This study investigated the structural basis for the difference in conductance between fetal and adult acetylcholine receptors (AChRs) in rat muscle. The researchers found: 1) The y- and e-subunit genes that encode the fetal and adult AChR isoforms each have 12 exons, with the four hydrophobic segments (M1-M4) that form the pore encoded by separate exons. 2) Exchanging all four hydrophobic segments (M1-M4) between the y- and e-subunits completely reversed the conductance difference between the fetal and adult AChR channels. 3) Mutational analysis identified the major determinants of the conductance difference as being

1. Journal of Physiology (1996), 492.3, pp.775-787
Structural determinants of channel conductance in fetal
and adult rat muscle acetylcholine receptors
S. Herlitze, A. Villarroel, V. Witzemann, M. Koenen and B. Sakmann*
Abteilung Zellphysiologie, Max-Planck-Institut fiur medizinische Forschung,
Jahnstrasse 29, D-69120 Heidelberg, Germany
1. The structural basis of the developmentally regulated increase in endplate channel
conductance in rat, where the y-subunit of the fetal muscle acetylcholine receptor (y-AChR)
is replaced by the c-subunit in the adult muscle receptor (C-AChR), was investigated by
analysing the structure of y- and e-subunit genes and by expressing recombinant AChR
channels of different molecular composition in Xenopus oocytes and measuring their single-
channel conductance.
2. The y- and c-subunit genes each have twelve exons. In both subunits, the four homologous
segments, designated Ml, M2, M3 and M4, which are thought to contribute to the
formation of the pore, are encoded by four separate exons, E7, E8, E9 and El 2.
3. Chimaeric e(y)- or y(e)-subunits were constructed from sequences derived from the parental
e- and y-subunits, respectively. Exchanging the four hydrophobic segments (M1-M4) of the
y-subunit for those of the c-subunit and vice versa completely reversed the difference in
conductance between y-AChR and e-AChR channels.
4. Effiects of single- and multiple-point mutations in M1-M4 segments of y- and c-subunits
indicate that the major determinants of the difference in conductance between fetal and
adult endplate channels are located in the M2 segment. The key differences are the
exchange of alanine/threonine (y-subunit) for serine/isoleucine (e-subunit) in M2, and the
lysine (y-subunit) for glutamine (e-subunit) exchanges in the regions flanking the M2
segment.
Acetylcholine receptors (AChRs) in mammalian muscle are
expressed in two isoforms, a fetal and adult form, which are
developmentally regulated in their time of appearance and
cellular localization (Mishina et al. 1986). The two isoforms
differ in that 'elementary' endplate currents in adult muscle
are of larger amplitude and of shorter duration than in fetal
or denervated muscle. In adult muscle the receptor is
composed of a-, ,B-, d- and c-subunits, presumably with an
a2flc8 stoichiometry (Mishina et al. 1986). This receptor
replaces postnatally the fetal isoform, which is composed of
a-, ,B-, y- and 8-subunits and which is only transiently
expressed around birth (Mishina et al. 1986; Gu & Hall,
1988; Witzemann et al. 1990). The functional difference
between the two isoforms offers the possibility of
identifying structural determinants of AChR channel
conductance by expressing recombinant AChRs of different
subunit composition (y-AChR and e-AChR) and selecting
differences in the amino acid sequence between the y- and
the c-subunits as targets for site-directed mutagenesis.
Previous analyses of recombinant AChRs support the view
that the hydrophobic M2 segment contributes, at least in
part, to the formation of the wall of the ion channel (Imoto
et al. 1986; Leonard, Labarca, Charnet, Davidson & Lester,
1988; Charnet et al. 1990). Charged residues bordering the
extra- and intracellular portion of this segment are major
determinants of channel conductance (Imoto et al. 1988).
The selectivity filter and the narrowest portion of the
channel are also formed by residues of the cytoplasmic
portion of the M2 segment (Villarroel, Herlitze, Koenen &
Sakmann, 1991 a; Villarroel, Herlitze & Sakmann, 1991 b).
To examine the contribution of different subunit domains to
the conductance properties of AChR channels, segments
were exchanged between y- and e-subunits by generating
chimaeric y(e)- and e(y)-AChR subunits. The segments
exchanged were selected according to motifs of gene
structure and the hydrophobicity profile of the amino acid
chain. The results indicate that for a complete switch of
conductance properties between y-AChR and e-AChR, the
* To whom correspondence should be addressed.
4896 775

2. S. Herlitze and others
Table 1. Generation of y(e)-chimaeric constructs
y(c)-Chimaeric constructs generated by PCR cloning
Mutated Template
Mutant region for PCR
y(e)-2 Ml y-wt
y(e)-4 M2 +'K/Q'sites y-wt
y(c)-5 M3 y-wt
y(e)-8 M2-M3 bend y-wt
y(C)-1I Ml + M2 y-wt
y(c)-12 Exon 8 y-wt
y(c)-13 M2 + M3 y-wt
y(c)-10 P bend M3-M4 y-wt
y(e)-6 M4 y-wt
y(e)-3 Ml +QQ(e)-2
y(c)-16 Ml + M2 + M3 y(e)-11
y(c)-14 Ml to exon 9 y(c)-16
y(e)-18 5' end to M3 e-wt for 5',
+ y-wt for 3' end
Y(,6)-9 M3-M4 bend e(y)-18 for 5',
+ y-wt for 3' end
y(e)-7 5'end e-wt for 5',
+ y-wt for 3' end
Restriction of
PCR product
EcoRV/Bsu36I
EcoRV/Bsu36I
EcoRV/Bsu36I
EcoRV/Bsu36I
EcoRV/Bsu36I
EcoRV/Bsu36I
EcoRV/Bsu36I
BstEII
BstEII
EcoRV/Bsu361
EcoRV/Bsu36I
EcoRV/Bsu36I
AatII/Bsu36I
BstEll
NsiI/NaeI
Vector
for ligation
y-pSPOoD
y-pSPOoD
y-pSPOoD
y-pSPOoD
y-pSPOoD
y-pSPOoD
y-pSPOoD
y-pSPOoD
y-pSPOoD
y-pSPOoD
y-pSPOoD
y-pSPOoD
y(e)-7
y(c)-18
y-wt; insertion
of c-wt
HindIII/Nael
fragment
y(e)-Chimaeric constructs generated by ligation *
Mutant Mutated
region
y(c)-15
y(c)-17
y(e)-l
M2 + M4
Ml + M2 + M4
Ml +M2+M3+M4
*PstI fragments excised from y(e)-6 carrying the
construct indicated on the right.
four segments Ml-M4 have to be exchanged. However, the
amino acids responsible for the major part of the difference
are located in the M2 segment, in particular, lysine or
glutamine residues bordering the M2 segment at the extra-
and intracellular funnel of the channel, and those amino
acids which are located close to the hydroxyl ring positions
forming the channel constriction.
METHODS
Structure of rat y- and e-subunit genes
The fragments of the y- and c-subunit genes (AACRe1 and
AACRy25), isolated from a genomic rat liver DNA library
(Witzemann, Barg, Nishikawa, Sakmann & Numa, 1987) were
subcloned into Bluescript plasmid vectors for sequencing. DNA
sequence analysis was performed using the Sequenase kit, version
2.1 (United States Biochemical Corporation, Cleveland, OH, USA),
to identify the exon/intron structure of the rat AChR y- and
c-subunit genes. The nucleotide sequence of the y-subunit gene was
obtained as follows. The ArACRy25 clone (Witzemann et al. 1987)
yielded a 4-1 kbp BamHl fragment that contained sequences from
Mutant-vector
construct for ligation
y(6)-4
y(,6)-1
y(6)-16
c-M4 region were ligated into the mutant-vector
exon 7 to exon 12. The larger BamtH1 fragment of the
ArACRy25 clone was cleaved to yield a 6 kbp XbaI/EcoRl
fragment that included exon 1 to parts of intron 4. The region
between exon 5 and exon 7 was amplified from the ArACRy25
DNA and subeloned for sequencing. The sequence of the e-subunit
gene was obtained from ArACRe1 (Witzemann et al. 1987).
HindIII fragments of 0-8 and 4-8 kbp contained the complete
protein coding region. A 2-4 kbp BamHl fragment served in
addition to sequence the region between exon 5 and exon 8. The
complete genomic sequences of the two subunit genes were
delivered to the EMBL data bank under the accession numbers
(X94365) and (X94364). Exon-intron borders were identified by
comparison with the respective rat muscle cDNA sequences
(Witzemann et al. 1990).
Construction of y- and e-AChR subunit mutants
Complementary DNAs encoding the rat muscle AChR subunits
(Witzemann et al. 1990) were cloned into pSPOoD (Villairoel et
al. 1991 a), a derivative of pSP64T (Melton, Krieg, Rebagliati,
Maniatis, Zinn & Green, 1984). Oligonucleotide-directed
polymerase chain reaction (PCR) mutagenesis, as described by Ho,
Hunt, Horton, Pullen & Pease (1989) and Herlitze & Koenen
(1990) was used to generate y- and c-subunit mutants. Primers
776 J Phy8iol.492.3

3. Developmental AChR isoforms
Table 2. Generation of e(y)-chimaeric constructs
e(y)-Chimaeric constructs generated by PCR clonin
Mutated
Mutant region
Ml
M2 +'K/Q' sites
M3
M2-M3 bend
Ml + M2
Exon 8
M2 + M3
P bend M3-M4
M4
Ml +Q
M1+ M2+ M3
5' end to M3
M3-M4 bend
5'end
g
Template
for PCR
6-wt
c-wt
e-wt
e-wt
c-wt
c-wt
e-wt
e-wt
e-wt
e(y)-2
c(y)-l 1
y-wt for 5',
+ e-wt for 3' end
y(6)-18 for 5',
+ c-wt for 3' end
y-wt for 5',
+ c-wt for 3' end
Restriction of
PCR product
AatII/Bsu36I
AatII/Bsu36I
Aatll/Bsu36I
AatIl/Bsu36I
AatIl/Bsu36I
AatII/Bsu36I
AatII/Bsu36I
BamHI/PvuII
BamHI/PvulI
AatlI/Bsu36I
AatIl/Bsu36I
EcoRV/SalI
EcoRV/SalI
HindII/Satl
Vector
for ligation
e-pSPOoD
c-pSPOoD
c-pSPOoD
e-pSPOoD
c-pSPOoD
c-pSPOoD
c-pSPOoD
e-pSPOoD
c-pSPOoD
c-pSPOoD
e-pSPOoD
,6(y)-7
y(C)-18
e-wt; insertion
of 5' HindIIJ
y-wt fragment
e(y)-Chimaeric constructs generated by ligation *
Mutant Mutated
region
6(y)-15
6(y)-17
e(y)-l
M2 + M4
Ml + M2 + M4
Ml +M2+M3+M4
*BamHI/PvuII fragments excised from y(e)-6
mutant-vector construct indicated on the right.
carrying mutation sequences for a single or two adjacent sites
contained at the 5' end at least nine, and at the 3' end eighteen,
complementary nucleotides. When more than two amino acids had
to be altered the mutation oligonucleotides carried the 'mutation
sequence' of up to fifty nucleotides at the 5' end while the fifteen to
eighteen nucleotides at the 3' end were complementary to the
starting sequence of the template DNA used for amplification. The
wild-type (wt) y- and c-subunit (y-AChR and c-AChR,
respectively) cDNA sequences were used for amplification. The
resulting y-subunit PCR fragments with single-site exchanges
were cleaved by EcoRV/Bsu361 and ligated into pSPOoD carrying
the corresponding y-wt sequence (y-pSPOoD). The e-subunit PCR
fragments were digested with AatII/Bsu36I and subcloned into
the pSPOoD carrying the corresponding c-subunit sequence
(c-pSPOoD). The chimaeras are designated as the parental subunit
with the inserted sequence in parentheses, thus y(e) refers to a
chimaera in which c-subunit sequences are inserted into the
parental y-subunit sequences. The construction of y(e)- and e(y)-
chimaeric subunits is summarized in Tables 1 and 2, respectively.
The nomenclature of mutants is such that the amino acid (in single
letter code) in one wild-type subunit is followed by its position
number according to Witzemann et al. (1990), followed by the
mutant amino acid. The mutant amino acid is that which is
Mutant-vector
construct for ligation
e(y)-4
e6(y)-1 1
6(y)-16
carrying the y-M4 region were ligated into the
present at the equivalent position in the other subunit. Multiple
mutations have an asterisk separating the position numbers.
Expresion of recombinant AChRs in XenopuA oocytes
Wild-type and mutant AChR subunit-specific cRNAs were
synthesized in vitro with SP6-polymerase (Krieg & Melton, 1984)
using SalI-cleaved pSPOoD plasmids carrying the respective
cDNAs as templates.
Xenopus laevis frogs were anaesthetized by immersion in water
containing tricaine methanesulphonate (6 g F') for 20 min until
they were unresponsive to touch. Oocytes were removed under
sterile conditions, and the abdomen was closed with sutures.
Following surgery, the frogs were washed with water and carried
back to the aquarium.
Xenopus laevis oocytes were injected with cRNAs encoding the
different subunits of wild-type and mutants of AChR channels (a-,
,-, d- and either y- or e-subunit cRNA in a molar ratio 2 :1:1:1)
and incubated for 2-5 days at 19 + 1 °C (Methfessel, Witzemann,
Takahashi, Mishina, Numa & Sakmann, 1986).
Current recordings
AChR channel expression was determined 2 days after cRNA
injection by measuring the current elicited in response to bath
c(y)-2
c(y)-4
c(y)-5
e(y)-8
e(y)-1 2
,6(y)-13
,6(Y)-10c(y)-lO
e(y)-6
c(y)-3
6(y)-16
,6(y)-18
6(y)-9
J Physiol.492.3 777

4. 778
application of 1 ,UM ACh in normal Ringer solution (135 mm NaCl,
5 4 mm KCl, 1P8 mm CaCl2, 5 mm Hepes, pH 7 2 adjusted with
NaOH) during conventional two-microelectrode voltage clamp
(membrane potential, -70 mV). The current-passing electrode had
a resistance of 0 1-0 2 MQ when filled with 3 M KCl.
Single-channel currents were measured in inside-out patches
isolated from the oocyte membrane (Methfessel et al. 1986). The
bath solution contained 100 mm KCl, 10 mm Hepes and 10 mm
EGTA and the pH was adjusted to 7-4 with KOH. The pipette
solution was the same as the bath solution, except that ACh was
present in a concentration of 0 5-1P0 /M. All measurements were
performed at 18 + 1 'C. Current records were sampled at 10 kHz
after low-pass filtering at 3 kHz (-3 dB). The amplitudes of
J Physiol.492.3
single-channel currents were measured using a semi-automatic
procedure (Imoto et al. 1986). To construct a single-channel
current-voltage (I- V) curve about 1000 single-channel currents
were measured, pooled in 2 mV bins and averaged. The
conductance values given represent chord conductances at
-100 mV using the interpolated reversal potential obtained from
5th-order polynomials fitted to the I- V relations.
For comparison of recombinant y-AChR and c-AChR with native
AChRs in muscle, single-channel I- V relations were also measured
in the cell-attached configuration with a pipette solution containing
150 mM NaCl, 1 mm BaCl2, 10 mm Hepes (pH 7 2) and 05,uM
ACh, as in Witzemann et al. (1987).
A
El E2 E3 E4 E5 E6 E7 E8 E9 E10 Eli E12
y-Gene
El E2 E3 E4 E5 E6 E7 E8
e-Gene
500 bp
I- I
Ml
EINGEWAIRHRPAKMLLDPVTPAEEAGHQKVVFYLLIQRK PLFYVINIIVPCVLISSVAILIYFLPA (J A-
E NGEWAIDYCPGMIRHYEGGSTEDPGETDVIYTLIIRRK PLFYVINIIVPCVLISGLVLLAYFLP Q A-
E9 El0 El1 E12
B
Exon 7
M2
IGGQK CTVATNVLLAQTVFLFLVdKKVPETSQAVPLIS K-
IGGQKJCTVSINVLLAQTVFLFLIA QKIPETSLSVPLLGj-R-
M3
YLTFLMVVTILIVVNSVVVLNV SLRSPHTHSMARGVR
YLIFVMVVATLIVMNCVIVLNV SLRTPTTHATSPLQ
* ** * * * i
M4
G4NEEWLLVGRVLDR VCFLAMLSLFICGTAGIFL MAHYNQVPDLPFPGDPRPYLPLPD
EAIELSDWVRMGKALDN VCFWAALVLFSVGSTLIF GGYFNQVPDLPYPPCIQP
Figure 1. Structure of the rat y- and e-AChR subunit genes and putative transmembrane
folding of these subunits
A, schematic representations of y- and c-subunit exon-intron structure are shown in the upper panel. The
lower panel shows the amino acid sequence of hydrophobic segments in the coding region of exons E7, E8,
E9 and E12, with single-letter notation of amino acids. Both subunits comprise four hydrophobic
segments designated as Ml-M4 (boxes in lower panel). Upper seqaence represents the y-subunit, lower
sequence the e-subunit. Differences in amino acid sequence of these putative transmembrane segments are
indicated by asterisks. Differences in the charged amino acids near the cytoplasmic and extracellular end
of this segment are also indicated by asterisks. B, putative transmembrane arrangement of hydrlophobic
segments M1-M4 (Numa, 1989). Individual exons are shown in black, and their hydrophobic sequences
are highlighted in white.
S. Herlitze and others
Exon 8
El 0
Exon 9
Exon 12

5. Developmental AChR isoforms
RESULTS
Exon structure of y- and e-subunit genes and
construction of chimaeric subunits
A comparison of the amino acid sequences of rat muscle
AChR subunits indicates that y- and e-subunits are those
most closely related (Witzemann et al. 1990). Figure 1A
shows the genomic structure of the rat y- and e-subunit
genes (see Methods). Each rat gene has twelve exons, as in
the chick and human y-subunit (Nef, Mauron, Stalder,
Alliod & Ballivet, 1984; Shibahara, Kubo, Perski,
Takahashi, Noda & Numa, 1985) and the mouse e-subunit
(Buonanno, Mudd & Merlie, 1989). Each of the putative
transmembrane segments, MI, M2, M3 and M4, is encoded
by a separate exon E7, E8, E9 and E12, respectively, in
both genes (Fig. 1A). The obvious structural determinants
responsible for the conductance differences between the two
AChR subtypes may reside in the hydrophobic segments
which presumably are membrane spanning (Fig. 1B) and
could form the walls of the pore.
We examined the conductance of recombinant AChR
channels assembled from a-, /1- and S-subunits with either
the y-subunit (y-AChR), the c-subunit (e-AChR) or a
chimaeric subunit, the sequence of which is derived partly
from the y- and partly from the e-subunit. The Ml, M2, M3
and M4 transmembrane segments, combinations of these
segments, or other motifs, were exchanged between the y-
and e-subunits (Fig. 2A). The y-AChR was used as a
template to reconstruct a functional c-AChR channel by
replacing amino acid segments of the y-subunit by the
corresponding segments of the c-subunit and vice versa.
The y(c)-chimaera refers to a y-subunit template with an
e-subunit insert and vice versa in e(y)-chimaeras. Of the
nineteen y(e)- and e(y)-chimaeric constructs, when co-
expressed in oocytes with a-, ,1- and S-subunits, sixteen
produced functional AChR channels (Fig. 2B and C). In
most cases whole-cell current responses did not decrease
significantly in amplitude as long as ACh (1 ,UM) was
applied. Some constructs, however, formed 'aberrant'
channels which showed rapid ACh-induced desensitization
(i.e. y(c)-7, y(e)-9, y(e)-15, y(e)-17 and e(y)-5, e(y)-9,
c(y)-15).
Exchange of M1-M4 segments reverses the
conductance difference between y- and c-AChR
channels
For a direct comparison between the conductance of native
fetal and adult AChR channels and recombinant y- and
c-AChR channels in Xenopus oocytes, the single-channel
slope conductance of the two recombinant channels in
oocytes was measured under ionic conditions comparable to
those used for conductance measurements in the native
A Chimaeric constructs B y(e)-Chimaeras C e(y)-Chimaeras
M1 M2 M3 M4 0-1
+
II
Lr rr.d.I-r
iI b, I
::. I:X
:.:::
rt'' it
r- I*.XI------4----4-- ---
.., X
10 100
01 1 10
Whole-cell current (pA)
01 1 10 100
c,-AChR
E(y)-l
e(y)-3
(y)-4
e-(y)--6
c.(y)-7
e;r-8
e(y)-19
e(y)-l 2
e;(y)-l 3
c(y)-l 5
e(y)-l 6
e;(y)-l 7
c(y)-l 8
100 0.1 1 10
Whole-cell current (uA)
100
Figure 2. Expression in Xenopus oocytes of recombinant parental y-AChRs or c-AChRs and
hybrid AChRs carrying y(e)- or e(y)-chimaeric subunits
A, schematic representation of y(e)- or c(y)-chimaeric subunit constructs. Parental subunit is represented
on top with four hydrophobic segments M1-M4. Sequence motifs which were exchanged between the two
parental (y- and e-) subunits are represented by grey boxes. The sequences of the parental subunit used as
a template are represented by open boxes. The sequence region(s) exchanged between the two subunits are
mirror images of each other. B and C, whole-cell current amplitudes of oocytes expressing recombinant
AChRs carrying y(e)-chimaeric subunits (B), or e(y)-chimaeric subunits (C) as shown in A. Chimaeric
subunits were co-expressed with a-, ,B- and S-subunits. The mean of peak inward currents at -70 mV in
response to bath application of 1 FM ACh is represented by each bar (measurements with 3-10 oocytes for
each construct).
7-7-71 -7
i
I I I - . I
'.-:
j. ...
.=::.
:.:.:.:.::::, ....'. .....,.
71T
-T-l-
J Physiol.492.3 779

6. S. Herlitze and others J Physiol.492.3
y-AChR
-", L % l_
e-AChR
+100 MyV -,-
y.
+50 mv_
0 mv ,_
-50 mv _ J- ,
20 -T
cra
10-
-200 -100
y-AChR
e6o
&e ,AChR D
-1 00 mv-OI
0
100 200
Voltage (mV)
8 pA
20 ms
B
20--
y(c)-l
NH2
liI
.... ...
~ ......
e(y)-l
NH2
cc
C:
a)
-200 -100
c(y)-l1
10t
I
100 200
Voltage (mV)
lo-+
y(e)-l
-20
Figure 3. Conductance of recombinant AChR channels carrying chimaeric subunits in which
the four hydrophobic segments M1-M4 were swapped between the y- and e-subunits
A, single-channel currents (left panel) activated by ACh (1 /LM) at different membrane potentials and
single-channel current-voltage (I- V) relations (right panel) showing the difference in conductance
between y- and c-AChR channels. Currents were recorded from inside-out membrane patches isolated
from Xenopus oocytes expressing y- or c-AChR; inward current is downward. The c-AChR (0) has a
conductance of 104 + 2 pS (n = 7) at -100 mV; the y-AChR (0) has a conductance of 69 + 2 ps (ni = 10)
at -100 mV. Records were obtained in 100 mm symmetrical K+ solution, in the absence of divalent
cations. B, schematic representation of y(c)-1 and c(y)-1 chimaeric constructs and single-channel currents
are shown in the left panel; the I- Vrelations of channels comprising these chimaeric subunits are shown
in the right panel. In the left panel the sequence derived from the e-subunit is represented by thin lines
and open M1-M4 segments. The sequence derived from the y-subunit is represented by grey M1-M4
segments and bars. Calibration bars are the same for A and B.
780
A
-zu-_ _

7. Developmental AChR isoforms
muscle membrane (Witzemann et al. 1987). The slope
conductances (measured between -30 and -130 mV) were
43±+pS (n=3) for the y-AChR and 63+1pS (n=3)
for the e-AChR; these values were indistinguishable from
those measured in muscle (42 + 3 pS and 63 + 3 pS,
respectively; Witzemann et al. 1987).
For the conductance analysis of mutagenized y- and
c-AChRs, simpler ionic conditions were used with 100 mM
K+ and no divalent cations on either side of the membrane.
Under these conditions the y-AChR and the e-AChR
channels have larger conductances of 69 + 2 pS (n = 10)
and 104 + 2 pS (n = 7), respectively (Fig. 3A). Figure 3B
illustrates the conductance difference between channels
carrying the chimaeric y(e)-l or e(y)-l subunits in which
the sequences of the M1-M4 segments were exchanged. The
y(e)-chimaeric subunit confers an increase in conductance
(102-7 + 2-5 pS, n= 5) to the y-AChR, the conductance of
which is indistinguishable from that of the e-AChR channel.
The chimaeric subunit e(y)-l confers a lower conductance
(67-2 + 1.1 pS, n = 5) to the e-AChR and has a similar
conductance to that of the y-AChR channel. Hence, the
differences in conductance between the y-AChR and the
e-AChR are located in the Ml, M2, M3 and M4 segments,
which are encoded by exons E7, E8, E9 and E12.
Functionally important domains in the M1-M4
segments
Construction of recombinant y-AChRs comprising chimaeric
y(e)-subunits in which different transmembrane segments
were exchanged identified shorter stretches of amino acids
which could determine an increased channel conductance
(Fig. 4A). Exchange of the MI segment in the y(e)-2
construct confers a slightly larger conductance to the
y-AChR (72-9 + 0 45 pS, n = 7). The M2 segment,
including the adjacent six amino- and three carboxyterminal
amino acid residues exchanged by the y(e)-4 construct,
produced a large increase in conductance (95-1 + 1-2 pS,
n = 5) and is responsible for about 74% of the conductance
difference between the y- and e-AChR. The M3 segment in
y(e)-5 also contributes to the conductance difference, the
increase being only slightly larger than that observed
when the Ml segment is exchanged, the conductance of
the former being 76-3 + 2-37 pS (n = 7). The conductance
of the recombinant channel carrying the chimaera y(e)-I1,
which swaps the Ml and M2 segments, did not differ
significantly from that carrying only the M2 segment in the
chimaera y(e)-4. Similar results were found for the chimera
y(e)-16 containing sequences from Ml to M3 (Fig. 4A).
Thus, the most important determinant of the larger
conductance contributed by c-subunit residues resides in
that part of the e-subunit segment containing the M2
region, plus the adjacent bends that carry glutamine (Q)
residues instead of the lysines (K) in the y-subunit. The
conductance increase observed with these y(e)-chimaeras,
however, is incomplete, indicating that all four trans-
membrane segments are required for an c-AChR channel
conductance.
Single-channel conductances of recombinant channels
carrying different e(y)-constructs, made to identify those
domains that confer the lower conductance, are
summarized in Fig. 4B. Exchange of the MI segment in
c(y)-2 barely reduced the conductance in comparison with
the parental e-AChR. Similarly, the exchange of either the
M3 segment alone in e(y)-5 or M4 segment alone in e(y)-6
did not result in a significant conductance decrease. The
e(y)-4 construct, however, which comprises the M2 segment
and the lysines of the flanking bends, conferred a strong
(68%) decrease in channel conductance (80'2 + 3'6 pS,
n = 4). The combined exchange of the Ml and the M2
segments in the chimaeric subunit e(y)-1 produced
channels that had a smaller conductance than channels
where only the M2 segment was exchanged. A similar
result was found for exchanges of the M2 and M4 segments
in e(y)-15. The chimaeric subunit e(y)-16, which comprises
the Ml -M3 region, generated a channel with a conductance
indistinguishable from the y-AChR. The chimaera e(y)-16
containing the Ml, M2 and M4 segments also produced a
complete reduction in channel conductance. Thus, the
'minimal' exchange required for abolishing the conductance
difference between c-AChR and y-AChR with
c(y)-chimaeric subunits includes the Ml and M2 segments
in combination with either the M3 or M4 segment.
Figure 4 also illustrates that in y(e)- and e(y)-chimaeric
subunits the following changes did not affect conductance:
the exchange of the N-terminal portion, the exchange of
the extracellular loop between M2 and M3 and the presence
or absence of a putative intracellular phosphorylation site
in the y(e)-10 and e(y)-10 constructs.
Charged and polar amino acids bordering the M2
segment
Sequence comparison of the M1-M2 region shows that
charged or polar residues are located in the region
bordering the M2 segment (Fig. 5A and B). In Torpedo
electroplax AChR, these amino acids form a cytoplasmic
and an extracellular ring at the channel entrances and
mutations at these positions strongly affect the channel
conductance (Imoto et al. 1988; Konno et al. 1991). The rat
y-subunit carries a lysine residue (yK268) whereas the
c-subunit carries a glutamine (eQ267) at equivalent
positions in the cytoplasmic portion (Fig. 5A). Similarly, in
the extracellular bend, the y-subunit carries a lysine
residue (yK293) whereas the e-subunit carries a glutamine
(eQ292). We refer to these positions as the intracellular and
extracellular K/Q positions.
Figure 5C and D shows the effects of mutants on channel
conductance where the charge differences at the K/Q
positions between the two subunits was reversed. In the
781J Phy8iol.492.3

8. S. Herlitze and others
y(e)-Chimaeras
Ml M2 M3
III i
11 1
- I__ -
III
M4
- I - I - I _
- - I. I-I~~DE
- - . I -
- .. .. r
- I -
1i1
- I
Y 6
-
e(y)-Chimaeras
Ml M2 M3
t I1
11
50 Conductance (pS)
M4
iii
r ~
11 In I;
I
11
11 I
IT11
1~ 11 m
U_11I_
rL 11
-~~~~ -
y-AChR
d
01
0t 0
c
110
6-AChR
9
~010I
0
0I
0
°l
ol
II
0
)
y 6
- 50 Conductance (pS)
Figure 4. Amino acid motifs of y- and e-subunits which determine differences in channel
conductance
A, conductance of recombinant AChR channels generated by co-expressing different y(e)-subunit
chimaeras together with a-, ,1- and s-subunits. Schematic representation of chimaeric constructs (left
panel) and mean single-channel conductance (at -100 mV) of AChRs carrying these chimaeric subunits
(right panel). In the schematic representation of constructs, the filled and open segments represent y- and
6-subunit-derived sequences, respectively. In the right panel, the left vertical dashed line indicates
conductance of the y-AChR channel, the right line the conductance of the e-AChR channel.
B, conductance of channels generated by co-expressing e(y)-subunit chimaeras with a-, ,B- and
s-subunits. Schematic representation of e(y)-chimaeric constructs (left panel) and conductance
(at -100 mV) ofchannels carrying c(y)-chimaeras (right panel). Both vertical dashed lines as in A.
A
y-AChR
y(e)-i
y(e)-2
y(e)-3
y(e)-4
y(c)-5
y(e)-6
y(e)-7
y(e)-8
y(6)-i 0
y(6)-i1
y(c)-l 2
y(e)-l 3
y(e)-l 6
y-AChR
0
I.
10
0
*;
e-AChR
*I
S
- amumim
B
e-AChR
6(y)-i
c(y)-2
6(y)-3
e(y)-4
6(y)-5
6(y)-6
6(y)-7
e(y)-8
6(Y)-9
6(y)-i 0
6(y)-l1
6(y)-i 2
6(y)-i 5
6(y)-i 6
e(y)-17
6(y)-l 8
110
782 J Phy8iol.492.3

9. J Physiol.492.3
A
Y P A
e P A
Developmental AChR isoforms
B
P E
P E
6(y)-4
Extracellular ring
Hydroxyl ring
Cytoplasmic ring
D
y-AChR
y(c)-4l
yK268Q*K293Q
yK268Q
yK293Q *
yA277S*T2781I
yA277S 0l
yT2781 *
e-AChR
e(y)-4l Ol
eQ267K*Q292K
eQ267K
eQ292K
eS276A*1277T
eS276A
e1277T
el290V*Q292K
20 Tr
a1)
cL10
-200 -100
I
y-AChR-_
< ~~-10
yK268Q*K293Q 2
eQ267K*Q292K
o
o>-1
8 1 1 1
100 200
Voltage (mV)
50 Conductance (pS) 110
Figure 5. Single amino acids in the M2 segment and the adjacent bends as conductance
determinants
A, amino acid sequences of M2 segments, indicated by the box labelled M2, and of adjacent bends
(M1-M2 and M2-M3), which are included in the larger box, showing the sequences exchanged in the
y(e)-4 and e(y)-4 constructs. Differences in amino acids which are located at equivalent positions are also
boxed. Numbers refer to amino acid positions in the two subunits according to Witzemann et al. (1990).
B, schematic representation of the M2 segment assuming ac-helical structure to illustrate the location of
amino acids that were mutated. Differences in charged or polar amino acids adjacent to the K/Q positions
(at cytoplasmic and extracellular ring positions) are also indicated. Numbers indicate amino acid positions.
C, conductance of channels generated by co-expressing y- (0) or e-subunit (0) mutants together with a-, ,-
and 8-subunits. The single-channel conductance at -100 mV represents the mean of measurements from
3-10 patches of different oocytes. Conductance was measured in 100 mm symmetrical K+ in the absence
of divalent cations. Left vertical dashed line indicates the conductance of the parental y-AChR; right
vertical dashed line indicates the conductance of the parental e-AChR. D, current-voltage relations of
channels carrying double mutants at K/Q positions (yK268Q*K293Q and 6Q267K*Q292K). For
comparison, the I- V relations of parental y-AChR and e-AChR channels are also shown.
783
I ~~~~~~M2
268 278 293
i A G G Q K C T V AiN V L L A Q T V F L F L VA K V
QA G G Q K C T VILLIN V L L A Q T V F L F L LIA Q KI
267 277 292
y-M2 c-M2
C

10. 784 S. Herlitze and others J Physiol.492.3
A y(e)-2 Ml
257
Y G H]Q[K]V V F Y L L I QR K P L F Y V I N I I V P C V L I S LV A I LIY F L P A K
e GET[V I Y T L I IJR K P L F Y V I N I I V P C V L I S[G L V L[LAY F L P A Q
256
e(y)-2
y-AChR *
y(c)-2
yH228E*K230D
yS257G*V258L*A259V*1260L
e-AChR Q
e(y)-2 0
cE227H*D229K 0
eR237Q 0
eG256S*L257V*V258A*L2591
50 Conductance (pS) 110
B y(e)-5 M3
316
Y P E T S AV P L[ KY L TF[LM V V[ IL I V[VNSVVV L N V S L R
e P E T S LV P L LGRY LI FV M V V AL I V MN CV IV L N V S L R
315
e(y)-5
y-AChR t
y(e)-5
yQ300L*A301 S*1305L*S306G*K307R *l
yT31OII
yT316A*1317T I
yT316A *
yI317T
yV321 M*S323C*V3251 I*
e-AChR Q
,6(y)-5 O
eL299Q*S300A*L3041*L305S*R306K 0
cL312V 0
eA315T 0 1
eM320V*C322S*1324V
50 Conductance (pS) 110
Figure 6. Effects of single- and multiple-point mutations on channel conductance in the Ml (A)
and in M3 (B) segments of y- and e-subunits
A, amino acid sequence of y- and e-subunits in the region comprising the Ml segment and adjacent bends.
Large box delineates Ml segment amino acids. Smaller boxes indicate amino acid positions where 'reverse'
mutants were generated. The graph below shows the conductance of Ml segment mutants (0, y-subunit
mutants; 0, c-subunit mutants); mean values of single-channel conductance at -100 mV in symmetrical
K+ (100 mM) solution. Numbers refer to amino acid positions according to Witzemann et al. (1990).
B, amino acid sequence of y- and c-subunits in the region comprising the M3 segment and adjacent bends.
Large box delineates the amino acids of the M3 segment. Smaller boxes indicate differences in amino acid
sequences that were reverse mutated. The graph below shows conductances of M3 segment mutants as in
A (0, y-subunit mutants; 0, c-subunit mutants); mean values at -100 mV in symmetrical K+ (100 mM)
solution.

11. Developmental AChR isoforms
e-subunit the exchange of glutamine for lysine residues
decreases the channel conductance (88-6 + 3-2 pS, n = 5),
whereas in the y-subunit, the exchange of lysines for
glutamines increases the conductance (76-3 + 1P6 pS, n = 6).
As expected, the conductances of the two mutants are still
different from those of the corresponding wild-type
channels. Comparison of single-point mutants indicates
that the exchange at the intracellular K/Q position in the
c-subunit produces a smaller change in conductance than a
substitution at the extracellular K/Q position (Fig.5C). The
different contribution to conductance of these two sites is
even more pronounced when comparing the effect at the
two y-subunit K/Q positions, where the intracellular
yK268Q mutant channel has a similar conductance to the
y-AChR channel.
Neutral amino acids in the M2 segment
Within the M2 segment the amino acid sequences differ at
three positions and point mutations in the M2 segment of
y- and e-subunits were made to investigate the effect of
removing these differences. At two of these positions
amino acids are located close to the narrow region of the
pore formed by the residues of the hydroxyl ring (Fig. 5A
and B; Villarroel et al. 1991a) where the amino acids differ
in hydrophobicity and polarity (A277 and T278 in the
y-subunit and S276 and I275 in the e-subunit). The double
mutation yA277S*T2781 produces a channel with a higher
conductance (76-8 + 1-37 pS, n= 4; Fig. 5C), while the
double mutant eS276A*I277T results in a channel with a
lower conductance (98'6 + 1 1 pS, n = 4; Fig.5C). Addition
of only one polar residue in the yA277S mutant subunit
fails to increase the channel conductance (66-8 pS, n = 2)
and replacement of the polar serine in the eS276A mutant
subunit is also without effect (103fi7 + 1 pS, n = 3). The
conductance changes of the double mutants are thus mainly
due to the changes in yT278 and cI277 residues since the
mutation cI277T decreases, whereas the mutation yT278I
increases, channel conductance (Fig. 5C). In addition to
the AT-SI exchange, there is also a difference between the
y- and e-subunit at a third position, at yV291 and cI290,
respectively. When cI290V is combined with eQ292K an
additional reduction in conductance is observed. The
conductance of the channel containing the mutant eQ292K
is 94-6 + 09 pS (n = 5), whereas that of the double mutant
eQ292K*I290V is 903 + 03 pS (n = 3). Therefore, most
of the differences in the amino acid composition of the M2
segment and the adjacent bends (Fig. 5B and C) are
determining the difference in conductance between the
y-AChR and e-AChR.
Amino acid differences in the MI and M3 segments
Point mutations in the Ml segment and its extracellular
border, summarized in Fig. 6A, show that conductance
differences are not changed significantly by exchange
substitutions. The conductances of mutant channels
yH228E*K230D and cE227H*D229K are 69-3 + 1-6 pS,
(n = 4) and 103X5 + 1.0 pS, (n = 3), respectively. Even
though a change in the charge of the amino acid residue is
involved, both values are not significantly different from
the values of y- and e-AChRs. A cluster of four amino
acid residues at positions 256-259 in the c-subunit is an
exception to the high degree of homology between y- and
e-subunits in Ml. The conductance of mutant
yS257G*V258L*A259V*I260L is only slightly higher
(71-2 pS, n = 2) than that of the y-AChR while the mutant
eG256S *L257V*V258A*L259I has a conductance of
100-2 + 07 pS (n = 3), which is slightly lower than the
c-AChR.
Point mutations in the M3 segment reversing differences
between the subunits are summarized in Fig. 6B. No effect
is observed after mutating the amino acids of the putative
intracellular bend bordering the M3 segment. A similar
negative result was obtained for most amino acid
exchanges in the M3 segment itself or with e-chimaera
e(y)-5 where the complete M3 sequence is swapped. An
exception is position yI317 which generates a channel with
a higher conductance in yI317T and the double mutant
yT316A*I317T, and thus may be responsible for the
slightly increased conductance mediated by the y(e)-5
construct.
DISCUSSION
Genomic organization and functional domains of
y- and e- subunits
Analysis of the genomic nucleotide sequence of the y- and
c-AChR subunit genes reveals that each subunit is encoded
by twelve exons. Both subunits are built in a modular form
with each of the four hydrophobic segments M1-M4, which
presumably span the membrane, encoded by a separate
exon. The exchange of the four hydrophobic segments in
the y(e)-1 and e(y)-1 chimaeric subunits was sufficient to
completely reverse the conductance properties of the
respective parental y- or e-AChR channel. This effect on
conductance was, however, not fully symmetrical because
replacement of all four M segments was necessary to obtain
the higher conductance channel, whereas the low
conductance channel could be produced by exchange ofonly
Ml and M2 together with either the M3 or the M4 segment.
Thus, the segments encoded by exons E7 and E8 are the
major determinants of the difference in conductance
between the fetal and adult muscle AChR channel.
Charged amino acids of M2 and channel conductance
Comparable with previous reports on recombinant AChR of
Torpedo californica electroplax (Imoto et at. 1988), on
hybrid recombinant AChR channels constructed from
mouse muscle and Torpedo subunits (Yu, Leonard,
Davidson & Lester, 1991) and fetal and adult mouse
chimaeric AChR (Bouzat, Bren & Sine, 1994), the number
of net negative charges in the cytoplasmic and extracellular
channel entrances (at the equivalent K/Q positions of M2)
confers differences in channel conductance. In the rat
J Physiol.492.3 785

12. S. Herlitze anid others
muscle AChR, channel the y-subunit contains positively
charged lysines (K) whlich in the c-subunit are replaced by
neutral glutainines (Q). The removal of the positive charges
in the mutant yK268Q*K293Q increases channel
conductance whereas including positive charges in the
mutant cQ267K* Q292K decreases channel conductance.
The positive charge of lysine could decrease the surface
potential at both clhannel entrances and repel cations.
Alternatively, charged amino acids in the cytoplasmic and
extracellular entrances may either keep neighbouring
subunits separated, and increase the pore size of the
channel, or a combination of positive and negative charges
may cause subunits to attract each other and decrease the
pore size.
Neutral amino acids of M2 and channel conductance
The a-helical, membrane-spanning structure of the M2
segment was initiallv deduced from the hydrophobicity
analysis of the subunit amino acid sequence (for review
see Numa, 1989). Mutagenesis experiments (Imoto et at.
1988; Charnet et at. 1990; Villarroel et at. 1991a,b) and
photolabelling with non-competitive inhibitors (Hucho,
Oberthiir & Lottspeich, 1986; Revah, Galzi, Giraudat,
Haumnont, Lederer & Changeux, 1990; Pedersen, Sharp,
Liu & Cohen, 1992; White & Cohen, 1992) have identified
amino acids within, and bordering, the M2 segment which
supposedly line the channel lumen. Interestingly, all these
amino acids, when replaced by cysteine, were accessible to
a positively charged, hydrophilic sulfhydryl-specific
reagent (Akabas, Kaufmann, Archdeacon & Karlin, 1994).
These exposed amino acids align in the y-AChR subunit
writh Q272 at the intracellular end, T275, N279, L282,
A283, V286, F289 within, and K293 at the extracellular
end of M2, assuming a symmetrical structural arrangement
of the five subunits (Unwin, 1995).
The residues at equivalent positions in the M2 segment,
yV291 and cI290, as well as yT278 and e1277, which
probably do not face the lumen of the channel (Akabas et al.
1994), contribute to the difference in conductance between
the y- and c-AChRs. This unexpected dependence of
channel conductance on the volume of the side-chain, where
increasing the side-chain of a lumen-exposed residue
should reduce conductance and vice versa, suggests that
c1277 is located in a region where the c-subunit is in contact
with its neighbouring subunit. The larger side-chain of
c1277 may prevent a tight packing of subunits when the
clhannel is open. The smaller side-chain of yT278, on the
other hand, may enable more compact packing, leaving less
space for ion movement.
Implications for AChR channel structure
The difference in ion conductance of the fetal and adult
AChR is determined to almost 70% by the M2 region,
flanked by the intra- and extracellularly located K/Q
position amino acids. This result is consistent with a
form the narrow portion of the open ion channel (Unwin,
1995). The amino acids at the K/Q position, located at the
carboxy-terminal end of M2, may contribute to the
formation of the 'extracellular funnel' of the channel
leading to the narrower constriction lined by amino acids
which are arranged on an a-helical structure. Complete
conversion between y-AChR and c-AChR, however,
requires the exchange of additional, presumably
membrane-spanning, domains indicating that the position
of amino acids lining the pore depends on interactions
between M2 and the other hydrophobic segments. Since
only the sequences contained in Ml, M3, and M4 contribute
to conductance differences, this supports the view that
these segments are located within the membrane. The Ml,
AM3, and M4 sequences, having either fl-sheet structure
(Unwin, 1995), or partial a-helical structure (Blanton &
Cohen, 1994; G6rne-Tschelnokow, Strecker, Kaduk,
Naumann & Hucho, 1994), may form a scaffold around the
central pore, and could interact with the a-helical M2 rods.
The extracellular amino-terminal ends of the y- and
c-subunits have little effect on conductance and are
apparently not involved in the conformational changes
which the receptor undergoes after agonist binding.
However, M3 and M4 with the putative intracellular
connecting bend, contain elements that influence gating
properties (Bouzat et al. 1994).
Conclusions
The results indicate that the difference in conductance
between the fetal and adult rat muscle AChR is determined
by structural differences in all four amino acid segments,
Ml1-M4, of the y- and e-subunits. The fact that each of the
M1-M4 segments is encoded by a separate exon supports
the view that the y- and c-subunit genes were generated
during evolution by gene duplication from a coinmon
precursor subunit gene, possibly the 8-subunit gene which
has the highest homology to both. The contribution of
different M segments to the differences in channel
conductance indicates that the major determinant is the M2
segment, encoded by exon E8. The dependence of other
differences in channel properties on this segment, such as
the fractional Ca2P current as well as the gating kinetics of
the channel, remains to be elucidated.
AKABAS, Xl. H., KAUFMIANN, C., ARCHDEACON, P. & KARLIN, A.
(1994). Identification of acetylcholine receptor channel-lining
residcues in the entire AI12 segment of the a sulbunit. Aeuroti 13,
919-927.
BLANTON, Ml. P. & COHEN, J. B. (1994). IdenitifyiIng the lipid-pr-otein
interface of the Torpedo nicotiniic acetyi lcolinie receptor: Secondary
structure implications. Biochemttistry 33, 2859-2872.
BOUZAT, C., BREN, N. & SINE, S. Xl. (1994). Structural basis of the
different gating kinetics of fetal and adult acetylcholine receptors.
channel structure where the amino acids of the M2 seginent
786 J Physiol.492.3
A'euroii 13, 1395-1402.

13. BUONANNO, A., MUDD), J. & MERLIE, J. P. (1989). Isolation and
characterization of the ,B and c subunit genes of mouse muscle
acetyicholine receptor. Journal of Biological Chemistry 264,
7611-7616.
CHARNET, P., LABARCA, C., LEONARD, R. J., VOGELAAR, N. J.,
CZYZYK, L., GOUIN, A., DAVIDSON, N. & LESTER, H. A. (1990). An
open-channel blocker interacts with adjacent turns of a-helices in
the nicotinic acetylcholine receptor. Neuroni 2, 87-95.
G;6RNE-TSCHELNOKOW, U., STRECKER, A., KADUK, C., NAUMANN, D.
& HuCHO, F. (1994). The transmembrane domains of the nicotinic
acetylcholine receptor contain a-helical and , structures. EMBO
Journal 13, 338-341.
Gu, Y. & HALL, Z. W. (1988). Immunological evidence for a change in
subunits of the acetylcholine receptor in developing and denervated
rat muscle. Neuron 1, 117-125.
HERLITZE, S. & KOENEN, M. (1990). A general and rapid mutagenesis
method using the polymerase chain reaction. Gene 91, 143-147.
Ho, S. N., HUNT, H. D., HORTON, R. M., PULLEN, J. K. & PEASE, L.
R. (1989). Directedc mutagenesis by overlap extension using the
polvrmerase chain reaction. Gene 77, 51-59.
HuCHO, F., OBERTHCR, XV. & LOTTSPEICH, F. (1986). The ion clhannel
of the nicotinic acetylcholine receptor is formed by homologous
lhelices MIII of the receptor subunits. FEBS Letters 205, 137-142.
1Luo3ro, K., BUSCH, C., SAKMANN, B., MISHINA, Al., KONNO, T.,
NAKAI, J., BuJo, H., AIORI, Y., FUKUDA, K. & NUMA, S. (1988).
Rings of negatively charged amino acids determine the
acetylcholine receptor channel conductance. Nature 335, 645-648.
IMOTO, K., METHFESSEL, C., SAKMANN, B., MISHINA, M., MORI, Y.,
KONNO, T., FUKUDA, K., KURASAKI, Al., BuJo, H., FUJITA, Y. &
NUMA, S. (1986). Location of a 8-subunit region determining ion
transport through the acetylcholine receptor channel. Nature 324,
670-674.
KONNO, T., BUSCH, C., VON KITZING, E., IMOTO, K., M ANG, F.,
NAKAI, J., MISHINA, M., NUMIA, S. & SAKMANN, B. (1991). Rings
of anionic amino acids as structural determinants of ion selectivity
in the acetylcholine r-eceptor channel. Proceedinigs of the Royal
Society B 244, 69-79.
KRIEG, P. A. & MELTON, D. A. (1984). Functional messenger RNAs
are produced b)y KSP6 ine vitro transcription of cloned cDNAs. Nucleic
Acids Research 12, 7057-7070.
LEONARD, R. J., LABARCA, C. G., CHARNET, P., DAVIDSON, N. &
LESTER, H. A. (1988). Evidence that the M2 membrane-spanning
region lines the ion clhannel pore of the nicotinic receptor. Scienice
242,1578-1581.
MELTON, D. A., KRIEG, P. A., REBAGLIATI, M. R., MANIATIS, T.,
ZINN, K. & GREEN, M. R. (1984). Efficient in vitro syntlhesis of
biologically active RNA and RNA hybridisation probes from
plasmids containing a bacteriophage SP6 promoter. Nucleic Acids
Research 12, 7035-7056.
METHFESSEL, C., WITZEMANN, V., TAKAHASHI, T., MISHINA, Al.,
NUMA, S. & SAKMANN, B. (1986). Patch clamp measurements on
Xenopus laevis oocvtes: currents through endogenous channels and
implanted acetylcholine ieceptor and sodium channels. Pfliigers
Arch iv 407, 57 -588.
AIISHINA, A., TAKAI, T., IMIOTO, K., NODA, Ml., TAKAHASHI, T., NUNIA,
S., METHFESSEL, C & SAKMANN, B. (1986). Mlolecular distinction
between fetal and adult forms of muscle acetylcholine receptor.
Nature 321, 406-41 1.
NEF, P., MAURON, A., STALDER, R., ALLIOD, C. & BALLIVET, M.
(1984). Structure, linkage, and sequence of the two genes encoding
the a and y subunits of the nicotinic acetvlcholine receptor.
787
NUMA, S. (1989). A molecular view of neurotransmnitter receptors and
ionic channels. IHarvey Lectures 83, 1 21- 1 65.
PEDERSEN, S. E., SHARP, S. D., Liu, W. S. & COHEN, J. B. (1992).
Structure of the non-competitive antagonist-binding site of the
Torpedo nicotinic acetylcholine receptor. [3H]-Meproadifen mustard
reacts selectively with a-subunit Glu-262. Journal of Biological
Chemistry 267, 10489-10499.
REVAH, F., GALZI, J.-L., GIRAUDAT, J., HAUMONT, P. Y., LEDERER, F.
& CHANGEUX, J.-P. (1990). The non -competitive blocker
[3H]-chlorpromazine labels three amino acids of the acetylcholine
receptor y subunit: Implications for the ox-helical organization of
regions MII and for the structure of the ion channel. Proceedings of
the National Academy of Sciences of the USA 87, 4675-4679.
SHIBAHARA, S., KUBO, T., PERSKI, H. J., TAKAHASHI, H., NODA, Al. &
NUMA, S. (1985). Cloning and sequence analysis of human genomic
DNA encoding y subunit precursor of muscle acetylcholine receptor.
European Journal of Biochemistry 146,15-22.
UNWIN, N. (1995). Acetylcholine receptor channel imaged in the open
state. Nature 373, 37-43.
VILLARROEL, A., HERLITZE, S., KOENEN, Al. & SAKMANN, B. (1991 a).
Location of a threonine residue in the a-subunit M2 transmembrane
segment that determines the ion flow thlough the acetvlcholine
receptor channel. Proceedings of the Royal Society B 243, 69-74.
VILLARROEL, A., HERLITZE, S. & SAKMANN, B. (1991 b). Amino acids
responsible for differences in conductance between adult and fetal
forms of the acetylcholine receptor. Biophysical Journal 59, 34a.
XVHITE, B. H. & COHEN, J. B. (1992). Agonist-induced chaniges in the
structure of the acetvlcholine receptor Al2 regions revealed by
photoincorporation of an uncharged nicotinic non-competitive
antagonist. Journal of Biological Chemistry 267, 15770-15783.
WXITZEMANN, V., BARG, B., NISHIKAWA, Y., SAKMANN, B. & NUMA, S.
(1987). Differential regulation of the y- and c-subunit mRNAs of
muscle acetvlcholine receptor. FEBS Letters 223, 104-112.
XVITZEMANN, V., STEIN, E., BARG, B., KONNO, T., CRIADO, M.,
KOENEN, A., KUES, W., HOFMANN, Al. & SAKMANN, B. (1990).
Primary structure and functional expression of the a-, ,-, y-, 5-
and c-subunits of the acetylcholine receptor from rat muscle.
European Journal of Biochemnistry 194, 437-448.
Yu, L., LEONARD, R. J., DAVIDSON, N. & LESTER, H. A. (1991). Single
channel properties of inouse Torpedo acetylcholine receptor hybrids
expressed in Xenopus oocytes. Molectilar Braini Research 10,
203-21 1.
Received 9 August 1995; accepted 30 Nlovemiber 1995.
Proceeding-s of the Nationial Academiy of Sciences of the USA 81,
7975-7979.
J Physiol.492.3 Developmental AChR isoforms