Note: Descriptions are shown in the official language in which they were submitted.
3~V7~3
This invention relates to an improvement in
an enzyme reaction in an aqueous phase. More specifi-
cally, this invention relates to a method of preventing
the activity of an enzyme from being degraded by a
substrate in a reaction utilizin~ the enzyme, in which
by adding an organic solvent misci~le with the substrate
but immiscible with water to the reaction system, the
reaction is carried out while maintaining the concen-
tration of the substrate in the aqueou~ phase below
that concentration at which the ac~ivity of the en~yme
is inhibited (the so-called inhibitory concentration~.
In a reaction utilizing an enzyme, water is
the best reaction medium for the en2yme, and therefore,
such a reaction is usually carried out in an aqueous
phase. Some substrates, however, have low solubility
in water. In the case of using such a substrate~
various measures are taken in order to secure an
amount of the substrate required for the reaction in
an aqueous phase containing an enzyme (to be sometimes
referred to as an enzyme-containing liquidl and carry
out the reaction smoothly. For example, it is known
~o perform the reaction using an organic solvent.
Japanese Patent Publication No. 341S3~1972 discloses
that ~oluene, butanol, ethaol or acetone is used as a
reaction promoter in a process for producing L trypto
phan from indole and serine by utilizing a bacterium
of the genus Aerobacterium. British Patent Publication
No. GB-13134 discloses that in the oxidation of decane,
a water-insoluble substrate, in the presence of a
decane-oxidizing enzyme, dimethylformamide miscible
with water and the substrate is usedO
In these methods, the organic solvents are
used to aid in dissolving the substrates in the enzyme-
containing liquid and to promote the reactions.
In reactions utilizing enzymest the efect of
the concentration of a substrate on an enzyme should
be considered. If in an enzyme reaction, the concen-
tration of a substrate in the reaction phase is hi~h,
the activity of the enzyme may be markedly reduced to
retard the reaction. It is known, for example, that
in the production of tryptophan in the presence o~ an
enzyme using anthranilic acid as a precursor, the
concentration of anthranilic acid in the aqueous ph~se
affects the degradation of the activity of the enzyme
used in the reaction tU. S. Patent Specification No.
4,363,875); in the production of L-DOPA from L-serine
and pyrocatechol by utilizing Erwinia herbicolal the
presence of pyrocatechol as one substrate in a high
concentration in the aqueous phase reduces the activity
of the enæyme (Agric. Biol. Chem., Vol. 37, 493-499,
725-735, 1973); and in the production of tryptophan
from indole and serine, indole as one substrate reduces
the activity of tryptophan synthetase (U. Behrendt, The
First European Congress on Biotechnologyt 2/186-~/189,
1978).
Accordingly, i~ in a reaction utilizing an
enzyme, a substrate is likely to reduce the activity
of the enzyme, the reaction should be carried out
S while maintaining the concentration of the substrate
in the enzyme-containing liquid below that concen-
tration at which the activity of the enzyme is degraded
(the inhibitory concentration) in order to make the
reaction smooth even when the substrate is fully
soluble in the enzyme-containing liquid. In order to
industrialize reactions of this type, a continuous
analyzer for continuously analyzing the substrate
concentration during the reaction must be developed in
order to maintain the concentration of the substrate
in the enzyrne~containing liquid below the inhibitory
concentration. Furthermore, to feed the substrate
continuously, a material feeding system directly
connected to the continuous analyzer is required.
Thus, the apparatus and the reaction operation inevit-
ably become complex. If in reactions of this type alarge excess of an enzyme is used and the speed of
consuming the substrate in the reaction is made larger
than the speed of mixing of the substLate with water,
the reaction can of cours~ be carried out at a substan-
tially lower substrate concentration than the inhibitoryconcentration. Since, however, the amount of the
enzyme used in the reaction is very large in practice,
7~D
-- 4 --
such a procedure is of no industrial significance.
Tryptophan obtained from indole by an enzyme
reaction or a fermentation reaction should desirably
be free from the unreacted indole when used in medicines
or feed additives, because the unreacted indole gives
off an inherent dispieasing odor~
Heretofore, steam distillation has been known
as a method for removing the unreacted indole from the
reaction solution in the production of L-tryptophan
from indole by an enzyme reaction (Japanese Patent
Publication No. 800~1982). But because the vapor
pressure of indole is lower than that of water, a
large quan~ity of steam is required for removing
indole by distillation. This adds to the cost of
energy, and such a method is not commercially feasi
ble. It has been strongly desired therefore to solve
this problem.
It is an object of this invention to pxovide
an improved method of preventing the activity o~ an
enzye from being degraded by a substrate in an enzyme
reaction.
According to this invention, ~his object is
achieved by a method of preventing the activity of an
enzyme from being degraded by a substrate in an en2yme
reaotion in an aqueous phase9 which comprises causing
an organic solvent immiscible with water but miscible
with the substra~e to be present in the reaction
3~ 3
system thereby to reduce the concentration of the
substrate in the aqueous phase below khat concentration
at which the activity of the enzyme i~ substantially
inhibited.
As embodiments o the afore~aid method, the
present invention provides
(A) a method which comprises performing an
en~yme reaction using indole as one substrate in an
aqueous phase in the presence of an organic solvent
immiscible with water but miscible with indole, re-
covering the organic solvent from the reaction mixture
and removing the enzyme, extracting the unreacted
indole from the resulting aqueous solution of L~trypto-
phan with the organic solvent recover2d, and re-using
the extract in the reaction; and
(B) a method which comprises performing an
enzyme reaction using indole as one substrate in an
aqueous solution in the presence of an organic solvent
immiscible with water but miscible with indole, and
separat.ing the unreacted indole as an organic solvent
layer from the reaction mixture.
The basic principle of this invention is that
the concentration of a substrate in an aqueous phase
is maintained below a certain fixed concentration in
: 25 accordance with the distribution ratio of the substrate
between an organic solvent phase having the substrate
dissolved thereln and the aqueous phase.
Generally, in a reaction in which the activity
of an enzyme is reduced by a substrate, the concentration
of the substrate in an aqueous pha~e must be maintained
always low, and for this purpose, a method is used in
5 which by some measure the substrate is added little by
little either continuously or intermittently to the
reaction system containing the enzyme. This method,
however, requires a complex operation and apparatus
for controlling the reaction such that when the sub-
strate is added to the reaction system, its concen-
tration in the enzyme-containing liquid is maintained
low~
In contrast, according to the method of this
invention, an organic solvent miscible with a substrate
but immiscible with water is added to an en~yme-contain-
ing liquid to dissolve the substrate almost completely
in the organic solvent phase. As a result, the concen-
tration of the substrate in the agueous phase can be
maintained substantially below the inhibitory concen-
tration in accordance with the distribution ratio ofthe substrate between the organic solvent phase and
the aqueous phase. At this time, it does not ~atter
whether the reaction product is dissolved or precipi~
tated in the organic solvent phase or the enzyme-con-
taining phase.
According to the method of this invention,the substrate in the enzyme-containing liquid phase is
continuously supplied from the organic solvent phase
in accordance with the distribution ratio of the
substrate between the two phases as the reaction
proceeds and consumes the substrate. The organic
solvent for dissolving the substrate may be added at a
time, or intermittently or contin~ously so long as the
concentration of the substrate in the aqueous phase
can be maintained lower than the inhibitory concen-
tration. If required, the speed of movement of the
substrate may be changed by changing the area of
contact between the organic phase and the aqueous
phase through such an operation as stirring.
In the method of this invention, enzymes may
be used singly or as a mixture of two or moreO But at
least one enzyme used should be of a type whose activity
is reduced by a substrate when it is used in an enzyme
reaction carried out in accordance with a conventional
method.
The enzymes used in this invention need not
always to be pure, and may be crude ones. For example,
the enzymes may be living cells collected from a
culture broth of an enzyme~producing microorganism by
centrifugation, etc~, cells obtained by reezing or
drying the living cells. treated products of these
2S cells obtained by such treatments as grinding, self-
digestion and ultrasonication, extracts of these
cells, and enzymes obtained rom the extracts.
'7~
The organic solvent used in the method of
this invention is miscible with substrates but immisci-
ble wi~h water. In practice~ it is necessary to
selectt accordiny to the enzymes u~ed, those organic
solvents which do not reduce the activities of the
enzymes under the reaction conditionsO
An organic solvent adaptable to a given
substrate and an enzyme is selected~ and its amount is
determined as shown below. Specifically, an enzyme
and a substrate to be reacted are used, and the in-
hibitory concentration of the substrate in an enzyme-
containing liquid under the reaction conditions is
measured~ Then, the distribution ratio of the sub-
strate between the enzyme-containing liquid and the
organic solvent is determined, and the concentration
of the substrate in the organic solvent which is lower
than the inhibitory concentration measured above is
prescribed. Once the concentration of the substrate
in the organic solvent to be used in the reaction is
determined as above~ the amount of the enzyme-contain-
ing liquid and the amount of the organic solvent used
can be determined from the amount of the substrate
used according to the concentration of the desired
reaction product accu~ulated in the reaction mixture.
Hence, so long as the concentration of the
substrate in the enzyme-containing liquid can be
maintained lower than the inhibitory concentration,
.
the organic solvent can be freely selected, and its
amount can be freely determined.
The method of this invention is described
below with regard to the production of L-tryptophan
using indole and L- or DL-serine as substrates~ Since
in the production of L-tryptophan from indole and
L-serine using tryptophan synthetase, indole reduces
the activity of tryptophan synthetase; the reaction
must be carried out while maintaining the concen
tration of indole in water low. For this purposeV
nonionic surface-active agents or adsorbent resins
have been used ( ehrendt, The First European Congress
in Biotechnology, 2/186 189, 1978). The reaction in
the presence of a nonionic surfactant, however, differs
lS in mechanism from the reaction in accordance with the
method of this invention which is carried out substan-
tially between two phases because indole is in the
form of a micelle in the aqueous phase by the effect
of the surfactant. Moreover, this method is not
commercially feasible because it is difficult to
separate the surfactant from the reaction produ~t
after the reaction. The method involving the use of
adsorbent resins is industrially disadvantageous
because with some adsorbent resins, indole adsorbed
thereby do not completely separate, or the reaction
product is likely to be adsorbed on these resins.
In contrast, when a solution of indole in an
organic solvent is added to an aqueous .solution contain-
ing serine and the enzyme and the reaction is carried
out in two phases, the concentration of indole dissolved
in the aqueous phase can be maintained low, and indole
is automatically supplied from the organic solvent
phase as it is consumed by the reaction. Hence, the
reaction proceeds smoothly~
In a liquid containing a mutant strain of
Escherichla coli, the inhibitory concentration of
indole is about 800 ppm. Examples of organic solvents
which distribute indole in a lower concentration than
the inhibitory concentration into the aqueous phase
are toluene, chlorobenzene, ethyl citrate, methyl
isobutyl ketone and anisole. For example~ if a 20% by
weight toluene solution of indole is used, indole
dissolves in a concentration of less than 720 ppm in a
liquid containing a mutant strain of Escherichia coli9
and the reaction can be earried out while maintaining
the concentration of indole lower than the aforesaid
inhibitory concentration. The concentration of indole
in an enzyme-containing liquid under the reaction
conditions can be maintained lower than ~00 ppm if the
concentration of indole in other svlvents is 40% by
weight for ethyl citrate, 50% by weight for methyl
isobutyl ketone, 30% by weight for anisole and 20~ by
weight for monochlorobenzene.
According to one typical embodiment of the
'7~
method of this invention, L-tryptophan can be produced
by cultivating a mutant strain of Escher_ ia coli by
subjecting a culture medium containiny a tryptophanase-
deficient mutant strain of L-tryptophan-requiring
S Escherichia _oli, a carbon source a nitrogen source
and inorganic salts to aerobic conditîons at a tem-
perature of 28 to 40QC and a pH of 6 to 8, suspending
the microbial cells either as in the culture broth or
after separating from the culture medium in a solution
containing pyridoxal phosphate, L-serine and inorganic
matter, then adding a solution of indole in an organic
solvent miscible with indole but immiscible with water
at a time or continuously at a pH of 7.S to 9~5,prefer-
ably 8 to 9, and carryi.ng out the reaction at a tempera-
ture of 20 to 40C.
Examples of the organic solvent which can beused at this time include aromatic hydrocarbons and
their derivatives such as benzene, toluene, chloro-
benzene, nitrobenzene and acetophenone; aliphatic
esters having at least 6 carbon atoms such as n-butyl
acetate, isoamyl acetate, ethyl butyrate and isobutyl
acetate; aliphatic ketones having at least 6 carbon
atoms ~uch as methyl isobutyl ketone, diisobutyl
ketone~ diisopropyl ketone, methyl n-amyl ketone and
di-n-propyl ketone; citric acid esters such as acetyl~
triethy~ citrate~ ace~yltributyl citrate~ triethyl
citrate and tributyl citrate; tartaric acid esters
- 12 -
malic acid esters; and ethers such as anisole.
The amount o the organic solvent used varies
depending upon its kind because it is determined by
the diskribution ratio of indole between an enzyme-
containing liquid phase and an organic phase under thereaction conditions and the amount of indole used.
Usually, it is determined such that the concentration
of indole in the enzyme-containing liquid is lower
than 800 ppm, industrially preferably lower than 750
ppm.
L~tryptophan which is the reaction product
precipitates as crystals in the enzyme-containing
liquid. Under the reaction conditions employedt these
crystals do not at all affect the proceeding of the
reaction.
In the above method, DL-serine may be used
as the substrate. In this case, by using cultivated
cells of Pseudomonas putida tMT-10182) or Pseudomonas
punctata (MT-10243) as a serine racemase, L-typtophan
can be obtained in high yields without degradation of
the activity of the enzyme by the organic solv~nt.
It is of much commercial significance that
L-tryptophan can be produced without any problem from
chemically synthesized DL-serine and indole in the
presence of two types of enzymes using easily avail-
able organic solvents.
The reaction mixture obtained by the above
~Z~
- 13 -
method contains the resulting L~tryptophan, the unre-
acted raw material, the organic solvent, the enzyme,
etc. Preferably, this reaction mixture is treated by
the metho~ according to embodiment (A) or (B) described
hereinabove.
According to embodiment tA), the following
method of recovery is employed. First, the organic
solvent is recovered from the reaction mixture and the
enzyme is removed. If the reaction mixture is vigorously
1~ stirred while the enzyme and the organic solvent are
present together therein, the interface between the
aqueous phase and the organic solvent phase becomes
obscure, and it sometimes becomes difficult to separate
the organic solvent having the unreacted indole extracted
and dissolved therein. To avoid this phenomenon, it
is the usual practice to recover the organic solvent
from the reaction mixture in the first place. This
- can be achieved by u~ual distillation/ for exampleO
Then, the enzyme contained in the reaction mixture
~om which the organic solvent has been recovered i~
removed.
There is no particular limitation on the
method o~ removing the enzyme used in the reactlon
from the reaction mixture. One industrially effective
method comprises adding a mineral acid to the reaction
mixture from which the organic solven~ has been removed,
thereby adjusting the pH of the reaction mixture to 2
'7~
- 14 -
to 5, heating it as required to promote flocculation
of the enzyme, and removing the flocculated en~yme by
such means as filtratio~.
As required, the reaction mixture from which
the enzyme has been removed may be concentratedO To
extract and separate the unrea~ted indole efficiently
from the reaction mixture, it is preferred to maintain
L-tryptophan in the dissolved state without precipitat-
ing it as crystals. When L-tryptophan is crystallized
from agueous solution, it involves the unreacted
indole having a similar molecular structure and causes
its simultaneous crystallization. For this reasonr
according to the treating temperature, the concen~
tration of the reaction mixture to be extracted by the
organic solvent should be adjusted to one at which
L-tryptophan is in the dissolved state,
The unreacted indole is extracted by an
organic solvent from the reaction mixture from which
the enzyme has been removed. The organic solvent may
be properly selected from those which are used in the
reaction. Usually, ~he same organic solvent as used
in the reaction is employed. Hence, the recovered
organic solvent may be used for this purpose. How-
ever, the solvent need not to be the same as the one
used in the reaction, and solvents effective for the
tryptophan-producing reaction and the extraction o
the unreacted indole respectively may be selected~
t;~
- 15
The above recovering method is especially
useful in the production of L-tryptophan by an enzyme
reaction using indole as one substrate because the
recovered indole can be re-used. Generally, in the
production of L-tryptophan by an enzyme reaction, the
enzyme, etc. used in the reaction is removed rom the
reaction mixture by a general method, and then L-tryp~o-
phan is isolated~ The use of the recovered solution
containing the unreacted material as such freqtlently
inhibits the activity of the enzyme~ and this gives
rise to a serious problem in industrial practice.
Howevez, when the unreacted indole is extracted from
the reaction mixture with the organic solvent in
accordance with the aforesaid recovering method! ~he
content of indole in L-tryptophan crystals can be
reduced, and the indole as extracted with the organic
solvent in the form of the solvent solution can be
reused in the next reactios~ without isolating the
extracted indole from the solution (when it is reused,
the reaction proceeds without inhibiting the activity
o the enzyme).
The amount of the organic solvent used in
extracting the unreacted indole varies dependiny upon
: the distribution ratio of indole between the organic
solvent phase and the aqueous phase and also upon the
reaction conversion of indole. However, since enzymes
having fixed specific activities are commercially
- lS -
available and the reaction conversion is con~tant, the
a~ount of the organic solvent used in extraction can
be easily determined.
There i5 no particular restriction on the
operation of extracting the unreacted indole~ and it
may be carried out by a conventional method. Generally,
a suitable amount of the organic solvent is added to
the reaction mixture from which the enzyme has been
removed, and the mixture is stirred to contact the
aqueous phase fully with the organic phase. A~ stated
hereinabove, it is preferred to maintain the reac ion
mixture in such a state that L-tryptophan crystals do
not peecipitate. Industrially, it is effective to
carry out the extraction in a heated condition in
lS order to increase the solubility of L-tryptophan in
water. After uniformly contacting the two phases as
above~ the mixture is left to stand to separate it
into an aqueous phase and an organic solvent phase~
The extracting operation ~ay be carried out batchwi^~e,
~0 but industrially~ it is carried out continuously by a
countercurrent extracting method.
The organic solvent containing the extracted
- unreacted indole is re-used in the next reaction after,
as required, adding a fresh supply of the solvent~ or
mixing it with another solvent, or suppplying fresh
indole.
This method is of great industrial significance
- 17 -
since the unreacted indole can be efficiently recovered
even when the conversion of indole is varied by fluctu-
ations of the specific activity of the enzyme which
often occur in en~yme reactions. Moreover, even when
the indole as extracted and recovered is reused as
such, the reaction can be carried out without inhibit-
ing the activity of the en~yme.
If the aforesaid recovering method is utili2ed,
the unreacted indole can be effectively extracted and
recovered also from an L-tryptophan-containing reaction
mixture obtained by a conventional method.
According to the embodiment (B), the reaction
mixture containing the resulting L-tryptophan~ the
unreacted material, the organic solvent, the enzyme is
first separated into an organic solvent phase and an
aqueous phase. The organic solvent phase containing
the unreacted material is recovered. Then, the enzyme
is removed from the aqueous phase, i~e. the reaction
mixture containing L tryptophan to obtain L-tryptophan.
~hen the enzyme and the organic solvent are present
together in the reaction mixture~ the interface between
the aqueous phase and the organic solvent phase some-
times becomes obscure, and their separation may become
difficul~. In such a case, the sepaeation ma~ be
carried out by a known method9 for example by subject~
ing the reaction mixture to a centrifugal separator.
If the extraction and separation are repeated using
~'7~
- 18
the organic ~olvent used, the effect of removing
indole increases.
In the separating procedure, L-tryptophan in
the aqueous solution may precipitate as crystals, but
the effect of extracting the unreacted indole is
better if it is in the dissolved state.
Desirably, the temperature at the time o
separation is below the boiling point of the organic
solvent, and if water and the organic solvent fvrm an
as20trope, below its boiling point. The separated and
recovered indole solution may be directly used in the
next reaction without any problem.
There is no particular restriction on the
method of removing the enzyme from the reaction mixture
after the organic solvent phase containing the unreacted
indole has been sepaeatedO One industrially effective
removing method comprises adding a mineral acid to the
reaction mixture left after removal of the organic
solvent phase to adjust the pH of the mixture to 2 to
5, heating it as required to promote flocculation of
the enzyme, and removing the flocculated mass by such
means as filtration. L-tryptophan can be obtained by
concentrating the reaction mixture from which the
enzyme has thus been removed.
By the above method, the unreacted indole in
the reaction mixture moves effectively to the organic
solvent. The or~anic solvent containiny the unreacted
7~3
- 19 -
indole can be used in the next reaction after~ as
required, adding a fresh supply of the solvent or a
fresh supply of indole.
This recovering method is al~o particularly
useful in the production of L-tryptophan by an enzyme
reaction using indole as one substrate for the same
reason as in the embodiment (A).
The method of this invention is of great
industrial significance becaus~ L-tryptophan of high
purity free from indole can be obtained, and the
startirlg indole can be recovered and reu~ed.
The following Examples illustrate the method
of this invention more specifically~
ExamPle 1
A 300 ml flask equipped with a stirrer was
charged with 9.27 g of L-serine, 3 g of ammonium
sulfate, 10 m~ of pyridoxal phosphate and 82.5 g of
waterr and they were well stirred. The pH of the
mixture was adjusted to 8.5 with coneentrated aqueous
ammonia, and the tempera~ure was raised to 35C~
Three grams of a wet cream cake o~ cultivated cells
of Escherichia coli (MT-10242) containing tryptophan
synthetase was added. Then, 68~9 g of an acetyltri-
butyl citrate solution of 6.89 g of indole was addedO
25 The reaction was carried out for 48 hours, and the
reaction mixture was diluted to a total volume of 300
ml with a 5% aqueous solution o~ sodium hydroxide to
- 20 -
dissolve the resulting L-tryptophan completely. The
solu~ion was separatad by a separating funnel into an
aqueous phase and an organic solvent phase. A part of
the aqueous phase was centrifuged by a centrifugal
~eparator o sediment the mlcrobial cells. The clear
supernatant liquid was collected, and the concentza ion
o$ L-tryptophan was analyzed by iiquid chromatography,
and its amount yielded was determined~ The yield of
the product based on indole was 96.~ mole%.
The above procedure was repeated except that
the amount of the organic solvent used was varied and
the distribution ratio in the aqueous phase was variedO
The results are shown in Table 1.
When the above reaction was carried out
without adding the organic solvent, the concentration
of indole in the microorganism-containing liquid was
3,000 ppm, and the yield of L-tryptophan was 47 mole~.
It is seen from this experimental fact that for the
production of L-tryptophan, it is important to add an
organic solvent immiscib.le with water and maintain the
concentration of indole in the microorganism~containing
liquid low. Specifically, when the reaction was carried
out in water without adding an oryanic solvent~ the con-
centration of indole in the aquoues phase became 3 ~ 000
ppm, and ~he ~ield of ~-trypto~han was 47 mole% based
on indole. ~n contrast, when butyl citrate was added
as the organic solvent and the concentration of indole
7~
in the aqueous phase was maintained at 790 ppm, the
yield of L-tryptophan increased markedly to 95 mole%
based on indole. When the concentration of indole in
'che aqueous pha e was maintained at 1070 ppm (higher
than 800 ppm), the yield of L-tryptophan decreased by
about 10% f rom that obtained when the indole concen-
tration was 800 ppm.
- 22 -
Table 1
Amount of Confentration Concentration Yield of
the organic of indole in of indole in L-tryptophan
solvent(*l) the organic the enzyme- ~mole~ based
used ~g~ solvent (*l) containing on indole~
solutionliquid (ppm)
_ (wt.%~ 2)
68.9 10 1~0 9~.5
34.~S 20 400 96.3
27.56 25 ~2U 9~,8
22.97 30 660 ~6.2
19.69 35 7~0 ~5.0
15.31 45 1070 ~.3
- (*3) - 3000 ~7
(*1): Acetyltributyl citrate
(*2): In the early stage of the reaction.
(*3) No organic solvent was added.
Exa~ple 2
6.8 g (solids content 1.7 g) of a wet cream
cake of E~cherichia coli ~IT-~0242) containing trypto~
_
phan synthetase and 3.4 g (solids colntent 0.85 g~ of
Pseudomonas putida (MT-10182) containing serine racemase
were suspended in water, and the total volume o the
suspension was adjusted to 20 ml.
An aqueous solution consisting of 11.3 g of
Drl-serine~ 6.0 g of ammonium sulfate, 10 mg of pyridoxal
phosphate and 66 9 of water was fed into a 300 ml flask,
and the pH of the aqueous solution was adjusted to 8.5
- 23 ~
with concentrat~d aqueous ammonia an~ th~ microbial
cell ~usp~nsion pr~par*d previously was fed into
flask.
The inside o the fla~k was kept at 35C,
and 57.2 g of a ~oluene solution o 11 9 5 9 of indole
was added, and the reaction was carried out at this
temperatllre or 48 hours. At th~ start of the reaction,
the concentration of indole in the aqueous phase was
720 ppm.
The reaction product was analyzed by the same
procedure as .in Example 1. It was found that L-trypto-
phan was obtained in a yield of 8g~3 mole% based on
indole.
In order to determine the effect of the concen~
tration of indole in the aqueous phase, the above
procedure was repeated except that the amount of
toluene used was variedO The yields of L-tryptophan
in these runs are shown in Table 2.
It is seen from the results ~hat the inhibi
tory concentration of indole in the aqueous phase is
about 800 ppm. As shown, when the reaction was started
at an indole concentration of not more than 720 ppm,
the yield of L-tryptophan was about 90 mole%~ but it
decreased to 82mole% when the reaction was started at
: 25 an indole concentration of 920 ppm.
'7~
24 -
Table 2
Amount of Concentration Concentration Yield of
the toluene of indole in of indole in L-tryptophan
solution the toluene the aqueous (mole~ based
(g) solution phase in the on indole)
(wt D ~ ) early stage
of the re-
action (p~m)
115 10 45~ 91.2
S7.5 20 720 89.3
3803 3~ 920 82
Example 3
By the same procedure as in Example 2, DL
serine and indole were reacted in the presence of
cultivated cells of Escherichia coli (MT-10232) and
cultivaterd cells of Psuedomonas punctata (MT-10243).
At this time, methyl isobutyl ketone was used
as the organic solvent, and the reaction was carried
out so that the concentration oP indole in the aqueous
phase at the start of the reaction was adjusted to 760
ppm.
After performing the reaction at 35C Por
48 hours, the yield oP L-tryptophan was 98.6 mole~
15 based on indoleO
To determine the ePfect of the indole concen-
15 tration in the aqueous phase~ the amount of methylisobutyl ketone was varied, and changes in the yield
: oP ~-tryptophan were examined. The results are shown
in Table 3.
7~
- ~5 -
It is seen that by maintaining the concen-
tration of indole in the aqueous pha~e lower than 800
ppm by u~ing methyl i~obutyl ketone, L-tryptophan can
be obtained in a yield of mor~ than 99 mole%.
Table 3
Amount of Concentrati~n Concentratio~ Yield of L
the ketone of indole in of in~ole in tryptophan
solution the ketone the aqueou~ ~mole~ based
~g) solution phase in the on indole)
(wt.~ the early
stage of the
_ _ _ _ _ reac ion ~m)
115 10 8~ ~9.8
57.5 20 190 g~9
38.3 30 3~0 99.8
28.75 4~ S10 99.6
23.0 50 720 9~O~
Example 4
By the same procedure as in Example 2, ~-
serine and indole were reacted in the presence of
~scherichia coli lMT-10242) and Pseudomonas ~
(MT-10182). By using anisole as a solvent, the con-
centration of indole in the aqueous phase was main-
tained lower than 3~0 ppm.
Aft~r the reaction at 35C for 48 hours, the
yield sf L-tryptophan based on indole was 9S mole%.
Exa ple 5
Twenty grams of DL-serine, 1.0 g of ammonia
asetate, 0~2 9 of sodium bisulfate, 0,1 9 of DETA and
26 -
a culture broth of Erwinia herbioola (ATCC 21434) were
fed into a 200 ml reaction vessel.
The culture broth had been obtained by cultivat-
ing Erwinia herbicola under aeration at 28C for 28 hours
_ _
in a culture medium consisting of 0.2% o~ L-tyrosine~
0~2~ Of ~2HP04~ 001% of MgSO4O7H2O, 2 ppm of FeSO4.7~2,
0.01~ of pyridoxine hydrochloride, 0.S~ of glycerol,
0~5~ of succinic acid, 0.1% of DL-methionine, 012~ of
DL-alanikne, 0.05% of glycine, 0.1~ of L-phenylalanine
and 12 ml o~ a soybean protein hydroly~ate.
A toluene solution containing 0.7 9 of pyro-
catechol was added, and the concentration of pyrocal'cechol
in water was adjusted to below 3,000 ppm. Under these
conditions, the reaction was carried out at a temperature
of 37C and a p~ of 8 for 48 hours~ The amount of
L-DOPA accumulated was 30 g/liter.
Example_6
Microbial cells containing ERcherichia coli
were cultivated at a temperature of 30C and a pH o
20 7 in the presence of monopotassium pho~phate, dipotassium
phosphate, ammonia sulfate, calcium chloride, iron
sul~ate, yeast ex~ract, and polypeptone while blowing
air into the culture medium and adding glucose and
indole. Furthermore, microbial cells containing
Pseudomonas putida were cultivated in a simila culture
.
medium at a tempera~ure of 30~C and a pH of 7 while
blowing air into it and adding glucose. In 40 hours,
7~
- 27 -
these microbial cells were formed in a concentration
of 30 to 35 g/liter. By a superhigh speed centrifugal
separator, they were obtained as cream cakes having a
water content of 85 to 85%.
DL-serine (77~3 g), 10.5 9 of ammonium sulfate
and 486 g of water were put into a flaslc, and the p~
of the mixture was adju~ted to 8.5 with 29% aqueous
ammonia. Furthermore, 51.2 g of the cream cake of
scherichia coli and 23.2 q of the cream cake of
Pseudomonas e~ida were added, and the entire miture
was stirred well. Furthermore, 392 9 of a toluene
solution of 78.4 g of indole was added, and the re-
action was earried out at 35C for 40 hoursO The
content of L-tryptophan in the reaction mass was found
to be 129,~ g by liquid chromatography~, Its yield sYas
95.0~ based on indole~
Toluene wa~ removed by distillation, and ~h*
reaction mixture was diluted with water to adjust the
concentration o L-tryptophan to 4~2~ by weight. The
pH of the mixture was ad~usted to pH 4.0 with 98~
: sulfuric acid, and 27 g of activated carbon was added.
The temperature was raised to 95C, and the mixture
wa~ maintained at 95 to 98C for 1 hour and hot-
filtered at this temperature to remove the cell~ to-
gether with activated carbon. At 80C, the recovered
toluene was added to the filtrate, and they were well
stirred. Stirring was then stopped~ and the mixture
~ 2~ -
was left to standO It separated into an upper toluene
layer and a lower aqueous layer. The toluene layer
was removed, and the same amount of toluene as above
was again added to the aqueous layer, and extraction
was repeated.
The concentration of indole in the aqueous
layer was 55 ppm after the first extraction and 4 ppm
after the second extraction.
The aqueous layer was concentrated to an
L-~ryptophan concentration of 10% by weight, and
cooled to 20C The precipit~ated L-tryptophan
crystals were collected by filtration, washed with
water~ and dried. The amount of L-tryptophan i~olated
was 100 9. It had a purity of 99.5~ and an indole
content of 0 ppm. When the recovered toluene solution
of indole was used in the next reaction after supply-
ing additional toluene, there was no effect on the
yield of L-tryptophan, and the reaction proceeded in
good oonditionO
~he result~ are shown in Table 4.
7~3
- 29 ~
Tabl~ 4
Number of recyclings Yield of L-
of the extracted tryptophan
toluene layer [msle~ based
on indole~
,
O ~5.0
1 98.3
2 9806
3 9~.0
4 95.4
98.7
316 9 of each of the reaction mixtures con-
taining 500 ppm and 2000 ppm of unreacted indol~ was
extracted with 27.6 9 of toluelle at 80C, and the
relation between the number of extractions and the
concentration of indole remaining in the aqueous phase
was examined. The results are shown in Table 5~
'7f~
- 30 -
Table 5
Concentration of Number of Concentration
indole in the exteactions of indole
aqueous phase in water afte~
in the early extraction
stage of the (ppm)
reaction ~Ppm)
200~ l ~99
299 2 54
5~ 3 918
500 l 45
2 l~
18 3 3
Example_7
L-tryptophan was produced by reacting L-serine
and indole in water and diisobutyl ketone at a te1npera~
ture of 35C and a pH of 8.5 in the presence of
Escherichia coli cells cultivated by the same procedure
-
as in Example 6. The yield of L-tryptophan was 9~O3%
based on indole.
The reaction mixture was distilled to remove
diisobutyl ketone, and then treated on a centrifu~al
separator to obtain a wet cream cake containing L-
tryptophan and the microbial cells~
The cream cake was ~ischarged into water and
diluted with water to an I. tryptophan concentration of
0.8% by weight. The L tryptophan crystals were dis-
solved in water, and the solution was passed through
an ul~rafiltration membrane at room temperature to
- 31
remove the microbial cells. At this time, the concen-
tration of indole in the aqueou-~ phas~ was 90 ppm.
Diisobutyl ketone in an amcunt one-fith of the amount
of the aqueous phase was added, and the mixture was
S s~irred. The mixture was left to stand to separate it
into two layers~ The diisobutyl ketone layer was
removed. Then, the same extracting operation wa5
repeated using the same amount of diisobutyl ketone a~
in the previous operation.
The concentration of indole in the aqueous
layer was 32 ppm after the first extraction, and 2 ppm
after the second extraction.
The diisobutyl ket.one layer containing indole
was used in the next reaction after supplying a required
amount of indole. No problem arose in the next reaction.
Example 8
L-tryptophan was produced by reacting DL~serine
and indole at a pH of 8.5 and a temperature of 35C
for 48 hours in water and benzene in the presence of
20 Escherichia coli and Pseudomonas ~unctata cultivated
and collected by the same procedure as in Example 6.
The yield of 1~tryptophan was 93.7 % based on indole.
The reaction mixture was then distilled to
remove benzene, and then diluted with water to an
L-tryptophan concentration o 4.2% by weightO The
diluted mixture was adjusted to p~ 3.5 with hydro~
chloric acid~ and 15~ by weiqht, based on the resultang
0'7~
- 32 -
tryptophan, of activated carbon, was added. The
mixture was heated at 95 to 98C for 1 hour. It was
press-filtered at the same temperature for 1 hour. To
the filtrate was added benzene at 80C The mixture
was well stirred and then left to ~tand to separate it
into layers. The benzene used was the one recovered by
distillation as above. The same extracting operation
was repeated three times using the same amount of
fresh benzene. The concentration of indole in the
aqueous phase was 1850 ppm before the extraction,
265 ppm after the first extraction, 43 ppm after the
second extraction, and 4.6 ppm after the third extrac
tion. When the reaction mixture was concentrated and
crystalli~ed, L-tryptophan having a purity of 99.7%
was obtained in a yield of 69~8% based on indole~ Its
indole content was 2.1 ppm.
The recovered benzene solution of indale was
used in the next reaction after supplying a required
amount of fresh indole. No problem arose in the nex~
reaction.
Example 9
An aqueous solution composed of 11.3 g of
DL-serine, 6.0 g of ammonium sulfate, 10 mg of pyridoxal
phosphate and 66 g of distilled ~ater was well stirred
at 35~C in a 300 ml flask, and 29% aqueous ammonia
was added to adjust ~he pH of the solution to ~.5.
Then, 4.0 g of a cell cream cake containing
Escherichia coli cultivated by the same opeeation as
in Example 6 and 2.5 g of a cell cream cake containing
Pseudomonas putida cultivated by the same opeeation as
in Exam21e 6 were added, and well stirred at 353C to
disperse them.
Furthermore, a solution of ll.S g of indole
in 26.8 9 of diisobutyl ketone was added, and the
mixture was stirr~d at 35C and 90 rpm for 40 hours7
Af~er the reaction, the stirring was stopped.
The reaction mixture was left to stand to separate it
into two layers. The upper diisobutyl ketone layer
was removed. Diisobutyl ketone was added in the same
amount as above, and the mixture was stirred at ~0 rpm
for 30 minutes. The mixture was then left to ~tandt
and the diisobutyl ketone layer was recoverd in the
same way as above.
It was found that the unreacted indole remain~
ing in the reaction mixture at the end of the reaction
was recovered to an extent of 85~ by the first extrac~
tion with diisobutyl ketone, and to an extent of 99%
by the second extraction. ThP quantity of indole was
analyzed by gas chromatography.
When the recovered diisobutyl ketone was used
in ~he next reaction after supplying a required amount
~5 of indole, no problem arose.
Example 10
The same reaction as in Example 9 was carried
- 34 -
out at a rotating speed of 640 rpm using 45.7 9 of
toluene. After the reaction, the reaction mixture was
subjected to a centrifugal separator to make he
interface between two layers distinct because lt was a
uniform emulsion.
The upper toluene layer was removed. Then,
toluene wa~ added in the same amount, and the extra-
ction of the unreacted indole was repeated.
It was found that the unreacted indol@ contained
in the reaction mixture at the end of the reaction was
recovered to an extent of 87% by the first extraction,
and to an extent of 99.9% by the se~ond extract on.
The recovered toluene containing indole was
used in the next reaction after supplying a required
amount of indole. No problem arose in the next re-
action.
~E~
The same reaction as in Example 9 was carried
out using 1403 g of ethyl citrateO After the reaction,
the reaction mixture in the form of a uniorm emulsion
was subjected to a continuous centrifugal separator to
separate it into an aqueous layer and an ethyl citrate
layerO To the aqueous layer 7 the same amount of ethyl
citrate as used in the reaction was added, and the
mixture was again separated into an aqueous layer and
an organic layer by a continuous centrifugal separator.
A part of the aqueous layer wa6 sample~; and
3'7~9
, ~
- 35 -
the resulting tryptophan was analyzed by high perfGr-
mance liquid chromatography. Its yield was found to
be 99.7% based on indole.
~he reaction mixture was diluted with water
to an L-tryptophan concentration o 4.2~ by weight and
adjusted to pH 3.5 with sulfuric acid. Powdery activated
carbon was added in a proportion of ~5% by weight
based on L-tryptophan~ and the mixture was heated at
90C or 1 hour. After the L-tryptophan crystals
completely dissolved~ the solution was suction-iltered
at the same temper~ture to remove the microbial cells
together with activated carbon.
The hot-filtrate was concenteated to an
L-tryptophan concentration of 10% by weight, and
cooled to 10C. The pH of the solution was then
adjusted to 5.9, and the precipitated scale-like
crystals were separated by filtration, wa~hed with
water, and driedO
The yield of L-tryptophan isolated was ~4~5
mole% based on indole. It had a purity of 9g~4~ and
an indole content of 1,2 ppm.
The recovered ethyl citrate containing indole
was used in the next reaction aftes supplying a required
a~ount of indole. No e~ect was exerted on the yield
of L-tryptophan in ~he next reaction~ and the reaction
proceeded in good condition.
Example 12
- 3
Cells containing Escherich~a coli and cell~
containing P~eudomonas ~utida were each cultivated in
the same way as in Example 9 and subjected to a super-
centrifugal separator to obtain cell cream cakes of
5 the two microorganisms each having a water content of
75~.
An aqueous solution composed o 77.3 g of
DL-serine, 10.5 g of ammonium sulfate and 486 g o
water wa~ adjusted to pH 8.5 with 29% a~ueous ammonia,
and 51.2 g of the cell cream cake of _cherichia coli
and 23.2 9 of the cell cream cake of Pseudomonas
~utida obtained as above were added. They were dis-
persed by thorough stirring, and 392 g of a toluene
solution of 78.4 9 of indole was added. The reaction
was carried out at 35C for ~8 hoursO
After the reaction, a part of the reaction
mixture was taken, and separated into a toluene layer
and an aqueous layer by a centrifugal separatorO
L-tryptophan was dissolved by adding an alkali to the
2û aqueous layer in which crystals of L-tryptophan were
precipitatedO Then, the aqueous layer was passed
through a membrane filter to remove the microbial
cells. The filtrate was subjected to high-performance
liquid chromatography to analyze L-tryptophanO The
yie~d of L-tryptophan in the reaction mixture was
found to be 99.3 mole% based on indole.
~ nother poetion of the reac~ion mixture was
- 37 -
taken, and separated into an agueou~ layer and a
toluene layer by a centrifugal separator. ~he concen-
tration of the unreacted indole contained in toluene,
as measured by gas chromatography, was ~.3 ppm.
s When the toluene solution containing indole
was used in the next reaction after adding a required
amount of fresh indole, no effect wa~ exerted on the
yield of L-tryptophan, and the reaction proceeded in
good condition.
The results are shawn in Table 6.
Table 6
Number of recyclings Yield of L-
of the recoveredtryptophan
toluene layer ~mole% based
on indole~
0 99.3
1 38.5
2 9809
3 ~9~2
98.2
99.4
The aqueous layer recovered by ~entrifugal
~eparation was diluted with water to an L-tryptophan
concentration of 4.2% by weight and adjusted to p~ 4~a
with 38% sulfuric acid. Powdery activated carbon ~27
9~ was added, and the temperature of ~he mixture was
raised to 98C. The mixture was ~hen maintained at
98 to 100C for 1 hour to dissolve the precipitated
- 38
crys~als of L-~ryp~ophan.
The microbal c211s were removed together wlth
activated c~rbon by hot filtration~ The filtrate wa~
concentrated to an L-tryptophan conc~ntration of 15%
S by weight to precipita~e ss~le-like cryetal~ of L~
tryptophan. The crystals w~re separated by filtration
at 10C, washed with water and dried to give pale
y~llow scale like crystals of L~teyptophan having a
purity of 99.2% in a yield of 81.3 mole% based on
indole~ The resulting L-tryptophan crystals had a
specific rotation of -31.8, a heavy metal content
o~ less than 20 pm, an ignition residue of 0.02~ by
weight and an ammonium content of 0~01~ by weigh~.
