JOURNAL OF
`PHARMACEUTICAL
`SCIENCES @
`
`DECEMBER 1982
`VOLUME 71 NUMBER 12
`
`RESEARCH ARTlCL ES
`
`Effects of Surfactants on the Aqueous Stability and
`Solubility of P-Lactam Antibiotics
`
`AKIRA TSUJI *x, ETSUKO MIYAMOTO $, MUNEAKI MATSUDA *,
`KEIKO NISHIMURA *, and TSUKINAKA YAMANA 8
`Received June 8,1981, from the *Faculty of Pharmaceutical Sciences and +Hospital Pharmacy, Kanazauia University, Takara-machi,
`Kanazari'a 920, Japan, and the 'School of Pharmacy, Hokuriku University, Kanagawa-machi, Kanazaula 920-11, Japan.
`Accepted for
`publication February 4, 1982.
`
`Abstract 0 Studies were undertaken to elucidate the interaction be-
`tween (3-lactam antibiotics and surfactant micelles and to examine the
`effects of surfactants on their aqueous stability and solubility. The ap-
`parent binding constant of the micelle-antibiotic complex was deter-
`mined as a function of the solution pH at 37' and p = 0.15 by the dynamic
`dialysis method and hydrolysis study. In the interaction with nonionic
`and anionic micelles of polyoxyethylene-23-lauryl ether (I) and sodium
`lauryl sulfate (111, large differences were noted in the binding constants
`between the undissociated and ionized species of penicillins. However,
`the cat ionic surfactants, cetyltrimethylammonium bromide (111). showed
`no significant difference in the binding constants for both species. Acid
`degradation of penicillins was protected in micellar solutions of I and 111
`but was facilitated in micelles of 11. The surfactants exerted no influence
`011 the neutral degradation of the antibiotics used. The solubilization of
`penicillin V acid by micelles of I was studied at pH 2.0 and 35". The sol-
`ubility increased threefold in the presence of 10 mM I.
`Keyphrases Antibiotics, /3-lactam-effects
`of surfactants on the
`aqueous stability and solubility, interactions with surfactant micelles
`0 Surfactant micelles-effects on the aqueous stability and solubility
`of @-lactam antibiotics 0 Binding constant-interaction
`between
`8-lactam antibiotics and surfactant micelles, aqueous stability and sol-
`ubility
`
`The interaction of surface-active agents with drugs is
`of theoretical and practical importance, since such sur-
`factants represent one of the most important groups of
`adjuvants in pharmaceutical preparations. Surfactants
`incorporated in the drug dosage form are able to influence
`the drug stability and dissolution as a result of drug-sur-
`factant micellar interactions.
`So far there have been only a few reports on the inter-
`action (1-3) between P-lactam antibiotics and surfactant
`micelles. Recently, a catalytic effect of cationic surfactants
`on the degradation of cephalexin at neutral pH by en-
`trapment of the antibiotic micelles was described. No ef-
`fects on the cephalexin stability, however, were observed
`in anionic micelles (2).
`
`The aims of the present study were to elucidate the en-
`trapment of penicillin and cephalosporin antibiotics into
`the micelle of various types of surfactants as a function of
`the solution pH, and to investigate the effects of the anti-
`biotic-micelle interaction on the stability and solubility
`of these antibiotics under a gastric pH environment. A
`preliminary report has already been published (3).
`
`EXPERIMENTAL
`Materials-Antibiotics-The
`following (3-lactam antibiotics were
`used as supplied: propicillin potassium' (993 pg/mg), penicillin V po-
`tassium2 (1490 U/mg), and cefazolin sodium:l(966 Fg/mg). Free acid of
`penicillin V was obtained from a commercial source4.
`ether (I), sodium lauryl
`Surfactants-Polyoxyethylene-23-lauryl
`sulfate (II), and cetyltrimethylammonium bromide (111) were obtained
`from commercial sources and used without further purification except
`11. Compound I1 was recrystallized according to the literature (4).
`Chemicals-All other chemicals employed were of reagent grade and
`used without further purification except imidazole. Imidazole was re-
`crystallized from benzene followed by a thorough washing with ether.
`and penicillin V were determined
`Analytical Procedures-Propicillin
`by the spectrophotometric method developed previously (5). No influence
`of surfactants in this assay was observed. The concentration of the an-
`tibiotic in the samples was calculated from a calibration curve prepared
`daily. Cefazolin was analyzed in the stability experiment by reversed-
`phase high-performance liquid chromatography (HPLC). The liquid
`chromatograph5 was equipped with a UV detector6 set at 254 nm. The
`stationary phase was octadecylsilane chemically bonded on totally porous
`silica gel, prepacked into a 125-mm stainless steel column7 (4.6-mm i.d.).
`The mobile phase was 10% (v/v) acetonitrile-0.01 M ammonium acetate.
`The instrument was operated at ambient temperature and at a flow rate
`
`~~
`
`Takeda Chemical Industries, Osaka, Japan.
`* Banyu Pharmaceutical Co., Osaka, Japan.
`Fujisawa Pharmaceutical Co., Osaka, Japan.
`Sigma Chemical Co., St. Louis, Ma.
`Model FLC-A700, Japan Spectroscopic Co., Tokyo, Japan.
`Model UVIDEC-100, Japan Spectroscopic Co., Tokyo, Japan.
`SC-01, Japan Spectroscopic Co., Tokyo, Japan.
`
`0022-3549/82/1200- 13 13$01.00/0
`@ 1982, American Pharmaceutical Association
`
`Journal of Pharmaceutical Sciences I 1313
`Vol. 71, No. 72, December 1982
`
`Opiant Exhibit 2309
`Nalox-1 Pharmaceuticals, LLC v. Opiant Pharmaceuticals, Inc.
`IPR2019-00685, IPR2019-00688, IPR2019-00694
`Page 1
`
`

`

`0.21
`
`I
`2
`
`4
`
`I
`8
`
`10
`
`6
`HOURS
`Figure 2-First-order plots for the dialysis of propicillin in the presence
`of surfactants of various pH ualues at 37" and = 0.15. Key: 1 (control,
`M II, pH 6.50); 3 (2 X
`M I , pH 6 SO); 4 (3 X
`pH 6.50), 2 (4 X
`10-2MII,pH4.00);.5(2.4X 10-3MIII,pH4.00);6(4 X 1 0 - 2 M I , p H
`M I , pH 3.00); 8 (6.9 X
`4.00); 7 (1.5 X
`M III, pH 6.50).
`
`Figure I-Apparatus
`for the dynamic dialysis experiment. Key: (A)
`control motor; (B) sampling hole; (C) thermostated water bath; (0)
`jacketed beaker; (E) dialysis membrane; and (F) stirring shaft.
`
`of 1.0 ml/min, and then samples were injected through a 100-pl injectorR.
`The peak heights were used for quantification.
`Procedure-Dynamic Dialysis-The method employed was essen-
`tially the same as that described previously (6). The apparatus used in
`this study is illustrated in Fig. 1. The system consisted of a jacketed
`beaker (500 ml) set in a thermostated water bath. Three hundred milli-
`liters of buffer solution (ionic strength 0.15) was placed in the beaker. A
`cellulose tubeg was knotted at one end to form a bag (length 10 cm) and
`attached with a rubber band to the glass tubing with a stirring shaft. Eight
`milliliters of antibiotic buffer solution with or without surfactants was
`placed into the bag. The bag attached with a stopper was fitted on the
`beaker. Both the inner and outer solutions were stirred. All experiments
`were carried out at 37 f 0.1" and at various pHs with phosphate, acetate,
`and citrate buffer systems maintained at an ionic strength of 0.15. At
`appropriate time intervals, aliquots (10 ml) of the outer solution were
`withdrawn and 10 ml of drug-free buffer solution preheated at 37" was
`added. The samples were analyzed by the spectrophotometric method
`or HPLC described in the previous section. The concentration of the
`outer solution samples was corrected as follows:
`5 (cl1)i-l
`+ (vs/v11)
`( c I I ) n =
`,=1
`
`where (CII), and (CII):~~ represent the true and observed concentrations
`of t.he nth sample from the outer solution, respectively, and V,$ and V11
`represent the sampling volume and volume of the outer solution, re-
`spectively.
`Degradation Kinetics-Unless otherwise stated, kinetic studies were
`carried out at 37 f 0.1' and an ionic strength of 0.15. Each antibiotic was
`dissolved in hydrochloric acid-potassium chloride aqueous solution with
`or without surfactant to give a final antibiotic concentration of 6 X 10-4
`M . A saturated solution of the antibiotic was sometimes used because
`of limited solubility. At appropriate time intervals, aliquots were with-
`drawn, cooled, and analyzed. The pseudo first-order rate constants, kdeg,
`were calculated by least-squares analysis of the slopes of plots between
`the logarithm of the antibiotic concentration and time.
`Solubility Measurement-An excess of penicillin V (acid form) was
`added to the hydrochloric acid-potassium chloride solution (pH 2.0 and
`ionic strength 0.5) in a glass-stoppered flask. The flask was placed in a
`thermostated water bath at 35 f 0.1" and shaken mechanically until the
`antibiotic concentration in the solution showed an equilibrium value. A
`sample was taken through a 0.45-pm membrane filterlo and, if necessary,
`assayed after appropriate dilution with distilled water. The pH of the
`sample solution was measured" before use and at the end of the experi-
`ment; no significant change was observed.
`Determination of Critical Micelle Concentration-The determination
`of the critical micelle concentration (CMC) was accomplished by deter-
`
`(Eq. 1)
`
`8 Model LP1-350, Japan Spectroscopic Co., Tokyo, Japan.
`9 Visking dialysis membrane, Union Carbide Corp., Chicago, 111.
`lo Sartorius-memhranfilter, GmbH, 34 Gottingen, West Germany.
`l1 PHM26 pH-meter, Radiometer, Copenhagen, Denmark.
`
`1314 I Journal of Pharmaceutical Sciences
`Vol. 71, No. 12, December 1982
`
`mining the concentration at which the break in the log concentration
`uersus surface tension plot occurs. The surface tension of surfactant so-
`lutions of ionic strength of 0.15 containing various concentrations of I,
`11, or I11 and the antibiotic at the concentration used for dialysis and
`degradation studies was determined at 37" by a Du Nouy tensiom-
`eterI2.
`
`RESULTS
`Kinetics of Dynamic Dialysis-For quantification of the interaction
`between a drug molecule and surfactant micelles, various methods are
`available such as equilibrium dialysis (7), dynamic dialysis (8), micellar
`solubilization (9), the potentiometric titration method (lo), molecular
`sieve (ll), and micellar catalysis kinetics in the drug degradation (9).
`Among these, the dynamic dialysis method provides quick information
`for the existence of the interaction with macromolecules by utilization
`of marked difference of the permeation rate through a dialysis mem-
`brane.
`According to Fick's first law of diffusion, the rate of drug dialysis can
`be expressed by:
`
`(Eq. 2)
`where (CI)T and (CII)T represent the total concentration of the drug in
`the inner and outer solutions of the dialysis bag, respectively. Since the
`present experiments were carried out under the sink condition, (CI)T -
`(CII)T was assumed to be equal to ( C ~ ) T . The apparent first-order dialysis
`rate constant, kdia, therefore, was calculated from:
`
`where Co represents the initial concentration of the drug solution in the
`dialysis bag. The value for ( C I ) T was calculated from the mass balance
`equation as follows:
`
`(Eq. 4)
`where VI and VII represent the volume of the inner and outer solutions,
`respectively. Figure 2 shows typical semilogarithmic plots of the molar
`fraction of propicillin remaining in various surfactant solutions in the
`bag uersus time, and it indicates that the dialysis rates follow first-order
`kinetics in conformity with Eq. 3. Plots of the kdia for propicillin uersus
`the concentration of surfactants are given in Fig. 3 and show a marked
`decrease of kdie with increase in the surfactant concentration and a ten-
`dency to reach constant rate constants. The results apparently indicate
`the occurrence of entrapment of propicillin in the micelles, which are
`difficult to be dialyzed.
`If the drug is incorporated into the micelle, there then will be an
`equilibrium between the drug in solution and that in the micelle. The
`apparent binding constant, Kap,,. can be expressed as:
`
`l2 Du Nouy tensiometer, Shimadzu, Co., Kyoto, Japan.
`
`Opiant Exhibit 2309
`Nalox-1 Pharmaceuticals, LLC v. Opiant Pharmaceuticals, Inc.
`IPR2019-00685, IPR2019-00688, IPR2019-00694
`Page 2
`
`

`

`30
`50
`40
`20
`10
`CONCENTRATION OF SURFACTANT, l o 3 M
`
`Figure 3-Plots of the pseudo first-order rate constant, kdia, versus total
`surfactant concentration for the dialysis of propicillin at 37" and p =
`0.15. Key: (0) I (pH 3.50); (A) II (pH 4.40); (0) I l l (pH 6.50). The points
`are experimental values. The solid curues were generated from Eq. 6
`using the parameters in Table II.
`
`where (Cr), and (CI), represent the concentration of the drug free and
`bound with the surfactant micelle, respectively, and C,g is the total
`concentration of surfactant. When it is assumed that only free drug can
`permeate through the dialysis membrane, Eq. 6 is obtained from Eqs.
`3 and 5:
`
`1
`1 + Kapp (Cn - CMC)
`kdta = ko
`where ko represents the first-order dialysis rate constant of the drug in
`the absence of surfactant. Rearrangement of Eq. 6 gives:
`
`(Eq. 6)
`
`(Eq. 7)
`
`Equation 7 predicts that plots of (kolkdi, - 1) uersus (CO - CMC) passing
`through the origin are linear. The values of CMC used for the calculation
`are 0.092 mM, 0.46 mM, and 0.32 mM for I, 11, and 111, respectively, which
`were determined in this laboratory in the presence of the ant.ibiotic at
`37" and p = 0.15. As illustrated in Fig. 4, the results obtained for propi-
`cillin with various surfactants revealed a linear relationship in accordance
`with Ey. 7. The apparent binding constant, Kspp, was calculated from
`the slopes and the values are listed in Table I. During the periods of the
`dialysis experiments, there was negligible degradat,ion of propicillin.
`pH-Dependency of t h e Apparent Binding Constant in Micelle-
`Antibiotic Interactions-Penicillins have a pKa value of 2.7-2.9 (12)
`and exist in aqueous solutions in undissociated and ionized forms. The
`respective forms may yield different binding behavior in surfactant mi-
`cellar solutions. Figure 5 shows the pH-dependency of Kapp for propicillin
`in solutions of I, 11, and 111 as determined by the dynamic dialysis method.
`Some of the data were those determined in a stability kinetic study, which
`will be described. In the solutions of I and 11, the values of K,,,
`for pro-
`picillin decreased markedly as the pH increased approaching a constant
`value. It is supposed that a considerable difference exists in the micellar
`interactions between undissociat,ed species of propicillin and its ionized
`form. The relationship between the hydrogen ion activity of the bulk
`solution and the apparent binding constant can be represented by (see
`Appendix):
`
`Ka
`+KA-
`O H
`Kdpp = K H A ___
`a ~ + K a
`a H + K a
`where K H A and K A are the binding constants for the undissociated form
`of propicillin and its ionized form, respectively. Incorporation at pKa 2.76
`of propicillin gave parameters of K H A = 489.0 f 27.0 M-' and K A = 40.4
`f 2.7 M-' for I as the best fit to the data using a NONLIN computer
`
`(Eq. 8)
`
`30
`20
`1o3ccD - CMC), M
`Figure 4-Plots according to Ey. 7for the dialysis of cefazolin (Q) and
`propicillin (other symbols) in the presence of I (cirles), I! (triangles),
`and 111 (squares) at various pHs, 37'. and p = 0.15, Key: (0) pH 6.50;
`(m) pH 4.00; (0) pH 3.00; (e) pH 3.50; ( 8 ) pH 4.00; ( 0 ) pH 5.00; ( 0 )
`pH 6.50; (A) pH 6.50; (A) p H 4.00; (A) pH 4.40; (a) pH 3.40 and pH
`6.50.
`
`4%
`
`program (13). The curves in Fig. 5 were generated for I and I1 from Eq.
`8 by the use of these parameters as listed in Table 11.
`In the interaction of propicillin and the micelles of 111, the apparent
`binding constant was virtually independent of the bulk solution pH. This
`indicates no significant difference in the magnitude of KHA and K A , and
`analysis of the data gave K H ~ = K A = 810.1 M-' as the mean of experi-
`mental data at all pH values.
`A similar experiment was also carried out for the interaction between
`cefazolin [pKa = 2.54 (14)) and the micelle of I. The values of Kapp were
`extremely low, being -3 M-' in the wide pH range of 3-7, indicating
`negligible entrapment of the undissociated and ionized cefazolin into the
`micelles of I (Fig. 4).
`Effect of Surfactants on the Stability of the Antibiotics-The
`acid-catalyzed degradation of fl-lactam antibiotics was examined in the
`surfactant solutions of I, 11, and 111 at 37 f 0.1" and an ionic strength of
`0.15. The degradation followed first-order kinetics with regard to the
`antibiotic concentration in all surfactant solutions. Typical results for
`propicillin obtained by linear semilogarithmic plots of the residual molar
`fraction of antibiotic uersus time are shown in Fig. 6.
`As illustrated in Fig. 7, the pseudo first-order rate constant for the
`degradation of propicillin at acidic pH was increased significantly by the
`
`Table I-Apparent Binding Constant, Kapp, between Propicillin
`and Various Surfactants *
`I
`
`111
`
`..
`
`I1
`
`~~
`
`1.10
`4.00
`6.50
`
`856.0h
`793.8
`780.4
`
`175.0b
`1.61
`459.36
`1.10
`119.0h
`2.60
`404.9*
`1.92
`91.3b
`3.00
`247.2
`3.00
`28.5
`3.50
`115.1
`3.50
`20.7
`4.00
`59.5
`4.00
`13.2
`5.00
`42.5
`5.00
`4.0
`6.50
`42.0
`6.50
`0 The values were calculated from the experimental data according to Eq. 7 by
`These values were determined
`the least-squares treatment at 37' and p = 0.15
`in the stabilitv study
`
`~~
`
`~
`
`Journal of Pharmaceutical Sciences 1 1315
`Voi. 71, No. 12, December 1982
`
`Opiant Exhibit 2309
`Nalox-1 Pharmaceuticals, LLC v. Opiant Pharmaceuticals, Inc.
`IPR2019-00685, IPR2019-00688, IPR2019-00694
`Page 3
`
`

`

`I
`4
`
`I
`5
`
`I
`6
`
`1
`
`I
`2
`
`I
`3
`HOURS
`Figure 6-First-order plots for the degradation of propicillin in the
`presence of surfactants at uarious pHs, 37", and p = 0.15. Key: 1 (3 X
`M I I , pH 1.61);2 (control, pH 1.10);3 (4.5 X
`M II, pH 2.50);
`3 ( 1 X 1 0 - 2 M I , p H 1 . 1 0 ) ; 5 ( 8 . 9 X 10-2MIII,pH1.10);6(3.1 X l O - '
`M I , pH 1.92).
`M III, pH 1.10); 7 (1 X
`
`Within an experimental period of <1 day, there was no influence of
`the surfactants on the neutral degradation of the antibiotic used in this
`study.
`Effect of Surfactants on the Antibiotic Solubility-The saturable
`solubility of penicillin V, C,, at pH 2.0 and 35' increased with concen-
`tration of I as shown in Table 111. The data indicated that the aqueous
`solubility of penicillin V increased threefold in the presence of 10 mM
`I at pH 2.0 and 35', showing that penicillins are solubilized by surfactant
`micelles.
`
`DISCUSSION
`Solutions of penicillin G are highly unstable at a gastric pH, the half-
`lives being 1 min at pH 1 and 7 min at pH 2 (15,16). Such chemical in-
`activation of penicillin G in the gastric fluid has been reported to be re-
`sponsible for the poor bioavailability of this antibiotic. The acid degra-
`dation rates of penicillin derivatives are known to depend on their 6-
`sidechain nature due to the rearrangement initiated by the attack of the
`sidechain amidocarbonyl on the P-lactam to produce the corresponding
`penicillenic acid and penillic acid (16).
`Considerable efforts have been made to stabilize acid-labile penicillins.
`Previous authors (17) succeeded in stabilizing potassium salts of penicillin
`G and penicillin V in simulated gastric juice by coating with cholesteryl
`acetate, yielding 1.6- and twofold higher urine levels, respectively, after
`oral administration of these pharmaceutical preparations to humans.
`The previous (3) and present studies revealed a marked stability of
`penicillins in acid solutions with both cationic and nonionic micelles. Four
`kinds of derivatives, penicillin G, penicillin V, phenethicillin, and pro-
`picillin, can be stabilized to maximal extents of 6-, lo-, 8-, and 10-fold
`by the micelles of 111 and of 4-, 6-, 7-, and 13-fold by the micelles of I,
`respectively (3). These stabilization effects are attributed to incorporation
`of the penicillin molecules into both types of micelles. As is apparent from
`these results (3), the apparent binding constant between the penicillins
`and micelles increased with increasing lipophilic character of the peni-
`cillins, as expressed in terms of their octanol-water partition coefficients,
`P (12). This suggests that hydrophobic binding is involved in the inter-
`action between the cationic or nonionic micelle and undissociated species
`of the penicillins. These strong interactions resulted in protection of the
`p-lactam ring sterically and/or electrostatically from intramolecular and
`nucleophilic attack of the sidechain amidocarbonyl oxygen. However,
`it is probable that due to the localized hydrogen ion activity surrounding
`the negatively charged micelle, the anionic micellar state by I1 leads to
`an increase in the rate of degradation. The maximal acceleration of the
`P-lactam cleavage of propicillin by micelles of I1 was predicted to be
`60-fold at pH 1.6 and 37'.
`In contrast to penicillins, the acid degradation of cefazolin, a relatively
`acid-unstable cephalosporin (14), was not influenced by the presence of
`any type of surfactant, like cephalothin described previously (3). This
`was due to the fact that cefazolin was not sufficiently bound to the mi-
`celles, the apparent binding constant being confirmed as almost negligible
`by the dynamic dialysis method (Fig. 4). The very weak interaction of
`
`1
`
`0
`
`1
`
`2
`
`4
`
`5
`
`6
`
`7
`
`
`
`3
`PH
`Figure 5-Plots of the apparent binding constant, K,,,,
`of propicillin
`versus t h e bulk solution pH at 37' and p = 0.15. Key: (0,e) I; (A,&
`I I ; (o,.) I l l ; (open symbols, dynamic dialysis; closed symbols, sta-
`bility).
`
`addition of anionic surfactant (11); whereas, it decreased on increasing
`the concentration of both nonionic and cationic surfactants (I and 111).
`For other penicillins, similar results have been reported (3). In all cases,
`the rate constants first increased or decreased rapidly and then ap-
`proached a constant value above the CMC of the surfactants, suggesting
`the formation of penicillin-micelle complexes.
`According to the literature (9), the apparent first-order degradation
`rate constant thus, is expressed by:
`ko t k,(Cn - CMC)
`1 + Kapp (Cn - CMC)
`kdey =
`Rearrangement of Eq. 9 gives:
`
`(Eq. 9)
`
`Equation 10 predicts that plots of l / ( k o - kdeg) uersus 1 / ( c ~ - CMC)
`should give a straight line from which it should be possible to obtain k ,
`and Kapp values.
`Plots of Eq. 10 for the degradation of propicillin in the presence of I
`and 111 are shown in Figs. 8 and 9, respectively. The values of Kapp for the
`various reaction systems are given in Table I.
`
`Table 11-Binding Constants for Undissociated and Ionized
`Prooicillin with Various Surfactants a
`K H A . M-'
`Surfactant
`I
`489.0 f 27.0
`171.8 f 29.9
`I1
`810.1 f 40.3b
`111
`(I The binding constants were calculated by nonlinear regression program,
`NONLIN, at 37' and g = 0.15. This value is the mean f SD of the experimental
`data.
`
`K a , M-'
`40.4 f 2.7
`4.5 f 1.3
`810.3 f 40.3b
`
`~~~~
`
`1316 I Journal of Pharmaceutical Sciences
`Vol. 71, No. 72, December 7982
`
`Opiant Exhibit 2309
`Nalox-1 Pharmaceuticals, LLC v. Opiant Pharmaceuticals, Inc.
`IPR2019-00685, IPR2019-00688, IPR2019-00694
`Page 4
`
`

`

`3r
`
`1.75
`
`L5
`
`1.25
`
`I
`I
`1
`50
`40
`30
`2 0
`10
`CONCENTRATION OF SURFACTANT, lo3 M
`
`Figure 7-Plots of the pseudo first-order rate constant versus total
`concentration of surfactant for the degradation of propicillin at 37" and
`p = 0.15. Key: (a) I (pH 1.10, left scale); (A) II (pH 3.00, right scale);
`(H) Ill IpH 1.10, left scale). The points are experimental ualues. The
`solid curiies were generated from Ey. 9.
`
`the cefazolin molecule with surfactant micelles undoubtedly is due to the
`low lipid solubility of the antibiotic itself.
`The partitioning behavior of p-lactam antibiotics was investigated (12),
`both in n-octanol-water and isobutyl alcohol-water systems, as a function
`of the aqueous phase solution pH and showed that both the undissociated
`and ionized species could be partitioned into the oil phase, although the
`
`200
`I
`
`400
`1
`
`600
`I
`
`800
`I
`
`1000
`
`1.
`
`0
`
`L z -
`
`m
`
`; . P O
`
`r
`0
`
`0 c - .
`
`&
`
`, pH 1.92
`
`gH
`I11 , pH 1.10
`
`0
`
`reciprocal plots according to Eq. 10 for the degra-
`Figure 8-Double
`dation of propicillin in the presence of I and Ill at 37" and p = 0.15. Key:
`upper and right scales for pH 1.10; loioer and left scalesfor p H 1.92.
`
`15 -
`
`PH 3.00
`
`-
`
`10
`L r
`0 c
`I
`m
`
`U c - . -
`
`5 -
`
`pH 1.61
`i
`
`8 8
`
`4
`6
`4
`2
`6
`2
`1/(cD - CMC), 10-'M-'
`reciprocal plots according to Eq. I0 for the degra-
`Figure 9-Double
`dation of propicillin in the presence o/lI at 37O and p = 0.15.
`
`former was far more lipophilic than the latter. As shown in Fig. 5, the pH
`dependency of the Kapp values in the propicillin interaction with I was
`parallel to that of the apparent partition coefficients in oil-water systems.
`The interactions of propicillin with the micelles of I1 and I11 showed a
`marked contrast in the dependencies of their respective Kal,,, values on
`the pH of the bulk aqueous solution. In the case of the micelle of 11, Kapl,
`clearly decreased as the pH increased, while in the micelle of 111, K,,,
`exhibited independency in the wide pH range betwen 1 and 6. These re-
`sults can be explained on the basis of the participation of electrostatic
`forces between the ionized species of the antibiotic and the ionized sur-
`factant micelles in addition to the hydrophobic contribution.
`Electrical repulsive forces may play a significant, role between ionized
`propicillin and the anionized surfactant micelle to produce the deduced
`K.4 value, as seen from the KHAIKA ratio being -100. Unlike the anionic
`case, the attraction due to the electrostatic forces between the ionized
`form of propicillin and cationic surfactant micelle may lead to an effective
`cancellation of charge within the molecule, resulting in K H A = K.4.
`However, the magnitude of KA is dependent on the differences both 01
`the hydrogen ion activity and the acid dissociation behavior between
`aqueous and micellar phases (see Appendix).
`Based on the present study, it should be emphasized that the acid-
`labile penicillins were significantly stabilized and solubilized by incor-
`porating the unionized species into nonionic surfactant micelles, and the
`penicillin molecules entrapped in the micelle could t,hen be easily released
`at neutral pH values by reducing the force in the interaction between the
`
`Table 111-Solubility of Propicillin in t h e Presence of I a
`Concentration
`Solubility of
`of I,
`Propicillin,
`103 M
`l @ ' M
`0.72
`0.0
`1.21
`3.0
`1.55
`5.0
`7.0
`1.89
`10.0
`2.40
`20.0
`-1.09
`" Values at 37". pH 2.00, and p = 0.5.
`
`Journal of Pharmaceutical Sciences I 13 17
`Vol. 71, No. 12, December 1982
`
`Opiant Exhibit 2309
`Nalox-1 Pharmaceuticals, LLC v. Opiant Pharmaceuticals, Inc.
`IPR2019-00685, IPR2019-00688, IPR2019-00694
`Page 5
`
`

`

`ionized species and micelles. Cephalosporins, which have a much lower
`lipophilicity than penicillins (12), are not incorporated into nonionic
`surfactant micelles, so that there is no significant influence on the
`chemical stability and solubility in aqueous solution.
`
`corporation of the undissociated and ionized forms of drugs into the
`micelles can be calculated, without knowledge of a ~ , , and K,,,,
`from
`the dependence of the bulk aqueous solution pH upon the Kapp
`values.
`
`APPENDIX
`Alteration of the acid dissociation equilibrium of the drug in si:rfactant
`micelles would be in terms of local alteration in hydrogen ion concen-
`tration in the micelles. It is to be expected that the hydrogen ion activity
`(aH,,,,) will be greater near the surface of the anionic micelles and lower
`near the surface of the cationic micelles than that ( a H ) in the bulk
`aqueous phase.
`Assuming that both species of undissociated and ionized drugs which
`exist in the aqueous phase are incorporated into the micellar phase, the
`various equilibria can be described as:
`
`where H A and A refer to the undissociated and ionized species, respec-
`tively, and the subscripts aq and m refer to the aqueous and micellar
`phases, respectively. The other parameters are the same as described in
`the text. It is clear that the KHA/KA = (Ka/Ka,m) from Eqs. AlLA4. The
`apparent binding constant, Kapp, between drugs and micelles is given
`by:
`
`By use of the equilibrium constants defined above, Eq. 8 can be arrived
`at from Eq. A6. Therefore, the binding constants, KHA and K A , for in-
`
`REFERENCES
`(1) F. Alhaique, C. BotrC., G. Lionetti, M. Marchetti, and F. M. Ric-
`cieri, J. Pharm. Sci., 56, 1555 (1967).
`(2) M. Yasuhara, F. Sato, T. Kimura, S. Muranishi, and H. Sezaki,
`J. Pharm. Pharmacol., 29,638 (1977).
`(3) A. Tsuji, M. Matsuda, E. Miyamoto, and T. Yamana, ibid., 30,
`442 (1978).
`(4) J. L. Kurtz, J . Phys. Chem., 66,2239 (1962).
`(5) H. Bundgaard and K. h e r , J. Pharm. Pharmacol., 24, 790
`(1972).
`(6) M. C. Meyer and D. E. Guttman, J . Pharm. Sci., 57, 1627
`( 1968).
`(7) N. K. Patel and N. E. Foss, ibid., 54,1495 (1965).
`(8) A. Agren and R. Elofsson, Acta Pharm. Suec., 4,281 (1967).
`(9) J. H. Fendler and E. J. Fendler, “Catalysis in Micellar and Mac-
`romolecular Systems,” Academic, New York, N.Y., (1975).
`(10) M. Donbrow and C. T. Rhodes, J. Chem. Sci., 1964,6166.
`(11) M. Donbrow, E. Azaz, and R. Hamburger, J. Pharm. Sci., 59,1427
`(1970).
`(12) A. Tsuji, 0. Kubo, E. Miyamoto, and T. Yamana, ibid., 66,1675
`(1977).
`(13) C. M. Metzler, “NONLIN, A Computer Program for Parameter
`Estimation in Nonlinear Systems,” Technical Report 7292/69/7297/005,
`The Upjohn Co., Kalamazoo, Mich.
`(14) T. Yamana and A. Tsuji, J. Pharm. Sci., 65,1563 (1976).
`(15) T. Yamana, A. Tsuji, and Y. Mizukami, Chem. Pharm. Bull.
`(Tokyo), 22,1186 (1954).
`(16) M. A. Schwartz, J. Pharm. Sci., 58, 643 (1969).
`(17) S. P. Patel and C. I. Jarowski, ibid., 64,869 (1975).
`
`ACKNOWLEDGMENTS
`The present results were presented in part at the APhA Academy of
`Pharmaceutical Sciences, 29th National Meeting, San Antonio, Texas,
`November, 1980.
`The authors acknowledge the gifts of P-lactam antibiotics from Takeda
`Chemical Industries, Banyu Pharmaceutical Co., and Fujisawa Phar-
`maceutical Co. The computer analysis was performed on the digital
`computer, FACOM M-160 at the Data Processing Center, Kanazawa
`University.
`
`13 18 I Journal of Pharmaceutical Sciences
`Vol. 71, No. 12, December 1982
`
`Opiant Exhibit 2309
`Nalox-1 Pharmaceuticals, LLC v. Opiant Pharmaceuticals, Inc.
`IPR2019-00685, IPR2019-00688, IPR2019-00694
`Page 6
`
`