Extracellular matrix-associated gene expression in adult sensory neuron populations cultured on a laminin substrate

Abstract

Background: In our previous investigations of the role of the extracellular matrix (ECM) in promoting neurite growth we have observed that a permissive laminin (LN) substrate stimulates differential growth responses in subpopulations of mature dorsal root ganglion (DRG) neurons. DRG neurons expressing Trk and p75 receptors grow neurites on a LN substrate in the absence of neurotrophins, while isolectin B4-binding neurons (IB4+) do not display significant growth under the same conditions. We set out to determine whether there was an expression signature of the LN-induced neurite growth phenotype. Using a lectin binding protocol IB4+ neurons were isolated from dissociated DRG neurons, creating two groups - IB4+ and IB4-. A small-scale microarray approach was employed to screen the expression of a panel of ECM-associated genes following dissociation (t=0) and after 24 hr culture on LN (t=24LN). This was followed by qRT-PCR and immunocytochemistry of selected genes.

Results: The microarray screen showed that 36 of the 144 genes on the arrays were consistently expressed by the neurons. The array analyses showed that six genes had lower expression in the IB4+ neurons compared to the IB4- cells at t=0 (CTSH, Icam1, Itgβ1, Lamb1, Plat, Spp1), and one gene was expressed at higher levels in the IB4+ cells (Plaur). qRT-PCR was carried out as an independent assessment of the array results. There were discrepancies between the two methods, with qRT-PCR confirming the differences in Lamb1, Plat and Plaur, and showing decreased expression of AdamTs1, FN, and Icam in the IB4+ cells at t=0. After 24 hr culture on LN, there were no significant differences detected by qRT-PCR between the IB4+ and IB4- cells. However, both groups showed upregulation of Itgβ1 and Plaur after 24 hr on LN, the IB4+ group also had increased Plat, and the IB4- cells showed decreased Lamb1, Icam1 and AdamTs1. Further, the array screen also detected a number of genes (not subjected to qRT-PCR) expressed similarly by both populations in relatively high levels but not detectably influenced by time in culture (Bsg, Cst3, Ctsb, Ctsd, Ctsl, Mmp14, Mmp19, Sparc. We carried out immunohistochemistry to confirm expression of proteins encoded by a number of these genes.

Conclusions: Our results show that 1B4+ and IB4- neurons differ in the expression of several genes that are associated with responsiveness to the ECM prior to culturing (AdamTs1, FN, Icam1, Lamb1, Plat, Plaur). The data suggest that the genes expressed at higher levels in the IB4- neurons could contribute to the initial growth response of these cells in a permissive environment and could also represent a common injury response that subsequently promotes axon regeneration. The differential expression of several extracellular matrix molecules (FN, Lamb1, Icam) may suggest that the IB4- neurons are capable of maintaining /secreting their local extracellular environment which could aid in the regenerative process. Overall, these data provide new information on potential targets that could be manipulated to enhance axonal regeneration in the mature nervous system.

Figure 1

IB4⁺ DRG neurons do not respond to a LN substrate with neurite growth. DRG neurons were separated using isolectin B4 (IB4) coated magnetic beads (as described in the Methods) to produce two populations of neurons - IB4 selected (IB4⁺) and IB4^-. Each cell group was plated on PL (A, C) and LN (B, D) coated culture dishes for 24 hr at 37°C. Cells were immunostained for βIII tubulin (green), integrin β1 (red) expression and IB4 lectin binding (blue) labeling. Note that while the larger IB4- neurons show extensive neurite growth, the IB4⁺ cells have no detectable growth. Scale bar-20 μm.

Figure 2

Expression of selected genes detected as being differentially expressed between IB4⁺ and IB4^- neurons by microarray analyses. The graphs present the mean normalized spot density for each condition + SEM. Statistical significance was determined with ANOVA and Tukey post-hoc testing; * - p<0.05, + p<0.10.

Figure 3

qRT-PCR analysis of the 9 selected genes. The graphs present gene expression relative (+SEM) to the IB4- t=0 condition (1); relative expression was calculated as outlined in the Methods. Statistical significance was determined with ANOVA and Tukey post-hoc testing; * - p<0.05, + p<0.10 using ANOVA; τ−p<0.05student t-test.

Figure 4

Plaur protein expression in DRG neurons. Dissociated DRG neurons were immunostained for Plaur (green), neuron-specific βIII tubulin (red) and IB4 lectin labeling (blue) at t=0 (A-D) and t=24LN (E-H). While both IB4⁺ and IB4^- neurons express Plaur, densitometric analyses (see Supp. Fig 2) showed that the expression was higher in the IB4⁺ cells. Scale bar-A-D, 50 μm; E-F, 20 μm.

Figure 5

Integrin β1 protein expression in DRG neurons. Dissociated DRG neurons were immunostained for integrin β1 (red), neuron-specific βIII tubulin (green) and IB4 lectin labeling (blue) at t=0 (A-D) and t=24LN (E-H). Both IB4⁺ and IB4^- neurons express Integrin β1 and densitometry indicated that the IB4⁺ cells had somewhat lower levels of expression. Scale bar-20 μm.

Figure 6

DRG neurons express LN. Dissociated DRG neurons were immunostained for laminin (red), neuron-specific βIII tubulin (green) and IB4 lectin labeling (blue) at t=0 (A-D) and t=24 hr after plating on poly-lysine (E-H). Neurons cultured on poly-lysine were used to detect the LN expression, as it was not possible to detect the cellular LN signal above the stained LN substrate. ICC quantitation showed that IB4⁺ cells had lower levels of LN expression at t=0 compared to the IB4^- cells. Scale bar-20 μm.

Figure 7

FN protein expression in DRG neurons. Dissociated DRG neurons were immunostained for fibronectin (green), neuron-specific βIII tubulin (red) and IB4 lectin labeling (blue) at t=0 (A-D) and t=24 hr after plating on LN (E-H). Both populations of cells express FN with no obvious difference detected by densitometry at t=0, although after 24 hr on LN, an increase in expression was observed for both populations. Scale bar-30 μm.

Figure 8

Expression of OPN, CTSH, AdamTs1 and Icam. ICC was used to detect the expression of osteopontin (A, C, green), AdamTs1 (E, G, green), CTSH (I, L, green), Icam (M, O, green); merged images show βIII tubulin (red, B, D, F, H, J, L, N, P) and IB4 binding neurons (blue) at t=0 and t=24LN. Densitometric analyses are presented in Supp Fig 2. Scale bar-A-L, 30 μm; M-P, 20 μm.

Figure 9

Immunohistochemisty of selected proteins in adult rat DRG cryosections. IHC and confocal microscopy was used to investigate expression of proteins encoded by a number of genes noted to be expressed at robust levels in the dissociated neurons, including those that have not been previously described in DRG neurons. Panels A – AdamTs1; B - Emmprin (Basigin); C – CTGF; D – CTSH; E - Fibronectin; F – Icam; G – Integrin β1; H – Laminin; I – MT-MMP1 (MMP19); J – Osteopontin (Spp1); K – RT-1 (RT-AWs); L – Plaur; M – SPARC; N – tPA (Plat); O – Peripherin; P – p75 neurotrophin receptor. Scale bar – 50 μm.