvement in VEGF secretion, by 721% after 7 days and 405% after 14 days.

Figure 4.

Vascular endothelial growth factor (VEGF) secretion as determined by hVEGF ELISA based on incubation of 3,000 CP clusters for 24 h in fresh supernatant. Values represent average ± SE. Filled bars: 7 days, open bars: 14 days. ∗p < 0.05.

The relative abundance of transthyretin and VEGF were calculated in relation to β-actin expression using real-time PCR. As shown in Figure 5, values were relatively consistent between suspension culture groups, with ENDO-SFM producing a 110% increase in transthyretin expression and 275% increase in VEGF expression compared to the RPMI-CPN group after 2 weeks in culture. This finding is consistent with the increase in secreted VEGF measured after 2 weeks (240%). However, groups cultured as epithelial shells in Matrigel showed dramatic increases relative to suspension culture groups, on the order of 1000% after 2 weeks.

Figure 5.

Relative expression of transthyretin and VEGF in suspension or 3D culture. Black bars: suspension culture (RPMI-CPN); open bars: suspension culture (ENDO-SFM); light gray bars: 3D culture (Matrigel/RPMI-CPN); dark gray bars: 3D Culture (ENDO-SFM). ∗∗p < 0.01.

Gene Microarray

Porcine whole genome microarray was used to characterize changes to individual genes following exposure to the culture conditions under study. Tables 2 and 3 summarize the genes with statistically significant changes (>2 or <2, p < 0.05) in expression levels between CP clusters cultured in RPMI-CPN and ENDO-SFM (Table 2) or Matrigel-HC in RPMI-CPN (Table 3). In a comparison against CP clusters cultured in RPMI-CPN, the CP cluster transcriptome in ENDO-SFM exhibited 3,093 statistically significant changes in gene expression after 7 days and 4,691 after 14 days (p < 0.05). Similarly, in comparison to RPMI-CPN suspension cultures, CP capsuloids formed in Matrigel-HC in RPMI-CPN showed 4,890 changes in expression after 7 days and 1,828 after 14 days (p < 0.05).

Table 2.

Log₂ Gene Expression Values of CP Clusters in Suspension Culture Obtained From Affymetrix Porcine Microarray

	Day 7				Day 14
	ENDO-SFM	RPMI-CPN	Fold Change	p-Value	ENDO-SFM	RPMI-CPN	Fold Change	p-Value
2′,5′-Oligoadenylate synthetase 1, 40/46 kDa					7.50	5.60	3.74	0.023
Acyl-CoA synthetase long-chain family member 4	10.97	9.83	2.20	0.001	10.86	9.79	2.10	0.005
Allograft inflammatory factor 1	10.39	9.30	2.12	0.026
Alpha-1,3-galactosyltransferase	10.10	8.41	3.21	0.022	9.90	8.15	3.38	0.002
Alpha-actinin-2-associated LIM protein	9.23	7.46	3.41	0.043	9.83	7.57	4.78	0.002
Annexin A1					9.43	10.45	-2.02	0.035
Antithrombin III					10.15	11.73	-2.99	0.026
ATP-binding cassette, sub-family B (MDR/TAP), member 1					6.48	5.46	2.03	0.036
ATP1A2 mRNA, 3'UTR, allele 1	13.31	9.62	12.90	0.006	13.34	8.66	25.68	0.005
Bestrophin 1					7.29	5.91	2.60	0.042
Beta-2-microglobulin	8.38	6.31	4.20	0.014
Calpain, small subunit 1	8.35	7.20	2.23	0.000
Calponin 1, basic, smooth muscle	8.20	7.19	2.01	0.009	8.76	7.70	2.08	0.016
Carboxypeptidase A1 precursor	7.89	9.21	-2.50	0.010	8.12	9.65	-2.89	0.026
Cathepsin K	11.93	10.42	2.83	0.006	12.25	10.45	3.50	0.001
Cathepsin L	10.51	9.47	2.05	0.015
Caveolin 1	10.79	8.70	4.26	0.008	10.12	7.88	4.72	0.008
CD74 antigen					9.23	8.20	2.04	0.033
Chemokine (C-X-C motif) ligand 2	10.25	9.08	2.25	0.011	10.03	8.30	3.32	0.008
Collagen, type I, alpha 1	7.67	6.01	3.16	0.015
Collagen, type III, alpha 1	10.39	8.00	5.22	0.002	9.91	6.59	9.98	0.037
Collagen, type VIII, alpha 1					9.65	8.51nt	2.21	0.013
Complement C4					9.45	8.23	2.32	0.001
Complement component 1, s subcomponent	10.06	8.83	2.35	0.045	10.69	8.78	3.76	0.001
Complement component 5	9.54	7.50	4.11	0.013	9.11	6.96	4.44	0.009
Connexin 43	9.50	8.15	2.55	0.003	9.72	7.74	3.95	0.003
CXCL14 chemokine	10.79	8.96	3.54	0.019	10.19	8.20	3.95	0.034
Cyclin-dependent kinase inhibitor 3	8.46	7.45	2.01	0.030	8.71	7.66	2.07	0.039
Dickkopf homolog 3					8.78	7.54	2.37	0.003
DnaJ (Hsp40) homolog, subfamily A, member 4					10.64	11.68	-2.06	0.012
Early growth response protein 1(Egr-1, NGFI-A)					7.60	8.97	-2.58	0.006
Estrogen sulfotransferase	8.53	10.54	-4.03	0.006	8.25	10.33	-4.21	0.049
Fatty acid binding protein 3, muscle and heart	8.70	7.55	2.21	0.032	9.67	8.08	3.02	0.003
Fatty acid binding protein 4, adipocyte	10.70	6.16	23.21	0.001	7.17	5.56	3.05	0.014
Fatty acid binding protein 5	11.51	9.58	3.80	0.018	11.77	10.00	3.43	0.002
Fibrinogen-like 2	5.96	7.28	-2.48	0.026	9.16	10.85	-3.22	0.002
Fibroblast growth factor 2	8.79	7.00	3.47	0.042	8.81	6.85	3.88	0.006
Fibronectin	9.94	8.45	2.80	0.021	9.51	7.31	4.59	0.003
Galectin-1	11.50	8.63	7.35	0.010	11.77	8.33	10.82	0.027
Gelsolin	13.45	12.45	2.00	0.017	13.62	12.19	2.69	0.001
Glutaminase					9.80	8.78	2.02	0.012
Glutamine-fructose-6-phosphate transaminase 1	6.44	7.66	-2.33	0.053	6.61	7.91	-2.47	0.025
Group VII phospholipase A2	9.76	7.41	5.11	0.001	9.36	7.52	3.57	0.016
Haptocorrin	10.10	4.66	43.16	0.008	6.64	4.64	4.00	0.011
High-affinity copper uptake protein					11.99	12.99	-2.01	0.023
Histone H1.3-like protein					10.58	9.07	2.85	0.002
Inhibin, beta A (activin A, activin AB alpha polypeptide)	12.52	11.41	2.16	0.012	12.52	11.41	2.15	0.025
Integral membrane protein					8.08	9.61	-2.88	0.008
Interleukin 13 receptor, alpha 1	8.56	9.77	-2.32	0.001	11.36	12.39	-2.05	0.015
Isocitrate dehydrogenase 2 (NADP+), mitochondrial	12.89	11.83	2.10	0.029
Kruppel-like factor 4					7.10	8.10	-2.00	0.005
Lactotransferrin					5.84	7.26	-2.68	0.018
Laminin, beta 1					7.20	6.01	2.29	0.035
Lipoprotein lipase	8.70	6.11	5.99	0.019	8.40	6.65	3.37	0.008
Low density lipoprotein receptor					11.34	10.16	2.26	0.001
Lysozyme	5.91	7.00	-2.14	0.008
Matrix Gla protein	9.13	6.73	5.30	0.026	9.03	6.16	7.29	0.021
Membrane-associated protein 17	9.55	8.29	2.39	0.008	9.85	8.23	3.06	0.001
Metallothionein-III	10.85	9.60	2.38	0.014	10.99	9.34	3.14	0.007
MHC class I PD6-glycoprotein					10.35	11.44	-2.12	0.001
Monocyte chemoattractant protein 1					7.56	6.47	2.12	0.022
Na,K-ATPase alpha 2 subunit	8.05	6.10	3.86	0.000	7.85	6.15	3.26	0.031
Neuron-derived orphan receptor-1 alfa					6.57	9.62	-8.28	0.017
Nexin-1					8.49	7.04	2.74	0.010
Non-histone protein HMG2					9.36	10.87	-2.85	0.041
Occludin	9.63	8.44	2.29	0.002	11.80	10.68	2.18	0.005
Odd homeobox 1 protein	12.49	10.72	3.41	0.001	12.48	10.76	3.30	0.003
Phorbol-12-myristate-13-acetate-induced protein 1					9.42	8.20	2.33	0.004
Phosphogluconate dehydrogenase	10.54	11.81	-2.41	0.018
Phosphogluconate dehydrogenase					10.59	11.73	-2.21	0.002
Pigment epithelium-derived factor	11.07	8.89	4.55	0.002	11.04	8.32	6.60	0.002
Plasminogen activator					7.61	6.45	2.23	0.019
Polo-like kinase 2 (Drosophila)	7.60	8.90	-2.46	0.029	7.73	9.38	-3.14	0.004
Polymerase (RNA) II (DNA directed) polypeptide B, 140 kDa					9.66	8.45	2.32	0.002
Porcine inhibitor of carbonic anhydrase	8.86	10.30	-2.71	0.002	8.93	10.57	-3.13	0.011
Pro-melanin-concentrating hormone	11.11	9.98	2.19	0.031
Procollagen-proline, 2-oxoglutarate 4-dioxygenase (proline 4-hydroxylase), alpha	10.70	9.65	2.06	0.050
Propionyl Coenzyme A carboxylase, beta polypeptide					10.19	8.91	2.42	0.006
Putative aldo-keto reductase family 1 member C4	7.18	5.99	2.29	0.009	7.49	6.39	2.15	0.019
Pyridoxal (pyridoxine, vitamin B6) kinase					10.13	9.11	2.02	0.000
RAN, member RAS oncogene family					7.29	8.74	-2.74	0.004
Renin binding protein					9.57	8.39	2.26	0.014
Rho family GTPase 3					10.12	11.17	-2.07	0.006
Ribonuclease, RNase A family, 4					8.84	10.12	-2.42	0.000
Scavenger receptor for phosphatidylserine and oxidized low density lipoprotein	8.94	7.92	2.03	0.044
Somatostatin	10.03	8.48	2.94	0.003
Tenascin C					7.90	6.82	2.12	0.032
Thioredoxin interacting protein	11.93	10.34	3.01	0.018
Tissue inhibitor of metalloproteinase-2					11.81	10.64	2.24	0.008
Tissue inhibitor of metalloproteinase-3					9.85	8.81	2.05	0.001
Transforming growth factor, beta 2	9.24	8.08	2.25	0.009	11.34	10.18	2.23	0.029
TrkB protein	12.01	10.21	3.49	0.000	12.14	9.95	4.58	0.011
Tropomodulin 3 (ubiquitous)	10.21	7.14	8.37	0.002	10.43	7.16	9.68	0.013
Tropomyosin 3					8.22	6.50	3.29	0.012
Tubulin alpha	8.75	6.58	4.50	0.031	10.14	8.99	2.22	0.003
Tumor-associated calcium signal transducer 1	6.65	8.50	-3.59	0.006	6.50	8.82	-5.00	0.014
Tyrosinase-related protein 1	7.17	4.50	6.37	0.005	8.65	4.75	14.98	0.011
Ubiquitous PKC-potentiated PP1 inhibitor					9.11	7.99	2.18	0.020
UDP-Gal:beta-GlcNAc beta-1,3-galactosyltransferase 3	8.15	6.72	2.69	0.013	8.56	6.96	3.02	0.001

Data are sorted alphabetically. Blank entries represent conditions in which there was no statistically significant change.

Table 3.

Log₂ Gene Expression Values of CP Clusters Cultured in 3D Within a Matrigel-HC Matrix, Obtained From Affymetrix Porcine Microarray

	Day 7				Day 14
	Matrigel-HC/RPMI-CPN	RPMI-CPN	Fold Change	p-Value	Matrigel-HC/RPMI-CPN	RPMI-CPN	Fold Change	p-Value
3-hydroxy-3-methylglutaryl coenzyme A reductase/HMG-CoA reductase	8.14	9.54	-2.64	0.026
5-hydroxytryptamine (serotonin) receptor 1F	5.69	4.55	2.21	0.051
5-hydroxytryptamine (serotonin) receptor 2C	5.77	7.97	-4.61	0.006
Activated leukocyte cell adhesion molecule	7.07	8.07	-2.00	0.041
Adrenomedullin	10.33	9.09	2.37	0.046
Alpha-1,3-galactosyltransferase	7.33	8.41	-2.13	0.033
AMP-activated protein kinase alpha 2	9.48	11.17	-3.22	0.030
Amylase, alpha 2B (pancreatic)	9.81	13.59	-13.80	0.029
Angiopoietin-like 4	7.73	6.69	2.07	0.022
Antithrombin III	13.06	11.08	3.94	0.014
Aquaporin 1	7.70	9.13	-2.71	0.010
Arginyl-tRNA synthetase	5.19	7.75	-5.88	0.012
ATP synthase, H+ transporting, mitochondrial F0 complex, subunit G	5.56	6.81	-2.38	0.019
Calcium/calmodulin-dependent protein kinase (CaM kinase) II delta	10.40	11.87	-2.78	0.027
Calmodulin	8.26	9.28	-2.03	0.050
Calpain, small subunit 1	8.10	9.37	-2.42	0.053
Carbonyl reductase 1					5.99	3.90	4.26	0.004
Carboxypeptidase A1 precursor	11.29	9.21	4.22	0.003
Caspase 3, apoptosis-related cysteine peptidase	6.34	8.57	-4.72	0.020
Caveolin 1	5.00	7.61	-6.14	0.050
Cell division cycle 42	8.71	10.10	-2.61	0.027
Chemokine	9.98	8.96	2.02	0.046
Chemokine (C-X-C motif) ligand 2					9.96	8.60	2.58	0.033
Collagen, type I, alpha 1	4.80	6.01	-2.31	0.038
Collagen, type III, alpha 1					9.99	6.59	10.51	0.032
Collagen, type VIII, alpha 1					9.53	8.51	2.02	0.006
Complement C1qC	5.75	9.09	-10.20	0.003	9.07	6.56	5.71	0.022
Complement component 1, q subcomponent, A chain					8.34	6.75	3.00	0.026
Complement component 1, q subcomponent, B chain	6.51	9.35	-7.17	0.019	9.37	6.99	5.21	0.004
Coxsackie-adenovirus-receptor homolog	8.57	10.20	-3.08	0.004
Cytochrome P450 1A1	8.44	6.85	3.02	0.031
Cytochrome P450 2C36	5.74	4.69	2.07	0.050
Dickkopf homolog 3					9.43	7.54	3.71	0.020
EFNA1	7.95	6.82	2.19	0.017
Elastase 2, neutrophil	8.31	7.23	2.11	0.033
Endothelial PAS domain protein 1	8.52	9.97	-2.75	0.008
Estrogen sulfotransferase	6.60	5.51	2.13	0.039
Eukaryotic translation elongation factor 1 alpha 1	4.96	6.90	-3.83	0.052
Eukaryotic translation elongation factor 1 gamma	9.77	13.30	-11.60	0.004
Exportin 1 (CRM1 homolog, yeast)	4.38	5.88	-2.82	0.000
Fatty acid binding protein 5	10.63	9.58	2.07	0.055
Fatty acid synthase					8.97	10.06	-2.12	0.006
Fc fragment of IgG, low affinity IIIb, receptor (CD16b)					8.19	6.86	2.52	0.045
Glutaminase	7.16	8.87	-3.28	0.024
Glutamine-fructose-6-phosphate transaminase 1	6.45	7.66	-2.32	0.021
Glutathione S-transferase	8.76	6.43	5.02	0.042
GTP binding protein overexpressed in skeletal muscle	10.28	7.71	5.93	0.002
Guanine nucleotide binding protein (G protein), alpha inhibiting activity polypeptide	6.40	7.45	-2.07	0.038
Heat shock protein 70	9.05	7.44	3.05	0.026
High-affinity copper uptake protein	11.28	12.63	-2.56	0.031
Histone H1.3-like protein	7.49	9.71	-4.67	0.017
HPCA	6.69	7.79	-2.15	0.048
Hypothetical protein LOC733658	7.69	9.11	-2.66	0.003
Hypoxia-inducible factor 1, alpha subunit (basic helix–loop–helix transcription)	5.54	8.09	-5.86	0.017
Inhibin, beta A (activin A, activin AB alpha polypeptide)	7.78	8.88	-2.15	0.049
Insulin induced gene 1	9.77	11.57	-3.48	0.038	7.00	9.08	-4.21	0.006
Interleukin 6 (interferon, beta 2)					10.16	8.76	2.64	0.005
Keratinocyte growth factor receptor	7.66	8.98	-2.49	0.002
Kruppel-like factor 4	8.54	7.06	2.77	0.013
Lipoprotein lipase	6.25	7.40	-2.21	0.015
Low-density lipoprotein receptor	8.41	9.53	-2.17	0.001
Low-density lipoprotein receptor					9.06	10.16	-2.14	0.055
Lysozyme	5.31	7.00	-3.23	0.004	7.51	5.58	3.82	0.026
Matrix Gla protein					9.70	6.16	11.64	0.008
Membrane-associated protein 17					8.81	7.75	2.08	0.048
Mothers against decapentaplegic homolog 1	9.02	10.16	-2.21	0.002
Motilin	7.83	6.52	2.47	0.002
Neuron-derived orphan receptor-1 alfa	8.74	6.85	3.70	0.005
Neuropeptide Y receptor Y1	9.69	11.44	-3.37	0.013
Nuclear factor of activated T-cells, cytoplasmic, calcineurin-dependent 1	7.14	10.00	-7.27	0.050
O-linked N-acetylglucosamine (GlcNAc) transferase (UDP-N-acetylglucosamine:polyp	9.10	10.17	-2.11	0.024
Olfactory receptor	6.24	5.16	2.11	0.043
Optineurin /// Tumor-associated calcium signal transducer 1	7.43	10.04	-6.08	0.035
Orosomucoid 1	10.06	8.94	2.17	0.028
Palate, lung and nasal epithelium carcinoma associated	7.32	9.27	-3.87	0.048
Peroxisome proliferator activated receptor, gamma, coactivator 1	7.78	8.93	-2.22	0.033
Plasminogen activator inhibitor I	8.35	6.88	2.75	0.004
Polo-like kinase 2 (Drosophila)	10.33	8.90	2.71	0.003
Polymerase (RNA) II (DNA directed) polypeptide B, 140 kDa	7.33	8.35	-2.02	0.028
Prepro-beta-defensin 1	11.30	9.20	4.31	0.010
Proteasome (prosome, macropain) activator subunit 3 (PA28 gamma; Ki)	7.04	8.15	-2.17	0.009
Protein kinase, cAMP-dependent, regulatory, type II, alpha	5.70	7.07	-2.59	0.015
Protein phosphatase 1, regulatory (inhibitor) subunit 12A	8.66	10.20	-2.91	0.053
Protein S	10.90	11.97	-2.09	0.048
Psoriasis susceptibility 1 candidate 2	9.06	7.97	2.13	0.040
Putative aldo-keto reductase family 1 member C4	7.68	10.24	-5.88	0.014
Putative membrane steroid receptor	11.74	12.84	-2.14	0.038
Quaking homolog, KH domain RNA binding (mouse)	8.14	9.42	-2.42	0.042
Ras-related protein Rab-11A	9.59	10.73	-2.20	0.034
Regulator of G-protein signaling 2, 24kDa	12.13	10.93	2.30	0.015
Ribose 5-phosphate isomerase-like protein	7.81	8.85	-2.06	0.049
Ribosomal protein L10	7.21	9.67	-5.47	0.045
Ribosomal protein S17	11.46	9.73	3.31	0.006	11.17	9.98	2.28	0.005
ROD1 regulator of differentiation 1	6.80	7.82	-2.03	0.055
Scavenger receptor class B member 2	6.70	8.17	-2.77	0.030
Secreted phosphoprotein 1 (osteopontin, bone sialoprotein I, early T-lymphocyte	10.13	11.51	-2.61	0.002
Solute carrier family 22 (organic anion transporter), member 8	8.80	9.93	-2.18	0.045
Solute carrier family 7 (cationic amino acid transporter, y+ system), member 1	9.44	10.77	-2.51	0.012
Solute carrier organic anion transporter family, member 1A2	5.37	8.49	-8.75	0.001
Sp1 transcription factor	6.98	5.78	2.29	0.041
Splicing factor, arginine/serine-rich 11	6.66	7.81	-2.22	0.003
Splicing factor, arginine/serine-rich 6	8.47	9.79	-2.50	0.010
Succinyl-CoA:alpha-ketoacid coenzyme A transferase	5.99	7.16	-2.25	0.015
T-cell receptor alpha chain mRNA C-region, 3′ end of cds	7.48	5.01	5.55	0.045
T, brachyury homolog (mouse)	8.63	7.55	2.12	0.023
Thioredoxin	6.33	8.01	-3.19	0.050
Transforming growth factor, beta 2	9.13	10.55	-2.66	0.053
Transmembrane protein 59	6.86	8.64	-3.42	0.004
TrkB protein	9.15	10.21	-2.09	0.002
Tropomodulin 3 (ubiquitous)	5.29	8.59	-9.86	0.019	8.94	6.92	4.08	0.006
Tubulin alpha	8.36	9.46	-2.14	0.011
Tyrosinase-related protein 1					8.96	4.75	18.57	0.007
v-myc myelocytomatosis viral oncogene homolog (avian)	10.04	8.40	3.11	0.022	9.95	8.42	2.89	0.032
Vascular cell adhesion molecule	6.07	7.50	-2.69	0.043
Vascular smooth muscle alpha-actin					12.57	8.38	18.26	0.021
Vitamin D₃ 25-hydroxylase	9.03	10.40	-2.59	0.009
Zinc finger protein 265	8.31	10.15	-3.58	0.028

Data are sorted alphabetically. Blank entries represent conditions in which there was no statistically significant change.

Table 2 shows selected gene changes in suspension culture (RPMI-CPN vs. ENDO-SFM) with magnitude greater than 2. The largest change after 7 days was in the gene for the cobalamin-binding protein haptocorrin, a vitamin B₁₂ carrier (15). Factors with significant changes having substantial therapeutic potential include TGF-β2, tissue inhibitor of metalloproteinases 2 (TIMP-2), PEDF, FGF-2, and galectin-1, among others. After 7 days in culture, these factors show variable expression, but by 14 days, they were all upregulated by greater than twofold. A number of extracellular matrix constituents also showed increased expression in ENDO-SFM, including nearly a 10-fold increase in the expression of collagen type III after 14 days and a 4.6-fold increase in fibronectin. Other notable changes after 2 weeks included a 15-fold increase in the expression of tyrosinase-related protein 1, a protein localized in the epidermis associated with melanocytes (25), a 9.7-fold increase in tropomodulin 3, a regulator of cell migration (9), and a 25.7-fold increase in ATP1A2, an Na⁺/K⁺ transporter (14).

Epithelial cells formed into capsuloids exhibited a differential expression profile that showed many significant changes after 7 days, but very few after 2 weeks. In this comparison, both culture systems included equivalent biomass in RPMI-CPN medium, with Matrigel-HC added to form fluid-filled shells. As shown in Table 3, the most abundant changes occurred after 7 days in culture, as many genes were downregulated by up to nearly 14-fold (amylase). Other genes of note with decreased expression (less than -2-fold, p < 0.05) included various components of the complement system (-7.2 to −10.2-fold), aquaporin 1 (2.7-fold), optineurin (6-fold), solute carriers (2.5 to 8.8-fold), tropomodulin 3 (9.9-fold), and others. Genes upregulated after 7 days (more than 2-fold, p < 0.05) included adrenomedullin (2.4), cytochrome p450 1A1 (3.0), fatty acid binding protein 5 (2.1), and several others. However, after 14 days, the system appeared to reach equilibrium, with only 21 statistically significant changes greater than 2-fold or less than -2-fold. These changes included upregulation of several cytoskeletal elements and extracellular matrix proteins including vascular smooth muscle α-actin (18.3), matrix GLA protein (3.8), and collagen type III (10.5), and decreased expression of fatty acid synthase (-2.1-fold) and low-density lipoprotein (LDL) receptor (-2.1-fold).

Discussion

Cultured choroid plexus epithelium is a malleable source of therapeutic proteins that has enormous potential in regenerative medicine. Its transcriptome is extremely diverse, encoding for secreted factors that are active across multiple pathways at biologically relevant levels of expression. In the current study, modulation of the cellular output was accomplished through manipulation of the local microenvironment, as CP epithelium formed into fluid-filled capsuloids expressed markedly higher levels of VEGF than CP clusters in suspension. Similarly, exposure to ENDO-SFM tissue culture medium enhanced the potency compared to the RPMI-CPN controls. A comparison of ultrastructure revealed that the change in expression was accompanied by a robust increase in the density of secretory vesicles, while other morphological features of healthy CP such as tight junctions were maintained. In capsuloid form, CP epithelium apparently reversed polarity, with cilia and microvilli oriented inward in a fluid-filled space, evident in real time occasionally entrapping motile CP clusters within. Presumably, this reversal of orientation was induced by an increased affinity of the basal epithelium for the extracellular matrix (ECM) proteins contained in the surrounding Matrigel.

The formation of capsuloids was accompanied early by a number of significant changes in the expression of genes shown in Table 3, which was dramatically reduced after 2 weeks. The ability of the CP epithelium to transport fluid through aquaporin channels (26), in conjunction with its sensitivity to ECM substrates, may combine to serve as the basis for the mechanism of capsuloid formation. As individual cells begin to bind to the 3D matrix, the movement of fluid through the epithelium into the hollow core could create radial forces that drive the growth of the capsuloid until equilibrium is reached, dictated by the number of cells present in the original cluster. The downregulation of aquaporin-1 in capsuloids after 7 days could reveal a shift in the equilibrium of fluid transport established in CP clusters initially cultured in RPMI-CPN.

The natural polarity of CP epithelium at the blood– cerebrospinal fluid interface serves a number of biological purposes, one of which is the regulation of growth factor composition at both sides of the interface. This is evident in the differential composition of factors that are secreted basally into the blood with respect to the composition in the cerebrospinal fluid regulated by apical secretions (19), as well as the localization of Na⁺,K⁺-ATPase to the apical membrane and GLUT-1, the facilitative glucose transporter, to the basolateral membrane (27). Certainly the upregulation of the Na⁺,K⁺-ATPase ATP1A2 by more than 25-fold after 2 weeks in the capsuloids in conjunction with increased protein secretion is indicative that this in vitro culture system has achieved biological regulation at the level of cellular trafficking.

Aside from the intuitively logical attraction of using CP as a secretory vehicle for multiple proteins effective in a number of human diseases, it is also worth considering CP as a tool for engineering 3D tissue constructs. 3D tissues require an extremely complex and difficult to achieve mixture of cells, vasculature, and ECM components to enable precise integration and proper physiological function. Together with previous studies (21), the data shown here support the CP as an active mediator of tissue assembly, with both autocrine and paracrine activity. The epithelium secretes a myriad of proteins with therapeutic potential as well as matrix and adhesion factors, and contains very responsive cytoskeletal components potentially capable of precise manipulation of cellular and extracellular niches. For instance, cultured CP expresses high levels of intercellular adhesion molecules (ICAMs), collagens, and integrins that could be used to control and modulate the extracellular niche of forming vascular networks, including secondary sources of exogenous cells such as endothelium. Likewise, the high abundance of matrix remodeling factors such as TIMP-2 and connective tissue growth factor could promote multicellular tissue assembly through modulation of ECM composition as well as migration and chemotaxis of fibroblasts and endothelial cells. With the high and controllable levels of VEGF secretion, as well as various other tissue-tropic factors, one could envision using CP as a critical component of 3D tissue construction for in vivo grafting or in vitro diagnostic and testing purposes.

The appeal of using the choroid plexus as a platform for cellular therapy has been extended to its role as a protein factory in vitro, collecting and purifying the supernatant for use as an acellular therapeutic (22). Developing a manufacturing process for such a product depends on a thorough identification and characterization of parameters that contribute to its composition, as well as the health and functionality of the cells themselves. In this manner, longitudinal process control can be achieved, generating acceptable ranges of culture variables that are used to manufacture the product, while also providing a means of manipulating the drug to potentially have crossover efficacy into alternative disease indications. The data presented here demonstrate this potential, as modulation of the transcriptome and secreted factor doses was possible through alteration of media composition as well as immobilized 3D ECM culture. A continued effort into deciphering the processes and factors involved in the CP as well as their mechanism of action may provide critical insight into regenerative mechanisms that drive tissue repair and neuroprotection.

Footnotes

Acknowledgments

C.T., B.B., M.G., K.B., and S.H. all have founder's shares in CytoSolv.

References

Borlongan

C. V.

; Skinner

S. J.

; Geaney

; Vasconcellos

A. V.

; Elliott

R. B.

; Emerich

D. F.

CNS grafts of rat choroid plexus protect against cerebral ischemia in adult rats. Neuroreport 15(10): 1543–1547; 2004.

Borlongan

C. V.

; Skinner

S. J.

; Geaney

; Vasconcellos

A. V.

; Elliott

R. B.

; Emerich

D. F.

Intracerebral transplantation of porcine choroid plexus provides structural and functional neuroprotection in a rodent model of stroke. Stroke 35(9): 2206–2210; 2004.

Borlongan

C. V.

; Skinner

S. J.

; Geaney

; Vasconcellos

A. V.

; Elliott

R. B.

; Emerich

D. F.

Neuroprotection by encapsulated choroid plexus in a rodent model of Huntington's disease. Neuroreport 15(16): 2521–2525; 2004.

Borlongan

C. V.

; Thanos

C. G.

; Skinner

S. J.

; Geaney

; Emerich

D. F.

Transplants of encapsulated rat choroid plexus cells exert neuroprotection in a rodent model of Huntington's disease. Cell Transplant. 16(10): 987–992; 2008.

Chodobski

; Szmydynger-Chodobska

Choroid plexus: Target for polypeptides and site of their synthesis. Microsc. Res. Techniq. 52: 65–82; 2001.

Emerich

D. F.

; Mooney

D. J.

; Storrie

; Rangasamy

S. B.

; Kordower

J. H.

Injectable hydrogels providing sustained delivery of vascular endothelial growth factor are neuroprotective in a rat model of Huntington's disease. Neurotox. Res. 17: 66–74; 2010.

Emerich

D. F.

; Thanos

C. G.

; Goddard

M. B.

; Skinner

S. J.

; Geaney

; Bell

W. J.

; Bintz

B. E.

; Schneider

P. A.

; Chu

; Babu

R. S.

; Borlongan

C. V.

; Boekelheide

; Hall

; Bryant

; Kordower

J. H.

Extensive neuroprotection by choroid plexus transplants in excitotoxin lesioned monkeys. Neurobiol. Dis. 23(2): 471–480; 2006.

Ericsson

; Liu

; Hart

R. P.

; Sawchenko

P. E.

Type 1 interleukin-1 receptor in the rat brain: Distribution, regulation, and relationship to sites of IL-1-induced cellular activation. J. Comp. Neurol. 361(4): 681–698; 1995.

Fischer

R. S.

; Yarmola

E. G.

; Weber

K. L.

; Speicher

K. D.

; Speicher

D. W.

; Bubb

M. R.

; Fowler

V. M.

Tropomodulin 3 binds to actin monomers. J. Biol. Chem. 281(47): 36454–36465; 2006.

10.

Gard

A. L.

; Gavin

; Solodushko

; Pennica

Cardiotrophin-1 in choroid plexus and the cerebrospinal fluid circulatory system. Neuroscience 127(1): 43–52; 2004.

11.

Greenwood

; Swetloff

; Wade

A. M.

; Terasaki

; Ferretti

Fgf2 is expressed in human and murine embryonic choroid plexus and affects choroid plexus epithelial cell behaviour. Cerebrospinal Fluid Res. 5: 20; 2008.

12.

Huang

; Ketova

; Fleming

J. T.

; Wang

; Dey

S. K.

; Litingtung

; Chiang

Sonic hedgehog signaling regulates a novel epithelial progenitor domain of the hindbrain choroid plexus. Development 136(15): 2535–2543; 2009.

13.

Mashayekhi

; Azari

; Moghadam

L. M.

; Yazdankhah

; Naji

; Salehi

Changes in cerebrospinal fluid nerve growth factor levels during chick embryonic development. J. Clin. Neurosci. 16(10): 1334–1337; 2009.

14.

McLean

W. J.

; Smith

K. A.

; Glowatzki

; Pyott

S. J.

Distribution of the Na,K-ATPase alpha subunit in the rat spiral ganglion and organ of corti. J. Assoc. Res. Otolaryngol. 10(1): 37–49; 2009.

15.

Morkbak

A. L.

; Poulsen

S. S.

; Nexo

Haptocorrin in humans. Clin. Chem. Lab. Med. 45(12): 1751–1759; 2007.

16.

Nadeau

; Rivest

Effects of circulating tumor necrosis factor on the neuronal activity and expression of the genes encoding the tumor necrosis factor receptors (p55 and p75) in the rat brain: A view from the blood-brain barrier. Neuroscience 93(4): 1449–1464; 1999.

17.

Salehi

; Mashayekhi

; Naji

; Pandamooz

Insulin-like growth factor-1 and insulin-like growth factor binding proteins in cerebrospinal fluid during the development of mouse embryos. J. Clin. Neurosci. 16(7): 950–953; 2009.

18.

Sharma

H. S.

; Tang

Z. H.

; Gho

B. C.

; Verdouw

P. D.

Nucleotide sequence and expression of the porcine vascular endothelial growth factor. Biochim. Biophys. Acta 1260(2): 235–238; 1995.

19.

Strazielle

; Ghersi-Egea

J. F.

Demonstration of a coupled metabolism-efflux process at the choroid plexus as a mechanism of brain protection toward xenobiotics. J. Neurosci. 19(15): 6275–6289; 1999.

20.

Szmydynger-Chodobska

; Strazielle

; Zink

B. J.

; Ghersi-Egea

J. F.

; Chodobski

The role of the choroid plexus in neutrophil invasion after traumatic brain injury. J. Cereb. Blood Flow Metab. 29(9): 1503–1516; 2009.

21.

Thanos

C. G.

; Bintz

B. E.

; Emerich

D. F.

Microencapsulated choroid plexus epithelial cell transplants for repair of the brain. Adv. Exp. Med. Biol. 670: 80–91; 2010.

22.

Thanos

C. G.

; Emerich

D. F.

; Bintz

B. E.

; Goddard

M. B.

; Mills

; Jensen

; Lombardi

; Hall

; Boekelheide

Secreted products from the porcine choroid plexus accelerate the healing of cutaneous wounds. Cell Transplant. 18(12): 1395–1409; 2009.

23.

Thanos

C. G.

; Schneider

P. A.

; Bintz

B. E.

; Jensen

; Bryant

; Bell

W. J.

; Hudak

; Emerich

D. F.

The in vitro expression and secretion of vascular endothelial growth factor from free and alginate-polyornithine encapsulated choroid plexus epithelium. Tissue Eng. 13(4): 747–756; 2007.

24.

Thouvenot

; Lafon-Cazal

; Demettre

; Jouin

; Bockaert

; Marin

The proteomic analysis of mouse choroid plexus secretome reveals a high protein secretion capacity of choroidal epithelial cells. Proteomics 6: 5941–5952; 2006.

25.

Tobin

; Quinn

A. G.

; Ito

; Thody

A. J.

The presence of tyrosinase and related proteins in human epidermis and their relationship to melanin type. Pigment Cell Res. 7(4): 204–209; 1994.

26.

Tradtrantip

; Tajima

; Li

; Verkman

S. Aquaporin water channels in transepithelial fluid transport. J. Med. Invest. 56: 179–184; 2009.

27.

Villalobos

A. R.

; Parmelee

J. T.

; Pritchard

J. B.

Functional characterization of choroid plexus epithelial cells in primary culture. J. Pharmacol. Exp. Ther. 282(2): 1109–1116; 1997.

Functional Modulation of Choroid Plexus Epithelial Clusters in Vitro for Tissue Repair Applications

Abstract

Keywords

Introduction

Materials and Methods

Choroid Plexus Isolation

Tissue Culture Media Screen and Formation of 3D Hollow Shells

Detection of VEGF

Isolation of RNA, Quantitative Real-Time PCR (QRT-PCR), and Microarray Analysis

Results

Morphology

VEGF Expression

Gene Microarray

Discussion

Footnotes

Acknowledgments

References