Cysteine Modification 41
lar solution, it was concluded that the closed channel gate was at the cytoplasmic
end of the channel. The strength of this conclusion was limited because these
experiments measured whether or not reaction had occurred rather than a com-
parison of reaction rates. To address this limitation a subsequent study measured
the MTS reagent reaction rates and reached similar conclusions that the gate was
at the cytoplasmic end of the channel (Pascual and Karlin 1998 ). However, this
study did not address the issues of mutation induced changes in spontaneous open
probability or weak partial agonist effects of the reagents. Unwin, using cryo-
electron microscopy images of the nACh receptor, inferred that the gate was in
the middle of the channel near the 9’ residue (Miyazawa et al. 1999 ). A SCAM
study in the homologous 5-HT 3 A receptor also concluded that the gate was in the
middle of the M2 segment (Panicker et al. 2002 ). Bali and Akabas showed that
picrotoxin could be trapped in the closed GABAA receptor channel suggesting
that a gate existed between the cytoplasmic end of the channel where picrotoxin
binds and the extracellular end of the channel (Bali and Akabas 2007 ). Other x-
ray crystal structures of GLIC and GluCl have shown that the narrowest region of
the channel is at the cytoplasmic end creating further uncertainty (Bocquet et al.
2009 ; Hibbs and Gouaux 2011 ; Hilf and Dutzler 2009 ). The ELIC channel struc-
ture, however, is narrowed by the presence of a phenylalanine in the extracellular
third of the channel (Hilf and Dutzler 2008 ), but the relevance of this constriction
to other Cys-loop receptor family members has been questioned (Gonzalez-Guti-
errez and Grosman 2010 ). The unresolved question is, which state of the channel
do these crystal structures represent, open, closed or desensitized (Akabas 2013 )?
One further piece of data regarding the structure of the closed channel is that in
the GABAA receptor, cysteines substituted at the extracellular end of M2, at the
20’ position, spontaneously form disulfide bonds (Horenstein et al. 2001 , 2005 ).
This suggests that in the closed state the extracellular ends of the channel-lining
M2 segments are mobile and the substituted cysteines can approach to within 2 Å
the distance necessary to form a disulfide bond (Careaga and Falke 1992b; Pakula
and Simon 1992 ; Krovetz et al. 1997 ). Thus, the protein dynamics and mobility
may be much greater than the fixed crystal structures suggest. At this point the
available data from functional and crystallographic studies does not provide a
consistent picture of the location of the Cys-loop receptor closed channel gate or
the structure of the closed channel. Further studies will be necessary to resolve
these issues.
SCAM experiments, combined with disulfide crosslinking studies, have pro-
vided insight into the conformational changes that occur during channel activation,
desensitization and during allosteric modulation. The channel in Cys-loop recep-
tors is surrounded by two rings of helices. The inner ring formed by the five M2
segments, one from each subunit, and the outer ring, formed by alternating M1
and M3 segments from the five subunits (Miyazawa et al. 2003 ; Unwin 2005 ). To
investigate whether conformational change occurred in the outer ring of helices we
conducted a SCAM study of the GABAA receptor α1 M3 segment. In the absence
of GABA, pCMBS reacted with cysteines substituted for two residues near the ex-