95KCNE β subunits. By subtracting the inhibited current traces from the pre-treated
traces, a hierarchy of KCNE modulation of KCNQ1 channels was determined:
KCNE3>KCNE1>>KCNE4. Although the thiol-reactive CTX-MAL was specifi-
cally designed to probe the assembly of KCNQ1-KCNE complexes, this technique
can be adapted to study other multi-subunit ion channels by merely changing the
toxin. Indeed the Du Bois lab applied this strategy to the NaV channel blocker, saxi-
toxin (Parsons and Du Bois 2013 ). Saxitoxin is particularly useful as its de novo
chemical synthesis (Andresen and Du Bois 2009 ) allows for the easy addition of
multiple chemically reactive functional groups (Ondrus et al. 2012 ).
In addition to probing channel function, a cleavable version of CTX-MAL pro-
vided a unique avenue to determine how many KCNE1 β subunits assemble with
KCNQ1 channels. In this version, CTX was chemically modified with a cleav-
able tether containing a maleimide (CTX-Clv) (Fig. 5b). Toxin sensitive complexes
were inhibited by CTX-Clv, allowing for maleimide labeling of a KCNE1 β subunit
(Morin and Kobertz 2008a). Upon tether cleavage, the toxin washed out leaving
behind one chemically-inactivated KCNE1 β subunit. Iterative rounds of CTX-Clv
application and cleavage resulted in a counting strategy that allowed the authors
to determine that the KCNQ1 channel stoichiometry was 4 channel subunits to 2
KCNE1 β subunits. This was the first time that the possibilities of mixed stoichi-
ometries could be ruled out, leading to renewed debate in the field (Nakajo et al.
2010 ). Questions regarding the accuracy of this counting technique were finally put
to rest with the Goldstein’s lab photobleaching studies, which confirmed the 4:2
stoichiometry (Plant et al. 2014 ).
Similar to the other chemically-reactive tethers discussed in this chapter, these
studies also required the use of an engineered cysteine. To get around this limitation,
Hua et al. utilized a membrane permeant, thiol-containing N-acetylmannosamine
derivative, which is metabolized by cells and incorporated into every glycosylated
ion channel or accessory subunits at the cell surface (Hua et al. 2011 ). These un-
natural sugars provided chemical handles on the ion channel complex of interest
without having to engineer non-native cysteines into the subunits. Using the afore-
mentioned CTX-MAL, Shaker K+ channels and toxin-sensitive KCNQ1-KCNE1
channel complexes were irreversibly inhibited because both Shaker and KCNE1
subunits are N-glycosylated (Santacruz-Toloza et al. 1994 ; Chandrasekhar et al.
2006 ). In contrast, cells metabolically-labeled with thiolsugar, but expressing un-
glycosylated, toxin-sensitive, homotetrameric KCNQ1 channels were only revers-
ibly inhibited by CTX-MAL, demonstrating that the reaction specifically occurred
with the unnatural thiolsugars attached to the ion channel subunits. To demonstrate
that ion channel subunits could be specifically labeled using this approach, the au-
thors used a cleavable derivative (CTX-Biotin) to deliver a biotin moiety to the
unnatural thiolsugars on the Shaker K+ channel. Given that most ion channel com-
plexes require N-glycans for proper function, this approach can be readily applied
to a wide variety of channels to specifically deliver molecular probes to wild type
ion channels subunits for use in subsequent imaging or biochemistry experiments.
Building on this delivery system, the Chambers lab developed a tri-functional
photo-cleavable probe (Fig. 5b) to fluorescently label non-NMDA glutamate recep-
tors (Vytla et al. 2011 ). This reagent contained a fluorescent reporter, a use-depen-
Bioreactive Tethers