Adaptation of sensory neurons to hyalectin and decorin proteoglycans

Abstract

Proteoglycans are abundantly expressed in the pathways of developing and regenerating neurons, yet the responses of neurons to specific proteoglycans are not well characterized. We have shown previously that one chondroitin sulfate proteoglycan (CSPG), aggrecan, is potently inhibitory to sensory axon extension in short-term assays and that over time, embryonic neurons adapt to aggrecan-mediated inhibition through the transcriptional upregulation of integrin expression (Condic et al., 1999). Here, we have compared the response of embryonic sensory neurons to structurally distinct CSPGs that belong to either the hyalectin (or lectican) family of large, aggregating proteoglycans or the decorin (or small leucine-rich proteoglycan) family of smaller proteoglycans. Both of these structurally diverse proteoglycan families are expressed in developing embryos and inhibit outgrowth of embryonic sensory neurons in short-term cultures. These results document a previously uncharacterized inhibitory function for the decorin-family proteoglycan biglycan. Interestingly, embryonic neurons adapt to these diverse proteoglycans over time. Adaptation is associated with upregulation of select integrin alpha subunits in a proteoglycan-specific manner. Overexpression of specific integrin alpha subunits improves neuronal regeneration on some but not all decorin-family CSPGs, suggesting that neurons adapt to inhibition mediated by closely related proteoglycans using distinct mechanisms. Our findings indicate that CSPGs with diverse core proteins and distinct numbers of chondroitin sulfate substitution sites mediate a similar response in sensory neurons, suggesting that increased integrin expression may be an effective means of promoting axonal regeneration in the presence of diverse inhibitory proteoglycan species in vivo.

Figure 1.

Structurally diverse hyalectin and decorin CSPGs are present during sensory axon outgrowth. A, Schematic diagram illustrates the range of core protein sizes and CS substitution patterns of hyalectin and decorin proteoglycans [adapted from Condic and Lemons (2002)]. Hyalectin proteoglycans are large proteoglycans with numerous CS substitution sites. In contrast, decorin proteoglycans are smaller and have fewer CS substitution sites. The two members of the decorin family expressed in the nervous system bear only one (decorin) or two (biglycan) CS side chains (Fransson et al., 2000). COOH, C terminus; NH2, N terminus. B-F, Adjacent lumbar cross sections of chick embryos at stage 23 (E3.5-E4; B, C, F) and stage 25 (E4.5; D, E) processed for immunohistochemistry. B, Anti-neurofilament-associated protein staining with the 3A10 antibody reveals the presence of sensory and motor axons. C, An anti-versican antibody (MY174) reveals staining in the skin and surrounding the notochord. Diffuse versican staining is also seen throughout the mesenchyme and in the dorsal spinal cord. D, Low-magnification view showing decorin immunoreactivity (labeled by antibody CB-1) in the skin, dermomyotome, and surrounding the notochord. E, Higher-magnification view showing decorin expression in the skin. F, Anti-aggrecan antibody S103L shows aggrecan is present in and surrounding the notochord.

Figure 2.

Monoclonal aggrecan antibodies reveal distinct patterns of immunoreactivity in the notochord and nervous system. A, Cross section of a stage 24 chick embryo (wild type) stained with monoclonal 2B12 shows immunoreactivity in the peripheral nerve, ventral spinal cord, oval bundle of His, and notochord. The oval bundle of His (top box) and ventral root (bottom box) are enlarged in the right panels. B, Western blot analysis indicates monoclonal 2B12 antibody recognizes a single band of appropriate molecular size for the aggrecan core protein (Ag) in preparations of brain membranes (Brain Memb) treated with chondroitinase ABC. L1 is also highly expressed in brain membrane preparations, but 2B12 does not recognize a band the same size as L1. The 2B12 antibody also recognizes a similar size band in chondroitinase ABC-treated urea extracts from the brain (Br). The 2B12 antibody identifies as lightly higher molecular weight smear in chondroitinase ABC-treated urea extracts from cartilage (Ct), consistent with the known keratan sulfate chains present on cartilage-derived aggrecan. C, Aggrecan message is detected by RT-PCR in samples of E4 notochord (E4 NC), spinal cord (E4 SC), and DRG (E4 DRG) as well as E11 brain (E11 Brn). D, E, Corresponding phase and fluorescent images of a cultured sensory growth cone that is lightly stained with 2B12 antibody. F-H, Wild-type (Wt) and nanomelic (Nm) embryos at stage 22 are immunostained with anti-aggrecan monoclonal antibodies 1G12 and 2B12. F, The 1G12 antibody reveals notochord-associated aggrecan staining (arrow) in wild-type animals, similar to the pattern seen with the monoclonal S103L antibody (see Fig. 1 F). G, In nanomelic animals, 1G12 immunostaining is not expressed, as is also the case for the S103L antibody (Domowicz et al., 2003). The arrow points to 1G12-negative notochord. H, In an adjacent cross section to that shown in G, the 2B12 antigen is detected in the ventral spinal cord, oval bundle of His (arrowheads), and ventral roots of nanomelic embryos but not in notochord (arrow). Agg, Aggrecan.

Figure 3.

CSPGs with diverse structure and unique core proteins inhibit the attachment of embryonic sensory neurons to laminin. Substrata were coated with the indicated concentrations of proteoglycan, followed by 20 μg/ml laminin. At all proteoglycan concentrations, laminin binding was constant (see Materials and Methods). Attachment of neurons after 3 h in culture is expressed as a percentage of control attachment to laminin. Inhibition of attachment to substrata containing versican (A), biglycan (B), and decorin (C) were similar to inhibition by aggrecan (black lines in all panels). Similar results were seen in at least three experiments, with data from a typical experiment shown.

Figure 4.

Extension of neurites on substrata containing laminin and proteoglycan is initially inhibited but recovers by 20-24 h in culture. Phase micrographs of neurons cultured on laminin alone or on substrata containing both laminin and proteoglycan. LM, Laminin; AGG, aggrecan/laminin; BIG, biglycan/laminin; DEC, decorin/laminin. Representative phase images are shown at 3 and 24 h in culture.

Figure 5.

Rate of growth cone migration is reduced in the presence of proteoglycans at 3 h but recovers by 20-24 h in culture. Neurons were cultured as in Figure 3, and the rates of axon extension were determined at 3 and 20-24 h (axis labeled 24 h) in vitro using video time-lapse microscopy. The rate of axon extension was suppressed on proteoglycan (PG)-containing substrata relative to laminin alone at 3 h but recovered completely by 20-24 h. ^*Significantly different from laminin controls (p < 0.001; t test).

Figure 6.

Laminin receptors containing integrins α1, α3, and α6 are upregulated by inhibitory proteoglycans. Neurons were cultured 20-24 h on laminin (LM) or laminin/proteoglycan substrata as in Figure 3. The proteoglycans tested were aggrecan (AG), biglycan (BG), and decorin (DC). Surface proteins were biotinylated and immunoprecipitated with antibodies specific for integrin α subunits. Biotinylated protein was detected on Western blots using HRP-conjugated avidin and a chemoluminescent detection reagent. Band intensity was compared with control (laminin) conditions, and ratios from at least three independent experiments are given below the gels. All three laminin receptors are upregulated at the cell surface in the presence of proteoglycans, with the exception of integrin α6 on biglycan. ^*Statistically different from a ratio of 1 (p < 0.01; t test).

Figure 7.

Increased integrin expression is sufficient for neurite extension in the presence of inhibitory proteoglycans. [The higher rates of neurite extension in these experiments compared with those of Fig. 5 are likely to reflect the longer period of time these cells were maintained in suspension before testing in short-term assays (see Materials and Methods).] A, Neurons transfected with control (pMESϕ) or human integrin α6-containing (pMESα6) constructs are readily detected by GFP expression in culture. Integrins containing human α6 (right) and α3 (data not shown) subunits were detected at the cell surface in cultures stained live with human-specific integrin antibodies. B, At 3 h in culture on substrata containing either biglycan or decorin in combination with laminin, neurons expressing the control construct extended growth cones at very slow speeds, whereas cells expressing human α3 showed rates of extension that are identical to control cells on laminin alone. In contrast, overexpression of human α6 improved neurite extension on decorin substrata, where this receptor is endogenously upregulated (Fig. 6), but not on biglycan, where α6 levels are not normally upregulated (Fig. 6). Similar results have been published for transgenic integrin expression on aggrecan/laminin substrata (Condic et al., 1999; Condic, 2001). ^*Statistically different from control constructs (p < 0.002; t test). Hu-α6, Human α6; Hu-α3, human α3.