Note: Descriptions are shown in the official language in which they were submitted.
~33L~3~
TITLE OF INVENTION
PROCESS E'OR THE PRODUCTION OF
N-FORMYL-ASPARTYL-PHENYLALANINE OR ITS METHYL ESTER
DETAILED DESCRIPTION OF THE INVENTION.
The present inven~ion relates to a process for
the production oE N-formyl-L~-aspartyl-L-phenyl-
alanine or its methyl ester, which is useful as an
intermediate for the preparation of a dipeptide
sweetener, N-o~-aspartyl-L-phenylalanine methyl ester.
More particularly, it relates to a process for the
production of said lnterme-1iate, which comprises
reacting aspartic acid with about stoichiometric
quantities of formic acid and acetic anhydride in the
presence of or in the absence of a catalyst, such as
metal oxides, metal hydroxides or their salts, to
form N-formylaspartic anhydride, and then directly
adding to the resulting reaction mixture phenyl
alanine or its methyl ester, so as to allow the two
compounds to condensate.
L~ aspartyl-l-phenylalanine methyl ester is
known to be a low caloric sweetener with a strong
taste closely similar to that of sucrose.
'
, 1 ~
' /~.X , ,
~31~3~
In the hitherto known processes, the dipeptide
ester has been prepared, for example, by reacting a
N-protected-L-aspartic anhydride with a L-phenyl-
alanine methyl ester in a solvent, followed by the
elimination of the protective group (see UOS. Patent
No. 3,7~6,039); or by reacting a N-protected-L-
aspartic anhydride with L-phenylalanine and then
eliminating the protective group to give L-~-
aspartyl-L-phenylalanine, followed by the esterifica-
tion (see Japanese Patent Publication No. 26,133/80
and U.S. Patent No~ 3,933,781). It is preferable,
from economic point of view, to use a formyl group
for the protection of the amino group in any of the
above processes.
Usually, in cases where N-formyl-L-aspartyl
anhydride is allowed to condense with L-phenylalanine
or its methyl ester to obtain Lt~-aspartyl-L-phenyl-
alanine methyl ester, there must be used ~~formyl-L-
aspartic anhydride purified from the dehydration
- reaction medium, so these cases have practical
problems. N-formyl-L-aspartic anhydride is usually
produced by adding L-aspartic acid to a large excess
of formic acid and acetic anhydride. In this case, a
large quantity of formic acid remains in the reaction
mixture even after the completion of the reaction.
' - 2 -
' ' . ' ' .
~: ~13~3~
The formic acid not only impedes the condensation of
the N-formyl-L-aspartic anhydride but also decreases
the ratio of the desired N-formyl-L--~-aspartyl-L-
phenylalanine (~-isomer) or its methyl ester to
N-formyl-L-~-aspartyl-L-phenylalanine (~-isomer) or
its methyl ester. Because of this, the N-formyl-L-
aspartic anhydride must be separated from the
reaction medium containing the residual formic acid,
for example, by adding aromatic hydrocarbons and/or
halogenated hydrocarbons to the reaction
(dehydration) mixture to precipitate the crystals of
the anhydride (see Japanese Patent Application
Laid-open No. 91,210/76) or by allowing the reaction
mixture to evaporate to dryness (see U.S. Patent No.
3,933,781). It would be obvious to those skilled in
the art that the same effect can be attained by
adding continuously or at a time a large quanti-ty of
acetic acid or aromatic hydrocarbons to the reactlon
mixture and then evaporating off the residual formic
: .
:,~ acid.
:~ E'rom the industrial viewpoint, these processes
are disadvantageous in that a large quantity of
energy is required for the cooling, separation or
evaporation and that the process becomes complicated
~- with a plant investment.
:` - 3 -
.
~31~3~8
In addition to the complication of the main
processes, there is also the disadvantage that a
fractionating plant must be constructed and operated
for the recovery and reuse of each component from the
mixture with formic acid and acetic acid or the
mixture with formic acid, acetic acid and aromatic
hydrocarhons, separated out of the reaction medium.
There is also proposed a process for producing
N-formyl-L-aspartic anhydride in which formic acid
and acetic acid are used in quantities stoichiometric
to L~aspartic acid (see Japanese Patent Application
Laid-Open No. 46,279/8~). However, this process too
has a problem when applied to the synthesis of
N-formyl-L-~-aspartyl-L-phenylalanine or its methyl
ester (N-formyl-~-dipeptide) involving condensation
with L-phenylalanine or its methyl ester. To be more
specific, a long period of time up to tens of hours
or more is required for the dehydration, and the
yield of N-formyl-L-aspartic anhydride based on the
starting rnaterial, i.e., L-aspartic acid, is lower
than in the process utilizing an excess of formic
acid. Consequently, the process results in an
- increase in the amount of impurities, including
unreacted starting materials t and requires the
separation of the crystals of the anhydride. It is
'
-- ~I --
' `' ' ' `
131~3~
therefore apparent that this process do not overcome
the disadvantage in the process utilizing an excess
of formic acid. In addition, this process is also
disadvantageous in that the loss of the anhydride
during the process of crystallization causes an
apparent decrease in the yield of the N-formyl-~-
dipeptide in the condensation step, calculated on the
basis of the starting L-aspartic acid.
As other known processes for producing
N-formyl-L-aspartic anhydride, there is one in which
fine powders of L-aspartic acid are employed, and one
wherein ultrasonic waves are irradiated during
dehydration (see Japanese Patent Application
Laid-Open No. 137,875/86). It is known that in these
processes the dehydration can be completed within a
short period of time even when formic acid and acetic
anhydride are used in small quantities. However,
these processes merely make it possible to shorten
the reaction time by means of the addition of such
equipment as a pulverizer. In other words, the prior
processes for producing N-formyl-L-aspartic anhydride
merely intended to produce the anhydride in high
purity and in high yield, and no considerations have
been paid to its use in the following step to produce
~:,
~'
` :'
- 5 -
~; ~
~31~5~
the N-formyl~-dipeptide~ Accordingly, the hither-to
known processes could not be satisfactory for the
purpose of the present invention.
In view of the above, the inventors have
conducted intensive investigations and, as a result,
have found that N-formyl-L~X-phenylalanine or its
methyl ester can be produced in a high yield by
directly adding phenylalanine or its methyl ester to
a reaction mixture obtained by reacting aspartic acid
with stoichiometric quantities of formic acid and
acetic anhydride in the presence of or in the absence
of a catalyst. The present invention, which makes it
possible to reduce the number of steps and can be
highly advantageous in the industrial production, has
been completed on the basis of the above finding.
The most characteristic feature of the present
invention lies in that the crystals of N-formyl-
aspartic anhydride are obtained at the end of the
first step (dehydration of aspartic acid) in the
state of a suspension in acetic acid, which is free
from residual formic acid and acetic anhydride. The
process of the present invention has the advantage
that -the reaction mixture from the first step can be
used as it is for the condensation in the second
step, without such steps as crystallization,
- 6 -
,
~3~5~
separa-tion, evaporation, etc. and with no loss of
N-formyl-aspartic anhydride formed. In addition, the
process of the present invention is also quite
advantageous in that the formation of the ~-isomer
(N-formyl-~-dipeptide) can be suppressed and the
yield of the cL-isomer can be markedly increased when
the condensation is carried out in acetic acid or in
a reaction medium containing acetic acid (see
Japanese Patent Application Laid-Open No. 113,841/76
and U.S. Patent No. 3,933,781). In the prior
dehydration processes for producing N-formyl-L-
aspartic anhydride, the acetic acid contained in the
reaction mixture must be once removed, and next,
acetic acid must be added again to the anhydride
obtained. The process of the present invention is
~;; free from such inefficiency.
In the process of the present invention, the
L-isomer of aspartic acid is used since the object of
the present invention is to produce the L,L-isomer of
the N-formyl-d~dipeptide. However, it may contain
its D-isomer in an amount not causing adverse effects
in the following steps.
In the process of the present invention, formic
; acid and acetic anhydride must be used in quantities
~ stoichiometric to aspartic acid. To be more
~` .
~ - 7 -
.. - ... , ~ , . .. . ..... . .
~31~3~
specific, it is most preferable to use one mole of
formic acid and 2 moles of acetic anhydride, per mole
of aspartic acid. However, good results can be
obtained by using from 0.9 to 1.1 moles of formic
acid and from 1.9 to 2.1 moles of acetic anhydride,
respectively, per mole of aspartic acid. If the
amount of formic acid used in the reaction is
insufficient, the formylation for protecting the
amino group of the aspartic acid proceeds only at a
low rate, and when it is used excessively, the
condensation in the second step proceeds only at an
insufficient rate. Similarly, when acetic anhydride
is used in an insufficient amount, the yield of the
dehydrated product will become lower, and the overall
yield of the desired will become lower when it is
used in an excessive amount. In order to maximise
the overall yield of the N-formyl~-dipeptide from
the starting material, i.e., aspartic acid, the raw
materials must be used in stoichiometric quantities
within the range described hereinabove.
,
Examples of metal compounds which may be used as
a catalyst for the dehydration include oxides and
hydroxides of a variety of metals, including alkali
metals, such as lithium, sodium, calcium, etc.,
alkaline earth metals, such as magnesium, calcium,
8 -
.
~3~35~
etc., elements of the copper group, such as copper,
e-tc., elements of the zinc group, such as zinc, etc.,
elements of the boron group, such as alurninum, etc.,
and elements of the iron group, such as iron, etc.,
as well as their salts with various acids, for
example, carbonates, carboxylic salts (e.g., salts
with acetic acid), hydrochloric salts (hydro-
chlorides) hydrobromic salts (hydrobromides),
~i nitrates, phosphates, sulfates and the like (see
Japanese Patent Application Laid-Open No.
175,484/84).
Although there are no particular limitations on
. ,
the amount of catalysts to be used, they are usually
used in an amount not causing adverse effects in the
following steps. The amount of catalysts to be used
depends on the kind of catalysts used. Good results
can be obtained by the use of even an extremely small
amount of catalysts, as shown, for example, in
Example 1, wherein 0.001 moles of magnesium acetate
is used per mole of L-aspartic acid. The amount of
catalysts to be used when the process of the
invention is practised in a commercial scale can be
determined without difficulty by those skilled in the
art by means of preliminary experiments. They are
usually added before the start of the dehydration,
,
,
. .
~ ~14358
however, it is also possible to add the catalysts to
the reaction medium during the course of the
reaction. In the process of the invention, the use
of catalysts is almost indispensable to minimize the
amount of by-products formed and the amount of the
starting materials that remain unreacted. However,
there is no need to use catalysts in cases where the
formation of by-products and the presence of
unreacted starting materials are permissible.
The dehydration is preferably carried out at a
temperature not higher than 100C, but not lower than
10C. It is most preferable to carry out the
reaction at a temperature of from 45C to 65C in
order to prevent the racemization of the reaction
product and to shorten the reaction time.
In the second step of the process of the
invention, the L-isomer of phenylalanine or methyl
ester is used, however, D-isomers may be contained
. ~, ,
therein in an amount not causing adverse effects in
the following steps. Phenylalanine and its methyl
~ ~ ester to be added to the reaction (dehydration)
;~ mixture can be either in the form of crystals or in
, ~
the form of a solution or a suspension in an
appropriate non-aqueous medium, such as acetic acid,
toluene, and the like.
'''.
~:
~,~ -- 10
,
., '~'` ': : , .
.,, ~, . ~ . ,
:. ~ , , :
,
,, :
131~3~g
There are no particular limitations on the
reaction temperature of the condensation. There is
no need of heating since the condensation proceeds
with a shor-t period of time even at room temper~ature.
It can be in the range of from -10C to 60C.
The reaction mixture containing the product of
the condensation, i.e., N-formyl-L~-aspartyl-~-
phenylalanine or its methyl ester can be used as it
is in the following deformylation step. Where
desired, the N-formyl-dipeptide or its methyl ester
can be subjected to a treatment for removing or
~ ~.
substituting the solvents by means of recrystalliza-
tion, evaporation, extraction, or -the like, before
being subjected to deformylation.
~In the case where the product formed-in the
: .~
second step is an ester (i.e., N-formyl-L-~-
~`iaspartyl-L-phenylalanine methyl ester), the desired
final product, L~-aspartyl-L-phenylalanine methyl
ester, can be readily obtained by subjecting the
ester to any of the known deformylation processes
(see, e.g., Japanese Patent Application Laid-Open No.
185,545/83).
In the case where the product of the second step
is a dipeptide having a free carboxyl group (i.e.,
N-formyl L aspartyl-L-phenylalanine), it can be
~f !~
-f~
: ' ' ~ .' ,'
~3~35~
deformylated and then esterified. It is also
possible to effect both deformylation and esterifica-
tion of the product at the same time (see U.S. Patent
No. 4,173,562).
The present invention will further be illustrated
by way of examples. It would however be understood
that the examples by no means restrict the scope of
the invention.
EXAMPLE 1
To 10.35 g (0.225 mol) of formic acid was added
45.99 g (0.450 mol) of acetic anhydride. Thereafter,
48 mg (0.00023 mol) of magnesium aceta-te (catalyst)
was added thereto with stirring, and the resulting
mixture was heated to a temperature of 55C. To this
was added 30.00 g (0.225 mol) of crystals of
L-aspartic acid, and the reaction was allowed to
proceed for 6 hours.
After the completion of the dehydration, the
reaction mixture was cooled to 30C, and 338 ml
(0.203 mol) of 0.6 mol/l solution of L-phenylalanine
methyl ester in toluene was dropped thereto. This
condensation of N-formyl-aspartic anhydride and
L-phenylalanine methyl ester was allowed to proceed
for 1 hour. The resulting reaction mixture was in
` the state of a slurry.
12 -
. ,
, ~ -
.
,:
-:
lL3~358
The reaction products were extracted with 500 ml
of water, and the aqueous solution obtained was
analyzed by high pressure liquid chromatography,
using a column of "Unizir QC" manufactured by
Gasukuro Industries Co., an analyzer of type "L~-3A"
manufactured by Shimadzu Corp., and an eluent of
phosphoric buffer. The products were de-tected at a
wavelength of 210 nm.
~; The yield of the desired product, N-formyl-Lt~-
aspartyl-l-phenylalanine methyl ester was 73.0%,
based on the amount of L-aspartic acid charged. The
ratio of the amount of the C~-isomer formed to that of
the -isomer formed as a by-product (~/~) was ~.85.
` EXAMPLE 2
The preparation of N-formyl-L-~-aspartyl-L-
phenylalanine methyl ester was repeated in the same
manner as in Example 1, except that the catalyst was
not used~
; The yield of the desired methyl ester was 53.8~,
based on L-aspartic acid charged. The ratio of the
~-isomer to the ~-isomer formed (d/~) was 3.83.
It would be understood from the above results
that when no catalysts are used in the dehydration
step, a lower yield is resulted and the formic acid
and
~C~ ` - 13 -
~' ~
.
,. .
13~43~8
acetic anhydride remain unreacted in larger
quanti-ties, and as a result, the yield of the
~-isomer in the condensation step also decreases.
EXEMPLE 3 AND COMPARATIVE EXAMPLE 1
The preparation of N-formyl-L-aspartyl-L-phenyl-
alanine methyl ester was repeated in the same manner
as in Example lj except that the molar ratios of
formic acid and acetic anhydride to L-aspartic acid
were changed to 1.1 and 2.1, respectively (Example
3), or to 1.2 and 2.2, respectively (Comparative
Example 1). Results obtained are shown in Table 1,
together with the results of Example 1.
The yields of the dehydrated products shown in
the table were determined in the following manner:
The dehydrations were repeated under the same
conditions as described above, and the slurries
obtained were condensed under reduced pressure, so as
to remove off the solvents by evaporation. Methanol
(100 ml) was added to the residue to dissolve it.
The amounts of N-formyl-~-aspartyl methyl ester and
N-formyl-L-~-aspartyl methyl ester were determined.
It is possible to know the yield of the N-~ormyl-
L-aspartic anhydride through the determination of the
- 13a -
. .
~3~43~
- quantities of the cL- and the ~-methyl esters since
the two ester are formed when the N-~ormyl-L-aspartic
anhydrides comes into contact with methanol.
'
~ ~ .
.
''
- 13b -
Xi ~
... . . .. . . ... . . . . . . . .. . . . . . . . . . ..
o~
3 ~ ~
~33~ I .
..
b
:~ o~ ~ ,.
e~ o O~ 0
:, I ~ ~
, .
~ C M ,- -
~ ~, .
~3~43~
It would be understood fro~ ~he above resul~s that
in the case where the molar ratios o ~ormic ~c~d and
acetic anhydride to aspa~tlc aoid a~e ~reater than 1.1
and 2.1, respectively, the yield of N~or~yl~ aspartyl-
L-phenylalan~ne methyl ester in the step of condensation
ls lowered considerably.
It would al~o be undesstood that ln the case where
ormic acid and acetic ~cid are used i~ strietly stoich~ometric
~antitles (Example 1), the highest yield can be attained
in the conden~ation step even i~ the yield attainable
in the dehydra~ion step is more or ~ess low~r.
When the reaction mixture o the dehydration in
Example 3 wa~ cooled to 5 C an~ the crystals of ~-formyl-
aspartic acid was collected by fil~ration, there was.
obtained an yield o~ 83% (~Yield in reaction~ x IYield
in cry~tallization3). ~t is a ~atter o~ co~rse that,
when the ~ame operation is applied to Example 1, the
yield o~ the anhydride wil} become lower because o~
the loss upon crystalllzation. The present ~n~ent~on
~a~es it possible to simp~ify the steps to a considerable
. e~tent in comparison ~ith the prior processe~ in~olving
the separation o the cr~stals of dehydrated products,
and yet ~o atta~n an yield as high ~s in the proior
proce~se~.
; ~XA~MP~ 4
: The dehydration step ~n Example 1 was repeated
: .. .
,
IOM ~j .
~ 3 ~ ~L 3 ~ 8
under th~ same conditions, The r-eaction ~ixture obtained
was heated to a temperature.of ~0 C, a~d 33.45 g (0.203 mol~
of crystal~ o~ L-phenylalanine was dropped ~heret4 over
a period o 45 minutes, The reaction was allowed to
proceed ~or additional 30 minute~, and the contents
o~ the thus obtained rea~tion mi~ture were analyzed.
The yield o the de~ired pxoduct, ~-~onmyl-L-a-
aspartyl-L-phenylalan~ne, wa~ 62.0 %, based on the L-aspartic
acid ch~rged. The ratio o~ the ~- ~o ~-isomers formed
) was 2.45.
COMæARATIVE ~XAMPLE ~
The dehydration process described in Japanese Patent
Publioation No. ~,133180 (U.S. Patent No. 3,933,781)
," ~ .
was employed.
To 84.8 8 (1.843 mol) of formic acid was added
44.05 ~ ~0.479 ~ol) of acetic anhydrideJ an~ the resultin~
mixturc was sti~red ~or 45 minute3 at 25 oc . Thereafter,
30,0 g (0.225 mol) o crystals of L-aspartic acid was
~ added thereto, and the reacLton ~dehydration) was allo~d
; to proceed ~or 3.S hour~
The temperature of ~he thus obtalned reaction mlxture
was raised to 50 ~C, and 35.33 g ~0.214 ~ol) o cryseals
o L~phenylalanine were addcd thereto over a period
o~ 45 minutes, The reac~ion ~condensation) was allowed
to proceed for additional 30 minutes,
~ e yield of the de ired product, N-for~yl--aspartyl
L-p~P~ylalanine, was 12.1Vf~, based on ~he aspar~ic acid
'
lo~
,".j
~31~3~
~c
charget. The ratio of t~e ~- to ~-iso~er~ ~ormed (~
was 1.83. - ~
, The above is a result of a ~arked ~ecrease in the
yleld of condensatlon, due to the presence of residual
~ormic acid used in an excessive quantity in the dehydration
s tep,
In the prior art descr~bed in U.S. Paten~ No. 3,933,781,
the reaction mixture is sub~ected) after the dehydration,
to evaporation to remove of the excess formic acid.
In order to remove off all the remaining formic acid,
all the solvents, including the acetic acid ormed from
acetlc anhydride, had to be evapo~ated off to giYe a
drled rasidue.
To this residue was a8ain added 3~6.3 g of acetic
acid, and the temperature of the resulting mixture was
raised to ~5 ~C. Therea~er, 35.33 g (0.214 mol) of
crystals o ~-phenylalanine w~s added there~o over a
period o 45 minute~, and the reac~ion was allowed to
-proceed ~or addltional 30 minutes. The desired ~-dipeptide
was obtalned i~ an yield of 64%, based on the aspartic
acid charged, and the ratio o~ thP ~- to ~-isomers or~ed
(al~) was 2.37,
. It would be easil~ ~nderstood from the results
that the present ~vention makes i~ possible to simpl~fy
t~e steps and to reduee the energy consumption to a
- 17- .
~3~43
. ' " . '
considerab:le e~ten~, throu~h the elimina~ion o the
- e~raporation of solvcn~ s without decrease ~n the yield
of ~he desired product.
.
.
-18~
