The Agrin Hypothesis proposes that the glycoprotein agrin plays an important role in the clustering of acetylcholine receptors (AChR) at the post-synaptic membrane of a neuromuscular junction (NMJ). In 1990, structural neurobiologist Uel McMahan of Stanford University was the first to suggest that agrin is synthesized and subsequently released into the basal lamina by the vertebrate motor neuron where it then interacts with its receptor on the membrane of the muscle fiber (McMahan 1990). The interaction of agrin and its receptor was thought to induce the aggregation of AChR and other components of the post-synaptic apparatus necessary for functional signal transduction from the motor neuron to the muscle cell. The hypothesis distinguishes between the role of agrin in the developing muscle, which was suggested to coordinate preliminary organization of the AChR, and its role in adult muscle, which was thought to maintain existing synaptic organization and direct formation of new synaptic structures for regenerating muscles. The hypothesis serves as a template for experimental clarification of the specific mechanisms by which the NMJ is organized. These experiments have confirmed that agrin is a crucial component of clustering of AChR, but later evidence showed that agrin is not responsible for initial formation of clustering, but is important in long-term maintenance of the NMJ by preventing the destabilization of AChR aggregates.
Agrin: structure and localization
Agrin (from the Greek word ageirein “to assemble”) is a ~200 kDa heparansulfate glycoprotein best known to be released by the axon terminals of motor neurons into the basal lamina of the NMJ, but has been found to be made in other neuronal and non-neuronal cells as well. The agrin gene contains three known sites of alternative splicing. One site of alternative splicing occurs in the N-terminus which results in two forms: a secreted form containing the N-terminal agrin (NtA) domain which binds to laminins in the extracellular matrix and is expressed by motor neurons and muscle cells, and a transmembrane form lacking the NtA that is mostly expressed by neurons in the CNS (Scotton et al. 2006).
The C-terminal domain of agrin contains the membrane binding and AChR clustering activity. This is also the location of the other two alternatively spliced sites. The first of these is a four amino acid insert in the second laminin globular domain (LG2) known as the A or y form (in chick and rodents respectively). It is expressed by neurons and glia in the CNS thus its function is not specifically addressed by the agrin hypothesis. The second site, known as the B or z form, occurs in the third laminin globular domain (LG3) where 8, 11, or 19 amino acids may be inserted to confer differing biological activities (Scotton et al. 2006). The 8 amino acid insert in motor neuron-synthesized agrin confers the highest level of AChR aggregation, approximately 1000 fold higher than the negligible AChR clustering activity of agrin synthesized by muscle fibers. Agrin binds to the muscle membrane at low concentration and clusters of approximately 200 AChR associate with one agrin molecule. This suggests that other molecules mediate agrin’s action to promote AChR aggregation.
Schematic representation of Agrin. Picture Source: Scotton et al. 2006
NtA: amino-terminal agrin domain;
TM: transmembrane segment;
FS: follistatin-like domain;
LE: laminin EGF-like domain;
S/T: serine/threonine-rich region;
SEA: sperm protein, enterokinase and agrin domain;
EG: epidermal growth factor domain;
LG: laminin globular domain
Early evidence for the agrin hypothesis
Many experiments during the mid-1980s to 90s validated the main concepts of the agrin hypothesis. Originally isolated from the basal lamina of the electric ray Torpedo californica (Godfrey et al. 1984; Nitkin et al. 1987), agrin was found to be present in motor neuron cell bodies (Magill-Solc and McMahan 1988) and transported down the axon (Magill-Solc and McMahan 1990). When agrin-containing extracts were applied to myotubes in vitro, clusters of AChR and other synaptic proteins were induced (McMahan and Wallace 1989; Wallace 1989). Additionally, anti-agrin antibodies applied to co-cultures of chick motor neuron-myotubes prevented the aggregation of AChR (Reist, et al. 1992), confirming that agrin was necessary for this organization of synaptic molecules.
In mice lacking agrin, experimenters saw a decrease in the number, size, and density of AChR clusters, substantiating the hypothesis that agrin was involved in post-synaptic AChR aggregation (Gautam et al. 1996). AChR aggregates that were present were often dispersed in areas not associated with the nerve terminal. The enzyme acetylcholinesterase also was not present on muscle from mice lacking agrin. However, the post-synaptic elements erbB3 and rapsyn and lamininβ2 on the extracellular surface were localized normally, which failed to implicate a role for agrin in their organization. Mice mutant for agrin died in the late embryonic stage and exhibited no physical movement, indicating the importance of agrin for sustained organismal development. Residual AChR clustering in the mutant mice left open the possibility that some other molecular components are also capable of inducing post-synaptic protein clustering.
Important molecular players in NMJ synapse organization
The agrin hypothesis suggests the existence of an agrin receptor through which agrin exerts its clustering activity. Localized at the post-synaptic membrane of the NMJ, the receptor tyrosine kinase MuSK (muscle-specific kinase) is thought to be the receptor for agrin because of its 1) upregulation during myotube differentiation 2) maintained expression in mature muscle solely at the NMJ 3) ability to stimulate agrin binding. The application of agrin to myotubes from wild-type mice induced tyrosine phosphorylation of MuSK within 1 min (Glass et al. 1996). Most importantly, mice lacking MuSK showed impaired AChR clustering similar to the agrin mutant phenotype (DeChiara et al. 1996). Furthermore, muscle deficient in MuSK failed to respond to agrin in culture. However, binding assays have failed to show direct interactions between agrin and MuSK. Moreover, while agrin activates MuSK in muscle, it fails to activate MuSK in fibroblasts (Glass et al. 1996). It has therefore been hypothesized that agrin may indirectly activate MuSK through an associated, muscle-specific, cell-surface molecule named MASC (myotude-associated specificity component) (Glass et al. 1997 and Dimitropoulou and Bixby, 2005).
It is thought that agrin activation of MuSK leads to tyrosine phosphorylation of AChR and subsequent clustering of AChR which has been shown to be dependent on an intracellular protein named rapsyn (receptor associated protein of the synapse). The association of rapsyn and AChR is necessary for proper post-synaptic clustering as AChR in mice lacking rapsyn fail to aggregate and mutant mice die hours after birth (Gautum et al. 1995). While AChR and other post-synaptic proteins have defective organization in rapsyn knock-out mice, the basal lamina components agrin, acetylcholinesterase, and s-laminin were expressed normally.
Postsynaptic clustering of AChR by agrin is also aided by neuregulin, an extracellular ligand that is similarly secreted into the basal lamina of the synaptic cleft by the pre-synaptic terminal. Neuregulin binds to and activates the receptor tyrosine kinases ErbB2 and ErbB3 on the post-synaptic membrane to induce a second messenger signaling cascase that leads to the upregulation of AChR. When added to myotubes in culture Ngo et al. 2004 found that neuregulin significantly potentiated AChR clustering in the presence of agrin versus agrin alone. No clustering was observed with neuregulin alone.
Agrin-induced aggregation of AChRs at the NMJ
Refining the role of agrin
The original agrin hypothesis submits that agrin is the factor mediating motor neuron-induced post-synaptic differentation of the NMJ. Several key experiments performed more than 10 years later now show that agrin is not necessary for the preliminary formation of AChR clusters, but rather, is essential for maintaining their proper organization.
Experiments by Lin et al. 2001 showed that primary post-synaptic differentiation was dependent on MuSK and rapsyn, but not on agrin or, surprisingly, the presence of the motor axon. The agrin hypothesis was based on the belief that the axon terminal induced differentiation (via agrin) in the motor end plate, yet Lin et al. detected AChR clusters at E14.5 that were not innervated by motor neurons. Additionally, agrin-/- mutant embryos at E14.5 showed comparable numbers of AChR aggregates. By E18.5, however, mutants showed a significant decrease in size and number of AChR clusters. This indicated that agrin was important for maintaining the aggregation of synaptic components, but not crucial for the initial formation of these clusters. Clusters of acetylcholinesterase and rapsyn followed a temporal pattern similar to that of AChR in agrin-/- mutant embryos. These experiments highlight the importance of investigating all stages in development for a complete understanding of a molecule's role in a structural organization.
The neurotransmitter acetylcholine (ACh) was identified as responsible for dispersal of AChR on the post-synaptic membrane. Mice lacking choline acetyltransferase (the enzyme that synthesizes acetylcholine) chat-/-, display larger AChR clusters than in wild-type animals (Misgeld et al. 2002). This hinted that ACh could be acting to disassemble the aggregates of AChR following neurotransmitter release from the motor neuron. The use of mice mutant for both agrin and chat enabled Misgeld et al. 2005 to determine that agrin is able to stablize the post-synaptic apparatus by counteracting the AChR dispersal effects of ACh. Mice lacking both agrin and ACh have normal NMJ synapse differentiation. In fact, the defects seen in agrin-/- mutants were rescued to a large extent by the choline acetyltransferase chat-/- mutant. This suggested that ACh is responsible for destabilizing the AChR clusters. To confirm this, the researchers stimulated CCh (a non-hydrolyzable cholinergic antagonist) release onto myotubes in vitro and saw that AChR aggregates were disassembled. In a subsequent experiment, the incubation of myotubes with agrin prevented this dispersal of AChR aggregates. Finally, they found that clustering of AChR became no longer dependent on agrin if neurotransmission was blocked by bungarotoxin, which binds to AChR and inhibits ACh sisgnaling.
It is not surprising that the original agrin-/- mutant studies suggested agrin has a role in the early development of NMJ synapses. The experiments described above clarify that the abnormal AChR clustering, defective post-synaptic differentation, and still-born phenotype in mice lacking agrin was not a result of complete failure to form AChR clusters and synaptic organization. Instead, agrin deficiency brought destabilization of the post-synaptic apparatus leading to disfunctional NMJs and premature death.
References and additional reading
DeChiara, TM et al., (1996). The receptor tyrosine kinase MuSK is required for neuromuscular junction formation in vivo. Cell 85, 501-512.
Dimitropouloua, A and Bixby JL, (2005). Motor neurite outgrowth is selectively inhibited by cell surface MuSK and agrin. Molec. Cell Neurosci. 28, 292-302.
Gautum, M et al., (1995). Failure of postsynaptic specialization to develop at neuromuscular junctions of rapsyn-deficient mice. Nature 377, 232-236.
Gautum, M et al., (1996). Defective neuromuscular synaptogenesis in agrin-deficient mutant mice. Cell 85, 525-536.
Glass, DJ et al., (1996). Agrin acts via a MuSK receptor complex. Cell 85, 512-523.
Glass, DJ et al., (1997). Yancopoulos, Kinase domain of the muscle-specific receptor tyrosine kinase (MuSK) is sufficient for phosphorylation but not clustering of acetylcholine receptors: required role for the MuSK ectodomain?, PNAS 94, 8848–8853.
Godfrey, EW et al., (1984). Components of Torpedo electric organ and muscle that cause aggregation of acetylcholine receptors on cultured muscle cells. J. Cell. Biol. 99, 615.
Ip, FC et al., (2000). Cloning and characterization of muscle-specifc kinase in chicken. Mol. Cell Neurosci. 16, 661-673.
Jones, G et al., (1996). Substrate-bound agrin induces expression of acetylcholine receptor epsilon-subunit gene in cultured mammalian muscle cells. PNAS 93: 5985-5990.
Kleiman, RJ and LF Reichardt. (1996) Testing the agrin hypothesis. Cell 85, 461-464.
Lin, W et al., (2001). Distinct roles of nerve and muscle in postsynaptic differentiation of the neuromuscular synapse. Nature, 410, 1057-1064.
Magill-Solc, C and UJ McMahan (1988). Motor neurons contain agrin-like molecules. J. Cell Biol. 107, 1825.
Magill-Solc, C and UJ McMahan (1990). Synthesis and transport of agrin-like molecles in motor neurons. J. Exp. Res. 153, 1.
McMahan, UJ (1990). The agrin hypothesis. Cold Spring Harbor Symp. Quant. Biol.
McMahan, UJ and BG Wallace (1989). Molecules in basal lamina that direct the formation of synaptic specializations at neuromuscular junctions. Dev. Neurosci. 11, 227.
Misgeld, T et al., (2002). Roles of neurotransmitter in synapse formation: development of neuromuscular junctions lacking choline acetyltransferase. Neuron 36, 635-648.
Misgeld, T et al., (2005). Agrin promotes synaptic differentiation by counteracting an inhibitory effect of neurotransmitter. PNAS, 102(31): 11088-93.
Nitkin, RM et al., (1987). Identification of agrin, a synaptic organizing protein from Torpedo electric organ. J. Cell. Biol. 105, 2471.
Ngo, ST et al., (2004). Neuregulin potentiates agrin-induced acetylcholine receptor clustering in myotubes. Neuroreport 15(16), 2501-2505.
Reist, NE et al., (1992). Agrin released by motor neurons induces the aggregation of acetylcholine receptors at neuromuscular junctions. Neuron 8, 865-868.
Rimer, M et al., (1998). Neuregulins and erbB receptors at neuromuscular junctions and at agrin-induced post-synaptic-like apparatus in skeletal muscle. Mol. Cell. Neurosci. 9, 254-263.
Scotton, P et al., (2006). Activation of MuSK and binding to dystroglycan is regulated by alternative mRNA splicing of agrin. J Biol Chem. Epub ahead of print.
Wallace, BG (1989). Agrin-induced specializations contain cytoplasmic, membrane, and extracellular matrix-associated components of the post-synaptic apparatus. J. Neurosci. 9, 1294.
Witzemann, V (2006). Development of the neuromuscular junction. Cell Tissue Res. 326, 263-271.
Recent updates to the site: