вход по аккаунту


Metal-Induced Dispersion of Lipid Aggregates A Simple Selective and Sensitive Fluorescent Metal Ion Sensor.

код для вставкиСкачать
Metal-Induced Dispersion of Lipid Aggregates:
A Simple, Selective, and Sensitive Fluorescent
Metal Ion Sensor**
D a r r y l Y. Sasaki, Deborah R. Shnek, Daniel W. Pack,
and Frances H. A r n o l d *
Lipid reorganization induced by specific ligand binding is a
feature intrinsic to biological processes such as signaling,['' membrane fusion.[21cell docking,[31and endocyt~sis.[~]
Often the ligand is a protein, which generates lipid reorganization through
multiple lipid-ligand interaction^.^^] Lipid reorganization, however, can also be driven with small
For example, calcium
can induce phase separation of phosphatidylserine lipids in vesicles of phosphatidylcholine lipids['] o r alter the packing phase of
cardiolipin from bilayer to inverted hexagonal."' In the course of
studying substrate binding to metal-complexing lipids and their
subsequent reorganization in membrane assemblie~,''~we discovered that metal ions can strongly affect the distribution of the
metal-chelating lipid. Lipid la, a pyrene-labeled lipid functionalized with iminodiacetic acid (IDA), forms aggregates when
placed in vesicles of distearoyl phosphatidylcholine (DSPC), as
evidenced by pyrene excimer formation in steady-state fluorescence measurements. Addition of divalent metal ions essentially
instantaneously results in a dramatic reversal of the ratio of
emission intensities of excimer and monomer, which is believed
to coincide with dispersion of the pyrene-lipid into the gel phase
of the matrix DSPC lipid. This process, which is highly sensitive
and selective for Cu2+.offers a particularly simple approach to
C u 2 + detection and quantification.
Fig. 1. Fluorescence emission spectra of a) 5 % lipid 1a/95'Y0 DSPC and b) 5 %
lipid la/95% SOPC in a MOPS buffer at 25°C. MOPS = 3-(4-morpholinyl)-lpropanesulfonic acid.
the monomer emission (maximum at 377 nm) and a broad featureless band with emission maximum at 470 nm, which is attributed to pyrene excimer. The large ratio of the fluorescence
intensities of excimer and monomer (E/M) reflects a high local
pyrene concentration in the bilayer assembly.['01 The E/M value
of vesicles containing 5 % lipid l a and 95% DSPC is 1.8
(curve a), but drops to only z 0.2 or less in vesicles composed
of 5 % lipid la in 1 : l DSPC-cholesterol (data not shown)
or SOPC (2-oleoyl-1 -stearoyl-sn-glycero-3-phosphocholine)
(curve b). At 25 "C SOPC and DSPC-cholesterol matrix
bilayers exist in the fluid phase, while DSPC is in the solid gel
phase. The markedly different E/M values suggest that lipid l a
aggregates or phase separates from the gel matrix of DSPC, but
is dispersed in the two fluid bilayers.
The pronounced effect of divalent metal ions on the E/M
value of lipid l a in DSPC vesicles is illustrated in Figure 2. With
increasing metal ion (Mn2+) concentration, the excimer emis-
Lipids 1 and 2 can be co-sonicated in aqueous solution with
various phosphatidylcholine lipids to prepare stable mixed bilayer vesicles. The steady-state fluorescence spectrum of 5 mol%
lipid l a in vesicles of DSPC shown in Figure 1 (curve a) includes
Prof. F. H Arnold, D. R. Shnek, D. W. Pack
Division ot' Chemistry and Chemical Engineering 210-41
California Institute of Technology
Pasadena. CA 91 125 (USA)
Telefilx: I n r . code + (818)568-8743
D. Y. Sasaki
Sandia National Laboratory
Albuquerque. NM 87185-0368 (USA)
This research was supported by the Office of Naval Research (N00014-92-J1 1 78) and the National Science Foundation (BCS-9108502). F. H. A. acknowledges an NSF PYI Award and a David and Lucile Packard Fellowship.
D. R . S. is supported by a predoctoral training fellowship from the National
Institute of General Medical Sciences, Pharmacology Sciences Program.
D. w.P. ia d Landau Fellow.
Fig. 2. Fluorescence emission spectra of 5 % lipid 1a:95% DSPC vesicles in a
MOPS buffer at 2S'C with increasing concentrations of MnCI,.
sion intensity decreases while the monomer emission increases.
Other metal ions have a similar effect, but at different concentrations, as shown in Figure 3. The E/M curves generally parallel the metal ion binding constants to IDA, with the exception
that Co2+ has a stronger affect than NiZ+.["] If the observed
changes in the E/M values correspond to metal complexation by
IDA, however, the metal binding affinity of the IDA headgroup
appears to be attenuated by a factor of 100-1000 when incorpo-
c [M"f]/M
1 0 - ~ lo-'
Fig. 3. E/M values (normalized by E/M values without added metal) for 5% lipid
la195 % DSPC vesicles. 0.11 mM total lipid. in the presence of various metal ions at
25 "C. The metals were added as the chloride salts in 0.1 M NaCI. Solid curves are
drawn in for clarity.
rated into vesicles, relative to that of free N-(2-methoxyethyl)iminodiacetate in solution." 21
Sensitivity to lower concentrations of metal ions can be obtained by using more dilute vesicles (lower lipid concentrations).
Figure 4 shows how titration with CuZ affects the E/M value
of 5 % lipid Ia/95% DSPC vesicle solutions at three lipid concentrations, the lowest of which is 6 5 0 n ~total phospholipid
10C [cUI']/M
( 3 5 n ~lipid la). This system is sensitive to Cu2+ at concentrations of 5nM (less than 1 ppb) in 0.1 M NaCl. It is conceivable
that even greater sensitivity can be achieved by using even lower
lipid concentrations and a pyrene-labeled lipid with a chelating
headgroup with a higher affinity for the metal ion than IDA
(e.g., nitrilotriacetate). The changes in E/M values shown in
Figures 2-4 occur within seconds of metal ion addition and are
completely reversible with the addition of excess ethylenediaminetetraacetate (EDTA) .
Owing to the strong affinity of IDA for Cuz+, this bilayer
system is highly sensitive and selective for Cu2 . Thus micromolar concentrations of Cu2+ ions can be detected in a 0.1 M NaCl
solution containing millimolar concentrations of Ca2 . Moreover, the fluorescence effects of submicromolar CuZ+concentrations can be detected by the naked eye in black light (365 nm).
value was observed for vesilipid lb, 2, or la before :left) and
cles containing l b and 2. Furafter addition of CuCL ( 5 m N
ion addition
{right). All measurements done in a
MOPS buffer at 25 "C.
to vesicles containing lipid l a
in a fluid matrix (50 YODSPC/
50 % cholesterol or SOPC) also induces only very small changes
in the fluorescence E/M value (although quenching is observed
for the more fluid SOPC matrix).
It appears that complexation of the metal ion by the IDA
headgroup induces the dispersal of lipid l a in the DSPC gel
matrix, which results in the dramatic reduction in the rate of
pyrene collisions, and therefore reduction of the E/M value.
This process is illustrated in Figure 6.
Fig. 4. E/M values (normalized by E/M values without added metal) for 5 % lipid
la195 % DSPC vesicles in the presence of Cu*+ at 25 "C: Total phospholipid 65 PM
(o), 6 . 5 m ~( o ) ,and 65011~(*) . The metal was added as the chloride salt in 0 . 1 ~
NaCI. Solid curves are drawn in for clarity.
Metal titration studies performed with the non-metal-chelating pyrene lipids l b and 2 in DSPC vesicles confirm that the
observed E/M change is a consequence of metal chelation by the
IDA headgroup, rather than fluorescence quenching or a metalinduced perturbation of the membrane. The fluorescence E/M
values of 5 % lipid 1 and 2 in DSPC vesicles are shown in the
form of a bar graph in Fig4.0
ure 5. While Cu2+ at 5mM
concentration induces a large
change in the E/M value of the
lipid la/DSPC vesicles, it has
very little effect on the E/M
values of vesicles prepared
with the nonchelating pyrene
M 1.0
lipids. Even at Cu2+ concentrations 1000 times higher
Verlagsgesellschufr mhH. 0-69451 Weinheim. 1995
Fig. 6. Schematic representation of metal-induced dispersion of lipid la in vesicles
of 5 % lipid la/95% DSPC. as observed through fluorescence measurements at
25 "C. Lipid la is believed to assemble into pyrene-rich aggregates which are dlspersed upon chelation of a divalent metal ion by the lipid headgroup.
The reorganization of a lipid labeled with a light-emitting
group induced by specific substrate binding provides a powerful
method for detection of substrates in solution. As a sensing
device, this extremely simple, two-component system offers
some attractive features: its sensitivity for Cu2 (nM) compares
favorably to that of ion-selective electrode^^'^] and other synthetic lipid-based sensors;['41 it is relatively insensitive to high
concentrations of weakly binding metal ions such as C a 2 + ;it
can be regenerated by washing with a strong chelating agent
(edta); and it has a rapid response time (seconds). Owing to its
simplicity and essentially instantaneous signal transduction, this
system could be readily adapted to continuous copper ion monitoring in a continuous flow fluorescence spectrometer. Studies
0570-0833/95/0808-0906 $10.00+ ,2510
Angew. Chem. Int. Ed. Engl. 1995, 34. N o . 8
to understand the mechanism by which metal binding induces
dispersion of the lipid aggregates and investigations of the generality of this type of sensor are underway.
Experimental Procedure
Stock lipid solutions were made by dissolving lipids and cholesterol (Sigma) in
CHCI, (HPLC grade). Lipid I or 2 and a matrix lipid (DSPC, SOPC, or I : 1
DSPC:cholesterol) were combined (5 mol% lipid I or 2 , 9 5 mol% matrix lipid) to
give 10 pmol total lipid in l 2 m L volumetric centrifuge tubes. The lipid mixture was
concentrated under aspirator vacuum, and 3 mL of MOPS buffer ( 2 0 m ~MOPS,
0.1 M NaCI. pH 7.5) was added to the tubes. The tubes were heated above 55°C and
sonicated with a probe tip (Heat Systems model 375) for 15 min at 25-35% power
under argon atmosphere in an ice bath. The vesicles were centrifuged a t 11 000 rpm
for 20 min to remove titanium particles. The phosphate concentration was determined for selected samples [15]. Vesicle sizes were measured by quasi-elastic light
scattering by using a Microtrac Ultrafine Particle Analyzer (Leeds & Northrop) at
25 C in phosphate buffer. The mean diameter was 49 nm (59 nm distribution width)
for unmetdlated vesicles. Vesicles were diluted 167-fold in MOPS buffer for metalbinding studies. Steady-state fluorescence measurements were performed at 25 "C
on a temperature-controlled Shimadzu RF-450 spectrofluorimeter, excitation at
346 nin. 5 n m excitation and emission slit width.
Received: October 13. 1994
Revised version: December 6, 1994 [Z 7397 IE]
German version: Angew. Chem. 1995, 107, 994
Keywords: fluorescent sensors . lipids . metal ion analysis . vesicles
1 1 1 a) K. Eichmann, Angew. Cliem. 1993, 105. 56; AngeM,. Chem. Inf. Ed. Engl.
1993.32.54: b) W. J. Fantl, D. E. Johnson, L. T. Williams. Annu. Ree. Bicichem.
1993, 62. 453.
[2] a) J Zimmerberg. S. S. Vogel. L. V. Chernomordrk, Annu. RPV. Biuphw.
Biomol. 3ruc.t. 1993. 22, 433; b) D. PapdhddjOpOUlOS. G. Poste, B. E. SchaefBioph~s.Acra 1974, 352. 10.
fer. W. J. V ~ i l Biochim.
[3] R. B. Kelly. Curr. Bid. 1993, 3, 474.
[4] S. C. Silverstein. R. M. Steinman, Z. A. Cohn. Annu. Rev. Biochem. 1977. 46,
[5] a) M. Ahlers, W. Muller. A. Reichert, H. Ringsdorf, J. Venzmer, Angew. Chem.
1990. 102. 1310: A n g w . Chum. I n f . Ed. Engl. 1990. 2Y, 1269; b) D. M. Haverstick. M. Glaser, Siopl7m J 1989, 55,677; c) P. Antes. G. Schwarzmann, K.
Sanhoff, Chem. P/i.ys. Lipids 1992, 62, 269; d) J. R. Wiener, R. Pal, Y. Barenh o k R. R. Wagner. Biochemistry 1985. 24, 7651; e) G. B. Birrell, 0. H.
Griftith, ibid 1976, 15, 2925.
16) a) H. Hauser. Chrm. Phvs. lipid^ 1991, 57. 309; b) K. Jacobson, D. Papahadjopoulos, Biuchrmisrrj 1975, 14, 152.
[7] S. Ohnishi. T. Ito. Biochrmistr~.1974. 13, 881.
[8] R. P. Rand, S. Sengupta, Biochim. Bioph13. Acru 1972, 255, 484-492; b) B.
deKruijff. A. J. Verkleij. J. Leunissen-Bijvelt, C . J. A. van Echteld. J. Hille, H.
Rijnbout, ibnf. 1982. 693. 1.
[9] D. R. Shnek, D. W. Pack, D. Y Sasaki, F. H. Arnold, Lungmuir 1994,10,2382.
[lo] a ) H.-.I.Galla. W. Hartmann. Chrm. f h w Lipids 1980, 27. 199; b) H.-J. Galla,
E. Sackmann. Biochhn. Biophy.7. Actu 1974, 339. 103.
[ l l ] The order of metal ion binding for N-(2-methoxyethyl)iminodiacetic acid in
solution is C u > > N i > C o > M n > C a : A. E. Martell, P. M. Smith, Crifical
Stahilitj Coiisrunts, Vol. 6 , Plenum, New York, 1974.
1121 The K,. for example. of methoxyethyl iminodiacetate is 3 x IO-'M for Mn'+,
whereas the EIM value of the vesicles begins to change at
M and shows a
midpoint at ca. 3 x W 4 M M n 2 + (Fig. 3).
[I31 a ) S. S. Kuan. G. G. Guilbault, Biosensors: Fundamental,$and Applications
(Eds : A. P. F. Turner, I. Karube, G. S. Wilson), Oxford University Press,
Oxford. 1987. p. 135; b) S. L. Belli, A. Zirino. Anal. Chrm. 1993, 65, 2583; c)
E. K. Quagraine, V. P. Y. Gadzekpo, Analysc 1992. 117. 1899; d) R. A. Durst.
Inn Se/<,ctivc Electrodes (Ed.: E. A. Durst), National Bureau of Standards
Publication. Washington, D. C.. 1969. p. 375.
[14] a ) M. Lerchi, E. Bakker, B. Rusterholz, \?! Simon, Anal. Chem. 1992. 664,
1534. b) M. Shimomura, T. Kunitake, J Am. Chem. Soc. 1982, /04,1757; c) A.
Singh, L:l. Tsao. M Markowitz, B. P. Gaber, Lungmuir 1992, R, 1570; d ) S.
Terrettaz. H. Vogel. M. Gritzel. J Elecrroanal. Chem. 1992, 326. 161: e) S.
Steinberg. I. Rubinstein, Lungmuir 1992, 8. 1183.
1151 W. R. Morrison, Anal. Biochrm 1964, 7, 218.
A n p + Chmni. Inr. Ed. Engl. 1995. 34, No. R
Solid-Phase Syntheses of Unnatural Biopolymers
Containing Repeating Urea Units**
Kevin Burgess,* D. Scott Linthicum,
and Hunwoo Shin
In medicinal chemistry, urea bonds have been used as critical
structural elements in enzyme inhibitors"] and as switching
points in retro-inverso peptidomimetics.['] The advent of "split
syntheses"[31in combinatonal methodology,[4. however, opens
an avenue for identification of many other urea-based compounds with potentially useful pharmaceutical properties. Consequently, there is a need for preparations of urea precursors,
and for solid-phase syntheses of compounds with several urea
functionalities incorporated in systematic fashion. This communication describes solid-phase syntheses of the two oligourea
compounds CH,GU. CH,F". CH,FU. CH,A". A-amide (8) and
YG. CH,G". CH,FU. L-amide (9).These labels are based on the
one letter codes for the amino acids from which the urea units
are derived (vide infra) .
Monomers for the oligourea syntheses were prepared by reducing the N-Boc protected amino acids to the corresponding
amino alcohols 1, which were then used as electrophiles in Mitsunobu displacements with phthalimide. Formation of regioisomeric products via undesired N-Boc aziridine intermediates is
known,[61and might have been problematic, but it was not.
Removal of the N-Boc group gave the required monoprotected
diamines 2.
ca 80 %
ca.70 %
Scheme 1 R = Me, CH,Ph, H a) 1) CICOOiBu, NEt,, 0 C. T H F , 2) NaBH,,
H,O. b) 1) Phthalimide (HNphth), PPh,, EtOCON=NCOOEt, THF, 25 C. 3 h , 2)
HCI,,,,, T H F
Trial reactions in solution were performed to determine
whether urea derivatives like 4 arise from isocyanate intermediates of type 3 formed in situ. To do this, the monoprotected
diamine from alanine (2a, R = Me) was treated with "triphosgene" (bis(trichloromethyl)carbonate)[71 and then with 1aminobutane. NMR analysis of the crude material indicated
that the only detectable product was the anticipated urea 4 a.
Scheme 2. a) 0.33 equiv (CCI,O),CO, NEt,, CH,CI,, 45 min. from 0-25°C. b)
BuNH,, 12 h.
Prof. K. Burgess, H. Shin
Department of Chemiustry
Texas A & M University, College Station, TX 77843-3255 (USA)
Telefax: Int. code + (409)845-8839
Dr. D. S . Linthicum
College of Veterinary Medicine, Texas A & M University (USA)
[**I Financial support for this work was obtained from the National Institutes of
Health and The Robert A. Welch Foundation. K.B. is a NIH Research Career
Development Awardee and a n Alfred P. Sloan Scholar
(C; VCH I/i.rlugsgesellschufl mhH. 0.69451 Weinheim, 1995
$ /0.00+ 2510
Без категории
Размер файла
487 Кб
simple, ion, induced, fluorescence, metali, selective, aggregates, sensore, sensitive, dispersion, lipid
Пожаловаться на содержимое документа