вход по аккаунту


Specific Recognition of a Minimal Model of Aminoacylated tRNA by the Elongation Factor Tu of Bacterial Protein Biosynthesis.

код для вставкиСкачать
a) B. Hasenknopf. J.-M. Lehn, B. 0. Kneisel, G. Braun, D. Fenske, Angew.
Chem. 1996, 108. 1987-1990; Angrn. Chem. Inr. Ed. Engl. 1996, 35, 18381840; 9. Hasenknopf, J.-M. Lehn, N. Boumediene, A. Dupont-Gervais. A. Van
Dorsselaer. B. Kneisel. D. Fenske, J Am. Chem. Sor. 1997, 119, in press; b)
R. W Saalfrank. S. Trummer, H, Krautscheid, V. Schunemann, A X.
Trautwein. S. Hien. C. Stadler, J. Daub, Angew. Chem. 1996, 108,2350-2352;
Anger<. Chem Inl. Ed. Engl. 1996. 35, 2206-2208.
a ) R. W. Saalfrank, R. Burak. S. Reihs, N. Low, F. Hampel, H.-D. Stachel. J.
Lentmaier. K. Peters, E. M. Peters, H. G. von Schnering, Angew. Chem. 1995,
107. 1085-1087; Angew. Chem. Inr. Ed. Engl. 1995, 34, 993-995; b) R. W.
Saalfrank, R. Burak, A. Breit. D. Stalke, R. Herbst-Inner, J. Daub, M. Porsch,
E. Bill, M. Miither, A. X. Trautwein, &id. 1994,106, 1697- 1699 and 1994,33.
1621 - 1623; c) T Beissel, R. E Powers, K. N. Raymond, ihid. 1996,108.11661168 bzw. 1996,35, 1084-1086.
R. W. Saalfrank. A. Dresel. V. Seitz. F. Hampel, M. Teichert, D. Stalke, Chem.
Eui J. 1997. 3. no. 12.
a) R. W. Saalfrank, 0 Struck, M. G. Davidson, R. Snaith, Chem Ber. 1994,
127, 2489-2492; b) R W. Saalfrank, 0. Struck, K. Peters, H . G . von
Schnering, {hid. 1993,126,837-840: c) R. W. Saalfrank, 0. Struck, K. Nunn,
C.-J. Lurz. R. Harbig. K. Peters. H. G. von Schnering, E. Bill, A. X. Trautwein,
ihid 1992,125,2331 -2335; d) R. W. Saalfrank, 0.Struck, K. Peters, H. G. von
Schnering, Inorg. Chim. Acta. 1994, 222, 5-1 1.
S . Mann, G. Huttner, L. Zsolnai. K. Heinze, Angew. Chem. 1996, 108, 29832984. A n g w Chem Inr. Ed. Engl. 1996, 35, 2808-2809. See also: D. Burdinski. F. Birkelbach, M. Gerdan. A. X Trautwein, K. Wieghardt, P Chaudhuri. J Chrm Soc. Chem. Commun. 1995,963-964; P. Chaudhuri, M. Winter,
P. Fleischhauer. N! Hasse, U Florke, H.-J. Haupt, hid. 1990, 1728-1730; C.
Krebs. M. Winter, T.Weyhermiiller, E. Bill, K. Wieghardt, P. Chaudhuri, ihid.
1995. 1913-1915.
X-ray crystal structure analyses: Nonius MACH3; Mo,, radiation,
i = 0.71073 A; the structures were solved by direct methods 1191 and refined
with full-matrix least-squares on F2 [20]. Hydrogen atoms were fixed in idealized positions using a riding model. An absorption correction was carried out
M , = 1986.74; trigonal, R3;
with Y scans. 3: C,,H,,CIFe6N6Na0,8-6CHCI,;
crystaldimensions0.30 x 0 30 x0.30mm';a = h = 19.177(3),c =17.484(4)
V = 5569(2) A3; T = 173(2) K ; Z = 3; pFnlod
= 1.777 Mgm-3; section of the
0 = 2.63-25.97";
reciprocal lattice - 2 0 < h < 0 , - 2 O t k t 0 , -21<1<21;
2648 reflections observed, 2421 unique; reflections with I>2u(I): 1735, R1
[I>2u(I)j. 0.0409, nlR2: 0.0939; R1 (all data): 00748. wR2: 0.1102; largest
difference peak and hole: 0.550/ - 0.633 e k ' . 4: C,,H,,CIFe,LiN,0,,.6CHC13; M , = 1970.69; tngonal, R I ; crystal dimensions 0.30 ~ 0 . 3 0 ~
0 . 2 5 m m 3 ; o = h =19.103(2),~=17.453(4)A; V = 5516(2)A'; T=173(2)K;
Z = 3; prDlid=1.780 Mgm-3 section of the reciprocal lattice - 2 2 < h < 0 ,
O < k < 19. 0 < / < 2 0 ; 0 = 2.64-4.98"; 1186 reflections observed, 1184 unique;
reflections with 1>2u(I): 988, R1 [I>2u(Z)]: 0.0493, n7R2: 0 1641, R1 (all
data). 0.0632, wR2: 0.1844; largest difference peak and hole: 0.830/
- 0.664 e k ' 5 . C,,H,,CIFe,N,CsO,;2CHCl,-2CH2Cl2; M. = 2193.08:
monoclinic, P2jn; crystal dimensions 0.30 x 0.20 x 0.20 mm'; a = 17.684(4),
h=13.034(3).c =19.359(4)A;h=90.53(3)", V=4462(2)A3; T=173(2)K;
Z = 2; pIIlud=1.632 Mgm-3, section of the reciprocal lattice O<h<20,
O<k<014, -22<1<22; 0 = 2.78-23.97"; 6969 reflections observed, 6969
unique; reflections with 1>2u(I): 3273, R1 [I>2u(I)]: 0.1059, wR2: 0.2610; Rl
(all data): 0.2219, wR2: 0.3120; largest difference peak and hole. 1.8511
- 1.219 e k 3 . I n the asymmetric unit there is one molecule of chloroform
and one molecule of dichloromethane that do not make contact with the iron
complex. The chloroform is disordered on two positions with a probability of
occupancy of 65: 35%. The recorded data allowed only an isotropic refinement
of the solvent molecules by using distance restrains for the C-CI bond. 7 .
M , = 1489.42; monoclinic, P2(l)/c; crystal
dimensions 0.15x0.15x0.10mm3; a =13.9130(9), b=18.701(5), c =
11.305(4)A; 8=106.04(2); V = 2827.1(14)A3; T=173(2)K; Z = 2;
=1.750Mgm-'; section of thereciprocal latticeO<h<16, - 2 2 < k < 0 ,
- 13<1<12: 0 = 2.33-24.97"; 5172 reflections observed,4960unique, reflections with />2u(I) 2664, Rl [ I > Z u ( I ) ] :0.0731, wR2: 0.1849; R1 (all data):
0.1745. wR2- 0.2285; largest difference peak and hole: 1 234/ -l.320eA-3.
Crystallographic data (excluding structure factors) for the structures reported
in this paper have been deposited with the Cambridge Crystallographic Data
Centre as supplementary publications nos. CCDC-100097 (3) and CCDC100472 (4. 5 and 7 ) . Copies of the data can be obtained free of charge on
application to The Director, CCDC, 12 Union Road, Cambridge CB2 lEZ,
UK (fax. int code f(1223) 336-033; e-mail:
For green-red stereorepresentation of 3 see
d = 2u - 2r,
Hitherto it proved impossible to isolate the complex [LicFe,{N(CH,CHzO)3)aICI
G. M. Sheldrick, SHELXS-96, A c f a Crystdlogr. Secf. A , 1990, 46, 467-473.
G. M. Sheldrick, SHELXL-96, Program for crystal structure refinement, Universitit Gottingen 1996.
Angeu Chem Int Ed Engl 1997.36, No 22
Specific Recognition of a Minimal Model
of Aminoacylated tRNA by the Elongation
Factor Tu of Bacterial Protein Biosynthesis**
Stefan Limmer,* Martin Vogtherr, Barbara Nawrot,
Rainer Hillenbrand, and Mathias Sprinzl
In the course of protein biosynthesis the aminoacylated transfer RNA (aa-tRNA), which is correctly "charged" with its
specific amino acid, has to be transported to the ribosome. In
bacterial biosynthesis this function is carried out by the elongation factor Tu (EF-Tu), which must be able to recognize all
aa-tRNAs as well as to reject uncharged elongator tRNAs and
the initiator tRNA-in
bacteria invariably met-tRNA
( M e t = formylmethionyl), which is formylated and interacts
with the initiation factor IF-2. Thus, the specific recognition of
the aa-tRNAs may depend neither on variations in the sequences and the tertiary structures, nor on the nature of the
attached amino acid itself.
The region common to all aa-tRNAs comprises the ubiquitous CCA terminus of the tRNA and the ester linkage at the 2'or 3'-OH groups of the ribose of the terminal adenosine 76."' In
solution the two isomers that have 2'- and 3'-ester bonds are
in a dynamic equilibrium.'21 It was shown13] that minihelices
consisting of at least ten base pairs and an aminoacylated,
single-stranded (A)CCA end are relatively tightly bound by
EF-Tu.GTP. The smallest conceivable unit as a minimal model
of aa-tRNA is made up of the terminal adenosine and an attached amino acid.12]Indeed, such aminoacylated adenosines
are able to compete sucessfully for the aa-tRNA binding site of
a specific inhibitor of EF-Tu, namely N-tosyl-L-phenylalanylchl~romethane.~~]
The stability of the ester bond between ribose and amino acid
is rather low under conditions which are optimal for EF-Tu
(pH =7.5). Hence, we used anthranilic acid as an amino acid
analogue, which has a higher stability with respect to hydrolysis
of the ester bond. Its suitability for binding to EF-Tu was
demonstrated for yeast tRNAPh"charged with anthranilic acid.[51
The transacylation rate of the anthranilic acid between the 2' and
3' ribose positions is drastically lower for anthraniloyladenosithan for corresponding aminoacylated adenosines; this
nes lC6'
Priv.-Doz. Dr. S . Limmer. DiplLChem. M. Vogtherr,
Dip].-Biol. R. Hillenbrand, Prof. Dr. M. Sprinzl
Laboratorium fur Biochemie der Universitat
D-95440 Bayreuth (Germany)
Dr. B. Nawrot
Centre of Molecular and Macromolecular Studies
Department of Bioorganic Chemistry
Polish Academy of Sciences, Lodz (Poland)
This work was supported by the Deutsche Forschungsgemeinschaft. We wish
to thank Norbert Grillenbeck for preparing EF-Tu, Karol Szkaradkiewicz for
supplying IF-2, and Melissa Brown for helping with the translation.
8 WILEY-VCH Verlag GmbH, D-69451 Weinhelm, 1997
0570-0833/97/3622-2485 $17 50f 50 0
correlates with the significantly increased stability of the ester
bond towards hydrolysis. Due to the lack of important regions
of the aa-tRNA-with which the EF-Tu interacts in its active,
GTP-bound formr7]-only a drastically diminished affinity
of 1 for EF-Tu can a priori be expected. This should give rise
to an exchange of 1 between free and bound forms that is
fast on the NMR time scale.[s391In this case the observed
values of chemical shift and linewidth represent the weighted
averages of the corresponding quantities in the free and bound
states. The linewidths of the signal for the small ligand molecule
should continuously increase with rising protein concentration.
Owing to the much larger molecular weight of the protein
(ca. 45 kDa) and the therefore distinctly faster T, relaxation,
the protein contributes merely a more or less pronounced,
poorly resolved background to the 'HNMR spectrum of the
In Figure 1 the 'HNMR spectra of the complexes of
EF-Tu. GTP from the eubacterium Thermus thermophilus with 1
in varying molar ratios are displayed. As expected for fast
exchange (that is, weak binding), the linewidth rises as the relative protein fraction increases from 1:20 to 1:2.5. The resonances for the 3'isomer are broadened to a distictly greater
extent than those for the 2' isomer, suggesting a preferred binding of the former. This confirms earlier
which an almost exclusive binding of the 3' isomer of aa-tRNA
to EF-Tu was deduced. By contrast, a weakened interaction of
both species with EF-Tu'GTP was found upon using aa-tRNA
variants in which the terminal adenosine was replaced by 3'- or
2'-deoxyadenosine.[' '1
All lines in the spectrum of 1 become broadened in the presence of EF-Tu (see Table 1 and Figure l ) , although not to the
same degree; the 2' and 3' ribose resonances of the 3' isomer are
Table 1. Linewidths[Hz][a] and chemical shifts (in parentheses) in the ' H N M R
spectrum of free and EF-Tu.GTP-bound[b] 1
3' Isomer
2.6 (8.409)
2.6 (8.292)
10.8 (6.257)
14.7 (5.156)
10.8 (5.598)
HI '
28.0 (8.356)
33.0 (6.207)
47.0 (5.121)
53.0 (5.598)
2 Isomer
2.6 (8.437)
2.5 (8.258)
9.9 (6.442)
12.9 (5.870)
11.4 (4.817)
13.0 (8.432)
17.3 (6.446)
19.3 (5.848)
16.5 (4.812)
[a] Linewidths at half-height, measured over the multiplet structure. [b] 1 mM 1,
0.18mM EF-Tu-GTP in borate buffer, pH =7.5, D,O/[D,,]methanol 7/1, 281 K.
[c] Strong overlap of the resonances for both isomers.
particularly broadened. From these differences, conclusions
concerning the contact sites of the ligand with the protein can be
derived: For the molecular regions of the ligand with the closest
contacts to the protein, the most significant alterations of the
relaxation behavior are due to additional relaxation pathways
involving protons of the protein. Such a difference in line broadening is often characteristic of a specific ligand-protein interaction.[l41
Remarkably, the integral ratio of the signals for 3'-1 and 7-1
shifts from about 2: 1 for free 1 towards 1:1 with rising protein
concentration. This means that with an increasing fraction of
bound 1, which preferably binds as the 3' isomer, the relative
population of the 2' isomer grows. The same effect is achieved
when the binding equilibrium is shifted by varying the temperature while keeping the concentrations of protein and ligand
0 WILEY-VCH Verlag GmbH, D-69451 Weinheim, 1997
Figure 1. Portions of the 'H NMR spectra with aromatic (top) and ribose 1'. 2'. and
3 signals (bottom) of 3'(2')-anthranyloyladenosine (1) in D,O/[D,]methanol(7/1) at
281 K. Concentration of 1 1 mM, concentration of EF-Tn.GTP a) 0.00, b) 0.05,
c) 0.08,d) 0.18, e) 0.4 mM. The signals of the 3' and 2' isomers are marked with o
and x, respectively (for the assignment, see Table 1). The baseline distortion is due
to the protein background spectrum.
constant. For 1 mM 1 and 0.18 mM EF-Tu.GTP, the ratio increases from about 1 : 1 (3':2') at 281 K to about 1.8:1 at 307 K.
Preferred binding of the 3' isomer alone is not sufficient to account for this behavior. One conceivable explanation is that,
0570-0833/97/3622-2486 $17 50f 50/0
Angeu Chem Inr Ed Engl 1997,36,No 22
compared to free 1, transacylation of 1 is accelerated upon binding to EF-Tu. Since EF-Tu binds the two isomers with different
strengths, an equilibrium is adjusted which reflects the different
affinities for the protein and leads to an accumulation of the
2’ isomer.
The observed line broadening (and hence ligand binding) is
independent of the guanosine nucleotide (GTP, GPPNHP,
GDP) bound to EF-Tu, which is different than in the case of
complete aa-tRNA.[I2]The rearrangement of the EF-Tu structure”. 1 6 a . b 1 associated with GTP binding is a prerequisite for
the binding of the tRNA in a cleft formed between domains I
and 11, in which stabilizing interactions between EF-Tu and
tRNA are enabled. Naturally, these contacts cannot take place
with the small ligand 1,which may explain both the diminished
affinity and the absence of any differences between the GTP/
GPPNHP- or GDP-complexed forms with respect to binding to
EF-Tu. For the specific interaction of the terminalA76, the
relative orientation of the three EF-Tu domains-which is determined by the nature of the bound nucleotide-plays only a
minor role.
To check the specificity of the complex formation, several
control experiments with structurally related compounds of low
molecular weight were carried out. In the case of the antibiotic
puromycin (2) the ester bond between ribose and amino acid is
replaced by an amide bond; in formylmethionyl-AMP (3) the
amino group of the amino acid is formylated. In both cases only
negligible increases in linewidth were measured under the conditions used for the 1.EF-Tu complex (1 mM ligand, 0.08 mM
EF-Tu - GTP, 281 K ; Figure 2a and b). Evidently, these molecules are not bound by the protein; with 3, neither of the two
isomers is bound.
The higher affinity of 3’-1 immediately suggests the question
of whether only the esterified 3’-OH group is required for recognition and binding, or if the 2’-OH group is also necessary. The
existence of the free 2’-OH group enables transacylation, which
could also be possibly relevant in the recognition process. To
check this, the binding of the 2’-deoxy derivative 4 was studied.
Under the same conditions as before, the resonances become
only slightly broadened yet again (Figure 2c), which indicates a
distinctly reduced binding as compared to 3’-1. This hints at an
important role of the 2-OH group both in 1 and in the terminal
adenosine of the aa-tRNA. This conclusion is corroborated by
a crystal structure analysis,[” according to which a hydrogen
bond involving a conserved glutamate 271 can be formed. Indeed, a drastically diminished aa-tRNA binding strength is
found for an EF-Tu variant, in which this glutamate was replaced by aspartate (E271D). The ‘HNMR signals of 1 (Figure 3) in the presence of the mutant protein E271D are broadened to a much lesser extent than with the wildtype potein, thus
suggesting a distinctly reduced binding strength. This substanti-
Figure 3. Ribose region of the ' HNMR spectrum (281 K) of 1 with 0.08 mM
a) EF-Tu.GTP wildtype, b) EF-Tu(E271D).GTP, and c) EF-Tu(C82AI
ates the specificity of the interaction, also on the part of the
protein, as the double mutant C82AIT394C gives rise to increases in linewidth that are comparable to those with the wildtype
EF-Tu (Figure 3c). Since intact tRNA is able to make further
contacts with EF-Tu,"] aa-tRNA variants with terminal 2'- and
3'-deoxyadenosines are also bound by EF-Tu. GTP, however,
with decreased affinity.["] To rule out nonspecific interactions
of 1 with proteins, in particular with those from the protein
biosynthesis cycle, the interaction of 1 with the initiation factor
IF-2 was investigated (again 1 mM ligand, 0.08 mM protein; Figure 2d). Here, too, no change in the spectrum as compared to
that of free 1 (Figure la) was detected; this indicates a negligible
interaction between 1 and IF-2.
The structure of free 1 was characterized recently.L6]The conformation of the EF-Tu-complexed molecule can be determined
with transfer-NOE (TrNOE) measurements.[' 141 This exploits
the different signs and magnitudes of the NOE for a small molecule like l (slightly positive) and for a protein-ligand complex
(strongly negative).[14]When the ligand (which is present in a
large excess) exchanges rapidly between enzyme-bound and free
forms, the NOE from the bound state is transferred into the free
state and exceeds the NOE due to the free ligand by far. The
measurement of the TrNOE thus permits a structure elucidation
of the bound ligand. The complex of EF-Tu with the extremely
slowly hydrolyzing GTP analogue GPPNHP was employed for
the conformation analysis, in order to preclude any effects of the
GTP hydrolysis and the concomitant conformational changes
of EF-Tu['. 16] during data acquisition. For technical reasons
the spectra were measured at 298 K. In the NOESY spectrum
(1 mM ligand, 0.05 mM EF-Tu) strong positive cross-peaks are
observed,@'14] which are more intense for the 3' isomer than for
the 2' isomer, due to the preferred binding of the former.
The structure of complexed 1 (Figure 4) was calculated using
the full relaxation matrix. It differs only slightly from the structure postulated for free 1.I6] In particular, the base is anti with
respect to the ribose ring, and the sugar ring adopts a 2'-endo
0 WILEY-VCH Verlag GmbH, D-69451 Weinheim, 1997
Figure 4. a) Structure ~n the EF-Tu.GTP bound state of 1 calculated from NMR
data (MOLSCRIPT [22] representation). b) Conformation of the terminal
adenosine in the aminoacylated tRNAPh' from the crystal structure [7] of the ternary
complex EF-Tu - GTP. Phe-tRNAPh"
puckering. The structure thus obtained can be compared to that
of the terminal adenosine of the aa-tRNA from the ternary
complex EF-Tu.GTP.Phe-tRNAPh" (Figure 4b).[71 The two
structures are in fair agreement, in particular with reference to
base orientation and ribose pucker. Moreover, since the amino
groups of the amino acid of the aa-tRNA and of 1 are positioned similarly, they are able to interact with histidine 273 and
asparagine 285 of EF-Tu.
The investigations presented here suggest that the specific
recognition of EF-Tu by aminoacylated elongator tRNAs most
likely occurs through contacts of the ribose of the terminal nucleotide of the tRNA, to which the amino acid is esterified.
Clearly, the isomer with the amino acid attached to the 3'-OH
position is bound much more strongly. The closest contact of
the terminal aminoacylated adenosine with the protein is
through the 2'-OH position of the ribose and the 3'-ester linkage. These findings could possibly serve as a starting point for
the design of compounds based on 1, which bind to EF-Tu even
better and hence could act as specific inhibitors of the aa-tRNA- EF-Tu interaction and thus the whole bacterial protein
Experimental Sect ion
The synthesis of 1 followed the method In ref. [S]. The preparation and purification
of EF-Tu [IS] and IF-2 1201 from Th. thermophilus overproduced in E. coli were
published previously. Compound 3 was synthesized according to ref. [19]. For solubility reasons, NMR samples of 1 and 2 were dissolved in a mixture of D,O buffer
(50 mM sodium borate pH =7.5; 50 mM KCI) and [DJmethanol (7/1). Methanol
was not required for 3, and [DJDMSO (9/l) was used instead of methanol for 4. For
final sample volumes of 500 pL, the protein was dissolved in 437.5 pL of the buffer
and, after lyophilization, redissolved in the same volume of D,O. A solution of 1 or
2 (8 mM, 62.5 pL) was subsequently added. Compounds 3 and 4 were treated in the
respective solvents.
All N M R spectra were measured at 500 MHz on a Bruker DRX-500 spectrometer.
The data was analyzed with the NDEE program package (Software Symblose
GmbH, Bayreuth, Germany). The NOESY spectra were recorded with 4 k data
points in t , and 512 I , increments with mixing times of 70 or 150ms. The HDO
signal was suppressed by presaturation.
0570-083319713622-2488 $17 50t 5010
Angen Chem Int Ed Ens/ 1997,36,No 22
The XPlor program package [17] was employed to calculate the structure of proteinbound 1 It cooled from 1200 to 50 K within 1 ps with both the geometric and a
relaxation matrix (correlation time for the complex 20 ns) [21] energy terms turned
on. The structures with the smallest R factor were selected.
Received: April 23, 1997
Revised version: July 14, 1997 [Z10376IE]
German version: Angew Chem. 1997, 109,2592-2596
Keywords: molecular recognition
nucleotides proreins R N A
NMR spectroscopy
[I] M. Sprinzl. C. Steegborn, F. Hubel, S. Steinberg. Compilarion of tRNA Sequmce.\ and Sequences ff rRNA Genes, Nucleic Acids Res. 1996, 24, 68-72.
[2] M Taiji, S. Yokoyama, S. Higuchi. T. Miyazawa, J Biochem. 1981, 90, 885888.
131 J. Rudinger. B. Blechschmidt, S Ribeiro, M. Sprinzl, Biochem. 1994,33,56825688.
14) a) J. Jonik, I . Rychlik, J. Smrt, A. HolL, FEES Left. 1979, 98, 329-332; b) J.
Jonik, J. Smrt. A Holy, I. Rychlik, Eur. J Biochem. 1980, 105, 315-320.
[5] L.Servillo, C . Balestrieri, L. Qudgholo, L. lorio, A. Giovane, Eur. J Biochem.
1993.213. 583 589.
[6] B. Nawrot, W. Milius. A. Ejchart, S. Limmer, M. Sprinzl, Nuclric Acids Res.
1997, 25, 94X -~954.
[7] P Nissen. M. Kjeldgaard, S. Thirup, G. Polekhina, L. Reshetnikova. B. F. C.
Clark, J. Nybory. Science 1995, 270, 1464-1472.
(81 0.
Jardetzky. G C. K. Roberts, N M R in Molecular Biology, Academic Press.
Orlando, FL, USA, 1981, pp. 115-142.
[9] L:Y. Lian, G. C. Roberts i n N M R ofMocromolecules(Ed.:G. C. K. Roberts),
IRL, Oxford, 1993, pp. 153-182.
[lo] M. Taiji, S. Yokoyama, T. Miyazawa, J Biochem. 1985, 98, 1447-1453.
I1 I ] M. Sprinzl. M. Kucharzewski, J. B. Hobbs. F. Cramer, Eur. J. Biorhem. 1977,
78. 5 5 ~61
[12] F. Janiak. V. A. Dell, J. K . Abraharnson, B. S. Watson, D. L. Miller, A. E.
Johnson, Blochem. 1990,29,4268-4277.
[I31 A. P. Campbell. B. D. Sykes, Annu. Reis. Biophys. Biomol. Strucr. 1990, 29,
99 122
I141 a ) F. NI. H A. Scheraga, Ace Chem Res. 1994, 27, 257-264; b) F. Ni. Prog.
N M R, 26, 517-606.
[15] D. Neuhaus. M. Williamson, The Nuclear Overhauser Effect in Strucrural and
Confor.mutiodAnalysis. VCH, New York, 1989.
[16] a ) H. Berchtold, L. Reshetnikova, C 0. A. Reiser, N. K Schirmer, M. Sprinzl.
R. Hilgenfeld. N a m e 1993, 36S, 126-132; b) M Kjeldgaard, P. Nissen, S.
Thirup. J. Nyborg, Sfructure 1993, 1 . 35-50.
[17] A. T. Brunger. X-Plor, Version 3.843, 1996
1181 S. Limmer, C. 0 A. Reiser, N. K. Schirmer, N. W. Grillenbeck. M. Sprinzl,
Biodii~m 1992. 32, 2970-2977.
[I91 A. V. Azhayev. S. V. Popovkina, N. B. Tdrussova, M. P. Kirpichnikov. V. L.
Florentiev, A. A. Krayevsky, M. K. Kukhanova, B. P. Gottikh, Nucleic Acids
Rc~.s.1979. 7. 2223 ~2234.
[20] H:P. Vornlocher. W.-R Scheible. H. G. Faulhammer, M. Sprinzl, Eur. J.
Bioc.hem. 1997. 243. 66- 71.
[21] C . R. Cantor. P. R. Schimmel, Biophysical Chemistr?. Part II, W. H. Freeman,
San Francisco. 1980, p. 461
1221 P J Krauhs. .I. A p p l Crysrallogr. 1991. 24, 946-950.
Electrostatic Layering of Giant Vesicles **
controlled. We report here experiments in which a giant vesicle,
an object of considerable recent attention,[’ 61 is successfully
coated with lipid bilayers. The layering is electrostatically driven
and detectable under phase-contrast microscopy. Controlled
layering now joins growth, fusion, undulation. excretion, and
healing15. as observable phenomena with giant vesicles.
Giant anionic vesicles (20-50 pm in diameter) were prepared
from a mixture of l-palmitoyl-2-oleoyl-sn-glycero-3-phosphocholine (POPC), the sodium salt of 1-palmitoyl-2-oleoyl-snglycero-3-[phospho-rac-(l-gIycerol)](POPG), and cholesterol
in a molar ratio of 79:9: 12. Giant cationic vesicles were prepared from POPC, didodecyldimethylammonium bromide
(DDAB), and cholesterol (molar ratio also 79 :9 : 12). Vesicle
formation was carried out with the following protocol: Lipid
mixtures in CHCIJMeOH were stripped of solvent, and the
resulting films vortexed with deionized water. Lyophilization
gave fluffy white powders. Afterwards, the powder (0.1 mg or
less) was smeared onto the microscope slide (within the confines
of a Teflon O-ring cemented onto the slide). The sample was
hydrated with water ( ~ 0 . mL)
for at least 30 min to give a
polydisperse array of vesicular structures.
Giant vesicles of opposite charge were brought into contact
according to the following procedure: A suitable anionic vesicle
was grasped with a polished holding pipet (under slight suction
provided by a picoinjector device) and guided into a vesicle-free
region of the microscope field. Cationic vesicles from another
preparation were then withdrawn into a capillary pipet with the
aid of a hand-controlled microsyringe. This pipet was emptied
of its contents adjacent to the lone anionic vesicle. Finally, the
held anionic vesicle was brought into contact with a “free”
cationic vesicle, and the interaction between the two membranes
was observed as a function of time under 400-fold magnification. Others have shown that light defracted from even a single
bilayer is visible by light micros copy,^'] and that vesicle adhesion
is amenable to micromechanical testing.[’]
Neutral vesicles composed only of POPC and cholesterol
failed to “stick” even when repeatedly pushed into contact; the
free vesicle immediately diffused away from the one held stationary by the pipet. In contrast, two vesicles of opposite charge
“snapped” together (Figure 1). The cationic vesicle formed a
Figure 1. A) A “free” cationic vesicle that adheres to an anionic vesicle held by a
micropipet. B) After about 9 0 s the cationic vesicle coats itself with anionic lipid
from the stationary vesicle and diffuses away. Bar = 25 pm
Fredric M. Menger* and Jason S. Keiper
In the Langmuir-Blodgett method, a vertical plate is dipped
up and down through a water-supported monolayer, thereby
depositing monolayers onto the plate one a t a time.“] Thus, the
thickness and constitution of a multilayered assembly can be
[*I Prof. F. M . Mcnger, J. S. Keiper
Department of Chemistry
Emory University
Atlanta. GA 30312 (USA)
Fax Int. code +(404)727-6586
[**I This work wiib supported by the U. S. National Institutes of Health
Angm Chm7 Inr Ed En@ 1997, 36. No 22
concavity on its surface in order to enhance its contact with the
neighboring anionic vesicle. The configuration was, however,
unstable. After approximately 90 s the cationic vesicle departed
while concurrently “peeling” lipid away from the anionic vesicle, as evidenced by a decided thinning of the anionic membrane. When the same two vesicles were forced into contact
following their separation, they were unable to adhere. Appai-ently, the transfer of lipid from anionic to cationic vesicle generated two multilamellar vesicles, both with outer anionic shells
that repelled each other.
WILEY-VCH Verldg GmbH, D-69451 Weinhelm, 1997
0570-0833/97/3622-2489 S 17 SO+ 50 0
Без категории
Размер файла
669 Кб
trna, factors, mode, specific, protein, recognition, elongation, aminoacylate, bacterial, minimax, biosynthesis
Пожаловаться на содержимое документа