Reasons for structure-acid resistance correlation in series of substituted phenylpenicillins

Article abstract of DOI:10.1023/A:1016172722206

The kinetics of acid hydrolysis of phenylpenicillins substituted in the benzene ring were studied at pH 3.0 and temperatures of 20, 25, and 30°C. A Hammett correlation analysis of the reaction was performed.

Full text of DOI:10.1023/A:1016172722206

Russian Journal of Applied Chemistry, Vol. 75, No. 2, 2002, pp. 281 285. Translated from Zhurnal Prikladnoi Khimii, Vol. 75, No. 2, 2002,
pp. 291 295.
Original Russian Text Copyright
2002 by Solovskii, Pek.
MACROMOLECULAR CHEMISTRY
AND POLYMERIC MATERIALS
Reasons for Structure Acid Resistance Correlation
in Series of Substituted Phenylpenicillins
M. V. Solovskii and G. Yu. Pek
Institute of Macromolecular Compounds, Russian Academy of Sciences, St. Petersburg, Russia
Shemyakin and Ovchinnikov Institute of Bioorganic Chemistry, Russian Academy of Sciences,
Moscow, Russia
Received July 2, 2001
Abstract The kinetics of acid hydrolysis of phenylpenicillins substituted in the benzene ring were studied
at pH 3.0 and temperatures of 20, 25, and 30 C. A Hammett correlation analysis of the reaction was
performed.
Antibiotics of the penicillin series significantly
electronic effects of substituents. Hence, significant
differ in stability in acid solutions. Among acid-re- differences in acid resistance must be expected for
sistant penicillins, which can be administered per-
orally, are phenoxymethylpenicillin, Oxacillin, Cloxa-
cillin, etc. Benzylpenicillin, p-hydroxybenzylpenicil-
lin, and Methicillin show poor acid resistance.
substituted phenylpenicillins too.
The substituted phenylpenicillins were prepared by
acylation of the core compound, 6-aminopenicillanoic
acid, with various substituted benzoyl chlorides in
aqueous acetone in the presence of sodium hydrogen
carbonate [4]. The properties of the resulting sodi-
um salts of substituted penicillins, which show high
biological activity, are listed in Table 1. The purity
Penicillins contain a common bicyclic system con-
sisting of fused -lactam and thiazolidine rings and
differ only in the structure of the side acyl group.
All the differences between penicillins are largely due
to structural features of this substituent, and, from this
standpoint, it is appropriate to discuss the difference
in acid resistance between penicillins.
Table 1. Sodium salts of phenylpenicillins
S
X
CH3
CH3
C NH CH CH C
Strukov et al. [1, 2] believe that the differences in
acid resistance of penicillins are associated with the
amide imidol tautomerism of their side chain; how-
ever, no explanations for this hypothesis are given.
Doyle et al. [3] showed that electron-withdrawing
substituents in the side aliphatic acyl group enhance
the acid resistance of penicillin, but they gave no
explanations for this fact either. Therefore, it seemed
appropriate to study the acid resistance of specially
chosen penicillin derivatives differing in the nature of
substituent in the aromatic ring of the side acyl group,
with the aim to reveal a correlation between the struc-
ture of the side acyl droup, on the one hand, and the
rate of penicillin acid cleavage, on the other, and also
to find the reason for this correlation.
O
C
N
CH COONa
O
Activity
X
iod-
optical D17, biological,
ometric,
g mg
C = 0.5%
in H2O
MSC,*
units ml
1
1
1
p-OCH3 49
740
820
970
850
860
820
850
+216
+232
+250
+282
+228
+222
+252
0.35
0.40
0.50
0.78
0.48
0.50
0.55
2
3
4
5
6
7
p-CH3
H
56
61
45
52
61
73
p-Cl
p-Br
m-NO2
p-NO2
As objects for our study we chose phenylpenicillins
substituted in the side chain. Their side acyl groups
are structurally related to benzoic acids, for which the
the dissociation constants strongly depend on the
*
MSC is the minimum suppressing concentration for
Staphylococcus aureus 60.
1070-4272/02/7502-0281 $27.00 2002 MAIK Nauka/Interperiodica

282
SOLOVSKII, PEK
Table 2. Methyl esters of phenylpenicillins
S1
X
C
8
6
5
2
b
C NH CH CH C CH3
a
7
4
N
3
H3C
O
C
CH C OCH3
O
O
N
S
Iodometric
activity,
Optical activity
Ester
no.*
mp,
Rf
calculated (%)/found** (%)
g mg
17, C = 0.5%
solvent
1
D
1
2
3
4
5
6
7
7.68/7.30
8.04/8.02
8.38/8.34
7.60/7.73
6.78/6.78
11.09/11.20
11.09/11.16
8.79/8.57
9.19/9.02
9.59/9.60
8.67/8.60
7.75/7.43
8.44/8.37
8.44/8.34
174 175
162 163
116 117
178 179
100 101
88 89
950
1000
960
+287
+252
+280
+256
+280
+204
+290
Cloroform:ethanol = 2:1 0.75
Cloroform:ethanol = 1:1 0.76
Ethanol
0.73
980
Cloroform:ethanol = 1:1 0.79
990
Ethanol
0.81
0.74
980
Cloroform
58 59
1000
Cloroform:ethanol = 1:1 0.77
*
**
Prepared from the respective compounds in Table 1.
Average of two determinations.
Table 3. Kinetic characteristics of acid inactivation of phenylpenicillins
1
K
103, min , at indicated temperature,
C
1
Comp. no.*
E, kcal mol
log(PZ)
20
25
30
1
2
3
4
5
6
7
70.0 0.6
40.0 0.9
18.0 0.6
7.0 0.1
81.0 0.9
47.9 0.5
28.0 0.6
11.3 0.3
10.5 0.1
1.48 0.05
1.19 0.02
117.0 2.0
65.0 0.3
42.2 0.3
17.5 0.2
16.7 0.2
2.60 0.06
2.18 0.04
5.1 0.1
7.8 0.5
14.2 0.2
16.2 0.3
16.0 0.5
20.3 0.6
21.5 0.4
0.83
2.63
7.10
7.23
7.03
10.38
11.09
6.7 0.07
*
Corresponds to Table 1.
phenylpenicillins were additionally identified in the
form of methyl esters. Furthermore, these derivatives
appeared to be more convenient for a H NMR study
of phenylpenicillins than the salts. The properties of
methyl esters of the phenylpenicillins substituted in
the benzene ring are listed in Table 2. It is seen that
these compounds were isolated pure.
of phenylpenicillins, according to iodometric analysis,
was 74 97%, which may be due to inactivation of
these compounds in the course of isolation and purifi-
cation. Purification of phenylpenicillins is very diffi-
cult because of their instability; therefore, substituted
1
The kinetics of acid inactivation of the chosen
penicillins were studied in a glycine buffer solution
(pH 3.0) at 20, 25, and 30 C. The initial concentration
1
of the antibiotic in solution was about 1000 units ml .
Under these conditions, hydrolysis of the amide bond
in the -lactam ring of substituted phenylpenicillins,
resulting in their inactivation, is a pseudomonomolec-
ular reaction and is described by a first-order equation,
as in the case of other penicillins [5 7].
The parameter log[C0/(C0 C )] is a linear func-
tion of (Fig. 1). The rate constants of acid inactiva-
tion of substituted phenylpenicillins and the activation
energies of the reactions are listed in Table 3. It is
seen that the rate constants of acid hydrolysis of
phenylpenicillins differ considerably.
Fig. 1. Acid inactivation of phenylpenicillins. 25 C,
pH 3.0; the same for Fig. 2. (A) Concentration of non-
inactivated substituted phenylpenicillin and ( ) time elapsed
from the start of the reaction. Figures at curves denote
compound nos. in Table 1.
RUSSIAN JOURNAL OF APPLIED CHEMISTRY Vol. 75 No. 2 2002

REASONS FOR STRUCTURE ACID RESISTANCE CORRELATION
The large difference between the activation ener-
283
gies of hydrolysis of p-methoxyphenylpenicillin
1
(5.06 kcal mol ) and p-nitrophenylpenicillin
1
(21.5 kcal mol ) shows that at the instant of cleav-
age of the amide bond in the -lactam ring the peni-
cillin molecule is strongly polarized. The probability
factors PZ also differ considerably. On the whole,
the rate constants of acid inactivation of substituted
phenylpenicillins vary in accordance with the polar
effect of the substituent.
Examination of the applicability of the Hammett
equation to the acid inactivation of phenylpenicillins
(Fig. 2) shows that the logarithms of the rate constants
of this reaction show a satisfactory linear correlation
Fig. 2. Correlation between the logarithms of the rate
constants K of acid activation of substituted phenylpenicil-
lins and the Hammett constants of substituents in the
benzoyl group. Figures at curves are compound nos. in
Table 1.
with the Hammett
constants in the temperature
range studied (in all cases we used common Hammett
constants [8]). Thus, despite complex structure of
penicillins, the description of their inactivation by the
Hammett equation is quite adequate. The coefficients
at 20, 25, and 30 C, calculated from the slopes of
the log K straight lines, are 2.00 (r = 0.956), 1.76
(r = 0.996), and 1.65 (r = 0.994), respectively. The
negative sign of shows that the inactivation is pro-
moted by donation of the electron density to the reac-
tion center. The coefficients are large in absolute
value, which means that acid inactivation of substi-
appreciable changes in the coupling constant of the
-lactam protons and in the chemical shift of C5H.
1H NMR data for methyl esters of phenylpeni-
cillins are given in Table 4. It is seen that for all the
compounds the coupling constants J(H5 H6) and the
chemical shifts
of the -lactam protons are very
similar. That is, there is no pronounced correlation
1
between the H NMR parameters of the lactam ring
and the structure of the side chain. The lack of such
tuted benzylpenicillins is highly sensitive to effects a correlation suggests that either there is no specific
of substituents in the benzene ring of the side acyl
group, despite their relatively large distance from the
reaction center (amide bond in the -lactam ring).
interaction in an inert medium between the amide
moiety of the -lactam ring and the side chain, or this
interaction is too weak to be detected by NMR. An
increase in the chemical shifts of the C5H protons
with increasing electronegativity of the side chain
(Table 4) suggests occurrence of a certain interaction,
but this trend is very weakly pronounced. The effect
of substituents in the benzene ring is apparently
leveled out by the amide group of the side chain,
which has a very low transmission factor = 0.12 [8].
To understand why the effect of substituents on the
cleavage of the relatively remote bond is so strong, it
seemed important to study how the substituents affect
the polarization of the amide bond in the -lactam
1
ring. Such information can be derived from the H
NMR spectra of the methyl esters of phenylpenicil-
lins. Indeed, if the electronic effect of substituents in
the benzene ring on the strength of the amide bond in
the -lactam ring were significant, it would alter the
hybridization of the nitrogen atom and hence cause
The high absolute values of for the acid inactiva-
tion of substituted phenylpenicillins, taking into ac-
count the very weak transmission of the inductive ef-
Table 4. 1H NMR data for methyl esters of phenylpenicillins (1 mol % in CDCl3)
Ester
no.
(Table 2)
a
b
OCH3
C3H
C5H
C6H
NH
J(H5 H6)
J(H6 NH)
C2 (CH3)2
, downfield from TMS, ppm
Hz
1
2
3
4
5
6
7
1.53
1.54
1.53
1.53
1.52
1.53
1.52
1.70
1.69
1.71
1.69
1.68
1.73
1.68
3.80
3.81
3.80
3.81
3.80
3.82
3.78
4.48
4.51
4.50
4.50
4.50
4.50
4.48
5.64
5.69
5.66
5.67
5.67
5.71
5.68
5.87
5.88
5.91
5.93
5.92
5.77
5.82
6.73
6.86
7.03
6.86
6.73
7.74
7.24
4.0
4.1
4.0
9.1
9.1
8.5
Not determined Not determined
4.0
4.0
8.5
8.0
Not determined Not determined
RUSSIAN JOURNAL OF APPLIED CHEMISTRY Vol. 75 No. 2 2002

284
SOLOVSKII, PEK
1
fects to the -lactam ring by the mechanism of suc-
cessive polarization of bonds, suggest direct partic-
ipation of the side chain in the limiting stage of the
acid inactivation:
(1 1.5 mg ml ) were applied to plates coated with
KSK silica gel (5 g per spot). Ascending chromatog-
raphy was performed in a closed chamber, with ethan-
ol as eluent. After separation, the chromatograms were
dried for 10 min at 20 C and for 15 min at 100 C, and
then developed with iodine vapor. The target com-
pounds gave rust-colored spots.
S
CH3
CH3
CH COOH
NH CH CH C
+
C
O
C
O
N
X
Phenylpenicillin sodium salt 3. A 0.2 g (0.00093 M)
portion of 6-aminopenicillanoic acid was dissolved in
a mixture of 20 ml of 0.1 M phosphate buffer solution
(pH 6.6) and 4 ml of 10% NaHCO3 solution. To the
resulting mixture was added dropwise, with stirring
and cooling on an ice bath, a solution of 0.197 g
(0.0014 mol; 1.5-fold excess) of benzoyl chloride in
10 ml of acetone. Acylation was performed for 20 min,
with addition of NaHCO3 solution to maintain the neu-
tral medium. After reaction completion, the unchanged
benzoyl chloride was extracted with diethyl ether and
discarded. The cooled aqueous solution was acidified
with 1 N HCl to pH 2.0, and the phenylpenicillic acid
formed in the process was extracted with 15 ml of
diethyl ether. The extract was treated with 12 ml of
0.05 N NaHCO3 to wash the acid back to the aqueous
phase in the form of the sodium salt. The product was
isolated by lyophilic drying and purified by reprecipi-
tation from aqueous acetone into dry acetone. Yield
of phenylpenicillin sodium salt 0.204 g (61%), iod-
S
+
CH3
CH3
HN CH CH C
H+
C
C HN CH COOH
X
O
O
In this case, the electronic effects of substituents
are probably transmitted by the field effect mechanism
via electrostatic interaction through space between
the oxygen atom of the carbonyl group of the side
chain and the carbonyl carbon atom of the -lactam
ring. In the process, the rate of formation of a new
O C
bond with simultaneous cleavage of the amide
bond in the -lactam ring is determined by the elec-
tron density ( charge) on the oxygen atom of the
side chain, which, in turn, strongly depends on the
substituent in the benzene ring. Just this factor is
responsible for the pronounced dependence of the rate
of acid inactivation of phenylpenicillins on the polar
effect of substituents in the benzene ring. The effect
is similar to that observed in the series of benzoic
acids, which is reflected in the linear correlation of
1
1
ometric activity 970 g mg . IR spectrum, , cm :
O C
1770 (C=O in -lactam ring); 1400 (
in
.
.
.
.
O
O
group); 1655, 1490 (C=O in
group).
C NH
O
C
log K with the Hammett
constants.
.
.
The increased charge on the amide oxygen atom
The sodium salts of phenylpenicillins substituted in
the benzene ring were prepared similarly.
of the side chain facilitates closure of the oxazolone
ring with cleavage of the -lactam ring; therefore,
electron-donor substituents in the benzene ring notice-
Phenylpenicillin methyl ester. A solution of
1.3 ml of diazomethane in 10 ml of diethyl ether was
added with stirring to a solution of 0.5 g of phenyl-
penicillic acid in 50 ml of diethyl ether. The reaction
was performed until the gas evolution ceased. Phen-
ylpenicillin methyl ester was isolated by vacuum
evaporation of the ether and purified by repeated re-
precipitation from ethyl acetate into petroleum ether.
Yield 0.208 g (40%); mp 116 117 C. IR spectrum, ,
ably accelerate the reaction. The negative sign of
,
suggesting the promoting effect of electron donors, is
consistent with this mechanism.
EXPERIMENTAL
The 1H NMR spectra were recorded at room
temperature, relative to internal TMS, on a JEOL
INM-C-60 spectrometer operating at 60 MHz. Samples
were taken as 1 2 mol % solutions in CDCl3. The IR
spectra were recorded on a UR-10 spectrometer (KBr
pellets). The specific optical rotation DT of substituted
phenylpenicillins and their methyl esters in solutions
was determined with an EPL-1 polarimeter. The iodo-
metric activity of phenylpenicillins and their methyl
esters was studied as described in [9]. The TLC study
of methyl esters of phenylpenicillins was performed
as follows: solutions of the compounds in chloroform
1
cm : 1770 (C=O in -lactam ring); 1655, 1520 (C=O
in
group); 1720 (C=O in
group);
C NH
C O
O
O
1170 (
).
O C
Found, %: N 8.34, S 9.60.
C16H18O4N2.
Calculated, %: N 8.38, S 9.59.
RUSSIAN JOURNAL OF APPLIED CHEMISTRY Vol. 75 No. 2 2002

REASONS FOR STRUCTURE ACID RESISTANCE CORRELATION
285
1
Iodometric activity 960 g mg . Methyl esters of
the other phenylpenicillins were prepared similarly.
the discussion of its results and to L.N. Samoilova for
determining the antimicrobial activity of phenylpen-
icillin sodium salts.
Dioazomethane was prepared by Arndt’s procedure
[10].
REFERENCES
CONCLUSIONS
1. Strukov, I.T., Antibiotiki, 1963, no. 11, pp. 963 976.
2. Maslova, G.A. and Strukov, I.T., Antibiotiki, 1965,
no. 11, pp. 1005 1010.
3. Doyle, F.P., Nayler, I.H.C., Smith, H., and Stove, E.R.,
(1) Phenylpenicillins substituted in the benzene
ring were prepared and identified in the form of meth-
yl esters.
Nature, 1961, vol. 191, pp. 1091 1092.
(2) The rate constants of acid hydrolysis of substi-
tuted phenylpenicillins correlate with the Hammett
constants of substituents in the benzene ring, with
4. Perron, Y.G., Minor, W.F., Holdrege, C.T., et al.,
J. Am. Chem. Soc., 1960, vol. 82, pp. 3934 3938.
5. Brandl, E. and Margreiter, H., Oster. Chem. Ztg.,
1954, vol. 55, pp. 11 21.
6. Kondrat’eva, A.P. and Bruns, B.P., Med. Prom st.
SSSR, 1957, no. 12, pp. 30 34.
high (in absolute value) sensitivity coefficients
:
2.05, 1.76, and 1.65 for 20, 25, and 30 C, respec-
1
tively. Contrasingly, the H NMR spectra of the lac-
tam moiety are weakly sensitive to the substituents in
the benzene ring.
7. Inozemtseva, I.I., Kleiner, G.I., Panina, M.A., et al.,
Antibiotiki, 1966, no. 6, pp. 497 500.
(3) The reaction mechanism, consistent with these
facts, was suggested, in which the side acyl group
takes an active part in the limiting stage.
8. Zhdanov, Yu.A. and Minkin, V.I., Korrelyatsionnyi
analiz v organicheskoi khimii (Correlation Analysis in
Organic Chemistry), Rostov: Rostov. Univ., 1966.
9. Alicino, I.F., Anal. Chem., 1961, vol. 33, pp. 648 649.
ACKNOWLEDGMENTS
10. Arndt, F., Organic Synthesis, Gilman, H., Adams, R.,
Clark, H.T., et al., Eds., New York: Wiley, 1946,
coll. vol. 2. Translated under the title Sintezy organi-
cheskikh preparatov, 1949, coll. 2, pp. 174 176.
The authors are sincerely grateful to E.F. Panarin
for initiation of this work and active participation in
RUSSIAN JOURNAL OF APPLIED CHEMISTRY Vol. 75 No. 2 2002