Biophysical Role of the NMDAR Transmembrane Domain in Channel Gating and Block

In the 1950s, Sernyl, more commonly known as phencyclidine (PCP), was created, and marketed as a surgical anesthetic. However, due to side-effects that mimic the symptoms of schizophrenia, PCP was removed from the market. In the 1980s, dizocilpine (MK-801) a derivative of PCP was introduced as a novel NMDAR channel blocker, however, MK-801 also exerts psychotomimetic (mimic a psychotic state) symptoms. Whereas PCP and MK-801 have psychotomimetic properties, other NMDAR open channel blockers, like memantine do not. In fact, while MK-801 has no clinical utility, memantine has moderate therapeutic use in the treatment of Alzheimer’s disorder. Despite the difference in their pharmacological actions, the mechanisms that distinguish the psychotomimetic properties of these compounds are unknown. We propose that understanding how these drugs interact with residues in the NMDAR channel pore is essential in distinguishing the psychotomimetic potential of open channel blockers. Our central hypothesis is that the psychotomimetic potential of a given NMDAR open channel blocker is defined by the degree to which the compound becomes trapped in the ion channel pore. Thus, we assert compounds like MK-801 are psychotomimetic because they become fully trapped. By contrast, compounds like memantine are not psychotomimetic because they do not become fully trapped in the NMDAR pore. I set out to determine the contribution of the residue at the threonine position of the highly conserved SYTANLAAF motif of NMDAR subunits in trapping MK-801 in the channel pore. Recently, this threonine residue has been identified as part of the binding pocket for MK-801 and other open channel blockers, however the contribution of this residue to trapping is unresolved. To determine how the amino acid at the threonine position contributes to MK-801 block, we used site-directed mutagenesis to substitute different amino acids and used whole-cell patch-clamp electrophysiology to determine how this impacted MK-801 block and recovery. Additionally, because the SYTANLAAF motif is part of the NMDAR gating machinery, we used whole-cell patch-clamp electrophysiology to assess NMDAR channel function of GluN1/GluN2A NMDARs. In Chapter 2, we substituted fourteen of the nineteen different naturally occurring amino acid residues in the GluN1 subunit and we found that any substitution at residue T648 caused MK-801 to dissociate more rapidly. Moreover, a subset of mutations that exhibited a rapid and complete recovery from MK-801 block also dramatically altered channel deactivation and desensitization. Curiously, two substitutions (threonine-to-leucine and threonine-to-isoleucine) profoundly impacted MK-801 block, but minimally impacted NMDAR function. In our analyses we demonstrate that these processes are greatly influenced by side chain polarity. In Chapter 3, we introduced the threonine-to-leucine mutation into the GluN2A subunit and demonstrate this mutation accelerates recovery from MK-801 block without influencing the percentage of block. However, unlike the GluN1 counterpart, the threonine-to-leucine mutation in the GluN2A subunit slows channel activation and deactivation kinetics. Taken together, the findings presented in this dissertation suggest that the threonine residues in the SYTANLAAF motifs of both the GluN1 and GluN2A NDMAR subunits are crucial for the slow dissociation of MK-801. Whereas the threonine residues in the SYTANLAAF motifs of GluN1 and GluN2A subunits have an equivalent role in determining the magnitude and duration of MK-801 block, they exert subunit-specific contributions to NMDAR gating.PhDPharmacologyUniversity of Michigan, Horace H. Rackham School of Graduate Studieshttp://deepblue.lib.umich.edu/bitstream/2027.42/178114/1/nijack_1.pd