|Home | About | Journals | Submit | Contact Us | Français|
The main immunogenic region on the α1 subunits of muscle nicotinic acetylcholine receptors provokes half or more of the autoantibodies in myasthenia gravis and its animal model. Many of these autoantibodies depend on the native conformation of the receptor for their ability to bind with high affinity. We mapped this region and explained the conformation-dependence of its epitopes by making chimeras in which sequences of human muscle α1 subunits were replaced in human neuronal α7 subunits or Aplysia acetylcholine binding protein. These chimeras also revealed that the main immunogenic region can play a major role in promoting conformational maturation, and, consequently, assembly of receptor subunits.
Nicotine acetylcholine receptors (AChRs) are formed from five homologous subunits organized like barrel staves around a central cation channel (Albuerque et al. 2009). The primordial AChR is thought to have been homomeric, with five identical subunits, which through gene duplication evolved into a variety of subunits with basically similar structures. Neuronal α7 AChRs are homomeric. There are five ACh binding sites found at interfaces between the α7 subunits. Muscle AChRs are heteromeric. Adult muscle AChRs have the subunit composition and arrangement (α1δ)(α1ε)β1, with two ACh binding sites, one between α1 and δ and the other between α1 and ε. ACh binding proteins (AChBPs) are water soluble homomeric proteins resembling in structure the extracellular domain of α7 AChRs (Brejc et al. 2001; Hansen et al. 2005). These are secreted by mollusc glia to modulate cholinergic transmission, but have been extremely valuable for structural studies because their crystal structures are easily obtained at high resolution, thereby defining the basic structures of the extracellular domains of subunits of AChRs and other receptors in their superfamily.
Myasthenia gravis (MG) is caused by an antibody-mediated autoimmune response to muscle AChRs (Lindstrom 2000; Conti-Fine et al. 2006). Experimental autoimmune MG (EAMG) can be induced by immunizing animals with AChRs from fish electric organs (Lindstrom 2000). Half or more of the autoantibodies to AChRs in MG and EAMG bind to the main immunogenic region (MIR) on α1 subunits (Lindstrom 2000; Tzartos et al. 1998). The MIR is defined by the ability of mAbs to compete for binding to this region (Tzartos and Lindstrom 1980; Tzartos et al. 1982). mAb 35 is a classic example of a mAb to the MIR (Tzartos et al. 1982). mAbs to the MIR can passively transfer EAMG and exhibit the two primary activities which make autoantibodies pathologically significant in MG and EAMG: 1) they fix complement, which causes focal lysis of the postsynaptic membrane resulting in loss of AChRs and disruption of the architecture of the synapse; and 2) they crosslink AChRs, which causes antigenic modulation, an increase in AChR turnover that reduces the number of AChRs (Lindstrom 2000). The MIR is located at the extracellular tip of α1 subunits (Beroukhim and Unwin 1995). Amino acids contributing to the MIR have been mapped to the region α1 66–76 (Tzartos et al. 1998). All AChR subunits have a homologous MIR loop (Albuerque et al. 2009). For example, some mAbs made to muscle AChRs, e.g. mAb 210, react not only with muscle α1 subunits but also with neuronal α3, α5, or β3 subunits (Wang et al. 1998; Kuryatov et al. 2008). For high affinity binding, most mAbs to the MIR depend on the α1 subunit being in its native conformation (Tzartos et al. 1998; Lindstrom 2000).
We made chimeras consisting of the α1 66–76 MIR loop alone or in combination with other α1 subunit extracellular domain sequences replacing the corresponding parts of α7 subunits or AChBPs (Lindstrom et al. 2008). These chimeras enabled our discovery that the antigenic structure of the MIR in the native conformation of α1 subunits depends on both the MIR loop and its interactions with the N-terminal α helix and adjacent regions. Further, we discovered that the α1 MIR greatly promotes expression of chimeric α1/α7 AChRs, overcoming the intrinsic inefficient expression of wild type α7 AChRs in Xenopus oocytes. And we found that the MIR influences AChR function.
The crystal structures of AChBP and α1 subunits indicate that the MIR loop at the extracellular tip of the subunit is paralleled by the N-terminal α helix on one side and the turn between β sheets 5 and 6 on the other (Brejc et al. 2001; Hansen et al. 2005; Dellisanti et al. 2007). This forms a structure angled out from the central axis of the AChR, as would be expected from the orientation of mAb 35 located bound to the MIR by electron crystallography and from the ability of mAb 35 and other mAbs to efficiently crosslink AChRs (Conti-Tronconi et al. 1981; Beroukhim and Unwin 1995).
Chimeras were constructed and characterized as described in Lindstrom et al., 2008. Examples of the data are shown in Figure 1. The chimera α1(66–76)/α7 expressed in Xenopus oocytes produced AChRs containing only the MIR loop. These could be immune precipitated by mAbs to the MIR such as mAb 6 or 210, which can bind both native and denatured α1. But this chimera bound poorly to mAb 198, which binds less well to denatured α1, and virtually not at all to mAb 35, which does not bind at all to denatured α1 on western blots. Increasing the loop region in the α1(60–81)/α7 chimera did not greatly increase binding of mAbs 35 or 198. A chimera with only the N-terminal αhelix of α1 subunits, α1(2–14)/α7 did not form functional AChRs. However, a chimera with both the N-terminal sequence and the MIR loop, α1 (2–14, 60–81)/α7 bound both conformation-dependent and conformation-independent mAbs to the MIR. Only longer sequences, α1(1–32, 60–81)/α7, allowed binding of a MG patient-derived mAb to the MIR (Luo et al. unpublished).
These results imply that the native structure of the MIR depends on both the MIR loop and its interaction with the N-terminal α helix. The N-terminal αhelix from α1 expressed alone in α7 presumably could not form mature AChRs because the α1 terminal region could not interact with the α7 MIR loop sufficiently to stabilize a conformation which would permit subunits to conformationally mature, assemble, and form functional AChRs. These results explain the conformation-dependence of binding of mAbs and autoantibodies to the MIR.
A similar α1(1–32, 60–81)/AChBP chimera also bound all rat mAbs to the MIR and both EAMG and MG sera, as shown in Fig. 2. We previously reported that it did not bind well to MG patient sera (Lindstrom et al. 2008). However, subsequent studies with this chimera revealed binding to MG patient sera, antisera from both cats and dogs with MG, and an MG patient-derived mAb to the MIR (Luo et al., unpublished).
These data reveal that rats with EAMG and human, dogs, or cats with MG recognize the MIR in slightly different ways. The epitopes to which antibodies bind can be determined by a small number of amino acids, and binding can be prevented by changing a single amino acid, even though the area on a protein obscured by the binding of an antibody can be very large (Amit et al. 1985; Konstantakaki et al. 2007). This explains why mAbs and serum autoantibodies can be mutually competitive for binding to the MIR while recognizing discrete epitopes and conformations within this region.
α1/α7 chimeras expressed much more efficiently in Xenopus oocytes than did native human α7 AChRs (Lindstrom et al. 2008). α1(60–81)/α7 expressed 10 fold better than α7, and α1(2–14, 60–81)/α7 expressed 26 fold more than wild type α7.
These data suggest that the α1 MIR in the chimeras increases expression by improving conformational maturation. Synthesis of α1 subunits requires only 1 minute, but conformational maturation prior to assembly requires 30 minutes (Merlie et al. 1983). This conformational maturation was detected by the simultaneous acquisition of the ability of α1 subunits to bind both α bungarotoxin and mAb 35. A further 60 minutes are required for assembly of subunits into mature AChRs in the endoplasmic reticulum. Yet another 60 minutes are required for transport of the mature AChRs to the cell surface. α7 AChRs require special chaperones for assembly (Millar 2008). The chaperone RIC-3 greatly increases expression of α7 AChRs in cells which would not otherwise express α7 AChRs, but even then expression is not altogether efficient. It seems likely that, in chimeras, interaction between the α1 N-terminal α helix and the α1 MIR loop nucleates much more efficient conformational maturation of α7 subunits than occurs with α7 subunits in oocytes, which permits assembly of these subunits into mature AChRs much more efficiently. This hypothesis is consistent with the observation that the N-terminal α helix is required for assembly of mature pentameric AChRs both for α7 and heteromeric AChRs (Castillo et al. 2009).
MIR chimeras also influence conformation changes in mature AChRs (Lindstrom et al. 2008). The EC50 for activation of α1(1–32, 60–81)/α7 AChRs by ACh, 17.8 μM, indicates that these are 10 fold more sensitive to activation than is wild type α7, EC50 = 182 μM. Conversely, the α1(2–14, 60–81)/α7 chimera, EC50 = 2353 μM, is 13 fold less sensitive.
These data suggest that the MIR takes part in global conformation changes associated with activation of AChRs. The resting conformation of the α1(1–2, 60–81)/α7 chimera may be less stable than wild type and more easily triggered into the open channel conformation by binding of ACh. Conversely, the resting conformation of theα1(2–14, 60–81)/α7 chimera may be stabilized and less easily triggered into the open channel conformation.
α1/α7 chimeras permit mapping of the antigenic structure of the MIR. These reveal interactions between the N-terminal α helix and the MIR loop which explain the conformation-dependence of the antigenic structure of the MIR.
Future crystallographic studies of mAbs bound to α1/AChBP chimeras might permit mapping of the antigenic structure of the MIR at atomic resolution.
The profound influence of the MIR on both the expression and function of α1/α7 chimeras suggests that the MIR at the extracellular tip of AChR subunits may have much greater roles in AChR conformation changes than previously appreciated. Interaction between the N-terminal αhelix and the MIR loop may nucleate conformational maturation of the subunit, and this maturation may permit assembly of mature AChRs. The MIR may also influence global conformation changes in mature AChRs associated with activation and desensitization.
This research was supported by grants from the National Institutes of Health to J. L. numbers NS052463 and NS11323.