close

Вход

Забыли?

вход по аккаунту

?

Quality Control of Peptide Drugs. Chiral Amino Acid Analysis versus Standard for Icatibant Acetate

код для вставкиСкачать
Full Papers
Quality Control of Peptide Drugs. Chiral Amino Acid Analysis versus
Standard for Icatibant Acetate
Joachim Ermer*, Jurgen Gerhardta), and Martin Siewert
Hoechst AG, Pharma Quality Control, 65926 Frankfurt / Main, Germany
a’
C.A.T. Chromatographie und Analysentechnik KG,Heerweg 10,72070 Tubingen, Germany
Key Words: chiral amino acid analysis, HOE 140, quality control, peptide drug, validation
Summary
A method of chiral amino acid analysis versus standard is pre-
sented. By treating the peptide sample and a chiral standard whose
intrinsic chiral composition is known in the same way, i.e. simultaneous hydrolysation and analysis, exact corrections can be made
for the sequence-related and many hydrolysation-related racemisations. The accuracy obtained in this way permits use of the results
for quality control of the chiral purity of peptide drugs. Using the
bradykinin antagonist Icatibant acetate (INNM) the method was
validated with respect to its precision, accuracy, specificity, limit
of detection. and robustness.
Introduction
The pharmaceutical development of peptide drugs is a
scientific challenge requiring special analytical effort and
sophisticated methods to achieve an adequate control of drug
quality
During the development of a new drug substance particular
attention is focused on potential and actually occurring impurities which cannot be avoided or further reduced by the
optimization of the chemical production process or which
increase by degradation during storage. Several international
guidelines have established orientational requirements for
registration purposes and very recently precise recommendations for “classical” synthetic pharmaceuticals have been
elaborated as part of an international harmonization process
[31. Since the proposed approaches are neither fully applicable
to nor valid for peptide drugs and since chiral aberrations
represent typical impurities for peptide drugs, a new proposal
for chiral purity analysis is presented.
Synthetic peptide drugs are often partially composed of
D-enantiomeric and/or non-naturally occurring amino acids.
Impurity peptide sequences containing “wrong” amino acid
enantiomers can occur by racemisation during synthesis
and/or from impurities in starting materials. Due to the very
similar physico-chemical properties of such stereoisomers,
commonly used reversed-phase HPLC offers only a limited
range of resolution. For a decapeptide, e.g. it is rarely possible
to separate all potential stereoisomeric impurities. Therefore,
an additional control test is needed for the determination of
the chiral purity.
To determine the optical purity of the amino acids constituting a peptide, the sample is usually hydrolysed and the
resulting free amino acids are analysed. However, the hydrolysation of a peptide results in the racemisation of the amino
acids to different degrees dependin on hydrolysation conditions and the peptide sequence ld.
In order to compensate
for this racemisation, a control test method for the new
peptide drug icatibant acetate (HOE 140) was established
using a defined chiral standard. By treating the sample and a
chiral standard in the same way, i.e. simultaneous hydrolysation and analysis, exact corrections for the sequence-related
and many hydrolysation-related racemisations can be made,
provided that the intrinsic chiral composition of the standard
is known.
NH
II
/\
I
NH*
0
II
II
NH
on
II
NH
Fig. 1: Structural formula of Icatibant acetate (HOE 140). D-Arginyl-L-arginyl-L-prolyl-L-[(4R)-4-hydroxyproly1]-glycyl-L-[3-(2-thienyl)alanyl]-L-selylD-(l,2,3,4-tetrahydroi~uinoline-3-yl-car~nyl)-L-[(3~,7~-oct~ydroindol-2-yl-c~bonyl]-L-~ginine
acetate. H-D-Arg-Arg-Pro-Hyp-Gly-Thi-Ser-DTic-Oic-Arg-OH.Molecular weight (base): 1304.6
Arch. Pharm.(Weinheim) 328,635639 (1995) 0 VCH VerlagsgesellschaftmbH, D-6945 1 Weinheim, 1995
0365-6233/95/0909-0635 $5.00 + .25/0
636
Ermer, Gerhardt, and Siewert
The control test is validated with respect to its precision,
accuracy, specificity, limit of detection, and robustness.
Validation of the Method
The method described was validated on the basis of regulatory requirements [13,141.
Specificity
The specificity of the GC-method is demonstrated by the
attached chromatograms (Fig. 2). The two enantiomers of
each amino acid present in icatibant are separated from each
other as well as from other amino acid enantiomers.
Limits of detection
The limit of detection of the GC method for each impurity
enantiomer was deduced from a signal-to-noise ratio of 3: 1
using typical chromatograms (Fig. 2, Table 4).
e
A
I
Fig. 2: GC chromatogram of derivatised amino acids.
Amino acid enantiomer: Retention times (min)
D-Pro: 6.53, L-Ro:6.68
D-Ser: 7.04. L-Ser: 7.68
L-Hyp: 10.50, D-Hyp: 11.96
D-Thi: 15.12, L-Thi: 15.61
D-Arg: 25.27, L-Arg: 25.71
Precision
The repeatability was determined retrospectively for 24
samples analysed as described.From the differencesbetween
the duplicate determinations, variances were calculated for
each minor enantiomer. Depending on the amino acid, the
standard deviations range between 0.10%and 0.24% (Table
2b). There are only small variations between the amino acids.
With the exception of D-Ser which is above the detection
limit in only 9 out of 24 samples, an average standard deviation of 0.19% is obtained. Therefore, the repeatability limit
for the determination of the enantiomeric content within one
series was set to 0.6% for all amino acids.
Repeatability limit: r = 2.83 x SD = 0.6 %
Reproducibility: Icatibant acetate (batch no. 2) was analysed three times. The mean values from three replicates in
each series were used to calculate the corrected values according to Eq. (1) (Table 3). They show good agreement
between the series. With the exception of L-Tic, the ranges
are even below the (intraserial)repeatability limit. In accordance with the large intraserial variances for L-Tic and D-Oic
(see Table 2a), a corresponding scattering of the corrected
values is observed for these amino acids.
Accuracy
When hydrolyzing a peptide sample in D;?O/DCl any
racemisation is accompanied by a deuterium exchange in the
a-C position. By monitoring the non-deuterated molecular
ions or suitable fragment ions with EI-SIM-mass spectroscopy, the amounts of D- or L-enantiomers originally present
in the peptide can be determined (“Intrinsic composition,
experimental”, see Tables 1 and 4 for chiral standard and
batch no.2, respectively). The lower sensitivity for D-Thi is
caused by a deuterium exchange in several positions and a
subsequently decreased response of the ion to be monitored.
In contrast, the limit of detection for D-Thi in the conventional GC method was determined as 0.2 %. Therefore, it is
justified to assume that the lowest confirmed content of D-Thi
is the upper limit of the intrinsic value. In 23 determinations
of the chiral standard, 0.6 % D-Thi was obtained 5 times.
Thus, the chiral standard was defined as given in Table 1
(“Intrinsic values, defined”).
The accuracy of the chiral amino acid analysis versus
standard was proved by comparing the corrected values for
icatibant acetate (batch no. 2) obtained according to the
control test with the intrinsic chiral composition (Table 3).
With respect to the differentlimits of detection of the methods
used and the possible error progression in the correcting
calculation there is very good agreement between the two
independent methods.
Robustness
The robustness of the control test method was investigated
by subjecting both the sample (batch no. 2) and chiral standard to thermal stress conditionsfor 3 and 14 d at 37 “C (Table
4). In an inspection of the experimental values of batch no. 2,
minor-to-moderate increases are observed in the amount of
the impurity enantiomer (an approximately 1 % increase for
D-Hyp, L-Tic, and D-Oic). In contrast, the corrected amount
of minor enantiomers of the stressed samples show neither
significant deviation from the unstressed ones nor from the
intrinsic values.
Results and Discussion
In order to make use of the results of a chiral amino acid
analysis as a quality parameter for the purity of a peptide drug,
a high level of accuracy and precision is needed. Commonly
performed methods are based on the hydrolysis of the peptide
sample and subsequent chiral separation of the amino acid
enantiomers [8-1 ‘I. The racemisation rate of each individual
Arch. Pharm.(Weinheim)328,635-639 (1995)
637
Quality Control of Peptide Drugs
Table 1: "Conventional" correction for racemisation of icatibant acetate (chiral standard)
% minor enantiomer
D-Arg
D-Pro
D-Hyp
D-Thi
L-Tic
D-Oic
0.2
1.7
2.0
1.2
0.3
2.6
2.9
33.6
-0.5
4.4
-0.2
0.0
-0.4
0.2
series 2
33.6
-0.4
-0.4
-0.3
0.0
-0.3
4.2
series 3
33.8
-0.2
0.4
0.3
0.3
0.2
0.3
series 4
33.2
-0.9
4.7
-0.5
0.1
0.2
-1.3
series 5
33.5
-0.7
4.4
0.5
0.6
0.1
series 6
33.2
-1.1
-1.1
-0.6
0.0
4.5
series 7
33.2
-0.9
-0.8
-0.4
0.0
1.2
experimental
33.3
<0.2
<0.2
4.5
0.2
<0.3
0.2
defined
33.3
<0.2
<0.2
<0.6
0.2
<0.3
0.2
L-Tic
D-Oic
Correction factors
Corrected values (Eq.2)
series 1
D-Ser
0.4
-1.1
-1
.o
Intrinsic values
Table 2: Statistical analysis.
% minor enantiomer
Variances of experimental values
D-Arg
D-Pro
D-Hyp
D-Thi
D-Ser
0.026
0.026
0.039
0.030
'
0.033
0.023
variances s:
0.106
0.141
0.332
0.313
'
0.829
0.679
F=s:/s;
4.13
5.41
8.49
10.35
#
25.37
0.22
0.16
O.lOw
0.24
0.23
a) 10 determinations of the chiral standard
Intraserial
variances s t
Interserial
tabulated value F
(P= 0.99;fi = 9;fz = 13): 4.2
b) 24 determinations of 18 different batches
Standard
deviations
#:
0.16
0.15
For D-Ser, no variances could be calculated. The values of only three series were above the detection limit.
#: Only the ranges of 9 duplicate determinations could be used other results at or below the detection limit.
Arch Phnnn (Weinheim)328,635639 (1995)
29.42
638
h e r , Gerhardt, and Siewert
Table 3: Chiral amino acid composition of icatibant acetate (batch no. 2, correction vs. standard, Eq.(l).
D-Arg
D-Pro
% minor enantiomer
D-Hyp
D-Thi
D-Ser
L-Tic
D-Oic
series 1
33.3
<0.7
<0.6
4.4
0.4
<0.3
0.1
series 2
33.0
<0.3
<0.5
4.5
0.4
<0.8
0.3
series 3
33.0
<O. 5
<0.5
4.4
0.3
<1.1
0.7
average
33.1
<0.5
<0.5
4.4
0.4
<0.7
0.4
Intrinsic composition
32.9
<0.2
<0.3
<1.0
0.3
0.3
<0.3
Table 4: Experimental and corrected chiral composition of thermally stressed icatibant acetate (batch no. 2).
% minor enantiomer
D-Arg
D-Pro
D-Hyp
D-Thi
D-Ser
L-Tic
D-Oic
0.2
0.2
0.2
0.2
<0.3
0.6
0.6
33.4
1.2
1.6
1.5
0.5
2.3
2.1
3 d at 37 “C
33.4
1.3
1.5
1.3
0.5
2.1
2.6
14 d at 37 “C
33.7
1.6
2.6
2.0
1.o
3.4
3.5
33.1
<0.5
<0.5
<1.4
0.4
<0.7
0.4
3 d at 37 “C
33.0
~0.4
<0.4
<0.8
0.4
<0.8
0.4
14 d at 37 “C
32.9
<0.2
<0.2
<1.0
0.5
<0.9
0.5
limit of detection
Experimental values
unstressed ”
Correction vs. standard (Eq.(l))
unstressed *)
” series 2
average from three series
2,
amino acid is not only dependent on the hydrolysation conditions but also on the amino acids which are linked to it, i.e.
from the sequence [41. Thus, the determined amount of an
enantiomeric amino acid represents the content originally
present in the peptide (intrinsic chiral composition) and also
the amount generated during the analytical procedure.
Correction factors using free amino acids or extrapolation
are not suited to cover sequence-dependent racemisation or
differing hydrolysation conditions. Therefore, the results obtained by this “conventional” correction (Eq.(2)) are not
satisfactory (see Table 1). Differences caused by the sequence
are indicated by frequent negative values.
The influence of hydrolysation conditions is demonstrated
by significant differences in the experimentally obtained
results between independent series of the same batch. For
Table 2a, the results of 10 independent series are statistically
analysed. Variances are calculated within (intraserial) and
between these series (interserial). Comparing the ratio of
intra- and interserial variances with the tabulated values
(F-test) [I2] shows significant differences for all amino acids
except Arg at a level of significance of 99 %. The large
interserial variances of L-Tic and D-Oic indicate that these
amino acids are especially sensitive to hydrolysis-caused
racemisation. Correcture by fixed factors does not influence
these differences. This is also reflected by the large range of
corrected values given in Table 1.
Both sequence-related and many hydrolysation-related
racemisations can be compensated by using a standard substance the chiral composition of which is known and which
is exposed to the same procedure as the sample to be analysed.
Assuming the same influence during simultaneous treatment,
sequence- and condition-related racemisation is identical for
the sample and chiral standard. Therefore, by comparing
experimental and intrinsic values of the chiral standard, the
racemisation-related amount of the corresponding enantiomer can be separated and the sample corrected (Eq. (1)).
Arch Pharm.(Weinheim)328.635439 (1995)
639
Quality Control of Peptide Drugs
Conclusions
The method described is well suited for quality control of
chiral peptide drug purity. Simultaneoustreatment of sample
and standard, the intrinsic chiral composition of which is
known, provides the possibility of an exact correction for
sequence-related and many hydrolysis-relatedracemisation.
In particular, the former contribution is not separable by the
use of “conventional” correction factors. The control test
method was demonstrated to determine the chiral composition of a sample precisely, accurately, and in a reproducible
manner.
Experimental
Icatibant Acetate (HOE 140, Internal Code)
Icatibant (Fig. 1) is a synthetic decapeptide with a similar structure to
bradykinin.It containsthree non-naturallyoccurringand two D-enantiomeric
amino acids which support both high affinity to the bradykinin Bz receptors
and resistance to enzymatic degradation. Icatibant is the most potent of the
Bz antagonists reported so far, and has virtually the same affinity to these
receptors as bradykinin itself ‘71. This drug substance contribute to the
investigation of the pathophysiological role of bradykinin. The reversal of
bradykinin-mediatedeffects may constitute a new approach to the treatment
of certain diseases.
temperature is increased to 200 “C at a rate of 4 “/min and maintained for
another 20 min. The peaks are identified by retention times.
Correcting Calculation
The experimental results are calculated on the basis of the computed peak
areas as a percentage of the minor enantiomerof each amino acid calculated
with reference to the sum of both enantiomers. With the exception of Arg,
the minor enantiomers of the other amino acids originate from impurities of
the drug substance or method-caused racemisation. Therefore, the terms
“minor” and “impurity” enantiomer are used as synonyms. The amount of
enantiomeric impurity originally present in the peptide sample is calculated
for each amino acid
Akorr = Akxp - AASexp + AASint
(1)
AAo = amino acid enantiomer of the sample, corrected
AGxp= amino acid enantiomer of the sample, experimental
AASexp= amino acid enantiomer of the standard, experimental
AASint= amino acid enantiomer of the standard, intrinsic
“Conventional” Correctionfor Racemisation
A A o = 50 - (50 - A k x p )x e2b
k = Rate constant [s-’1of inversion of L-amino acids
t = hydrolysis time [s]
References
Determination of the Intrinsic Chiral Composition of the Icatibant Standard
About 0.5 mg of the standard is hydrolysed in 0.5 ml of 6N DCl in Dz0
for 24 h at 110 “C. After the removal of excess reagent with a stream of
nitrogen, the standard is esterified with 0.4 ml of 4N hydrochloric acid in
n-propanol for 30 min at 110 “C. After cooling to about 50 “C, the vial is
opened and the reagent evaporated with a gentle stream of nitrogen at a
moderate temperature. For the determination of Arg, Pro, Hyp, Thi (3-(2thieny1)-alanine),and Ser (abbreviations of the proteinogenic amino acids
according to the three letter code of IUPAC), half of the residue is dissolved
in 0.15 ml of trifluoroacetic anhydrideand the vial is tightly closed and heated
for 10 min to 150 “C. Forthe determinationof Tic (1,2,3,4-tetrahydroisoquinoline-3-yl-carboxylic acid) and Oic ((3aS,7aS)-octahydroindol-2-yl-carboxylic acid), the other half of the residue is dissolved in 0.1 mi of pivaloyl
chloride and heated for 10min to 110 “C. After cooling to room temperature,
the excess reagents are removed with a stream of nitrogen. The residues are
dissolved in 0.15 ml of toluene.
The derivatized amino acids are separated by capillary gas chromatography (see below). EI-SIM-mass spectroscopy is used for detection. The
relative amounts of D- and L-enantiomers originally present in the sample
are determined by monitoring the non-deuterated molecular ions or suitable
fragment ions of both enantiomers. These are for Arg: 435 d e , Pro: 253 d e ,
Hyp: 365 d e , Thi: 309 d e , Ser: 352 d e , Tic: 218 d e , and Oic: 208 d e .
Chiral Amino Acid Analysis versus Standard
P.J.M. van den Oetelaar, P.S.L. Jansen, P.A.T.A. Melgers, G.N.
Wagenaars, P.B.W. ten Kortenaar, J.Contr.Release 1992,21, 11-22.
Salem, M.C. Bedmar, M.M. Medina, A. Cerezo, J.Liq.Chromatogr.
1993, 16, 1183-1194.
International Conference on Harmonization (ICH) of Technical Requirements for the Registration of Pharmaceuticals for Human use.
Impurities in New Drug Substances. Sixth Draft,March 15,1994.
W. Woiwode, H. Frank, G.J. Nicholson, E. Bayer, Chem.Ber. 1978,
111,3711-3718.
H. Frank, W. Woiwode, G.J. Nicholson, E. Bayer, Liebigs Ann. Chem.
1981,354-365.
J. Gerhardt, J. Maucher, Peptides 1992, Proceedings of the 22nd
European Peptide Symposium, Escom Leiden 1 9 9 3 ,4 5 7 4 8 .
F.J. Hock, K. Wirth, U. Albus et al., Br.J.Pharmaco1. 1991, 102,
169-773.
H. Frank, G.J. Nicholson, E. Bayer, Chromat.Sci. 1977, IS, 174-176.
E. Bayer, ZNaturforsch. 1983,38b, 1281-1287.
[ 101 J.Gerhardt, K.Nokihara, R.Yamamoto, Peptides Chemistry and Biology, Proceedings of the 12th APS (Eds. J.A.Smith, J.E.Riviers),
ESCOM 1992,531-532.
Approximately0.2 mg of the sample and 0.2 mg of the chiral standard are
each dissolvedin 0.5 ml6N hydrochloricacid and hydrolysedsimultaneously
for 24 h at 110 “C. Both sample and standard are hydrolysed in duplicates.
The derivatisation procedure is the same as described above.
[ l l ] J.G.Adamson, T.Hoang, A.Crivici, G.A.Lajoie, Biochem. 1992, 202,
2 10-214.
Capillaty Gas Chromatography
[ 131 Test on Validation of Analytical Procedures.Draft ICH TripartiteText,
The N(0,S)-trifluoroacetylamino acid esters are separated on a desactivated glass capillary coated with 0.2 pm of Chirasil-Val (dimensions 20 m
x 0.28 mm). The carrier gas is hydrogen (1.5 ml/min). Split-injection of the
sample is performed at a column temperature of 80 “C; after 3 min, the
Arch P h a m (Weinheim)328.635439 (1995)
[ 121 K.Doerffe1, Statistik in der analytischen Chemie, Dt. Verl. Grundstof-
find., Leipzig 1990,141-146.
Pharmeuropa 1993,5/4,343-345.
[14] USP XXII, General Information, <1225> Validation of Compendia1
Methods, 1710.
Received: March 1, 1995 [FPOO5]
Документ
Категория
Без категории
Просмотров
4
Размер файла
385 Кб
Теги
acid, chiral, drug, amin, analysis, versus, quality, peptide, acetate, standart, control, icatibant
1/--страниц
Пожаловаться на содержимое документа