ArticlePDF Available

Perturbation of insulin-receptor interactions by intramolecular hormone cross-linking. Analysis of relative movement among residues A1, B1, and B29

Authors:

Abstract

We have evaluated, by use of isolated canine hepatocytes, the importance of intramolecular hormone cross-linking (and of concomitant changes in molecular flexibility) to the interaction of insulin with its plasma membrane receptor. Cross-linked hormone analogs were prepared by reacting porcine insulin, N alpha A1-t-butyloxycarbonyl insulin or N alpha A1-t-butyloxycarbonyl [D-LysA1]insulin with various dicarboxylic acid active esters to obtain alpha-GlyA1/epsilon-LysB29-, alpha-PheB1/epsilon-LysB29-, and epsilon-D-LysA1/epsilon-LysB29-cross-linked insulins, respectively. In the aggregate, insulin analogs cross-linked by groups containing 2-12 atoms retained 1.4-35% of the receptor binding potency of native insulin. Analysis of our results suggests that: (a) loss of chemical functionality, steric interference, and restriction of potential intramolecular movement can all play roles in determining the receptor binding potencies of cross-linked insulin analogs; (b) restriction of intramolecular movement between residues A1 and B29 affects negatively the binding of insulin to its receptor (but accounts for only a fraction of the conformational change which insulin must undergo to achieve a high affinity state of ligand-receptor interaction); and (c) introduction of a cross-link between residues B1 and B29 (residues that are in fact in proximity in one crystalline form of the hormone) decreases markedly the receptor binding potencies of the corresponding analogs. The importance of these findings is discussed in relation to the potential structure of insulin when it is bound to its plasma membrane receptor.
THE
JOURNAL
OF
BIOLOGICAL
CHEMISTRY
Q
1989
by
The American Society for Biochemistry and Molecular
Biology,
Inc.
Vol.
264,
No.
1, Issue
of
January
5,
pp.
272-279 1989
Printed
in
iT.S.A.
Perturbation
of
Insulin-Receptor Interactions
by
Intramolecular
Hormone Cross-linking
ANALYSIS
OF
RELATIVE MOVEMENT AMONG RESIDUES Al, B1, AND B29*
(Received for publication, May 27, 1988)
Satoe H. Nakagawa and Howard
S.
TagerS
From the Department
of
Biochemistry and Molecular Biology, The Uniuersity
of
Chicago, Chicago, Zllinois 60637
We
have evaluated, by use of isolated canine hepa-
tocytes, the importance of intramolecular hormone
cross-linking (and of concomitant changes in molecular
flexibility) to the interaction of insulin with its plasma
membrane receptor. Cross-linked hormone analogs
were prepared by reacting porcine insulin, WA'-t-bu-
tyloxycarbonyl insulin
or
WA'-t-butyloxycarbonyl
[D-
LysA']insulin with various dicarboxylic acid active es-
ters to obtain a-GlyA'/t-LysB2"-, a-PheB'/r-LysB2"-, and
t-D-LysA'/e-LysB2"-cross-linked
insulins, respectively.
In the aggregate, insulin analogs cross-linked by
groups containing
2-12
atoms retained
1.4-35%
of the
receptor binding potency of native insulin. Analysis of
our results suggests that:
(a)
loss of chemical function-
ality, steric interference, and restriction of potential
intramolecular movement can all play roles in deter-
mining the receptor binding potencies of cross-linked
insulin analogs;
(b)
restriction
of
intramolecular move-
ment between residues
A1
and
B29
affects negatively
the binding of insulin to its receptor (but accounts for
only a fraction of the conformational change which
insulin must undergo to achieve a high affinity state of
ligand-receptor interaction); and
(c)
introduction of a
cross-link between residues
B1
and
B29
(residues that
are in fact in proximity in one crystalline form of the
hormone) decreases markedly the receptor binding po-
tencies of the corresponding analogs. The importance
of these findings is discussed in relation to the potential
structure of insulin when it is bound to its plasma
membrane receptor.
Recent studies on insulin and on the nature of insulin-
receptor interactions have emphasized the flexibility of the
peptide hormone and its propensity to undergo structural
change. Crystallographic analysis, in particular, has demon-
strated structural differences
(a)
between molecules I and I1
of the 2-zinc porcine insulin hexamer (yielding molecules with
propagated structural changes in several domains) (1, 2),
(b)
between molecules
I
and
I1
of the 4-zinc porcine insulin
hexamer (resulting in the usually extended NHz-terminal
B-
chain domain assuming a helical conformation contiguous
with the central B-chain helix in molecule
11)
(3, 4), and
(c)
between hagfish insulin (5)
or
bovine des-pentapeptide-(B26-
*
These studies were supported by Grants DK18347 and DK20595
from
the National Institutes of Health. The costs
of
publication of
this article were defrayed in part by the payment of page charges.
This article must therefore be hereby marked "aduertisement" in
accordance with
18
U.S.C. Section 1734 solely
to
indicate this fact.
$
To whom correspondence should be addressed: Dept.
of
Biochem-
istry and Molecular Biology, The University
of
Chicago, 920 East
58th St., Chicago, IL 60637.
B3O)-insulin (6) and porcine insulin (leading for example to
the NHz-terminal B-chain domain pointing away from the
molecule in a different direction). While packing forces clearly
play a major role in determining the crystallographic struc-
tures
of
insulin, these studies show that even a globular
peptide hormone like insulin can exist in multiple conforma-
tions and suggest that the complex interactions of insulin
with its receptor may involve important degrees of confor-
mational change (7,
8).
The proximity of the amino terminus of the insulin A-chain
and of LysBZ9 in the COOH-terminal domain of the insulin
B-chain (in all crystallographic forms of insulin studied to
date, and apparently in insulin in solution as well, Refs.
2
and
9) has led investigators to study the consequences of restrict-
ing the extent
of
potential conformational change in insulin
by preparing analogs cross-linked between a-GlF' and
r-
LysBZ9. Many analogs have been examined, and all show
markedly decreased potency during interaction with the in-
sulin receptor (to about 1-10% of that observed
for
the native
hormone, Refs. 10-12). In some cases, these cross-linked
insulins have been shown to exhibit only small deviations
relative to insulin in their circular dichroic spectra (10, 13),
to aggregate to dimers and higher forms in ways similar to
those seen for insulin (14), and to refold to the correct
structure as the result of reduction and subsequent reforma-
tion of disulfide bonds (15). Most important, the crystal
structure of one of these cross-linked insulins
(NaA1,NcBZ9-
2,7-diaminosuberoyl insulin)
is
very nearly identical to that
of
the native hormone (16). While the decreased receptor
binding potencies of these cross-linked insulins have been
attributed to limitations in molecular flexibility,
(a)
they all
contain modifications of the A-chain a-amino group, a group
which itself has been shown (by derivatization, replacement,
deletion, or extention) to play an important role in directing
insulin-receptor interactions (17, 18); and
(b)
chemical mod-
ifications have themselves the potential for inducing small
changes in insulin structure even at relatively distant sites
(7). Related analogs also containing modifications at the
amino group at GlJP' have been synthesized and include
N"A',NfB29-s~beroyl [D-AlaA']insulin (19), NaA',NmB'-dodeca-
nedioyl insulin (13), and "miniproinsulins" in which GlF' is
in peptide bond with LysBZ9 (in cyclized de~-Ala~~~-insulin
(20))
or
with Arg2* (in cyclized
des-octapeptide-(B23-B30)-
insulin (21)); these cross-linked, modified insulins have been
reported
to
have 9,
0.5,
<0.1, and 0.02% of the potency of
native insulin, respectively.
Since conformational changes in insulin may be important
in determining high affinity ligand binding to receptor (7, 8,
22, 23) and since related considerations most probably apply
to other hormone-receptor systems as well (24-26), we under-
took to examine further the relationship between potential
272
Insulin-Receptor
Interactions
273
relative mobility in the A-chain NH2-terminal and B-chain
COOH-terminal domains and the ability
of
insulin to bind to
its receptor. Through the semisynthesis of [D-Lys*']insulin
(an analog containing a well tolerated amino acid replacement
(27)),
we were able to prepare cross-linked derivatives which
limit relative molecular movement between the a-carbons
of
residues
AI
and
B29,
but which retain the critical a-amino
group at position
Al.
We explore, in addition, potential struc-
tural relationships between insulin residues PheB' and LysBZ9
(residues that are in close proximity in molecule
I1
of
4-zinc
insulin or in all molecules
of
monoclinic insulin, see Refs.
3,
4 and
28)
during insulin interactions with its receptor.
EXPERIMENTAL
PROCEDURES
Materials-Monocomponent porcine insulin and [[1251]iodo-Tyfl'4]
insulin were obtained from Novo Research Institute (Copenhagen)
and Eli Lilly (Indianapolis, IN), respectively. Bis-N-hydroxysucci-
nimidyl suberate and
ethyleneglycolbis(N-hydroxysuccinimidyl)
succinate were purchased from Pierce.
Bis-N-hydroxysuccinimidyl
succinate (29), bis-p-nitrophenyl oxalate (30), and N-hydroxysucci-
nimidyl acetate (31) were synthesized according to published proce-
dures.
Bis-N-hydroxysuccinimidyl
adipate (melting point
=
189-
191 "C) was synthesized from adipic acid and N-hydroxysuccinimide
by use of
N,N'-dicyclohexylcarbodiimide.
Fmocl-N-hydroxysucci-
nimidyl ester and Na-Boc,N'-Fmoc-lysine (both
D-
and L-enantio-
mers) were purchased from Bachem (Torrence, CA). The N-hydrox-
ysuccinimidyl esters of the
D-
and L-enantiomers of N"-Boc,N'-Fmoc-
lysine were synthesized according to previously applied general pro-
cedures (32) (yield and melting point: 77% and 178-179 "C and
88%
and 178-179 "C, respectively).
General Methods-Unless noted otherwise, all procedures were
applied at room temperature. HPLC was performed by use of reverse-
phase C-18 columns (Altex, Ultrasphere, 5
p
particle size, 0.45
or
1.0
X
25 cm), a Perkin Elmer liquid chromatographic system, and a
solvent system consisting of acetonitrile and NaC104-containing tri-
ethylammonium phosphate buffer as described (33). Peptides were
detected by measuring optical absorbance at 214
or
276 nm and were
desalted (subsequent to HPLC purification and for other purposes as
well) by passage through columns of Bio-Gel P-4 (Bio-Rad) by use of
3
M
acetic acid as the solvent. Digestion of peptides by trypsin
occurred in 0.1
M
N-methylmorpholine buffer (brought to pH
8
by
the addition
of
acetic acid) containing 10 mM CaCL and
1
mM
ethylenediaminetetraacetic
acid by use of trypsin and peptides in the
ratio 7:lOO by weight. Oxidative sulfitolysis was performed as de-
scribed (34). Amino acid analyses were performed by use of a Beck-
man Model 119CL amino acid analyzer subsequent to peptide hy-
drolysis (6
N
HCl, 110 "C, 24 h). End group analysis was performed
by use of 2-dimethylaminonaphthalene-1-sulfonyl chloride (35), acid
hydrolysis, and thin layer chromatography (36). All insulin deriva-
tives were purified by use of preparative HPLC; purity was confirmed
by use of analytical HPLC. All derivatives exhibited the expected
amino acid compositions and NHZ-terminal residue(s). Details of the
preparation of isolated canine hepatocytes and of their use in binding
studies have been published (22,37). Cells were incubated in duplicate
with
[
['251]iodo-TyrA14]insulin
(about 20
PM)
and unlabeled insulin
or
insulin analogs at concentrations ranging from 0.01 nM to
1
pM
as
described (22, 23). Complete data sets were obtained for each analog
on several different occasions, and relative receptor binding potency
was determined by reference to the potency of insulin as measured in
the same experiment. Representative data are illustrated in Fig. 2,
and quantitative data are provided in Table I.
Preparation
of
a-G1fl'/e-LysSz9
Cross-linked Insulin.-The homo-
bifunctional reagents bis-p-nitrophenyl oxalate and the bis-N-hy-
droxysuccinimidyl derivatives of succinic acid, adipic acid, suberic
acid, and
ethyleneglycolbissuccinic
acid (5.5 pmol dissolved in
1
ml
of dimethyl sulfoxide) were separately added (in 10 equal portions
over
1-2
h) to 10-ml aliquots of dimethyl sulfoxide which contained
5
pmol of insulin and 10 pmol of N-methylmorpholine. The respective
mixtures were stirred for
18
h and were then freeze-dried, the residues
were dissolved in 3
M
acetic acid, and the resulting solutions were
gel-filtered by use of columns of Bio-Gel P-4 (equilibrated with 3
M
'
The abbreviations used are: Fmoc,
9-fluorenylmethyloxycarbonyl;
Boc, tert-butyloxycarbonyl; HPLC, high performance liquid chro-
matography.
acetic acid) to separate hormone dimers from monomers. The pooled
monomer-containing fractions were then divided into several por-
tions, and the cross-linked insulins were purified by HPLC. Material
in the major respective peaks was pooled, desalted, and freeze-dried
to obtain the product in each case. Yields of the final products ranged
from 20 to 24% of that expected, based on the amount
of
insulin used.
Preparation
of
e-D-Ly@'/c-LysBZ9-
and
~-Ly@~/c-Ly~~~-cros~-linked
Insulin.-N"A1-Phenylisothiocarbamyl
insulin was prepared by re-
acting insulin
(40
pmol) with phenylisothiocyanate (120 pmol) in a
mixture of methanol
(40
ml) and triethylamine (10 ml) adjusted to
pH 9.3 with HCl for
2
h (38). The major product was purified by
HPLC in 38% final yield. The N"-GlyA'-protected insulin (14 pmol)
was dissolved in 3 ml of dimethylformamide, and Fmoc-N-hydroxy-
succinimidyl ester (280 pmol) and N-methylmorpholine
(56
pmol)
were added. After 3 h of incubation with stirring, the product was
precipitated with diethyl ether and was dried. The resulting
NnA1-
phenylisothiocarbamyl,NaB',iVfBzg-bis-Fmoc
insulin was treated with
trifluoroacetic acid
(10
ml,
1
h), desalted, and freeze-dried to yield
the corresponding des-GlyA' insulin derivative in 85% yield. Portions
of the derivative (13 pmol) were reacted during 3 h with N"-Boc,N"
Fmoc-D-lysine N-hydroxysuccinimidyl ester
(or
its L-lysine counter-
part) (130 pmol) in 2.5 ml of dimethylformamide containing
N-
methylmorpholine (26 pmol). The product was precipitated by use of
diethyl ether, dried, and treated with
piperidine:dimethylformamide
10:90 (6 ml, 15 min,
4
"C) (39). The resulting solution was diluted
with ice-cold 3
M
acetic acid and was desalted. Peptide-containing
fractions were freeze-dried, and the crude product was purified by
HPLC. Based on the amount
of
NmA1-phenylisothiocarbamyl
insulin
used, the final yields of NUA1-Boc [D-LysA']insulin and N"*'-Boc
[LysA']insulin were 51 and 41%, respectively.
A small amount of each product was treated with trifluoroacetic
acid to obtain the corresponding unprotected insulin analogs. To
obtain the corresponding cross-linked analogs, the a-N-Boc-pro-
tected, substituted insulins
(2
pmol) were separately treated with
various homobifunctional reagents as described earlier for a-GlP'/e-
LysBm-cross-linked insulins. The resulting products were treated with
trifluoroacetic acid to remove the protecting group and were subse-
quently purified by HPLC.
e-D-LysA'/t-LysB29-cross-linked
insulins
bearing the oxaloyl, succinoyl, suberoyl,
or
ethyleneglycolbissuccinoyl
linkages were obtained in 13-32% yield. NfA1,NfBm-Suberoyl [LysA1]
insulin was isolated in 17% yield.
Preparation
of
a-Phes'/c-Ly~2s-cross-linked
Insulin.-Insulin (20
pmol, dissolved in a mixture of dimethyl sulfoxide (10 ml) and
N-
methylmorpholine (40 pmol)) was reacted with 2-Boc-oxyimino-2-
phenylacetonitrile (25 pmol) during
3
h with stirring. The product
was freeze-dried, desalted, and purified by HPLC to yield NaA1-Boc
insulin in 29% yield. To obtain cross-linked analogs, portions of the
protected insulin (2.7 pmol) were separately treated with various
homobifunctional reagents as described earlier for a-Gl~'/c-LysBZg-
cross-linked insulins. The Boc group was then removed by treatment
with trifluoroacetic acid, and the crude peptides were purified by
HPLC. The isolated yields of the succinoyl, suberoyl, and ethylene-
glycolbissuccinoyl derivatives were 6.7, 17, and 13%, respectively.
Preparation
of
Bisacetyl Insulin.-Insulin (20 pmol, dissolved in a
mixture of dimethylformamide (2.4 ml), water (1.2 ml), and
1
M
NaHC03 (0.4 ml), see Ref. 40) was reacted with 2-Boc-oxyimino-2-
phenylacetonitrile (25 pmol dissolved in
1
ml of dimethylformamide)
during 3 h with stirring. The reaction mixture was diluted with 3
M
acetic acid and was desalted. Peptide-containing fractions were pooled
and freeze-dried, and the resulting crude material was purified by
HPLC. The yield of NmB1-Boc insulin was 37%. NeB1-Boc insulin,
NUA'-Boc [D-LysA']insulin (see earlier),
or
NUA'-Boc insulin (see ear-
lier) were converted to their corresponding bisacetyl derivatives by
treating the respective, protective peptides
(1
pmol, dissolved in 0.5
ml of dimethyl sulfoxide) with N-methylmorpholine (5
pl)
and
N-
hydroxysuccinimidyl acetate (10 pmol) during
2
h. The resulting
derivatized peptides were precipitated with diethyl ether, dissolved in
trifluoroacetic acid to remove the protecting group (0.5 ml,
1
h,
4
"C),
and purified by HPLC. The yields of the final products ranged from
23 to 42%.
RESULTS
As illustrated in Fig.
1,
our approach to the semisynthesis
of
insulin analogs containing intramolecular cross-links de-
pended on well known procedures
for
the preparation of the
a-G1~'/~-LysBZ9-~ross-linked
analogs directly from insulin
274
Insulin-Receptor Interactions
a-Gly*/s-LysBZ9
cross-link
Insulin
+
.
N~-phenylthiocarbyl-insulin
cross-link
bis-Fmoc-insulin
insulin
#
cross-link
FIG.
1.
Diagrammatic scheme for the semisynthesis of
cross-linked insulin analogs.
Heavy arrows
indicate steps involv-
ing introduction of the cross-link itself (sometimes necessarily fol-
lowed by an additional step of deprotection);
light
arrows
indicate
various other chemical reactions. See "Experimental Procedures" and
"Results" for details. The figure shows
(a)
how insulin analogs cross-
linked between a-N-Glp and c-N-LysBm were synthesized directly
from insulin,
(b)
how insulin analogs cross-linked between a-N-PheB'
and c-N-LysBZ9 were synthesized from a protected intermediate, and
(c) how insulin analogs cross-linked between C-N-D-LYS~'
(or
e-N-
LysA') and c-N-LysBZ9 were synthesized via a series of intermediates
which correspond to various protected
or
otherwise derivatized pep-
tide forms.
group is used both as an a-amino protecting group and as an
activator for peptide bond cleavage during Edman degrada-
tion,
(b)
the
a-
and c-amino groups
of
PheB' and LysBZ9,
respectively, are protected by the Fmoc group to ensure mono-
derivatization of the relevant des-Glfl' analog by protected
D-
or
L-lysine-N-hydroxysuccinimidyl
esters, and
(c)
the
cross-linked product is deprotected quite simply to remove
the Boc group on
a-N-D-
or
CU-N-L-LYS.
Fig. 2a and Table I provide information on the receptor
binding potencies of insulin analogs that have been intramo-
lecularly cross-linked between a-GlyA' and t-LysBz9.
As
is
readily apparent, simple derivatization
of
the two amino
groups to yield N"A',N'B29-bisa~etyl insulin (Peptide
2)
de-
creases the apparent receptor binding potency of the uncross-
linked analog nearly 4-fold relative to native insulin.
It
is thus
clear that even in the absence
of
a cross-link, loss of the
a-
amino function at Glfl' is deleterious to insulin-receptor
interactions (2,12,31). Also as illustrated in Fig. 2a and Table
I, introduction
of
intramolecular cross-links between a-Glf'
(10,13,29,41), on simple protection at Glfl' for the semisyn-
thesis of the
a-PheB'/t-LysBzg-cross-linked
analogs, and on a
more detailed protocol for the semisynthesis of analogs cross-
linked between the t-amino groups of D-LYs*' and LysBZ9 in
[D-LysA']insulin. The low reactivity of the PheB' a-amino
group under the conditions used for cross-linking permitted
the synthesis of position A1-B29 cross-linked analogs in the
first and third cases without need for protection of PheB1. In
these cases, of course, some derivatization of PheB' occurred,
and in all cases some mono-derivatization, bis-derivatization
(but without intramolecular cross-linking), and dimerization
was observed. Nevertheless, the resulting side products were
readily removed by HPLC
on
reverse-phase columns, and the
major intramolecularly cross-linked products were readily
identified by end group analysis, by use
of
oxidative sulfitol-
ysis and gel filtration (as appropriate) to determine the num-
ber of insulin chains remaining after disulfide bonds were
broken and by digestion with trypsin and subsequent analysis
by HPLC to assess the succeptibility of the Ar.g22-GlyZ3 peptide
bond to enzymatic cleavage; in agreement with a previous
report (42), the apparent susceptibility of the Arg2'-GlyZ3
peptide bond to cleavage by trypsin was markedly reduced
or
was absent in all cross-linked insulins examined (data not
shown).
In the procedure outlined in Fig.
1
for the semisynthesis of
t-LysA'/t-LysBZ9-cross-linked
insulins (one adapted from pro-
cedures used by others to prepare insulin analogs with replace-
ments at position
A1
(38)),
(a)
the phenylisothiocarbamyl
g
u)
L
Log
molar
peptide
concentration
FIG.
2.
Inhibition of
[['aaI]iodo-TyrA14]insulin
binding to
isolated canine hepatocytes by cross-linked insulin analogs.
Analogs and radiolabeled insulin were incubated with canine hepa-
tocytes as described under "Experimental Procedures." Quantitative
information and further details are provided in Table
I;
identifying
numbers from Table
I,
rather than symbols, are used to indicate the
peptide under consideration.
Panel a,
insulin analogs cross-linked
between a-Glfl' and c-LysBB:
1,
insulin;
2,
NaA',NmB-bisacetyl in-
sulin;
7,
NaA1,NrBZs-~~aloyl insulin;
8,
N"A',NfB2g-s~~~inoyl insulin;
10,
N*1,NcBz9-s~ber~yl insulin;
11,
NaA',NtBZ9-ethyleneglycolbissuc-
cinoyl insulin. An additional member of this series (NaA',NfBm-adi-
poyl insulin, Peptide
9,
see Table
I)
has been omitted from the figure
cross-linked between c-D-L~s~' and c-LysBB:
1,
insulin;
6,
NfA1,NfBB-
in order to improve clarity of presentation.
Panel
b,
insulin analogs
bisacetyl [D-LysA']insulin;
12,
~A',NfB29-oxaloyl [D-LysA']insulin;
13,
NeA',NfBZ9-s~~~in~yl [D-LysA']insulin;
14,
NfA', NfBB-suberoyl [D-
LysA1]insulin;
15,
NfA1,
NfBB-ethyleneglycolbissuccinoyl
[D-LySA1]
insulin.
Panel
c, insulin analogs cross-linked between a-PheB' and
c-
LysBB:
1,
insulin;
3,
NmB',NBm-bisacetyl insulin;
17,
NmB1,NrBz9-s~~-
cinoyl insulin;
18,
NaB',NfBm-suberoyl insulin;
19,
NaB',NtBZ9-ethyle-
neglycolbissuccinoyl insulin.
Insulin-Receptor Interactions
TABLE
I
Identification and receptor binding potencies of insulin analogs
The semisynthetic insulin analogs prepared for this study and their receptor binding potencies (relative to the
potency of porcine insulin) are identified below. Details of the semisynthetic methods used and of the cell
preparation and incubation conditions are provided under "Experimental Procedures."
275
Receptor binding potency
Identifying number and peptide
Relative to insulin"
Uncross-linked insulins
1.
Insulin
2.
NnA',N'B29-Bisa~etyl insulin
3.
N"B',N'B2g-Bisacetyl insulin
4.
[LysA']Insulin
5.
[D-LysA']Insulin
6.
N'A',N'BZg-Bisacetyl [D-LysA']insulin
7.
N"A',~B29-0~aloyl insulin
8.
N"A',N'B29-S~~~inoyl insulin
9.
NaA',WBm-Adipoyl insulin
10.
NaA',N'B"-Suberoyl insulin
11.
N"A',N'BZg-Ethyleneglycolbissuccinoyl
insulin
12.
N'A',N'B2g-O~aloyl [D-LysA']insulin
13.
N'A',N'B29-S~~~inoyl [~-Lys~']insulin
14.
N'A',WB29-Suberoyl [D-LysA']insulin
15.
N'A'.N'B29-Eth~lenedscolbissuccinorl
Cross-linked insulins
100
27
f
6 (3)
84
f
10 (5)
24
f
1 (2)
91
f
11 (5)
88
*
13 (3)
3.3
f
1.2 (4)
1.4
f
0.5 (7)
3.8
f
0.8
(2)
9.5
f
3.3 (6)
11
f
3
(4)
15
f
0.5
(4)
17
f
4 (5)
24
2
1
(5)
35
f
3 (3)
Relative to parent- Length
of
cross-link'
derivatized
insulinb
12
14
35
40
17
19
27
40
36
5.2
5.4
2.6
13
[D-~ys~~linsulin
-
"
16.
N'A',N'B29-S~beroyl [LysA']insulin
8.7
f
1.7 (2)
17.
NQB',N'B29-S~~~inoyl insulin
4.5
f
1 (6)
18.
N"B',N'B29-S~beroyl insulin
2.2
f
0.8
(5)
19.
N"B',N'B29-Ethyleneglycolbissuccinoyl
insulin
11
f
2 (3)
a
Determined as ((concentration of peptide causing half-maximal inhibition of binding of [["61]iodo-Ty~'4]
insulin to
receptor)/(concentration
of insulin causing half-maximal inhibition of binding of
[
['T]i~do-Ty@'~]
insulin to receptor in the same experiment))
X
100.
All inhibitions were complete and all curves describing the
concentration dependence for the inhibition of radiolabeled insulin binding to receptor were parallel. Each value
represents the mean
+-
S.D.
of multiple determinations; the number of separate determinations (each with complete
data sets obtained in duplicate) is shown in parentheses. The concentration of insulin causing half-maximal
inhibition of radiolabeled insulin binding was
0.63
f
0.13
nM (n
=
17).
Since
510%
of the
20
pM radiolabeled
insulin became cell-associated in the experiments reported (even in the absence of competitor), the data were not
significantly affected by decreases in ligand concentration arising from hormone binding to receptor. The relative
receptor binding potencies reported in the table can therefore be considered to reflect relative binding affinities.
*
Determined as ((receptor binding potency of cross-linked peptide relative to insulin)/(receptor binding potency
of appropriate, derivatized parent insulin analog relative to insulin))
X
100.
Peptide
2
was considered to be the
parent of cross-linked Peptides
7-11;
Peptide
3,
of cross-linked Peptides
17-19;
Peptide
4,
of cross-linked Peptide
16;
and Peptide
6,
of cross-linked Peptides
12-15.
The peptide N'A',WB29-bisa~etyl [LysA']insulin was not available
for study, but most probably possesses a relative receptor binding potency indistinguishable from that of [LysA']
insulin; cf. results with Peptides
5
and
6.
e
The length of the chemical cross-link is given as the number of atoms occurring only in the inter-amino group
linkage. The length of the cross-link shown in parentheses corresponds to the number of atoms
(C,
N, and
0,
as
appropriate) which bridge the a-carbon atoms of the
2
relevant amino acid residues. For example, for Peptide
7,4
C
+
1
N
(from LysBZ9)
+
2
C
(from the oxaloyl group)
+
1
N (from Glfl')
=
8;
for Peptide
12,
4
C
+
1
N
(from
LysBZ9)
+
2
C
(from the oxaloyl group)
+
4
C
+
1
N
(from D-LYs~')
=
12.
and t-LysBZ9 by groups containing
2-12
atoms yielded analogs
(Peptides 7-11) with 1.4-11% of the receptor binding potency
of insulin. In all cases, introduction of the cross-link also
decreased receptor binding potency relative to the uncross-
linked bisacetyl derivative, but the decrease ranged from 2.5-
to 19-fold depending on the length of the cross-linker used.
Whereas the decrement in relative binding potency with
decreasing cross-linker length follows a progression for cross-
linkers decreasing in length from
12
to 4 atoms, the unex-
pectedly high affinity of the receptor for N"A',NLB29-o~aloyl
insulin
(or
perhaps the unexpectedly low affinity of the recep-
tor
for
N"A',N'B29-s~~~in~y1 insulin) has been noted before
(12).
All considered, it is clear that intramolecular cross-
linking between a-GlyA' and t-LysBZ9 decreases the ability of
the corresponding analogs to bind to receptor (whether the
comparison is made by reference to insulin or to the parent
bis-derivatized, but uncross-linked, bisacetyl insulin, see Ta-
ble I). The comparison is made more complicated, however,
by the major decrease in receptor binding potency that attends
acylation of the a-amino group at GlyA'.
To
approach the analysis of insulin analogs cross-linked
between residues A1 and B29, but simultaneously retaining
the position
A1
a-amino function, we considered the synthesis
of analogs in which GlF' was replaced by Lys and in which
cross-linking occurred between the t-amino groups at posi-
tions
A1
and B29. Table I provides information on the relative
receptor binding potencies of the two possible substituted
insulins: [LysA']insulin (Peptide 4) and [D-LysA']insulin
(Peptide
5).
As
expected from the results of others (27,
43),
replacement of GlyA' by the L-amino acid yielded an analog
with markedly decreased receptor binding potency (24% of
that of insulin), whereas replacement
of
G19' by the D-amino
acid yielded an analog which bound to the receptor with very
nearly the same affinity as insulin itself.
As
important (Table
I and Fig.
2b),
the bis-derivatized analog N'A',N'B29-bisa~etyl
[D-LysA']insulin (Peptide
6,
a peptide that might be consid-
2
76
Insulin-Receptor Interactions
ered to be the parent compound of analogs in which the
t-
amino groups of D-LYs~' and LysBZ9 are cross-linked) exhib-
ited
a
receptor binding potency essentially not different from
that of the underivatized substituted insulin. Thus, [D-LysA']
insulin represents a very favorable analog for ascertaining the
importance of potential restrictions in intramolecular move-
ment (by hormone cross-linking) on the affinity of the recep-
tor for insulin.
Fig.
2b
and Table I provide information on the relative
receptor binding potencies of insulin analogs in which residues
D-LYs~' and LysBZ9 are cross-linked via their e-amino groups
by linkages containing
2-12
atoms (Peptides
12-15).
The
analogs retain
1535%
of the receptor binding potency of
insulin and 17-40% of the receptor binding potency of
NfA',NfBZ9-bisacetyl [D-LysA']insu1in (Peptide
6)
(values little
different from each other since the bisacetyl derivative in this
case retains the critical a-amino group at position
A1
and
since acylation of the relevant t-amino groups at positions
A1
and
B29
has only a small effect on the ability of the corre-
sponding analogs to bind to receptor). Most important, the
relative receptor binding potencies of the t-D-LysA'/t-LysBZ9-
cross-linked insulins are considerably higher than those of
the
a-Glfl'/e-LysB29-~ross-linked
insulins
(cf.
Fig.
2,
a
and
b,
and Table
I),
and the decrement in relative binding potency
with decreasing cross-linker length is seen for t-~-Lp*'/t-
Ly~~~~-cross-linked insulins to follow a monotonic progression
for all cross-linkers decreasing in length from
12
to
2.
Fig.
3a
plots the relative receptor binding affinities of
a-
Glfl'/t-LysBZ9- and
t-D-LysA'/t-LysB29-~ross-linked
insulins
against the relative intramolecular distance between relevant
a-carbon atoms. Several relationships should be noted. First,
as described earlier, the relative potencies of the t-~-Lys"'/t-
Ly~~~~-cross-linked derivatives exceed those of the a-GlyA'/t-
Ly~~~~-cross-linked derivatives, but they do
so
by only
2-4-
40r"--
'8 12
16
20
"--A
24
8
Relative intramolecular distance
FIG.
3.
Relationship between length
of
cross-link and rela-
tive receptor binding potency among insulin analogs cross-
linked between residues
A1
and
B29.
Data are shown for analogs
cross-linked between cr-GlyA' and c-LysBZ9
(O),
and for analogs cross-
linked between c-D-LYs~' and e-LysBZg
(0).
Relative intramolecular
distance is considered to be the number
of
atoms occurring between
the a-carbons
of
the relevant cross-linked residues
in
both cases; that
is, the number of atoms introduced by application of the cross-linking
agent plus the number of atoms appearing in relevant sidechains or
functional groups.
Panel
a,
data for the analogs are plotted as percent
of control receptor binding affinity (that exhibited by insulin).
Panel
b,
data for the analogs are plotted as the logarithm of the percent of
control receptor binding affinity. See text, Table
I,
and footnotes to
Table
I
for further details.
fold for analogs containing the same number of atoms in the
intramolecular cross-link. Second, the effect of increasing the
length of cross-linker at longer cross-linker lengths is seen to
become blunted in the
a-GlyA'/t-LysBZ9-cross-linked
analogs,
whereas it remains robust in the case of the t-~-Lys~'/t-
LysBZ9-cross-linked analogs. Third, the effect of decreasing
the length of cross-linker at shorter cross-linker lengths is
seen apparently to differ considerably for each set of analogs.
The difference between a-Glfl'/t-LysBZ9- and t-~-Lys~'/t-
Ly~~~~-cross-linked analogs is emphasized more strongly by
plotting the logarithm of relative binding affinity
uersus
rel-
ative cross-linker length. Indeed, the linear relationship illus-
trated in Fig.
3b
for
t-~-Lys~'/t-Lys~~~-~ross-linked
insulins
(one presumably arising from the proportionality between the
free energy of ligand-receptor interaction and the logarithm
of the affinity constant) suggests for this case that changes in
receptor binding potency arise quite simply from changes in
cross-linker length and from concomitant changes in molec-
ular flexibility. Relationships among a-Gl~'/t-LysBZ9-~ross-
linked insulins are clearly more complex.
Notwithstanding the markedly decreased receptor binding
potency of [LysA']insulin (Peptide
4),
introduction of the
suberoyl linker between the t-amino groups
of
LysA' and
LysBZ9 in this low affinity analog (to yield N'A',N'B29-s~beroyl
[LysA']insulin, Peptide
16,
Table
I)
decreased by nearly
3-
fold the affinity of the insulin receptor for the resulting cross-
linked peptide. This value is little different from the corre-
sponding decrement recorded for the suberoyl derivative of
[D-LysA']insulin and suggests again that the introduction of
the cross-link, rather than details of molecular configuration,
determine the ultimate receptor binding potencies of insulin
analogs cross-linked between the t-amino groups of
D-
or
L-
LysA' and LysBZ9.
To complete our analysis of the importance of potentially
restricting intramolecular motion on the receptor binding
potencies of cross-linked insulin analogs, we synthesized
fi-
nally a series of analogs in which the a-amino group of PheB'
was cross-linked to the t-amino group of LysB2'. Information
provided in Fig.
2c
and Table I identifies that
(a)
the parent
bis-derivatized, but uncross-linked analog NuB',NfBZ9-bisace-
tyl-insulin (Peptide
3)
retains as much as 84% of the potency
of insulin during interactions with the receptor,
(b)
analogs
cross-linked between residues a-PheB' and t-LysBZ9 by linkers
containing 4-12 atoms exhibit only
2.2-11%
of the potency
of insulin, and
(c)
after correcting for the receptor binding
potencies of the corresponding bisacetyl-insulin derivatives,
the
a-PheB'/t-LysB29-~ross-linked
insulins in fact exhibit
receptor binding potencies considerably lower than those ap-
plicable to derivatives cross-linked between the amino groups
of a-Glf' and t-LysBZ9 (see Table I). It thus appears that,
notwithstanding the proximity of the a-PheB' and t-LysB2'
amino groups in molecule I1 of 4-zinc insulin
(3,
4),
cross-
linking of the
2
residues decreases substantially the affinity
of the insulin receptor for the corresponding analogs. It should
be noted that, despite several attempts, we were unable to
prepare either NuB1,N'B29-o~aloyl insulin
or
insulin analogs
cross-linked between the amino groups of a-Glfl' and
a-
PheB' by linkers containing from
4
to
12
atoms. Nevertheless,
the semisynthesis of insulin analogs cross-linked between
positions a-Glfl' and a-PheB' has been reported
(13).
DISCUSSION
As
illustrated in Fig. 4a, residues
A1
and
B29
are rather
close to each other in all structures of insulin determined
(1-
5, 28);
the proximity and functionality of Glfl' and LysBZ9
allow the
2
residues to be cross-linked readily and allow
Insulin-Receptor Interactions
277
FIG.
4.
Diagrammatic representations
of
the structures of
insulins that have been modified
by
the introduction of var-
ious cross-links. The
basic
view
of
the molecule is
taken
from
the
crystallographic
studies
of
others
(1-3)
and
corresponds to the
insulin
monomer viewed
in
a
direction perpendicular to the
3-fold
axis
in
the
2-zinc
insulin
hexamer. Pictorial
aspects
of
the structures
are
taken
from
Ref.
8
and
are
identified
as
follows:
bars
with horizontal
hatching,
helices
within the
insulin
A
chain;
bar with
uertical
hatching,
the
helix
within
the
insulin
B
chain;
open
bar,
the sheet-like region in
the COOH-terminal
domain
of
the
insulin
B
chain; light
lines,
other
portions
of
the insulin
molecule;
heauy
lines,
cross-links
introduced
into the structure
of
insulin.
a,
insulin
cross-linked between residues
a-GlP
and
t-LysB";
b,
insulin
cross-linked
between
residues
c-D-
LysA'
and
t-LysBZ9;
c,
insulin
cross-linked between residues a-PheB'
and
a-GlP;
d,
insulin
cross-linked between
residues
a-PheB'
and
t-
LysBZ9.
The positions
of
the
A-chain
helices
in
c
and
d
have
been
adjusted
somewhat (relative to those
in
a
and
b)
to account
for
movements that might
mirror
those
occurring
in
molecule I1
of
4-
zinc
insulin
(a
structure
in
which
the amino terminus
of
the B-chain
is
found
to
approach
the
amino
and
carboxyl termini
of
the
A-
and
B-chains,
respectively,
Refs.
3
and
4).
It
should
be
emphasized that,
with the exception
of
NnA1,NrB"-2,7,-diaminosuberoyl
insulin
(16)
(an
analog
having
the
general
form
illustrated
in
a),
the structures
of
these
cross-linked
insulins
have
not
been
determined. The figure
must therefore
be
regarded
as
being
very
approximate.
questions to be asked concerning relative molecular move-
ment between the A-chain amino terminus and the COOH-
terminal domain of the B-chain. The stage is best set by
reference to
NQA',NrBpg-2,7-diaminosuberoyl
insulin, a cross-
linked analog that has been shown to exhibit normal crystal-
lographic (10,16) and solution structures
(2,
lo), to aggregate
normally (14), and to refold subsequent to reduction of disul-
fide bonds (10). Nevertheless, the analog binds to the insulin
receptor with an affinity only about 5% of that exhibited by
insulin; the decrease (or at least a portion of it) has been
ascribed to limitations in molecular flexibility (16). Since
(a)
the a-amino group of Glfl' has been shown to be critically
important in insulin-receptor interactions,
(b)
this group is
modified by introduction of a cross-link between residues
G1P and LysB2' (in the above example and also in the analogs
described in Fig. 2a), and
(c)
modification of the group might
well affect interpretations based on restriction of potential
intramolecular movements, we undertook to construct related
cross-linked insulin analogs in which the a-amino group at
position
A1
remained free.
As
documented under "Results,"
analogs constructed by cross-linking the €-amino groups of
D-
LysA' and LysBZ9 in [D-LysA']insulin (illustrated in Fig. 4b
fulfill all necessary criteria. Importantly, chemical derivati-
zation of these two groups has little
or
no effect on the receptor
binding potency of the parent insulin. Effects of cross-linking
in these cases can therefore be attributed exclusively to those
of the cross-link itself.
As
presented under "Results," a-Glf'/t-LysBz9- and
t-D-
LysA'/t-LysB29-~ross-linked
insulins differ considerably in
terms of both their absolute and relative potencies for inter-
action with the insulin receptor and their relative dependen-
cies on cross-linker length. Analysis of the data suggests that
(a)
derivatization
of
the a-amino group of Glfl' in the former
set of analogs causes a major decrease in receptor binding
potency, a decrease which apparently accounts for up to
75%
of the decreased potency of
a-G1fl'/t-LysB29-~ross-linked
in-
sulins;
(b)
the additional small, but significant, decrement in
the receptor binding potency of analogs cross-linked between
a-Glfl' and t-LysBZ9 by long cross-linkers (see Table I) ap-
parently arises from structural alterations which do not in-
volve derivatization at Glp
per
se;
(c)
decreasing the length
of the cross-linker results in an additional and abrupt decrease
in receptor binding potency in
~~-G1fl'/t-Lys~~~-~ross-linked
insulins; and
(d)
insulin analogs cross-linked between
t-~-
LysA' and t-LysBZ9 retain at longer cross-linker lengths high
receptor binding potency, exhibit progressively decreased
receptor binding potencies as cross-linker length is decreased,
and retain as much as 12% of the receptor binding potency of
insulin at extrapolated short cross-linker lengths. Since the
effective length of the cross-link in
t-D-~ySA1/t-~ysBz9-CrOss-
linked insulins (relative to that in a-Glfl'/~-LysBZ9-~ross-
linked insulins) might actually be expected to be somewhat
less than that illustrated in Fig.
3
(due to the absence of a
direct path), the simple relationship shown by the upper
curves is all the more notable.
Findings discussed above suggest that the decreased recep-
tor binding potencies of cross-linked insulin analogs can arise
in various proportions from chemical modification of impor-
tant functional groups, steric interference, and restriction of
favorable molecular flexibility. While cross-linking might also
induce unfavorable conformations (as the result of intramo-
lecular strain that is dissipated over long distances), the
relative ease with which most cross-linked insulins are syn-
thesized indicates that residues
A1
and B29 can be very close.
Our results show, on one hand, that the decreased receptor
binding potencies of
a-Gl~/t-LysBZ9-cross-linked
insulins
seem to arise from chemical modification at Glfl', from steric
interference (as assessed by use of longer cross-linkers) and
from the restriction of molecular flexibility (as assessed by
the use of shorter linkers); in the last regard, it should be
noted that whereas the suberoyl group would apparently
fit
nicely between a-Glp and t-LysBS in insulin's crystallo-
graphic structure (16,44), shorter cross-linking groups would
not. Our results show, on the other hand, that the decreased
receptor binding potencies of
t-~-Lys~~~/t-Lys~~~-cross-linked
insulins most probably arise mainly as the result of restricting
intramolecular motion. Taken together, these findings suggest
that such motion with regard to the relative positions of
residues
A1
and B29 indeed seems to play a role in determin-
ing the ability of insulin to bind to its receptor. Nevertheless,
calculations based on the analysis of apparent association
constants suggests that it plays a role which could account
for only about 10% of the binding energy defining insulin-
receptor interactions. An additional 10-15% of the binding
energy may arise as the result of separate intramolecular
movements involving the disposition of the insulin mainchain
in the region of the TyrBZ6-Th? peptide bond (22, 23).
Whereas the significance of intramolecular movements in-
volving the COOH-terminal B-chain domain have long been
278
Insulin-Receptor
Interactions
postulated, related movements involving the NHz-terminal
domain of the B-chain have actually been observed. That is,
residues Bl-B4 appear in slightly to very substantially differ-
ent relationships relative to the rest of the molecule among
the crystallographically determined structures of des-penta-
peptide-(B26-B30)-insulin
(6), hagfish insulin
(5),
and mole-
cules
I
and
I1
of both 2-zinc and 4-zinc porcine or human
insulin (1,3,4,8). Molecule
I1
of 4-zinc insulin represents the
extreme example. In this case, the amino terminus of the B-
chain (a portion of the molecule rather distant from residues
A1
and B29 in other structures) is brought, by extension of
the main B-chain helix, into proximity with residues Gl9'
and LysB29
(3,
4). Most important, the necessary changes in
molecular structure can actually be seen to occur when crys-
tals of 2-zinc insulin are soaked in solutions that would give
4-zinc insulin (3, 45, 46), and insulin's structure in solution
can be modified by the presence
of
p-cresol or related com-
pounds to yield one with the expected character of molecule
I1 of 4-zinc insulin (47). It thus seemed possible that insulin
during its interaction with receptor might undertake a con-
formation similar to that of molecule
I1
of 4-zinc insulin, a
structure in which the amino terminus of the B-chain is
actually rather close to both the amino terminus of the
A-
chain and the carboxyl terminus of the B-chain. Although
preliminary data have suggested that insulins cross-linked
between residues
A1
and B1 (Fig. 4c) retain only very low
ability to bind to receptor (13), derivatization of the Gl9'
a-
amino group and induction of potential torsional strain in-
volving critical relationships within the NHz-terminal domain
of the A-chain, in this case, might well have resulted in
analogs with severely aberrant structures. Our efforts were
therefore directed toward the construction of analogs bearing
cross-links between residues a-PheB' and ~-Lys~", analogs
which would avoid the negative characteristics of the a-PheB'/
a-G1yA"cross-linked derivatives noted above. Fig. 4d illus-
trates diagrammatically the structure of insulin that might be
taken by analogs cross-linked between residues B1 and B29.
Data provided in Table
I
identify that the relative receptor
binding potencies of analogs cross-linked between residues
PheB' and LysBZ9 in sum are actually rather similar to those
that apply to analogs cross-linked between residues Gly"' and
LysBZ9. Reference to the receptor binding potencies of the
corresponding bisacetyl derivatives, however, demonstrates
that a major fraction of the decreased potency of a-GlyA'/t-
LysBZ9-cross-linked analogs arises from amino group deriva-
tization; the equivalent difficulty does not hold for analogs
cross-linked between residues PheB' and LysBZ9. Application
of the appropriate corrections (to consider the receptor bind-
ing potencies of cross-linked analogs relative to the potencies
of the derivatized parent insulins, Table
I)
indicates that
cross-linking
per
se
contributes to the decreased receptor
binding potencies of
a-PheB'/t-LysBZ9-~ross-linked
insulins
much more significantly than it does to the decreased receptor
binding potencies of
a-Gl~'/c-LysBZ9-cross-linked
insulins.
Given the lack of a simple relationship between the potencies
of
a-PheB'/t-LysBZ9-~ross-linked
insulins and cross-linker
length, however, we have no cause to imagine that decreased
potency in this case arises from restriction of potential mo-
lecular flexibility (as opposed to the induction of unfavorable
conformational change), or to guess that insulin when it is in
combination with its receptor exhibits a structure in which
PheB' and LysBZ9 are in close proximity.
Taken together, our assessment of insulin structure-func-
tion relationships has provided a basis for dissecting multiple
components that contribute to the potentially favorable
or
unfavorable interactions of related cross-linked hormone an-
alogs with receptor. While
(a)
conformational changes are
clearly implicated in the attainment of high affinity states of
ligand-receptor interaction for insulin (7,
22,
23, 44) and for
other hormones as well (24-26),
(b)
the scale of potential
structural changes which insulin can undergo is unexpectedly
large
(8,
44, 48), and
(c)
the intra- and intermolecular move-
ments of insulin in the crystal can be tightly coupled over
both short and long distances (8,44,48), it would appear that
the magnitude (but not necessarily importance) of confor-
mational changes which are necessary to cause high affinity
insulin-receptor interactions may well be rather small.
Acknowledgments-We
thank Dr. Yasushi Nakagawa for perform-
ing the amino acid analyses, Hector Escoffie for help in preparing
isolated canine hepatocytes, and Crystal Sherman-Jones for assist-
ance in preparing the manuscript.
1.
2.
3.
4.
5.
6.
7.
8.
9.
10.
11.
12.
13.
14.
15.
16.
17.
18.
19.
20.
21.
22.
23.
24.
REFERENCES
Blundell,
T.
L.,
Cutfield,
J.
F., Cutfield,
S.
M., Dodson, E. J.,
Dodson,
G.
G.,
Hodgkin, D. C., Mercola, D.
A.,
and Vijayan,
M.
(1971)
Nature
231,506-511
Blundell, T., Dodson,
G.,
Hodgkin, D., and Mercola, D. (1972)
Adu. Protein Chem.
26, 279-402
Bentley,
G.,
Dodson, E., Dodson,
G.,
Hodgkin, D., and Mercola,
D.
(1976)
Nature
261,166-168
Smith,
G.
D., Swenson,
D.
C., Dodson, E. J., Dodson,
G. G.,
and
Reynolds, C. D. (1984)
Proc.
Natl. Acad. Sci.
U.
S.
A.
81,7093-
7097
Cutfield,
J.
F., Cutfield,
S.
M., Dodson, E. J., Dodson,
G. G.,
Emdin,
S.
F., and Reynolds, C. D. (1979)
J.
Mol. Biol.
132,
Bi, R. C., Dauter, Z., Dodson, E., Dodson,
G.,
Giordano, F., and
Dodson, E. J., Dodson,
G.
G.,
Hodgkin, D. C., and Reynolds, C.
Chothia, C., Lesk,
A.
M., Dodson,
G.
G., and Hodgkin, D. C.
Zahn, H., and Meienhofer,
J.
(1958)
Makromol. Chem.
26, 153-
Brandenburg, D., Schermutzi, W., and Zahn,
H.
(1973)
Hoppe-
Freychet,
P.,
Brandenburg, D., and Wollmer,
A.
(1974)
Diobeto-
Glieman, J., and Gammeltoft,
S.
(1974)
Diabetologia
10,105-113
Brandenburg, D., Busse, W.-D., Gattner, H.-G., Zahn, H., Woll-
mer,
A.,
Gliemann, J., and Puls, W. (1972) in
Proceedings
of
the 12th European Peptide Symposium
(Hanson, H., and Jak-
ubke, H.-D., eds) pp. 270-283, Elsevier Scientific Publishing
Co., New York
Dodson,
G. G.,
Cutfield,
S.,
Hoenjet, E., Wollmer,
A.,
and Bran-
denburg,
D.
(1980) in
Insulin, Chemistry, Structure and
Func-
tion
of
Insulin and Related
Hormones
(Brandenburg, D., and
Wollmer,
A,,
eds)
pp.
17-26, Walter de Gruyter, New York
Brandenburg, D., and Wollmer,
A.
(1973)
Hoppe-Seyler'sZ. Phys-
iol. Chem.
354,613-627
Cutfield, J., Cutfield,
S.,
Dodson, E., Dodson,
G.,
Hodgkin, D.,
and Reynolds, C.
(1981)
Hoppe-Seyler's
Z.
Physiol. Chem.
362,
Pullen, R.
A.,
Lindsay, D.
G.,
Wood,
S.
P.,
Tickle,
I.
J., Blundell,
T.
L.,
Wollmer,
A.
Krail,
G.,
Brandenburg, D., Zahn, H.,
Gliemann, J., and Gammeltoft,
S.
(1976)
Nature
259,369-373
Rosen,
P.,
Simon, M., Reinauer, H., Friesen, H. J., Diaconescu,
C., and Brandenburg, D. (1980)
Biochem.
J.
186,945-952
Saunders, D., and Freude, K. (1982)
Hoppe-Seyler's
Z.
Physwl.
Chem.
363,655-659
Markussen, J., Jorgensen,
K.
H., Sorensen,
A.
R., and Thim,
L.
(1985)
Int.
J.
Pept. Protein Res.
26,
70-77
Kubiak, T., and Cowburn, D. (1986)
Biochem.
J.
234,665-670
Nakagawa,
S.
H., and Tager, H.
S.
(1986)
J.
Biol.
Chem.
261,
Nakagawa,
S.
H.,
and Tager, H.
S.
(1987)
J.
Biol. Chem.
262,
Sklar,
L.
A..
Bokoch.
G.
M., Button. D., and Smolen,
J.
E. (1987)
85-100
Reynolds, C. (1984)
Biopolymers
23, 391-395
D. (1979)
Can.
J.
Bwchem.
57,469-479
(1983)
Nature
302,500-505
166
Seyler's
Z.
Physiol. Chem.
354, 1521-1524
logia
10,
1-5
755-761
7332-7341
12054-12058
J.
Bwl. Chem.
262,135-139
..
25. Covarrubias, M., Prinz, H., Meyers, H.-W., and Maelicke,
A.
Insulin-Receptor Interactions
279
(1986)
J.
Biol. Chem.
261, 14955-14961 38.
Drewes,
S.
E., Magojo, H. E.
M.,
and Gliemann,
J.
(1981)
Hoppe-
Demeter, M., Lustbader, J., and Erlich,
P.
(1987)
J.
Biol. Chem.
39.
Carpino, L. A., and Han, G.
Y.
(1973)
J.
Org. Chem.
37, 3404-
27.
Geiger, R., Geisen,
K.,
Regitz, G., Summ, H.-D., and Lagner, D.
40.
Geiger,
R.,
Schone, H.-H., and Pfaff, W.
(1971)
Hoppe-Seyler’s
(1980)
Hoppe-Seyler’s
2.
Physwl. Chem.
361,
563-570
2.
Physwl. Chem.
362,1487-1490
28.
Smith, G. D., Swenson, D. C., Derewenda, Z., Dodson, E. J.,
41.
Brandenburg, D.
(1972)
Hoppe-Seyler’s
Z. Physwl. Chem.
363,
Dodson, G. G., Reynolds, C. D., and Sparks,
K.
(1988)
in
869-873
Peptides,
Chemistry
and Bwlogy, Proceedings of the 10th Amer-
42.
Wang, C., Chu,
S.,
and Brandenburg, D.
(1981)
Hoppe-Seyler’s
ican Peptide Symposium
(Marshall, G.
R.,
ed) pp.
44-46,
ES-
Z.
Physwl. Chem.
362,639-646
COM Science Publishers, Leiden
43.
Cosmatos, A., Cheng,
K.,
Okada,
Y.,
and Katsoyannis,
P.
G.
29.
Lindsay, D. G.
(1971)
FEBS Lett.
21, 105-108 (1978)
J.
Biol. Chem.
253,6586-6590
30.
Schnell,
J.,
and Zahn, H.
(1965)
Kolloid-Zeitschrift
Z.
Polymere
44.
Dodson, E. J., Dodson, G. G., Hubbard,
R.
E., and Reynolds, C.
31.
Lindsay,
D.
G.,
and Shall,
S.
(1971)
Biochem.
J.
121,737-745
45.
DeGraaf,
R.
A. G., Lewit-Bentley, A., and Tolley,
S.
P.
(1981)
in
32.
Anderson, G. W., Zimmerman,
J.
E.,
and Callahan,
F.
M.
(1964)
Structural Studies on Molecules of Biological Interest
(Dodson,
J.
Am. Chem SOC.
86,1839-1845
G., Glusker,
J.
P., and Sayre, D., eds) pp.
547-556,
Clarendon
33.
Shoelson,
S.,
Haneda, M., Blix, P., Nanjo, A., Sanke, T., Inouye, Press, Oxford
K.,
Steiner, D., Rubenstein, A., and Tager, H.
(1983)
Nature
46.
Renscheidt, H., Strassburger, W., Glatter,
U.,
Wollmer, A., Dod-
26.
Moyle, W. R., Pressey, A., Dean-Emig, D., Anderson, D. M.,
Seyler’s
2.
Physwl. Chem.
362,745-753
262,16920-16926 3409
203,27-32
D.
(1983)
Biopolymers
22,281-291
302,540-543
son, G.
G.,
and Mercola, D. A.
(1984)
Eur.
J.
Bbchem.
142,
34.
Bailey,
J.
L., and Cole,
R.
D.
(1959)
J.
Biol.
Chem.
234, 1733- 7-14
35.
Gray, W.
R.
(1967)
Methods Enzymol.
11, 139-151
36.
Zimmer, H. G., Neuhoff,
V.,
and Schulze,
E.
(1976)
J.
Chromatogr. Seyler
368,903-911
37.
Bonnevie-Nielsen, V., Polonsky,
K.
S.,
Jaspan,
J.
J., Rubenstein, Reynolds, C. D., and Vallely, D.
(1981)
in
Structural Studies
A. H., Schwartz,
T.,
and Tager, H.
S.
(1982)
Proc. Natl. Acad.
Sci.
U.
S.
A.
79,2167-2171
on Molecules of Bwlogkal Interest
(Dodson, G., Glusker,
J.
P.,
and Sayre, P., eds) pp.
527-546,
Clarendon Press, Oxford
1739
47.
Wollmer, A., Rannefeld, B., Johansen, B.
R.,
Hejnaes,
K.
R.,
Balschmidt, P., and Hansen,
F.
B.
(1987)
Bwl. Chem.
Hoppe-
124,120-122 48.
Cutfield,
J.
F.,
Cutfield,
S.
M., Dodson, E. J., Dodson, G. G.,
... According to the types of linkers used, there are two main stages of the SCI analogue study. During the first stage (mainly in the 1960s-1980s), various bifunctional chemical agents were used to connect the C terminus of the B chain to the N terminus of the A chain (Nakagawa and Tager 1989). Additionally, during this stage, various cross-linked insulin derivatives were constructed and mainly used to study the relationship between structure and function, such as A1-B1, in which the N termini of both chains were linked together (Freychet et al. 1974). ...
... This study drew attention to the important function of the backbone of the A chain in the insulin-IR interaction. In 1989, α-Gly A1 /ε-Lys B29 -, α-Phe B1 /ε-Lys B29 -, and ε-D-Lys A1 /ε-Lys B29 -cross-linked insulin derivatives were obtained by semisynthesis and were applied to analyze the effect of intramolecular cross-linking on insulin-IR interactions (Nakagawa and Tager 1989). These SCI analogues, which contained chemical linkers ranging from 2 to 12 atoms, exhibited 1.4-35% of the IR-binding ability of the natural hormone. ...
... These SCI analogues, which contained chemical linkers ranging from 2 to 12 atoms, exhibited 1.4-35% of the IR-binding ability of the natural hormone. These detailed findings suggest that the reduced IR-binding potencies of these insulin derivates may be due to different ratios from chemical alteration of crucial functional groups (such as the α-amino group at position A1), steric interference, or the limited favorable molecular intramolecular motion (Nakagawa and Tager 1989). Two years later (in 1991), α-Gly A1 /ε-Lys B29 insulin derivatives, which contained chemical cross-linkers ranging from 2 to 12 carbon atoms, were applied by the same research group to analyze the conformational stability/flexibility analysis. ...
Article
Full-text available
Insulin therapy remains the most effective method to treat diabetes mellitus (DM), and the demand for this valuable hormone has exceeded that of any other protein-based medicine as a result of the dramatic increase in the number of diabetic patients worldwide. Understanding the structure of insulin and the interaction with its receptor is important for developing proper formulations. As a result of the relatively low thermal stability of native insulin and its two-chain analogues, the application of single-chain insulin (SCI) analogues, which can be obtained relatively easily by recombinant DNA technology or chemical synthetic methods, represents a promising alternative approach. In this review, the basic knowledge of insulin (discovery, biosynthesis, and structure) and the current model of the interaction with its receptor are outlined. Furthermore, we outline the strategies for the design and production of various SCI analogues and their reported applications.
... Interest in SCIs was stimulated by classical studies in which bifunctional chemical reagents were employed to tether the C-terminus of the B chain to the N-terminus of the A chain (50,51). Such nonstandard linkers most often connected the amino group of Lys B29 to the -amino group of Gly A1 , thereby mimicking a connecting peptide. ...
... Because D-amino acid substitutions at A1 are well tolerated (52), an alternative approach employed a D-Lys A1 -amino group, extending the effective linker length (51). The relative receptor-binding affinities of the resulting B29-A1 tethered insulin analogs reflected the number of atoms within the crosslink: the longer the tether, the stronger the insulin analog-IR binding interaction. ...
Article
Full-text available
Domain-minimized insulin receptors (IRs) have enabled crystallographic analysis of insulin-bound "micro-receptors." In such structures the C-terminal segment of the insulin B chain inserts between conserved IR domains, unmasking an invariant receptor-binding surface that spans both insulin A- and B chains. This "open" conformation not only rationalizes the inactivity of single-chain insulin (SCI) analogs (in which the A and B chains are directly linked), but also suggests that connecting (C) domains of sufficient length will bind the IR. Here, we report the high-resolution solution structure and dynamics of such an active SCI. The hormone's closed-to-open transition is foreshadowed by segmental flexibility in the native state as probed by heteronuclear NMR spectroscopy and multi-conformer simulations of crystallographic protomers as described in a companion article. We propose a model of the SCI's IR-bound state based on molecular-dynamics simulations of a micro-receptor complex. In this model a loop defined by the SCI's B and C domains encircles the C-terminal segment of the IR α-subunit (αCT). This binding mode predicts a conformational transition between an ultra-stable closed state (in the free hormone) and an active open state (on receptor binding). Optimization of this switch within an ultra-stable SCI promises to circumvent insulin's complex global cold chain. The analog's biphasic activity, which serendipitously resembles current premixed formulations of soluble insulin and microcrystalline suspension, may be of particular utility in the developing world.
... Several insulin derivatives or analogs were also reported, among them the single-chain insulin with a crosslink between the A1 and B29 residues was well-studied experimentally (152)(153)(154)(155). The reported single-chain insulin was biologically inactive Shown in tabulated data are PDB codes (where known; n/a, not available) and simulation conditions including water models SPC, TIP3P, TIP4P, and their variants. ...
Article
Full-text available
Insulin is a peptide hormone known for chiefly regulating glucose level in blood among several other metabolic processes. Insulin remains the most effective drug for treating diabetes mellitus. Insulin is synthesized in the pancreatic β-cells where it exists in a compact hexameric architecture although its biologically active form is monomeric. Insulin exhibits a sequence of conformational variations during the transition from the hexamer state to its biologically-active monomer state. The structural transitions and the mechanism of action of insulin have been investigated using several experimental and computational methods. This review primarily highlights the contributions of molecular dynamics (MD) simulations in elucidating the atomic-level details of conformational dynamics in insulin, where the structure of the hormone has been probed as a monomer, dimer, and hexamer. The effect of solvent, pH, temperature, and pressure have been probed at the microscopic scale. Given the focus of this review on the structure of the hormone, simulation studies involving interactions between the hormone and its receptor are only briefly highlighted, and studies on other related peptides (e.g., insulin-like growth factors) are not discussed. However, the review highlights conformational dynamics underlying the activities of reported insulin analogs and mimetics. The future prospects for computational methods in developing promising synthetic insulin analogs are also briefly highlighted.
... The B-chain detachment model envisaged that (a) the classical structure of insulin (wherein the B24-B28 -strand packs against the hormone's -helical core (41)) represents an inactive conformation and (b) Gly B24 or corresponding D-amino-acid substitutions destabilize this auto-inhibited state to enable native IR binding (45,52). Supported by the low activities of cross-linked (83) or single-chain (80,84) insulin analogs, this hypothesis received direct experimental support from crystallographic studies of insulin bound to fragments of the IR ectodomain (40,85). Indeed, the aromatic triplet was observed to pack within a groove between ectodomain elements L1 and αCT; their displacement indeed enables receptor engagement by conserved non-polar side chains in the central B-chain -helix and N-terminal A-chain -helix (for review, see (86)). ...
Article
Full-text available
Globular protein sequences encode not only functional structures (the native state), but also protein foldability, i.e., a conformational search that is both efficient and robustly minimizes misfolding. Studies of mutations associated with toxic misfolding have yielded insights into molecular determinants of protein foldability. Of particular interest are residues that are conserved, yet dispensable in the native state. Here, we exploited the mutant proinsulin syndrome (a major cause of permanent neonatal-onset diabetes mellitus) to investigate whether toxic misfolding poses an evolutionary constraint. Our experiments focused on an invariant aromatic motif (Phe B24 –Phe B25 –Tyr B26 ) with complementary roles in native self-assembly and receptor binding. A novel class of mutations provided evidence that insulin can bind to the insulin receptor (IR) in two different modes, distinguished by a “register shift” in this motif, as visualized by molecular-dynamics (MD) simulations. Register-shift variants are active but defective in cellular foldability and exquisitely susceptible to fibrillation in vitro. Indeed, expression of the corresponding proinsulin variant induced endoplasmic reticulum (ER) stress, a general feature of the mutant proinsulin syndrome. Although not present among vertebrate insulin and insulin-like sequences, a prototypical variant (Gly B24 -insulin) was as potent as wild-type insulin in a rat model of diabetes. Although in MD simulations the shifted register of receptor engagement is compatible with the structure and allosteric reorganization of the IR-signaling complex, our results suggest that this binding mode is associated with toxic misfolding and so disallowed in evolution. The implicit threat of proteotoxicity limits sequence variation among vertebrate insulins and insulin-like growth factors.
... Важно отме тить, что конформация может иметь большее значение, чем конкретный боковой аминокислотный остаток. Так, замена GlyA1 на L-аминокислоты (Ala, Val, Leu, Pro, Trp, Lys или Glu) приводит к уменьшению связывания с рецептором до 2-20% от нормального значения, однако замена GlyA1 на D-аминокислоты (D-Phe, D-Leu, D-Trp, D-Ala, D-Lys или D-Glu) не снижает биологическую активность аналогов [48][49][50][51]. Мутации, нарушающие сворачивание α-спирали A-цепи, также приводят к снижению биологической активности, что определялось по уменьшению молярной эллиптичности кругового дихроизма [52]. ...
... A prerequisite for a prodrug is the identification of a stable, inactive form of insulin from which the conversion to a fully active agonist could be engineered to occur at a predetermined rate under physiological conditions. The prospect of using molecular constraint is something that appears to hold promise and in this regard several chemically crosslinked insulin derivatives have been made [10][11][12][13][14]. These have been commonly prepared by crosslinking of two of the three primary amines by symmetrical bifunctional active esters. ...
Article
Background: Research has been directed at the optimization of insulin for medicinal purposes. An insulin analog that could be reversibly activated might provide more precise pharmacokinetic control and broaden the inherent therapeutic index of the hormone. The prospect of using intramolecular structural constraint to reversibly inactive insulin might constitute the first step to achieving such an optimized analog. Chemically crosslinked insulin analogs have been reported where two amines are covalently linked by reaction with symmetrical bifunctional active esters. There is little selectivity in this synthetic approach to molecular constraint with multiple derivatives being formed. Objective: To systematically evaluate the synthesis of covalently crosslinked insulin analogs by asymmetric methods and the biological consequences. Method: We report synthesis of amine crosslinked insulin analogs via a two-step procedure. The stepwise approach was initiated by amide bond formation and followed by second site alkylation to produce site-specific, cross-linked insulin analogs. Results: A set of unique insulin analogs crosslinked at the two of the three native amines were synthesized. They were chemical characterized and assessed by in vitro bioanalysis to result in a significant and reasonably consistent reduction in biological potency. Conclusion: We achieved an unambiguous two-step synthesis of several crosslinked insulin analogs differing in location of the chemical tether. Bioanalysis demonstrated the ability of the molecular constraint to reduce bioactivity. The results set the stage for in vivo assessment of whether such a reduction in potency can be used pharmacologically to establish a constrained hormone upon which reversible tethering might be subsequently introduced.
... Such pivoting unmasks an otherwise hidden nonpolar surface of the A chain (including Ile A2 and Val A3 ), thereby extending the hormone's primary receptor-binding surface (17,18). This conformational transition rationalizes a wealth of prior biochemical data (10,(19)(20)(21)(22), including early studies of insulin analogs modified by bifunctional reagents to cross-link the C-terminal segment of the B chain and N-terminus of the A chain (23,24). These studies demonstrated that short cross-links markedly impair IR binding whereas longer crosslinks could, at least in part, restore activity. ...
Article
Full-text available
Thermal degradation of insulin complicates its delivery and use. Previous efforts to engineer ultra-stable analogs were confounded by prolonged cellular signaling in vivo, complicating mealtime therapy and of unclear safety. We therefore sought an ultra-stable analog whose potency and duration of action on intravenous bolus injection in diabetic rats are indistinguishable from wild-type (WT) insulin. Here, we describe the structure, function and stability of such an analog: a 57-residue single-chain insulin (SCI) with multiple acidic substitutions. Cell-based studies revealed native-like signaling properties with negligible mitogenic activity. Its crystal structure, determined as a novel zinc-free hexamer at 2.8 Å, revealed a native insulin fold with incomplete or absent electron density in the C domain; complementary NMR studies are described in a companion article. The stability of the analog (ΔGu 5.0(±0.1) kcal/mol at 25 °C) was greater than that of WT insulin (3.3(±0.1) kcal/mol). On gentle agitation the SCI retained full activity for >140 days at 45 °C and >48 hours at 75 °C. Whereas WT insulin forms large and heterogeneous aggregates above the standard 0.6 mM pharmaceutical strength, perturbing the pharmacokinetic properties of concentrated formulations, dynamic light scattering and size-exclusion chromatography revealed only limited SCI self-assembly and aggregation in the concentration range 1-7 mM. These findings indicate that marked resistance to thermal inactivation in vitro is compatible with native duration of activity in vivo. Such a combination of favorable biophysical and biological properties suggests that SCIs could provide a global therapeutic platform without a cold chain.
Article
Insulin is the principal hormone involved in the regulation of metabolism and has served a seminal role in the treatment of diabetes. Building upon advances in insulin synthetic methodology, we have developed a straight-forward route to novel insulins containing a fourth disulfide bond in a [3+1] fashion establishing the first disulfide scan of the hormone. All the targeted analogs accommodated the constraint to demonstrate an unexpected conformational flexibility of native insulin. The bioactivity was established for the constrained (4-DS) and unconstrained (3-DS) analogs by in vitro methods, and extended to in vivo study for select peptides. We also identified residue B10 as a preferred anchor to introduce a tether that would regulate insulin bioactivity. We believe that the described [3+1] methodology might constitute the preferred approach for performing similar disulfide scanning in peptides that contain multiple disulfides.
Article
Full-text available
We analyzed the structural properties of the peptide hormone insulin and described the mechanism of its physiological action, as well as effects of insulin in type 1 and 2 diabetes. Recently published data on the development of novel insulin preparations based on combining molecular design and genetic engineering approaches are presented. New strategies for creation of long-acting insulin analogs, the mechanisms of functioning of these analogs and their structure are discussed. Side effects of insulin preparations are described, including amyloidogenesis and possible mitogenic effect. The pathways for development of novel insulin analogs are outlined with regard to the current requirements for therapeutic preparations due to the wider occurrence of diabetes of both types.
Article
The concept of a glucose-responsive insulin (GRI) has been a recent objective of diabetes technology. The idea behind the GRI is to create a therapeutic that modulates its potency, concentration or dosing relative to a patient's dynamic glucose concentration, thereby approximating aspects of a normally functioning pancreas. From the perspective of the medicinal chemist, the GRI is also important as a generalized model of a potentially new generation of therapeutics that adjust potency in response to a critical therapeutic marker. The aim of this Perspective is to highlight emerging concepts, including mathematical modelling and the molecular engineering of insulin itself and its potency, towards a viable GRI. We briefly outline some of the most important recent progress toward this goal and also provide a forward-looking viewpoint, which asks if there are new approaches that could spur innovation in this area as well as to encourage synthetic chemists and chemical engineers to address the challenges and promises offered by this therapeutic approach.
Article
Full-text available
Previous studies have suggested that the COOH-terminal pentapeptide of the insulin B-chain can play a negative role in ligand-receptor interactions involving insulin analogs having amino acid replacements at position B25 (Nakagawa, S. H., and Tager, H. S. (1986) J. Biol. Chem. 261, 7332-7341). We undertook by the current investigations to identify the molecular site in insulin that induces this negative effect and to explore further the importance of conformational changes that might occur during insulin-receptor interactions. By use of semisynthetic insulin analogs containing amino acid replacements or deletions and of isolated canine hepatocytes, we show here that (a) the markedly decreased affinity of receptor for insulin analogs in which PheB25 is replaced by Ser is apparent for analogs in which up to 3 residues of the insulin B-chain have been deleted, but is progressively reversed in the corresponding des-tetrapeptide and des-pentapeptide analogs, and (b) unlike the case for deletion of TyrB26 and ThrB27, replacement of residue TyrB26 or ThrB27 has no effect to reverse the decreased affinity of full length analogs containing Ser for Phe substitutions at position B25. Additional experiments demonstrated that introduction of a cross-link between Lys epsilon B29 and Gly alpha A1 of insulin decreases the affinity of ligand-receptor interactions whether or not PheB25 is replaced by Ser. We conclude that the negative effect of the COOH-terminal B-chain domain on insulin-receptor interactions arises in greatest part from the insulin mainchain near the site of the TyrB26-ThrB27 peptide bond and that multiple conformational perturbations may be necessary to induce a high-affinity state of receptor-bound insulin.
Chapter
This chapter discusses the formation and stability of dansyl chloride procedure. Dansyl chloride is used for providing fluorescent “handles” for the study of proteins. The reaction of the dye with α-chymotrypsin led them to propose its use as a means of studying the α-amino and other reactive groups of proteins and peptides. It is useful in determining the amino acid sequences of small amounts of peptides. Dansyl chloride is a typical aromatic sulfonyl chloride, and reacts with a wide variety of bases, forming derivatives of differing stabilities. Some recommended procedure for peptides discussed are (1) reagents, buffers, and marker mixtures, (2) labeling of peptide, (3) hydrolysis, (4) visualization of compounds on paper, and (5) interpretation of results. The increased requirement for dansyl chloride brings with it the necessity for removing the large amounts of DNS-OH formed by its hydrolysis. The removal procedures used are labeling, removal of salt, urea, DNS-OH, hydrolysis, and identification of end groups. Some of the advantages and limitations of the method are mentioned. The main advantage is the ease and rapidity with which end groups may be determined on very small amounts of peptides.
Chapter
Publisher Summary This chapter reviews the physical, chemical, and biological properties of insulin in the light of the atomic arrangement found in insulin crystals. It also describes the relation of the three-dimensional arrangement of the atoms in the molecule of 2-zinc insulin crystal to the solution properties of insulin (particularly its states of aggregation), to the chemical reaction and chemical modification of the molecule, and to its primary biological activity. Normally the insulin crystals contain two zinc ions to every six molecules of insulin—a hexamer. The slow solution of the crystals provides a method of delaying the action of insulin that closely parallels the methods adopted in the pancreas itself for the storage and release of insulin. Within many β granules, grains can be seen that almost certainly contain zinc insulin hexamers packed in a crystalline array, and in experimental animals diabetes has been induced by chelating agents, such as EDTA, perhaps simply by interfering with normal insulin storage. It, therefore, seems plausible that ready crystallization of insulin in the presence of zinc is a reflection of the storage processes in the β cell.
Article
Amorphes, zinkarmes Rinder-Insulin wurde mit Difluordinitrobenzol unter verschiedenen Konzentrationsverhältnissen behandelt und die Reaktionsbedingungen für weitgehend intramolekularen Umsatz ermittelt. Bei der papierchromatographischen Untersuchung der Hydrolysate des behandelten Insulins konnten die Reaktionsprodukte mittels der in der ersten Mitteilung beschriebenen Modellsubstanzen identifiziert werden. Die Ergebnisse bestätigen auf neuem, unabhängigen Weg: Die kleinste Einheit des Insulins ist das Zweikettenmolekül mit dem Molekulargewicht (Rinder-Insulin) M = 5733. Bei Aggregation lagern sich die Zweikettenmoleküle vermutlich mit ihren Aminoenden an einer Seite zusammen.Ox-insulin (amorphous with very little zinc) has been treated with difluorodinitrobenzene and conditions were found to provide an intramolecular reaction. By paper chromatography the hydrolysed products of the treated insulin were identified by comparison with the model substances as described in the first communication. The results prove in a new independent way that the smallest unit of insulin is the two chain molecule with a molecular weight 5733. The two chain molecules under aggregation lie side by side with their amino ends together.
Article
A new amino-protecting group, the 9-fluorenylmethyloxycarbonyl group (FMOC), which is stable toward acids and catalytic hydrogenation but readily cleaved under mildly basic, nonhydrolytic conditions, is reported. The FMOC group may be introduced by reaction of the amine with 9-fluorenylmethyl chloroformate. A number of protected amino acid derivatives were coupled with other amino acids or esters by use of the corresponding N-hydroxypiperidine esters. Deblocking of the FMOC group was carried out with liquid ammonia or at room temperature with piperidine, morpholine, ethanolamine, etc.
Article
A number of N-hydroxysuccinimide esters of acylamino acids have been synthesized. These compounds are crystalline solids which react readily with amino acids or peptides or their esters. Peptide formation under aqueous conditions goes well. Because of the water solubility of N-hydroxysuccinimide, these esters appear to be more generally useful than the analogous esters of N-hydroxyphthalimide.