вход по аккаунту


Enzymes Hosted in Reverse Micelles in Hydrocarbon Solution.

код для вставкиСкачать
Volume 24
Number 6
June 1985
Pages 439-528
International Edition in English
Enzymes Hosted in Reverse Micelles in Hydrocarbon Solution
By Pier Luigi Luisi*
Reverse micelles are spheroidal aggregates formed by certain surfactants in apolar media.
In contrast to normal micelles in water, the polar head groups of the surfactant molecules
are directed towards the interior of the aggregate and form a polar core which can solubilize water (the “water pool”); the lipophilic chains are exposed to the solvent. The water of
the water pool exhibits properties that (depending o n the mole ratio of water to surfactant)
differ from those of bulk water. Surprisingly, these reverse micelles are able to solubilize in
hydrocarbon solvents hydrophilic molecules, e. g., enzymes and even plasmids, that are
much larger than the original water-pool diameter. These biopolymer-containing reverse
micelles can be viewed as novel microreactors, whose physical properties can be controlled
through the water content. Remarkable is the ability of enzyme-containing micelles to react
with water-insoluble, hydrocarbon-soluble substrates, as in the example of lipoxygenase
with linoleic acid.
1. Introduction
This review presents the work carried out over the last
few years on enzymes solubilized in apolar solvents with
the help of reverse micelles. The basic properties of reverse
micelles will be described first, followed by a discussion of
the physical and chemical properties of enzymes solubilized in reverse micelles.
Some common surfactants that form such aggregates in
apolar solvents are summarized in Table 1. As is well
known, other types of aggregates can be found in aqueous
such as bilayers, vesicles, and liposomes; these,
however, will not be considered here. Figure I gives a n
idea of the structure of normal (aqueous) and reverse micelles, as well as of the surfactant AOT, which has been
mostly used in our studies. Most important is the fact that
reverse micelles exhibit relatively ordered structures, characterized by a definite (although average) radius, molecular weight, and packing density. Micelles are an ideal subject of investigation for those chemists who are interested
[*] Prof. Dr. P. L. Luisi
lnstitut fur Polymere der Eidgenossischen Technischen Hochschule
ETH-Zentrum, C H-8092 Zurich (Switzerland)
Angew. Chem. In(. Ed. Enyl. 24 (1985) 439-450
Table 1. Aggregation of amphiphilic molecules in hydrocarbon solvents.
n = average aggregation number [a].
Amphiphilic molecule
[“C] range
dodecylammonium propionate
dodecylammonium propionate
dodecylammonium benzoate
dodecylammonium benzoate
sodium 1,2-bis(2-ethylhexyloxycarbony1)- I-ethanesulfonate (AOT)
sodium 1,2-his(2-ethylhexyloxycarhony1)-1-ethanesulfonate (AOT)
lithium dinonylnaphthalenesulfonate
lithium dinonylnaphthalenesulfonate
a-monoglyceryl oleate
[a] Table adapted from [I].
0 VCH Verlagsgesellsehafr mbH. 0-6940 Weinheim. 1985
0570-0833/85/0606-0439 $ 02.50/0
Surfactant Molecule
lipophilic chain
polar head
normal micelles
reverse micelles
CH 2
Fig. I . An idealized reprebentation of normal (aqueous) and reverse micelles
in cross section. The structure of sodium 1,2-bis(2-ethylhexyloxycarbonyl)1-ethanesulfonate (AOT) is also shown. Typical conditions for reverse micelles: isooctane (2,2,4-trimethylpentane), 25- 100 mM AOT, 0.5-2% water.
Reverse micelles are fairly monodisperse, dynamic aggregates, which can solubilize relatively large (ca. looh) amounts of water. For mole ratios
wn=[HzO]/[AOTll>7,the term microemulsion should be used (see text).
in the spontaneous formation of ordered structures. Figure
1 should not, however, convey the idea that micelles are rigid structures: on the contrary, they are characterized by
several dynamic p r o c e s s e ~ . [ For
~ - ~example,
one important
characteristic of reverse micelles is their coalescence upon
collision.141This is assumed to take place through a “transient dimer” having a communication channel that permits rapid exchange of material. This is illustrated in Figure 2, along with the significance of this phenomenon for
the enzymatic activity cycle (see Section 4).
Of considerable importance is the interaction of reverse
micelles with water. Due to their polar core, reverse micelles can solubilize water in apolar solvents. This solubilized water is usually referred to as a “water pool”.121
The reverse micelles formed by sodium 1,2-bis(2-ethylhexyloxycarbony1)- I-ethanesulfonate (AOT) will now be
considered more closely. The system we will mostly be
dealing with is AOT/isooctane/water. An important parameter is the molar ratio of water to surfactant,
wo= [H,O]/[AOT]. This ratio, more than the absolute
amount of water o r surfactant present in the hydrocarbon
solvent, determines most of the structural and physical
properties of reverse m i ~ e 1 l e s . l The
’ ~ ~ water
~ ~ ~ in the water
pool is partly “bound” to the polar wall, and is relatively
“free” only above a certain critical concentration. In the
case of AOT, it appears that the molecules of water up to
perhaps ~ 0 ~ 6 - are
8 tightly bound to the head groups of
the ~urfactant~~.’]
and that only above this value does “free
water” exist in the water pool. In fact, reverse micelles
swell when the water content increase^.^^.^^ Figure 3 (solid
line)15.61illustrates this phenomenon with enzymes as guest
molecules (see Section 3).
Particularly at low wo values, where the micelles are
smaller and a relatively high amount of water is bound to
Fig. 2. Dynamic properties of reverse micelles. a) Exchange of materials between two reverse micellar solutions, which i s the prerequisite for the reaction between
A and B (from [Z]). Coalescence between micelles and exchange of reagents takes place through a transient dimer, a concept elaborated by Eicke et al. [4a]. b) Schematic representation of the site of contact of two reverse micelles prior to coalescence (from [4a]). c) The proposed steady-state cycle of an enzymatic reaction in reverse micelles (from (241).
Angew. Chem. Inr. Ed. Engl. 24 (198s) 439-450
2. Experimental Problems
Figure 4 shows the three techniques that have been applied until now to prepare enzyme-containing micellar solutions. The first method (Fig. 4a) is the one we first used
in our group for peptides”’.’‘] and defined as the “phase
transfer” method. According to this method, the protein is
present initially in an aqueous solution, which is covered
with a layer of the micellar solution of AOT. Upon gentle
stirring, a part of the protein is slowly transferred from the
aqueous phase into the hydrocarbon phase (the process
can be followed spectrophotometrically, for example at
280 nm). This method is relatively slow, but has the advantage that the final system is in thermodynamic equilibrium
and the final hydrocarbon micellar solution, containing
the biopolymer, is generally stable. This point is better understood if one compares the phase transfer method with
the so-called injection method (Fig. 4b), whereby a few mi-
5 c
LO -
20 -
Fig. 3. The radius r of the AOT reverse micelles as a function of wo=
[H20]/[AOTl. (-): “empty” micelles, i.e., micelles containing only water.
This curve corresponds to the data of Eicke et al. [S, 61 (from [24]). (----):
protein= ribonuclease (RNase); (-.-.): protein=liver alcohol dehydrogenase (LADH). All data in this figure and in the following figues refer to room
temperature if not otherwise stated.
the micelle walls, water in reverse micelles behaves “anomalously”-i.e., its physical properties are somewhat different from those of bulk water.[8-’01With increasing water
content, the physical properties of the water pool asymptosurfactant in
tically approach those of pure water. However, even at re- hydrocarbon
latively large w o values, small differences may remain.
The w o region where water becomes free is usually taken
protein in
as the dividing line between micelles and microemulsions.
Microemulsions can indeed be visualized as distinct droplets of water in oil (hydrocarbon), whereas micelles form a
homogeneous system, the water being used only to solvate
the surfactant molecules. This distinction is usually disregarded since we generally study a large w o range and it
would be too confusing to switch nomenclature depending
upon the w o value-furthermore, the dividing line is not so
Fig. 4. The three methods used u p to now to solubilize proteins in reverse miclear. However, the difference between micelles and micelles: a) “phase transfer” method: b) the most commonly used “injection
method”; c) method for water-insoluble proteins.
croemulsions should be kept in mind, because the physical
properties of the system (e.g., mass transport, dynamics,
reactivity and structure of guest molecules) may be different in the two w o ranges.
croliters of the concentrated stock solution of the protein
The discussion above suggests that the reactivity and
are injected into the hydrocarbon solution of AOT (no
conformation of biopolymers solubilized in reverse misonication is used in this or in the other solubilization
celles changes as a function of the micellar parameters, in
methods). If the hydrocarbon solution becomes oversaturated with protein, so that the micellar solution is metastaparticular as a function of wo.Since, with increasing wo,the
physical properties of the water pools approach those of
ble, a decrease of the protein absorption with time might
be observed, which is indicative of protein precipitating
the bulk water, the enzyme-containing micelle can be
viewed as a microreactor, whose physical properties can be
out of the micellar system. However, the method is so fast
cbntinuously changed by simply regulating the amount of
and simple that it has become the method of choice. In
water present in the micellar system. These new biopomost cases, we have been able to obtain stable micellar solymer aggregates may be of biotechnological use and may
lutions of the proteins u p to w o = 5-30 by simply injecting
provide chemists with some basic information about the
a n aqueous stock solution of the protein in the pH range
interaction of proteins with water and with membrane-like
7-10 into the hydrocarbon solution of the surfactant at
room temperature.
Angew. Chern. I n t . Ed. Engl. 24 (198s) 439-450
44 1
Generally, it is important to characterize the stability domain of the protein-containing reverse micellar system. A
simple, relatively rapid way to accomplish this is shown in
Figure 5 , which shows the regions of temperature and of
water content in which the enzyme micellar system is thermodynamically stable. We usually do not work with micellar solutions of proteins unless they are completely transparent: an operative criterion is to measure the scattering
at 330nm and to discard all solutions having an optical
density greater than 0.1.
Fig. 5 . Stability diagram o l AOT reverse micelles in isooctane with and without enzymes. The area underneath the curve represents the stability region
where hydrocarbon micellar solutions are clear. Outside this region, the preparation of micellar solutions results in cloudy and/or unstable systems. A
(-) Stock solution: 0 . 0 5 ~AOT with 5 0 m borate
buffer, pH 8.5; B (- .-.)
stock solution with lipoxygenase ( 0 . 1 6 ~ ~C) ;(----) stock solution with ribo) ; (. . . . ) s t o c k solution with rihonuclease ( 1 7 . 5 ~ ~ ) .
nuclease ( 1 7 5 ~ ~ D
The methods illustrated in Figures 4a and 4b are appropriate for water-soluble proteins. Of great advantage and
interest is the ability of reverse micelles to solubilize waterinsoluble proteins. This is accomplished according to the
method illustrated in Figure 4c, where the insoluble protein, as a powder, is gently stirred with the hydrocarbon
micellar solution containing already a certain amount of
water (for example, wo= 3-30). This solubilization phenomenon can perhaps be understood on the basis of the already mentioned fact that water in reverse micelles may
have different physical properties from those of bulk water, so that its solubilizing ability may be changed. This
method can be useful for water-insoluble membrane proteins, because the protein-containing micelles may mimic
to some extent the protein-containing lipid membranes.
This area of research, however, has not yet been investigated intensively, and there are to the best of our knowledge no systematic investigations on water-insoluble
membrane proteins in reverse micelles. Wuks, Carol et al.
(CNRS, Paris) have been able to solubilize the Folch-pi
proteolipid (or lipophyline) (Fig. 6) and are presently
studying the interaction of this protein with the water-soluble myelin basic protein.'"] Both proteins ensure the lamellar structure of myelin. By the use of reverse micelles it
is possible to study the interaction between the two proteins, one of which is soluble, the other insoluble in water.
AOT [ m ~ ]
Fig. 6. Solubilization 0 1 the tolch-pi proteolipid in isooctane-AOT-water reverse micelles. The overall concentration of the proteolipid in the assay was
20 VM (20 nmoles/mL). A micellar solution of this concentration corresponds to 100% solubilization. a) The percent of solubilization plotted
against w(,=[H20]/[AO'I'l at 300 (-) and 200 (---) mM AOT. b) The percent
of solubilization against AOT concentration at w g = 5.56 (the bars represent
experimental error) (from 1131).
Table 2 summarizes the main studies that have appeared
until now on proteins solubilized in reverse micelles. The
main groups active in this area at the present are those of
Murtinek in the USSR, Robinson in the UK, L a m e in the
Netherlands, and Bulasubrumaniun in India. Several
groups in France ( M . P. Pileni, M . Waks, J. F. Biellmunn)
are studying related problems. Table 2 will be referred to
again in Sections 4 and 5.
Although we have been able to solubilize almost all
kinds of proteins, difficulties are still encountered with the
heme-containing proteins : for example, hemoglobin and
myoglobin are not stable in AOT/isooctane reverse micelles, probably because the heme group dissociates rapidly once the protein is solubilized in reverse micelles.
Also, using AOT, we could not get reproducible data for
the activity of cytochrome P-450.
An interesting practical problem concerns the definition
and determination of the pH in the water pool. The difficulty stems from the fact that one cannot reliably use a
glass electrode in an isooctane solution containing as little
as 1 % water: even if this could be done, no reliable value
could be obtained, because the water of the water pool,
particularly at low w o values, is a novel solvent with unknown properties and for which no pH calibration is available. Problems are probably less severe at high water contents (cf. 114-'71). In general, it appears that the pH of reverse micelles cannot be determined with certainty. We
proposed"61 an empirical calibration based on 31P-NMR
data (measurement of the phosphate ion concentration on
the basis of its chemical shift)-but, to obtain p H values
from these experiments, we had to assume that the p K of
the phosphate ion in reverse micelles is the same as in water-an arbitrary, although quite reasonable, assumption.
Another practical problem concerns the definition of the
concentration of guest molecules inside reverse micelles.
Angew. Chem. Int. Ed. Engl. 24 (1985) 439-450
Table 2. A schematic representation of the activity of various groups in the
field of enzymes in reverse micelles.
Most important studies, main results
Martinek et al. (1978)
a-chymotrypsin and
peroxidase in AOT/
a-chymotrypsin and
other enzymes in
AOT/octane or
chymotrypsin in AOT/
activity studies, enzymes remain active
peroxidase in AOT/
activity of enzyme
much higher than in
substrate specificity
change of relative substrate specifity
kinetic studies, surfactant acts as substrate
Leoashou et al. (1980)
Martinek et al. (1981)
Martinek et al. (1982)
pancreatic lipase in
horse liver alcohol
deh ydrogenase
phospholipase A:, in
Malakhoua et al
(1983) I571
Misioro w k i . Welk
(1974) [58]
determination of k,,,
and K , , theoretical
considerations of enzyme kinetics in reverse micelles
ultracentrifuge studies
o n “filled” and “unfilled” micelles
a-chymotrypsin in
AOT and Cl2E4/octane [b]
kinetic studies with
various substrates, k,,,,
similar to bulk water
Kumar, Balasubramanian (1980) [60]
a-chymotrypsin and
bovine serum albumin
in various surfactants
Menger. Yamada
a-chymotrypsin in
spectroscopic studies
and activity measurements
investigation of activity as a function of
pH and w o
cryoenzymology in reverse micelles
activity comparable
to water, cryoenzymology
model for the protein
in reverse micelles
Fletcher et al. (1984)
(1979) [29]
Douzou et al. (1978)
Douzou et al. (1979)
cytochrome P-450 in
sorbitan tristearate
trypsin in AOT/heptane
Ramakrishnan et al.
(1983) 1631
Darszon et al. (1979)
Schonfeld et al. (1980)
Hilhorst et al. (1982)
Hilhorst, Laane, Veeger
(1983) [39]
rhodopsin in phospholipids/hexane
reaction centers from
hydrogenase in
hydrogenase in
Delahodde et al.
enzyme remains photochemically active
centers remain pho.
tochemically active
photosensitized hydrogen formation with
sensitizers in the interface
multi-enzyme system
with progesterone as
substrate in organic
in AOT/isooctane
solubilization of a water-insoluble protein
in reverse micelles,
spectroscopic studies
Pikeni (1981) [67]
cytochrome in AOT/
reduction of cytochrome by photoelectron transfer
Luisi et al. (1977-79)
142, 121
u-chymotrypsin, trypsin, pepsin, glucagon
in methyltrioctylammonium chloride/
transport of enzymes
into organic phase,
spectroscopic properties investigated, loss
of enzymatic activity
Wo/L Luisi (1979) (231
ribonuclease in AOT/
spectroscopic properties similar t o water,
enzyme remains active
Meier, Luisi (1980)
~ 5 1
horse liver
alcohol dehydrogenase
in AOT/isooctane
spectroscopic investigations and activity
Bonner et al. (1980)
various enzymes in
ultracentrifuge studies
and the proposal of
water-shell model
(1984) 1131
Ed Enyl 24 fIYN5) 439-450
Barharit. Luisi (1981)
a-chymotrypsin in
Grand; el al. (I98 I )
lysozyme in AOT/
activity and spectroscopic properties as
ftmction O f W~ and PH
[a] CTAB = hexadecyltrimethylammonium bromide. [b] C,*Ea= tetraethyleneglycol monododecyl ether.
For simplicity, let us consider one mole of a compound A
that is soluble only in water and is present in one liter of a
hydrocarbon reverse micellar solution containing 1 % water
(Fw=0.01). The “overall” concentration (relative to the total volume, i.e., hydrocarbon plus water) is [A],” = I M ; but
its water-pool concentration (relative to the water “microphase” alone) is [A],,=[A],,/F,
= 100 M (cf.
Velocities, and all other properties having dimensions of concentration, can be expressed by two different numbers. Which
one is physically relevant? This question is not simply semantic, but involves the physics itself of reverse micelles.
For example: is the concentration of N a + (the counterion
of the sulfonate group) in reverse micelles 50 mM or (so/
Fw) mM? This question is also important because it may
generate confusion when comparing different data in the
literature. In some simple instances, however, it is a moot
question and the difference in the final physical property
disappears when the proper normalization is consistently
used throughout. An example is the Michaelis-Menten
constant in enzyme kinetics.[361
Last but not least among the practical problems is the
purity of the surfactant. Three kinds of impurities are
usually present, to a greater or lesser extent, in most commercial preparations of AOT. There are one or more still
unidentified impurities that absorb in the ultraviolet region
(Fig. 7). In addition, commercial preparations of AOT
often contain an acid impurity. This may produce a significant decrease in the pH of the water pool with respect to
that of the stock aqueous solution used to prepare the micelle water pools. We have found two efficient ways to
considerably reduce the UV-absorbing species: HPLC and
a n extraction procedure.[*”’ The acid impurity can also be
eliminated,’341and certain commercial preparations are
much better than others to start with in this regard. Robinson et a1.[i81have pointed out another source of difficulty
with AOT preparations: AOT can partly hydrolyze at alkaline pH. Finally, commercial AOT preparations may
contain a n indeterminate amount of common salt, which is
an important determinant for the conformation of guest
b i o m ~ l e c u l e s . ~ ’In
” ~ principle, therefore, one should d o
routine conductometric measurements to check the ionic
strength of a standard aqueous AOT stock solution.
These questions of chemical purity are now more important than in the past, due to the increased level of sophistication of the research (spectroscopy, enzymatic activity,
conformational properties), and should be taken into account in order to make possible the comparison of data
from different laboratories. We would like to end this discussion with one example. Figure 8 shows the pH-activity
profiles of a-chymotrypsin in water and in two different
AOT micellar hydrocarbon solutions. The first pH-activity
profile is taken from our first investigation,[”] when we
were not yet completely aware of the acid impurities in the
1- 3.0
2 90
2 .o
Fig. 8. The effect of AOT purity on the pH profile of enzymatic reactions: achymotrypsin catalysis of the hydrolysis of N-glutarylphenylalanine-p-nitroanilide. The curve on the right ( w u = 13.5) is taken from our earlier work
1191, in which AOT was purified according to the existing literature procedure. The two curves on the left (w0=9 and M+!= 13.5) are obtained with
HPLC-purified AOT(prepared by P. Liifhr 1301: cf. 1201). Another such example is shown in [20]. pH,,=pH value of stock solution; c,.,/e,,=initial rate/initial enzyme concentration.
NaOH [pmol]
Fig. 7. UV absorption (a) and potentiometric titration (b) profiles of various
AOT samples. The spectra were obtained using 5OmM AOT solutions in UVgrade isooctane, and titrations were performed with 10 mL of a 1 : I (v/v) water-methanol solution containing ca. 2.2 mmol of AOT, using 0.1 N NaOH a s
the titrating agent. The extinction coefficient, E , was calculated as if the absorption were due only to AOT. Thus, an E of 1 L mol ' cm -', for example,
refers to an absorbance of 0.05 in a I-cm cell. ( 0 - 0 ) : commercial AOT
(SERVA); ( A - A ) : purified according to (341; -1
.): purified according
to [54]; (0-0): see text for purification procedure; ( A -A): purified by
HPLC (from [20]).
commercial preparations of AOT and only used the purification procedure described in the literature. In this case,
there is a sizeable shift of the pH-activity profile towards
more alkaline values, which can be explained (all or in
part) by the presence of an acid impurity. The other curve
is with an HPLC-purified AOT preparation;[201in this case,
the micellar solution gives about the same pH-activity profile as the aqueous solution. The result cannot, however, be
generalized : lysozyme and ribonuclease, for example, continue to exhibit a shift of the p H optimum even with the
purified AOT preparations.[20.2"In the case of a-chymotrypsin, the presence of acid impurities in our old AOT
preparations does not seem to affect the conformational
properties (e.g., enhancement of the helical structure) or
the peculiar enhancement of k,,, at low wo over the value
found in aqueous solution."']
3. Structural Aspects
Where and how is the biopolymer situated inside the reverse micelle-or more generally-what can be said about
the structure of the protein-containing reverse micelles?
This question is not completely clarified as yet. In fact, as
depicted in Figure 9, different models are conceivable, and
the answer to the above question may be different for different enzymes. We believe that most of the data for hydrophilic proteins can be interpreted on the basis of the socalled water shell model (see Fig. 9a), according to which
the protein is confined to the middle of the water pool and
is protected by a layer of water from the charged inner wall
of the micelle and from the external organic mi lie^.'^^.^^^
The evidence for this model is still indirect. For example,
the conformation and activity of enzymes depend on
w0,122-273which indicates that the enzyme is sensitive to the
amount of water surrounding it. This would not be so if
the enzyme adhered with its active site region to the internal wall of the reverse micelle, or if the enzyme were predominantly exposed to the hydrocarbon solvent. In the latter case, most enzymes would be denatured. Finally, circular dichroism data for some enzymes, for example, horse
liver alcohol dehydrogenase, ribonuclease, and lipoxygenase, indicate rather similar conformations in the reverse
micelle and in bulk water.'25,28,231
However, lysozyme and
other enzymes show marked difference~.''~.
Additional indirect evidence in favor of the water shell
model is provided by the fact that the kinetic behavior of
Angew. Chem. I n ( . Ed. Engl. 24 (1985) 439-450
Murtinek has presented a model, also based o n ultracentrifuge studies, according to which the uptake of a-chymotrypsin does not produce any significant increase in size of
the host reverse m i ~ e l l e . [ ~Our
~ , ~own
’ ~ studies using neutron scattering (carried out at Oak Ridge National Laboratory in collaboration with L. Magid) appear, on the contrary, to give dimensions for the lysozyme-containing micelles that are at least 30% larger than those predicted by
our original model. The situation is therefore still in a state
of flux.
4. The Reactivity of Enzymes in Reverse Micelles
Fig. 9. Possible models for a protein hosted in a reverse micelle (see text): a) the
water-shell model, in which the protein is located in the middle of the water
pool and is protected from the micelle wall by a layer of water; b) the protein
has a very lipophilic part, which tends to interact directly with the bulk hydrocarbon: c ) the protein is adsorbed to the micelle wall; d) the protein is
solubilired by several small micelles, whose hydrocarbon tails interact with
the hydrophobic portions of the protein.
enzymes in reverse micelles appears to be essentially the
same as in aqueous s o l ~ t i o n . [ Furthermore,
~ ~ - ~ ~ ~ the pH-activity profile for enzymes in reverse micelles, although
somewhat shifted with respect to bulk water, forms a very
similar bell-shaped ~ u r v e . I ~If~the
. ~ enzyme
~ . ~ ~ adhered
the inner wall of the micelle by electrostatic effects, one
would expect rather dramatic changes of activity when the
enzyme, becoming more negatively charged, dissociates
from the micelle wall. For a-chymotrypsin, further circumstantial evidence comes from fluorescence studies: the
fluorescence properties of a-chymotrypsin in reverse micelles for wo? 7 (i.e., when “free” water is present) are very
close to its fluorescence properties in bulk water,”’] indicating that the environment of the active protein is essentially aqueous.
Using the water shell model, we have developed a
scheme, partly based on ultracentrifuge experiment^,"^'
which enables us to approximate the dimensions of the
protein-containing micelles and to estimate the micellar
parameters characterizing such aggregates. This model was
based on several assumptions, the most important being
that the volume of the protein-containing micelle is the algebraic sum of the volume of the protein and the volume
of water molecules originally present in the “empty” micelle, and that w o does not change. The two dashed curves
in Figure 3 show the dimensions of the protein-containing
micelles as a function of wo according to our
Noticeably, for small proteins such as ribonuclease, there is
very little difference in size between “empty” (containing
only water) and “filled” micelles (containing the protein)
for w o 2 15. In contrast, for small values of wo, the protein
induces an enlargement of the reverse micelles (with the
consequent mass redistribution in the micellar solution). In
the case of a larger enzyme like horse liver alcohol hydrogenase (M,.= SOOOO), the micelles that contain guest molecules will always be larger, even at large w o values, than the
empty micelles. It is now doubtful whether the first approximation model is valid for all enzymes. The group of
Angen,. Chem. lnr. Ed. Engl. 24 (1985) 439-450
The preceding discussion provides the basis for understanding qualitatively which kind of micellar parameters
may affect the activity of an enzyme hosted in reverse micelles. When there is enough “free” water (namely at
w,>7-8), the activity should be very similar to that found
in bulk aqueous solution, and the steady-state cycle may
be represented simply according to Figure 2c: enzyme and
substrate molecules form an active complex that eventually
decays to products which dissociate from the enzyme. The
difference with respect to water solution is that in micellar
solution each bimolecular step requires micellar collision
and coalescence.[241
At lower wo values, when most of the water molecules
are bound to the inner surface of the AOT micelles, one
may expect some interaction between the protein surface
and the inner wall of the reverse micelles, which may affect
the mechanism. Also, at low wo values, deviations from the
normal activity profile can be expected because the water
in the water pool is a novel solvent. Another effect that
may influence the enzyme activity in this regard is the local
pH. The transport rate of the substrate into the reverse micelles or the transport rate of the product out of them can
also, in principle, play a role. Depending upon the structure of the micelle or upon the relative concentration of
micelles and substrate molecules, these events can possibly
become the rate-limiting step.
With all this in mind, we will now consider some of the
enzymes studied to date. Figure 10 shows the activity of lysozyme as a function of p H in water and in reverse micelles. Notice that the maximal velocity is at about w o= 10
and not at the maximal water content. This is surprising
and can be generalized to all hydrolases. Notice also the
shift of the pH-optimum with respect to water, which can
possibly be ascribed to a difference in p K of the amino
acid residues involved in the catalysis. The turnover number at the optimal p H is not much different from that in
water: it appears therefore that lysozyme, apart from the
pH-shift of the activity, behaves quite normally in reverse
micelles. However, things are not so simple if the conformation of the protein in reverse micelles is considered. As
shown in Figure 11, there are large differences in the circular dichroism (CD) spectra on going from water to micellar
As is well known, the circular dichroism
spectrum in the near-ultraviolet region reflects the environment of the aromatic amino acid residues of the protein,
while the CD spectrum in the far-ultraviolet region, particularly around 220 nm, reflects the secondary structure of
the protein. The near-UV region will be considered first.
O 1
, wo
Fig. 10. Degradation of chitin with lysozyme in micellar solutions prepared
from 50 mM AOT/isooctane; dependence of the rate constant k,,, on the water content (a) and p H (b). Curve A, for comparison, represents an aqueous
solution. Curve B refers to the micellar solutions. 0 : pH of the stock solutions used to prepare the reverse micelles (pH=pH,,); 0 : pH values measured directly in the reverse micellar solutions using "P-NMR at w g = 16.8
(from [16] and 1221).
Since the ultraviolet absorption spectra of lysozyme in this
region are the same in water and in reverse micelles,12z1the
changes observed in the C D spectrum can be safely ascribed to conformational changes. The orientation of several aromatic side chains of lysozyme is therefore probably
different in water and in reverse micelles. The C D spectrum in the lower wavelength region, where the ultraviolet
absorption spectra of micellar solutions also show an increase in intensity, can be interpreted in terms of an increase in the helical structure. Thus, the C D spectra of Figure 11 show that the enzyme has lost its conformational
specificity without loosing activity. In order to interprete
this apparent paradox, we have proposed that the changes
in the enzyme's conformation may only involve peripheral
regions[221rather removed from the active site. However,
one should recognize that lysozyme has three of its five
tryptophan residues adjacent to o r within the active site; it
is thus hard to conceive that the conformation of these aromatic residues is changed without affecting the overall active-site geometry. We are presently carrying out CD studies on the enzyme in reverse micelles in the presence of inhibitors and/or substrates in order to clarify this point".
[*I Note added in proof: The problem has now been clarified. The CD spectra of lysozyme in reverse micelles formed by AOT in isooctane d o not
change significantly with respect to water when NAG (N-acetylglucosamine) or NAG3 is bound. By inference, the conformation of the enzyme in
AOT reverse micelles remains the same (or nearly the same) as in aqueous solution when the substrate is present. Without the substrate, AOT
brings about a rapid denaturation of the enzyme (lysozyme appears to be
Fig. 11. Near- (a) and far- (b) ultraviolet circular dichroism spectra of lysozyme in water (A) and in AOT reverse micelles at w n = 3 (B), w,,=6 (C),
wn= 16 (D), and wo=25 (E).
the only enzyme where this happens thus far), with the CD spectrum
changing a s shown in Figure 11. I n other words, the lysozyme in the reverse micelles of Fig. 1 I is denatured. The reason why our earlier investigation failed to notice this is due to the fact that the enzyme was added t o
a reverse micellar solution already containing the substrate, so that the
enzyme did not have time to denature (B. Steinmann, P. L. Luisi, in preparation).
Angew. Chem. Int.
Ed. Engl. 24 (1985) 439-450
The situation with chymotrypsin is also very interesting.
As apparent from Table 2, this case has been studied by at
least four different groups. Menger and Y u r n u d ~ the
first to study this enzyme in reverse micelles, initially failed
to find activity with an ester substrate at a nominal p H of
7 . The reason was that the micellar environment had
markedly changed the pH-activity profile of the enzyme.
In fact, they found a very good activity in the alkaline p H
range. We found that, with N-glutarylphenylalanine-p-nitroanilide, the turnover number is larger at low wo values
than in water.[”] Although the effect is not extremely large,
it is certainly worthy of attention. Of particular interest in
this case is the relationship between the activity and the
conformation of the main chain of the enzyme, as determined again by circular dichroism. Here, too, there is an
enhancement of the ellipticity at 220 nm, which can again
be interpreted as an increase of helical content. The activity enhancement has been interpreted“’] by postulating
that in reverse micelles the conformational equilibrium of
a-chymotrypsin is shifted towards the more active form of
the enzyme. a-Chymotrypsin is not the only example of
“superactivity”. An increase has also been found for ribonuclease and lysozyme,[2’~23.241
again at low w o values. Murhave reported a large increase of activity
tinek et al.r27,351
for peroxidase (Fig. 12).
ported by the Russian group of Martinek: the relative specificity of alcohols in reverse micellar solutions appears to
be different from that found in aqueous solution.[351
Concerning the kinetic parameters, an important question is whether K , in reverse micelles is different from that
in water. We have already indicated[361that K , in reverse
micelles should be expressed as K,.,, in order to be properly compared with the values in aqueous solution. Table 3
lists typical K , values in water and reverse micelles. Generally, there is a close agreement between the K , values in
water and in reverse micelles. The efficiency of the enzymes (usually expressed by kcat/K,,,)is also not impaired,
therefore, when enzymes are solubilized in reverse micelles. The stability of enzymes in reverse micelles is important for kinetic studies and is also of direct relevance
for the biotechnological applications (see Section 5). To
date, there have been few systematic studies on this important subject. In the case of chymotrypsin,~’’]the time stability depends on w o ;at low wo,the enzyme can be more stable and more active than in water. Martinek et al. have
studied the stability of several enzymes in reverse micelles
formed by different surfactants and/or organic solvent~.[~~’
Table 3. Kinetic parameters for enzymatic reactions in reverse micelles
(isooctane/AOT/water). pH,, = p H value of the stock solution [a].
LADH in water [c], p H 7.1
LADH in water, pH 9
LADH in micelles,
~ 0 = 4 2 pH,,=8.8
LADH in micelles,
wg= 19, pH,,=9.2
lysozyme in water [d], p H 5.1
lysozyme in micelles,
w0=22.2, pH,,=7.6
a-chymotrypsin in water [el,
pH 7.9
a-chymotrypsin in micelles,
wg= 13.5, pH,,=9.8
a-chymotrypsin in micelles,
w0=9, pH,,= 11.8
1 I4
[a] Adapted from [53]. For further details see [ 19, 22, 251. [b] Expressed as
“overall” concentration. The numerical value of K,,,, depends on the water
content of the micellar system. [c] Horse liver alcohol dehydrogenase, with
acetaldehyde as substrate. [d] Chitin oligomers as substrate. [el N-glutarylphenylalanine-p-nitroanilideas substrate.
Surfactant [MI
big. 12. Dependence 01‘ the firat-order rate constant k,,,, for peroxidase oxidation of pyrogallol on the concentration of surfactant (from 121). (0- 0 ) :
AOT in octane, 0 . 1 imidazole
buffer; ( 0 - 0 ) :AOT in octane; ( A - A ) :
dodecylammonium propionate in diethyl ether/benzene (1 :2); (0- 0) egg
yolk lecrthin in methanol (2.5%) v/v)/pentanol (5% v/v)/octane. 2 6 T ,
wil= 13, 0.025 M aqueous buffer (phosphate-borate-acetate), pH 7.0. The kinetic assay was carried out as described in (371.
Dehydrogenases are particularly interesting from the
point of view of preparative bioorganic chemistry. As is
apparent from Table 2, horse liver alcohol dehydrogenase
has been investigated by several groups. In addition to the
reduction of acetaldehyde, the reduction of a highly lipophilic substrate (decanal) and of a steroid hormone has
also been
An interesting finding has been reA n y e w . Chenr I n r . Ed. Engl. 24 (1985) 439-450
5. Biotechnological Relevance
Until now we have considered hydrophilic enzymes with
their water-soluble substrates. A very interesting aspect of
enzyme-containing reverse micelles is their ability to accept and transform water-insoluble, hydrocarbon-soluble
substrates. An example is lipoxygenase. The enzyme, although lipophilic, is readily soluble in water (and insoluble
in hydrocarbons) and, as such, is probably hosted in the
water pool of reverse micelles. The fact that its kinetic behavior corresponds closely to that observed in water supports this view. The substrate linoleic acid is not, o r only
very sparingly, soluble in water at neutral or acidic pHs. It
is soluble, however, in isooctane, and can be added in its
pure liquid form to a micellar solution containing lipoxy447
genase. Spectrophotometric analysis of the solution reveals
that the substrate is enzymatically converted into the expected products.[2x1Typical data are shown in Figure 13.
The important point here is that the substrate dissolved in
m \
sistant, semipermeable, hollow fibers, in which the enzyme-containing reverse micelles are entrapped. Thus, the
product can be obtained in pure form and physically separated from the enzyme and reagents.
Although such reactions may be interesting in organic
synthesis, until now there has been no example of relevant
practical applications of enzymes in reverse micelles. As
apparent from Table 2, Hilhorst, Lame. and Veeger‘zy.4i1
Wageningen (Netherlands) have come closest to such a
goal. They were able to solubilize three enzymes in reverse
micellar solution at the same time; in a chain reaction,
these enzymes reduce regioselectively a ketosteroid. As
shown in Figure 14, molecular hydrogen is transferred by a
hydrogenase (H,ase) to methyl viologen (MV), which in
turn is able to recycle the system NAD/NADH with the
help of lipoamide dehydrogenase (LipDH). Finally,
NADH reduces the ketosteroid in a reaction catalyzed by a
specific steroid dehydrogenase (HSDH).
Fig. 13. pH-activity profile of the lipoxygenase reaction. Enzymes in reverse
micelles react with both water-soluble and water-insoluble (hydrocarbon-solublej substrates. An example is the oxidation of linoleic acid with lipoxygenase. [enzyme]=2.5x IO-*M, [substrate]=4x 1 0 - 4 ~ ,[AOTJ= l o x IO-’M,
isooctane. Overall concentrations are given (u,,Je,)as in Fig. 8).
20 p- h ydroxysteroid
organic phase
the bulk hydrocarbon finds its way to the enzymatic active
site in the reverse micelle, i.e., through the layer of water
that surrounds the enzyme. How this can happen is not yet
clear. It may be that the particular physical properties of
the water in the reverse micelles, for example, a lower dielectric constant, result in a better solubility of lipophilic
substrates. The other possibility, particularly in the case of
linoleic acid or analogous substrates, such as arachidonic
acid, is that they are incorporated into the micellar interface as cosurfactants and that the enzyme interacts dynamically with the interior micellar wall. The solubility of lipophilic substances in reverse micelles is, however, not restricted to fatty acids ; steroid^,^^^.^^^ long-chain aldehyde~,[~’]
or water-insoluble peptidesl4’I are also readily accepted as substrates by their respective enzymes in reverse
micelles. The fact that the mechanism of this uptake is not
yet clarified does not decrease the biotechnological relevance of the finding that water-insoluble, hydrocarbon-soluble substrates can be enzymatically converted into their
products (for a review on the few results available, see 12Ri).
This may open the way to a new application of enzymes in
organic chemistry and biotechnology. The interest is increased by the observation that almost all enzymes tested
until now can be solubilized in reverse micelles-it is
therefore a general method (by contrast, the use of enzymes in biphasic systems, for example, does not appear to
be so general, since many enzymes rapidly denature under
such conditions). Preliminary work involving the use of achymotrypsin for the synthesis of peptide bonds should
also be mentioned: acetylalanylphenylalanineethyl ester
and leucinamide are coupled enzymatically to yield the
water-insoluble, isooctane-soluble, protected tripeptide
AcAla-Phe-LeuNH2, which, after synthesis in the water
pool, is expelled into the bulk hydrocarbon. The reaction
is carried out in an enzyme reactor‘4o1with hydrocarbon-re448
Fig. 14. The achrinc. 0 1 H I / / I ~ I I Laurie.
and keeyer [3Y]: three enzymes in reverse micelles cooperate to selectively reduce a water-insoluble ketosteroid in
a chain reaction (see text).
The enzymatic conversion of lipophilic substrates is not
the only field where reverse micelles can be used for biotechnological applications. Another interesting aspect
stems, for example, from the phase transfer of proteins and
nucleic acids. Indeed, our interest in the whole field originated in the transfer of amino acids and peptides, and later
polypeptides and proteins, from an aqueous to a hydrocarbon solution containing the surfactant.” ’. 12.421 F’igure 15
shows the three possibilities of phase transfer (cf. I4‘l). In
the “forward transfer”, the protein is transferred from an
aqueous solution to the supernatant micellar solution. In
the “backward-transfer”, the enzyme, solubilized in hydrocarbon, is recovered as an aqueous solution. In the “dou-
big. 15. I’hrlbc. trandrr procc\>ca. a) In the “lorward-tran,ler“ the prolein IS
transferred from the water phase A into the micellar solution B; b) in the
“backward transfer,” the protein is transferred from the hydrocarbon micellar solution into the aqueous solution; c j in the “double transfer” method.
the protein is vectorially transported from a first water phase A, into a second water phase Az via reverse micelles present in the hydrocarbon bridge.
The vials are gently shaken during the experiment. The shaking is interrupted
periodically for the removal o fsmall aliquots to monitor the concentration of
the transported biopolymer.
Angew. Chem. Int. Ed. Engl. 24 (1985)439-450
ble transfer” experiments, the biopolymer is present initially in an aqueous phase, and is then transferred into a
second aqueous phase, which is separated from the first by
a hydrocarbon bridge containing the reverse micelles. In
all cases, the transfer of the protein is carried out by the
micelles. We have presented until now only preliminary
data, but it appears that the method can possibly be used
as a chromatographic separation technique for biopolymers. There are already several groups engaged with such
project^.[^^.^"' The separation of proteins and nucleic acids
should also be possible, since the transfer of D N A is much
slower than that of proteins.
The use of reverse micelles as a drug delivery system by
should also be mentioned (cf. ‘““I).
The biotechnological relevance is actively being explored, but we will have to wait a few years to see whether
and to what extent the present expectations are fulfilled.
hydrocarbon medium does not interfere spectroscopically
with water or nucleic acids. Studies in this direction are in
progress in our group.[”’
6. Nucleic Acids
The internal wall of AOT reverse micelles is negatively
charged, and although there is a stoichiometric amount of
positively charged sodium ions, the solubilization of negatively charged polyelectrolytes such as RNA or D N A was
somewhat of a surprise. In fact, the solubilization of nucleic acids in hydrocarbon micellar solutions occurs readi l ~ [ ~ and
’ ’ is actually easier at alkaline pHs. The mechanism of this uptake is not clear yet. We know that the solubilization of high-molecular-weight D N A can only take
place at relatively high w o values (for example, above 14).
Proteins, on the other hand, behave differently.
Does the structure of a high-molecular-weight nucleic
acid change upon insertion into the reverse micelles? The
solubilization is attended by a decrease of absorbance in
the 260-nm region, with a corresponding increase of ellipti~ity.’~’’
This can be interpreted as an increase in the basepair interactions (more hydrogen bonds are formed),
which brings about an increase in the UV hypochromic effect and an increase in the macromolecular conformational
rigidity. The picture is thus not unlike that proposed for
proteins. The effect is more pronounced for RNA, which is
more flexible in water solution, than for DNA, which exists already as a rigid double helix; it is not significant for
low-molecular-weight oligonucleotides.
Of particular interest is the observation that the plasmid
pBR 332 (M,=2.7 x 10“) can also be solubilized in isooctane/AOT/water reverse micelles. We are still investigating the structure of such aggregates. It is really not easy to
imagine how a small micelle can accept such a large molecule. Certainly, the uptake involves an enlargement of the
original micelle, with a redistribution of its components.
The interest in this solubilization experiment also stems
from the fact that the C D spectra of the plasmid-containing reverse micelles (Fig. 16) belong to the so-called \v family.‘“’’ Such spectra in turn are supposed to be diagnostic
for superpackaged DNA, like that present in vivo when the
long macromolecules are obliged to condense into a geometrically very restricted environment (e.g., a virus head or
chromatin). The spectroscopic study of such structures in
reverse micelles should be facilitated by the fact that the
Angew. Chem I n l . Ed. Engl. 24 11985) 439-450
k, [ n m l
Fig. 16. CD spectra of the DNA of the plasmid pBR 322 in AOT reverse micelles (y-spectra). ( . . . - . . .): borate buffer; (-):
wo= 18.5: (---): AOTiisooctane, wil= 14.8 [69].
7. Concluding Remarks
Biopolymer-containing reverse micelles are not only interesting because of their biotechnological relevance, but
also because such structures may exist in vivo in biological
m e m b r a n e ~ . ~ ~ ~Perhaps
the first hint of this derives from
the well-known hexagonal structure that is occasionally
formed by phospholipids. The group of de KrujyfS in
U t r e ~ h t ’ ~ ’ .has
~ ~ presented
very convincing arguments in
favor of the biological relevance of reverse micelles. However, the question of the biological relevance of reverse micelles in membrane structure remains to be definitly answered (cf. [381). The question of the possible existence of
reverse micelles in vivo bears upon the more general one
of whether and to what extent reverse micelles can be considered as proper models for biological membranes. The
term “biological model” is a misused one, as pointed out
elsewhere (cf. appendix to 13“]) and we d o not want to dwell
here further on this question.
Despite the uncertainty about the biological and biotechnological relevance of biopolymers in reverse micelles,
this review demonstrates the fascination of the field. It
stretches over quite different areas, from biotechnology to
pure biology, from basic enzymology to nucleic acid structures, and touches upon a variety of important physical aspects. These include the transfer of macromolecules across
interphases, or the transfer of low-molecular-weight materials across interfaces, as well as the structure and stability
of novel aggregates of biopolymers.
I thank my students Ergin Irnre, Peter Luthi, and Bettina
Steinrnann-Hoffrnann, as well as Dr. Hans Jackle and Prof.
Lee Magid, uisiting professor from the University of Tennessee. for very helpful discussions
Received: June 25, 1984 [A 536 IE]
German version: Angew. Chem. 97(1985) 449
J. H. Fendler: Membrane Mirneric Chemistry. Wiley, New York 1982.
F. M. Menger, J. A. Donohue, R. F. Williams, J. A m . Chem. Soc. 95
(1973) 286.
S. Harada, Z. A. Schelly, J. Phys. Chem. 86 (1982) 2098.
a) H. F. Eicke, T. C. W. Shepherd, A. Steinemann, J. Colloid Interface
Sci. 56 (1976) 168; b) B. H. Robinson, D. Steytler, R. Tack, J. Chem. Sac.
Faraday Trans. I 75 (1979) 481; c) R. Zana, J . Lang in K. L. Mittal, J. H.
Fendler: Solution Behauiour of Surfactants, Vol. 2, Plenum, New York
H. F. Eicke, J. Rehak, Helu. Chim. Acta 59 (1976) 2883.
M. Zulauf, H. F. Eicke, J . Phys. Chem. 83 (1979) 480.
M. Wong, J. K. Thomas, T. Nowak, J. Am. Chem. Sac. 99 (1977) 4730.
H. F. Eicke, P. Kvita in 191, p. 21.
P. L. Luisi, 8 . Straub: Reuerse Micelles - Technological and Biological
Releuance. Plenum, New York 1984.
L. Gierasch in 191. p. 265.
A. Dossena, V. Rizzo, R. Marchelli, G. Casnati, P. L. Luisi, Biochim. Biophys. Acta 446 (1976) 493.
P. L. Luisi, F. Henninger, M. Joppich, A. Dossena, G. Casnati, Biochem.
Biophys. Res. Commun. 74 (1977) 1384.
A. Delahodde, M. Vacher, C. Nicot, M. Waks, Arch. Biochem. Biophys.
231 (1984) 86.
0. A. El Seoud, R. C . Vieira, J . Colloid Interface Sci. 93 (1983) 289.
9 . Robinson in [9], p. 69.
R. E. Smith. P. L. Luisi. Helu. Chim. Acta 63 (19801 2302.
1171 0 . A. El Seoud in 191, p. 81.
[I81 P. D. I. Fletcher, A. M. Howe, 9. Robinson, D. C. Steytler in (91, p. 73.
[I91 S. Barbarit, P. L. Luisi, J. Am. Chem. Sac. lfl3 (1981) 4239.
1201 P. L. Luisi, P. Meier, V. E. Imre, A. Pande in 191, p. 323.
121) B. Steinmann-Hoffmann, P. L. Luisi, in preparation.
1221 C. Grandi, R. E. Smith, P. L. Luisi, J . Biol. Chem. 256 (1981) 837.
1231 R. Wolf, P. L. Luisi, Biochem. Biophys. Res. Commun. 89 (1979) 209.
[24] F. 1. Bonner, R. Wolf, P. L. Luisi, J. Solid-Phase Biochem. 5 (1980)
1251 P. Meier, P. L. Luisi, J . Solid-Phase Biochem. 5 (1980) 269.
(261 K. Martinek, Y . L. Khmelnitsky, A. V. Levashov, N. L. Klyachko, A. N.
Smenov, I. V. Berezin, Dokl. Akad. Nauk S S S R 256 (198 I ) 1423.
(271 K. Martinek, N. L. Klyachko, A. V. Levashov, I. V. Berezin, Dokl. Akad.
Nauk S S S R . in press.
1281 P. L. Luisi, P. Luthi, I. Tomka, A. Pande, Proc. VII Int. Enzyme Eng.
Conf., Ann. N . Y. Acad. Sci. 434 (1984) 549.
1291 F. M. Menger, K. Yamada, J. A m . Chem. Sac. I01 (1979) 6731.
[30] P. Liithi, Dissertation, ETH Zurich, in preparation.
1311 K. Martinek, A. V. Levashov, Y . L. Khmelnitsky, N. L. Klyachko, V. Y.
Chernyak, I. V. Berezin, Dokl. Akad. Nauk S S S R 258 (1981) 1488; Anal.
Biochem. 118 (1981) 42.
1321 C. R. Cantor, P. R. Schimmel: Biophysical Chemistry, Val. 2, Freeman,
San Francisco 1980.
133) M. Montal in 191, p. 221.
1341 C. A. Martin, L. J. Manid, J . Phys. Chem. 85 (1981) 3938
K. Martinek, A. V. Levashov, Y. L. Khrnelnitsky, N. L. Klyachko, I. V.
Berezin, Science 218 (1982) 889.
P. L. Luisi, L. Magid, Proc. XVlI Int. Solvay Conf., CRC Press, Boca
Raton, FL, in press.
K. Martinek, A. V. Levashov, N. L. Klyachko, V. 1. Pantin, I. V. Berezin,
Biochem. Biophys. Acta 657 (198 I ) 277.
J. Seelig in (91, p. 209.
R. Hilhorst, C . Laane, C . Veeger, FEES Left. IS9 (1983) 31.
P. Luthi, P. L. Luisi, J. Am. Chem. Sac. 106 (1984) 7285.
R. Hilhorst, R. Spruijt, C. Laane, L. Veeger, Eur. J. Biochem.. in press.
P. L. Luisi, F. Bonner, A. Pellegrini, P. Wiget, R. Wolf, Helu. Chim. Acra
62 (1979) 740.
K. van? Riet, Proc. 3rd Eur. Conf. Biotechnol., in press.
T. A. Hatton et al., in preparation.
P. P. Speiser in 1. T. Lingle, D. D. Breimer: Applied Eiol0g.v and n e r a peutics. Vol. 6 , Elsevier, New York 1983.
P. L. Luisi, V. E. Imre, H. J k k l e , A. Pande in D. D. Breimer, P. P. Speiser: Topics in Pharmaceutical Sciences, Elsevier, New York 1983.
V. E. Imre, P. L. Luisi, Biochem. Biophys. Res. Commun. I07 (1982)
L. S. Lerman in J. Duchesne: Physicochemical Properties of’Nucleic Acids.
Vol. 3, Academic Press, London 1973, p. 59-76.
8 . d e Kruijff, P. R. Cullis, A. J. Verkleij. TIES March 1980, p. 79.
P. R. Cullis, B. d e Kruijff, M. J. Hope, R. Nayar, S. L. Schrnid, Can. J.
Biochem. 58 (19x0) 1091.
I. Wirz, J. P. Rosenbush in 191, p. 231.
P. R. Cullis, M. J. Hope, Nature (London) 271 (1982) 672.
P. L. Luisi, R. Wolf in K. L. Mittal, J. H. Fendler: Solution Behaaiour of’
SurJactants, Val. 2, Plenum, New York 1982.
M. Wong, J. K. Thomas, M. Gratzel, J. Am. Chem. SOC. 98 (1976)
K. Martinek, A. V. Levashov, N. L. Klyachko, 1. V. Berezin, Dokl. Akad.
Nauk S S S R 236 (1978) 951.
A. V. Levashov, N . L. Klyachko, V. I . Pantin, Y. L. Khmelnitsky, K.
Martinek, Bioorg. Khim. 6 (1980) 929.
E. A. Malakhova, B. I. Kurganov, A. V. Levashov, 1. V. Berezin, K. Martinek, Dokl. Akud. Nauk S S S R 270 (1983) 474.
R. L. Misiorowski, M. A. Wells, Biochemistrl; 13 (1974) 4921
P. D. 1. Fletcher, R. B. Freedman, J. Mead, C. Oldfield, B. H. Robinson,
Colloids Surfaces 10 (1984) 193.
1601 C . Kurnar, D. Balasubramanian, J . Colloid Inleiface Sci. 74 (1980) 64.
523 (1978) I .
[61] P. Douzou, P. Debey, F. Franks, Biochim. B i o p h , ~ Acta
1621 P. Douzou, E. Keh, C. Balny. Proc Nail. Acad. Scr. USA 76 (1979)
1631 V. R. Ramakrishnan, A. Darszon, M. Montal, J . B i d . Chem. 258 (1983)
[64] A. Darszon, R. Strasser, M. Montal, Biochemisrry I8 (1979) 5205.
[65] M. Schonfeld, M. Montal, G. Feher, Biochemisrry 19 (1980) 1535.
1661 R. Hilhorst, C. Laane, C. Veeger, Proc. Null. Acad. Sci. USA 79 (1982)
1671 M . P. Pileni, Chem. Phys. Lett. 81 (1981) 603.
1681 V. R. Ramakrishnan, A. Darszon, M. Montal, J . Biol. Chem. 258 (1983)
[69] E. Imre, E. Battistel, P. L. Luisi, in preparation.
Angew. Chem. Int. Ed. Engl. 24 (I985) 439-450
Без категории
Размер файла
1 109 Кб
reverse, solutions, enzymes, hydrocarbonic, micelle, hostex
Пожаловаться на содержимое документа