International Journal of Pharmaceutics 489 (2015) 186–194
Contents lists available at ScienceDirect
International Journal of Pharmaceutics journal homepage: www.elsevier.com/locate/ijpharm
Pharmaceutical nanotechnology
Relationships between human intestinal absorption and polar interactions drug/phospholipids estimated by IAM–HPLC. Lucia Grumetto, Giacomo Russo, Francesco Barbato * Dipartimento di Farmacia, Università degli Studi di Napoli Federico II, Via D. Montesano, 49, I-80131 Naples, Italy
A R T I C L E I N F O
A B S T R A C T
Article history: Received 5 March 2015 Received in revised form 16 April 2015 Accepted 22 April 2015 Available online 23 April 2015
Phospholipid affinity indexes ðlogkW Þ for 15 structurally non-related basic, acidic, ampholytic, and neutral drugs were measured by HPLC on two different phospholipid stationary phases (immobilized artificial membrane – IAM). According to a method we previously proposed, polar and electrostatic forces
IAM
0
involved in drug/membrane interactions were quantified both as DlogkW and as D logkW . These values IAM
IAM
IAM logkW
Keywords: Intestinal absorption Oral bioavailability Phospholipids IAM–HPLC Membrane Lipophilicity
are the differences between the experimental and the values expected for a neutral compound having the lipophilicity value equal to either that of the neutral form of the analyte (log PN) or that of the mixture of charged and neutral forms of the analyte at jejunum pH 6.5 (log D6.5), respectively. Jejunum IAM
absorption values, log Peff, measured by the Loc-I-Gut technique, did not relate with logkW values. A moderate linear relationship was observed with log PN values for all the analytes and a weak parabolic relationship was observed with log D6.5 values, but only after the exclusion of two analytes. In contrast, a highly significant linear inverse relationship was observed with DlogkW values. Therefore, differently from the results of our recent studies on blood–brain barrier passage, the intestinal absorption data for IAM
0
not only bases and zwitterions but also for acids relate significantly with DlogkW and not with D logkW values. The results suggest that membrane passage at jejunum level can be described according to the “flipflop” model; indeed, the lipophilicity of the neutral forms (log PN) appears related to the passage through IAM
IAM
the non-polar inner moieties of phospholipids whereas DlogkW parameter appears related to the “trapping” forces at their polar surfaces. The method, easy to perform and at medium throughput, could be of use for preliminary screening of new drugs based on oral absorption potential. ã 2015 Elsevier B.V. All rights reserved. IAM
1. Introduction The extent of intestinal absorption of drugs in humans is usually determined by in vivo experiments. However, in the last years increasing ethical issues have been raised about in vivo pharmacological experiments on both animals and humans, making desirable the development of alternative in vitro methods. The methods based on the determination of physicochemical parameters, albeit ineffective when active transport mechanisms occur, are highly reproducible and at high/medium throughput; furthermore, they may be useful both to predict the oral bioavailability of drug candidates at the early stages of their development and to formulate mechanistic hypotheses useful for drug/prodrug design. Their use could avoid, or at least reduce, in vivo experiments, such as those
* Corresponding author. Tel.: +39 81 678627; fax: +39 81 678107. E-mail address: [email protected] (F. Barbato). http://dx.doi.org/10.1016/j.ijpharm.2015.04.062 0378-5173/ ã 2015 Elsevier B.V. All rights reserved.
based on the Loc-I-Gut method. The latter yields the effective intestinal permeability values, Peff, which, although actually accounting for the enterocyte apical membrane absorption, were demonstrated strictly related to the extent of human intestinal absorption (Lennernäs, 1997). Passive drug absorption of orally administered drugs is assumed to be related to drug lipophilicity (Liu et al., 2011), expressed as the logarithm of the n-octanol/water partition coefficient, log P, when referred to single species (neutral or ionized). Indeed, this parameter appears to describe adequately the partition coefficients of neutral compounds in the membrane, which, according to the Fick’s first law (as modified to take into account the existence of a barrier) is a driving force for the membrane passage. However, a large majority of drugs support at least one ionizable function (Comer and Tam, 2001) and their noctanol lipophilicity depends on their ionization degree. The values corrected for ionization, i.e., log D values, are the weighted average of log P values of the various forms, ionized and neutral,
L. Grumetto et al. / International Journal of Pharmaceutics 489 (2015) 186–194
existing in solution as a function of the pH value (Leo et al., 1971) and are always smaller than log P values of the related neutral forms. Membrane passage of ionizable compounds seems to be more realistically described by the so-called “flip-flop” model (Gurtovenko and Vattulainen, 2007; Krämer et al., 2009). According to this model, both neutral and ionized forms, in dynamic equilibrium, are involved in the passage of membrane phospholipid bilayer. Furthermore, comparisons between partition data in phospholipids and log D values demonstrated that the latter are often inadequate at describing the interactions actually occurring between ionizable analytes and membrane phospholipids (Amato et al., 2000; Barbato, 2006; Barbato et al., 1996, 1997a, b, 1998, 2004, 2005, 2007). A rapid and reproducible way to achieve partition data in phospholipids is the measurement of chromatographic retention factors, k, by HPLC (high performance liquid chromatography) on phosphatidylcholine-like stationary phases (so-called immobilized
187
artificial membrane – IAM). These values are a direct measure of phospholipid affinity (Barbato, 2006) and are usually reported as IAM
logkW , i.e., the logarithm of k values measured at, or extrapolated to, 100% aqueous mobile phase (Amato et al., 2000; Barbato, 2006; Barbato et al., 1997b, 2004, 2005; Taillardat-Bertschinger et al., 2003). In contrast with log D values, IAM–HPLC data indicate that the interactions between ionized forms and phospholipids can be even stronger than those of the corresponding neutral forms. This occurs because, differently from the interaction forces in noctanol/water partition systems, the recognition forces involved in the interactions with phospholipids include ionic bonds (Taillardat-Bertschinger et al., 2003; Vrakas et al., 2008). The latter can be estimated by interpolating lipophilicity values in n-octanol and phospholipid interaction data, according to a simple procedure we proposed in our recent works (Grumetto et al., 2012, 2013, 2014), 0
giving DlogkW and D logkW parameters. The soundness of these parameters was recently supported by studies indicating that data IAM
Scheme 1. Chemical structures of the compounds considered.
IAM
188
L. Grumetto et al. / International Journal of Pharmaceutics 489 (2015) 186–194
of penetration through the blood–brain barrier (BBB) inversely 0
related with either DlogkW (for bases) or D logkW (for acids) values (Grumetto et al., 2012, 2013, 2014), but did not with either IAM
IAM
IAM
log D or logkW values, in contrast with other authors (Ducarme ska et al., 2000; Péhourcq et al., 2004; Reichel et al., 1998; Kepczyn and Begley, 1998; Salminen et al., 1997). This suggests that, despite ionic bonds enhance partition in phospholipids, they act as “trapping” forces at the membrane level, hindering its permeation. On the basis of the results reported above (Grumetto et al., 2012, 2013, 2014), we decided to investigate about possible relationships between these parameters and intestinal permeation data. In fact, it was suggested that “pure passive membrane diffusion is universal for membranes with different physiological functions and physicochemical properties” (Lennernäs, 1997). In this work we considered fifteen structurally unrelated molecules, usually orally administered and supposed to be absorbed by mainly passive mechanism at the intestinal level. Indeed, an involvement of active transport mechanism was reported for the intestinal absorption of cefalexin, a beta lactam antibiotic, but its contribution to the total absorption would play an only minor role (Bretschneider et al., 1999). The set consisted of six bases (cimetidine, desipramine, propranolol, ranitidine, terbutaline and verapamil), five acids (fluvastatin, hydrochlorothiazide, isotretinoin, ketoprofen and naproxen), three zwitterions (amoxicillin, cefalexin and piroxicam), and one neutral compound (carbamazepine) (Scheme 1). Their phospholipid affinity data were experimentally measured by IAM–HPLC technique on two different phospholipid stationary
phases (IAM.PC.MG and IAM.PC.DD2) to reasonably exclude that the data were affected by secondary retention mechanisms. Comparisons between the scales of n-octanol lipophilicity and IAM data were preliminarily performed to highlight similarities and dissimilarities. Finally, possible relationships between jejunal absorption data and either n-octanol lipophilicity data or IAM data were investigated.
2. Materials and methods 2.1. Chromatographic conditions and equipment The analyses were performed according to a method we previously reported in the literature (Grumetto et al., 2012, 2013, 2014). The analytical HPLC columns were a IAM.PC.MG (4.6 mm 150 mm; Regis Chemical Company, Morton Grove, IL) and a IAM.PC. DD2 (4.6 mm 100 mm; Regis Chemical Company, Morton Grove, IL). Only one IAM.PC.MG and only one IAM.PC.DD2 columns were used throughout the present study. Chromatographic retention data are reported as log k (the logarithm of the retention factor), calculated by the expression: k = (tr – t0)/t0, where tr and t0 are the retention times of the drug and a non-retained compound (acetone), respectively. Direct measurements of k values in fully IAM
aqueous mobile phases ðkW Þ, i.e., 0.1 M phosphate buffer at pH 7.0, were only possible for the compounds eluting within 20 min,
Table 1 pKa values, ionization degrees at pH 6.5, and logarithms of lipophilicity values in n-octanol and of chromatographic retention factors on IAM phases for the compounds considered. Compound
pKa
[Ionized]/[unionized] ratio at pH 6.5
log PN
log D6.5
logkW
Amoxicillin Hydrochlorothiazide Ranitidine Cimetidine Cefalexin Terbutaline Carbamazepine Piroxicam Ketoprofen Propranolol Naproxen Verapamil Fluvastatin Isotretinoin Desipramine
2.44o/7.14o 7.90b 8.20d 6.80e 3.12o /6.84o 10.10f – 5.46i /1.86i 4.45b 9.42b 4.15b 8.92b 4.30n 4.76o 10.40b
p
1.22a 0.03c 0.27b 0.40b 0.65b 0.90b 2.19g 3.00j 3.12b 3.28c 3.34m 3.79b 4.17c 4.20b 4.90b
1.70a 0.05 1.44 0.08 1.00c,q 2.70 2.19g 1.20j,r 1.07 0.36l,s 0.99 1.37 1.97 2.45 1.38c,q
-0.920 0.540 1.130 1.030 0.220 0.662 1.039h 1.850k 1.120k 1.821e 1.260k 2.049 2.210 2.807 2.826h
a
0.04 50 2 p
3981 p p
112 832 224 263 158 55 7943
Winiwarter et al. (1998). Law et al. (2014). c Avdeef (2012). d Khan et al. (2007). e Grumetto et al. (2012). f Panigrahi et al. (2005). g Lombardo et al. (2000). h Grumetto et al. (2014). i Bernhard and Zimmermann (1984). j Tsai et al. (1993). k Barbato et al. (1996). l Barbato et al. (1990). m La Rotonda et al. (1983). n Van de Waterbeemd and Testa (2008). o pKa values calculated. p Not reported because either zwitterion or neutral compound. q log D7.4. r log D6.07. s log D6.7. b
IAM:MG
IAM:DD2
logkW
0.728 0.977 0.860 0.783 0.021 0.863 1.717h 1.767 1.360 2.480e 1.339 3.085 2.843 3.704 2.741h
L. Grumetto et al. / International Journal of Pharmaceutics 489 (2015) 186–194
189
2.2. Sample preparation
whereas, for the solutes requiring the addition of acetonitrile to the IAM
eluent, the kW values were calculated by an extrapolation method (Braumann et al., 1983). All reported log k values are the average of at least three measurements; for each log k value the 95% confidence interval associated with each value never exceeded 0.04. To avoid that the experimental measurements were affected by retention changes due to column aging, the retention times of five test compounds (amlodipine, p-nitroaniline, toluene, isradipine and ketoprofen) were checked weekly. No correction was done to the experimental retention values since no retention value of test compounds changed more than 4% during the study.
A
Each analyte was dissolved in the mobile phase or in methanol to ca. 104 M concentration. 2.3. Lipophilic and biological activity parameters The log PN values, i.e., partition coefficients n-octanol/aqueous phase of the neutral form of analytes, were from the literature (Avdeef, 2012; Law et al., 2014; La Rotonda et al., 1983; Lombardo et al., 2000; Tsai et al., 1993; Winiwarter et al., 1998).
3.5 3.0
desipramine
2.5
logk W IAM.MG
2.0
piroxicam
1.5 1.0 0.5 0.0
cefalexin
-0.5 amoxicillin
-1.0 -1.5 -2.0
-1.0
0.0
1.0
2.0
3.0
4.0
5.0
6.0
log PN bases
B
acids
neutral
4.0 3.5 3.0 desipramine
logk W IAM.DD2
2.5 2.0 piroxicam 1.5 1.0 0.5 cefalexin
0.0 -0.5
amoxicillin -1.0 -2.0
-1.0
0.0
1.0
2.0
3.0
4.0
5.0
6.0
log PN bases IAM:MG
IAM:DD2
Fig. 1. Relationships between either logkW (A) or logkW compounds (Taillardat-Bertschinger et al., 2002).
acids
neutral
(B) and log PN values for the 15 compounds considered in comparison to the plots of 36 neutral
190
L. Grumetto et al. / International Journal of Pharmaceutics 489 (2015) 186–194
The n-octanol/aqueous buffer at pH 6.5 partition coefficients (log D6.5) were calculated according to the following equations:
A
3.5 3.0
logD6:5 ¼ logPN logð1 þ 106:5pK a Þðfor acidsÞ
2.5
with the exception of (i) amoxicillin, whose log D6.5 value was taken from the literature (Winiwarter et al., 1998), (ii) cefalexin and desipramine, whose experimental log D7.4 (Avdeef, 2012) were assumed as a reasonable estimate of log D6.5 values, and (iii) piroxicam and propranolol, whose experimental log D6.07 and log D6.7, respectively, (Tsai et al., 1993; Barbato et al., 1990) were also assumed as a reasonable estimate of log D6.5 values. The pKa values were either calculated by the program ACD/labs (release 12.00) or taken from the literature (Bernhard and Zimmermann, 1984; Grumetto et al., 2012; Khan et al., 2007; Law et al., 2014; Panigrahi et al., 2005; Van de Waterbeemd and Testa, 2008). The log Peff values, measured at pH 6.5, were from the literature (Lennernäs, 2014).
2.0
log k W IAM.MG
logD6:5 ¼ logPN logð1 þ 10pK a 6:5 Þðfor basesÞ
1.5 1.0 0.5 0.0 -0.5 -1.0 -1.5 -2.0
-1.0
0.0
1.0
2.0 N
3.0
4.0
5.0
6.0
4.0
5.0
6.0
6.5
log P (log D ) bases
neutral
acids
B 4.0 3.5
2.4. Statistical analysis
3.0
3. Results and discussion
2.5
log k W IAM.DD2
Linear regression analysis was performed by a commercially available statistical package (Microsoft Excel 2003) for personal computer observing the requirements of significant regression analysis.
2.0 1.5 1.0 0.5 0.0
3.1. Selection of the physicochemical parameters Ionizable compounds show different lipophilicity values in noctanol at different pH of the aqueous phase (log D values), according to the abundance of the neutral and ionized forms in solution. The log D values theoretically calculated by the equations above reported in Section 2 do not take into account the contribution of the ionized forms to the partition. This implies that the theoretical values are close to the experimental ones only if the contribution of the ionized forms is negligible. Therefore, log D calculated at the pH values at which the fraction of the neutral form is 1% do not adequately reflect the actual lipophilicity values. For ten of the considered compounds that showed at pH 6.5 an appropriate ionized/neutral form ratio we took into account the calculated log D6.5 values. For the bases propranolol and desipramine we considered their experimentally determined log D6.7 and log D7.4, respectively. Since propranolol is already extensively ionized at pH 6.7 (ionized/neutral form ratio >500) and, even so more at pH 6.5, its log D6.7 is expected to be very close to its log D6.5 value. Analogously, log D7.4 value for desipramine, showing an ionized/neutral form ratio 1000 at pH 7.4, is expected to be very close to its log D6.5 value. The log D6.5 values for the zwitterionic ampholytes amoxicillin, cefalexin, and piroxicam cannot be theoretically calculated by the equations reported above. The value of log D6.5 for amoxicillin was taken from the literature (Winiwarter et al., 1998) whereas, to the best of our knowledge, no experimental log D6.5 value is reported in the literature for both cefalexin and piroxicam. Since zwitterions behave as “lipophilicity buffers” (Pagliara et al., 1997), we assumed the experimentally determined log D7.4 value for cefalexin (Avdeef, 2012) and the experimentally determined log D6.07 value for piroxicam (Tsai et al., 1993) as reasonable estimates of their log D6.5. Table 1 summarizes pKa values, the ratios between the concentrations of ionized and unionized forms at pH 6.5
-0.5 -1.0 -2.0
-1.0
0.0
1.0
2.0
3.0
log PN (log D6.5) bases
neutral
acids
IAM:MG
IAM:DD2
Fig. 2. Relationships between either logkW (A) or logkW (B) and the combination of log PN values for 10 compounds (bases, zwitterions and neutral) and log D6.5 values for five acids, in comparison to the plots of 36 neutral compounds.
(calculated by the Henderson–Hasselbalch equation), log PN, and log D6.5 values for the compounds considered, as well as their IAM:MG
IAM:DD2
IAM
and logkW values, i.e., logkW on IAM.PC.MG and logkW IAM.PC.DD2 stationary phases, respectively. As can be seen, the compounds considered span a very large range of log PN values (from 1.22 to 4.90). It is worth underlining that using eluents at pH 7.0, to maximize column stability and data reproducibility, does not have negative impact on the significance of the data as measures of membrane interactions occurring at slightly different pH values (e.g., pH 7.4 and pH 6.5, for the BBB passage and the jejunum absorption, respectively). Indeed, it was demonstrated that retention on IAM phases, even for ionizable compounds, is only negligibly affected by variations of the pH of the eluent within the range of 5.5–7.0 (Amato et al., 2000). IAM
Finally, logkW values were determined on two phospholipid stationary phases differing from each other in the end-capping of residual amino groups of the silica–propylamine core. IAM.PC.MG supports hydroxy groups (being end-capped by methyl glycolate) whereas IAM.PC.DD2 supports C10 and C3 alkyl chains (being endcapped by both decanoic and propionic anhydrides). Therefore, retention on IAM.PC.MG could be supposed to be affected by
L. Grumetto et al. / International Journal of Pharmaceutics 489 (2015) 186–194
191
Table 2 Values of the differences between observed and expected logarithms of retention factors on IAM.PC.MG and IAM.PC.DD2 stationary phases (DlogkW
IAM:MG
D
0
D
IAM:MG logkW ,
IAM:DD2 logkW
0
and D
IAM:DD2 logkW ,
and
respectively) and logarithms of the human effective jejunum permeability values for the compounds considered
(log Peff). Compound
DlogkIAM:MG W
IAM:MG D0 logkW
DlogkIAM:DD2 W
IAM:DD2 D0 logkW
log Peff
Amoxicillin Hydrochlorothiazide Ranitidine Cimetidine Cephalexin Terbutaline Carbamazepine Piroxicam Ketoprofen Propranolol Naproxen Verapamil Fluvastatin Isotretinoin Desipramine
0.778 1.296 1.648 1.445 0.003 0.681 0.038 0.206 0.619 0.045 0.653 0.221 0.361 0.216 0.319
1.158 1.309 3.001 1.823 1.304 3.532 0.038 1.632 1.008 2.268 1.209 1.697 1.384 1.600 2.469
1.294 1.888 1.491 1.292 0.297 0.905 0.555 0.152 0.671 0.299 0.898 0.428 0.169 0.664 0.953
1.743 1.904 3.087 1.737 1.838 4.268 0.555 1.529 1.247 3.027 1.299 2.690 1.889 2.297 2.335
4.50 5.06 4.57 4.58 3.81 4.52 3.37 3.18 3.06 3.54 3.07 3.17 3.62 4.00 3.35
secondary mechanisms (e.g., H bonds) involving hydroxy groups of methyl glycolate whereas retention on IAM.PC.DD2 could be supposed to be affected by secondary mechanisms, partition based, occurring at level of C10 and C3 alkyl chains. However, IAM:DD2
IAM:DD2
logkW
IAM:MG
¼ 1:106ð0:104ÞlogkW
þ 0:171ð0:170Þ
A -3.0
-3.5
(1)
n = 15 r2 = 0.896 s = 0.404 F1,13 = 112.23 F1,13 a,0.001 = 17.82 In this and the following equations, n is the number of data considered to derive the regression equation, r2 is the square of the correlation coefficient, s is the standard error of the estimate, F (the subscripts are the number of variables and the degrees of freedom) is the Fisher statistic of the regression, followed by the tabulated critical value for a = 0.001. The standard errors of the regression coefficients are reported in parentheses.
log Peff jejunum
IAM:MG
and logkW values were found strongly collinear logkW (Eq. (1)) and IAM data can be assumed as substantially reflecting the interactions between analytes and phospholipids, with secondary interaction mechanisms playing an only minor role.
does not exist as sole form at any pH of the medium (Barbato et al., 2007). The points of the less lipophilic compounds (log P < 1) are shifted upward with respect to the regression line. However, it
-4.0
-4.5
-5.0
-5.5 -2.0
-1.0
0.0
1.0
3.2. Relationships among the physicochemical parameters
IAM logkW
found that, differently from ionizable compounds, the values of structurally non-related neutral compounds relate unambiguously with n-octanol lipophilicity values in the log P range of 1.0–4.8. The relationships are expressed by Eqs. (2) and (3).
bases
¼ 0:792ð0:038ÞlogP 0:732ð0:105Þ
¼ 0:934ð0:038ÞlogP 0:883ð0:104Þ
6.0
acids
-3.0
-3.5
(3)
n = 36 r2 = 0.946 s = 0.246 F1,34 = 595.74 F1,34 a,0.001 = 12.90 IAM
5.0
(2)
n = 36 r2 = 0.926 s = 0.248 F1,34 = 422.40 F1,34 a,0.001 = 12.90 IAM:DD2 logkW
4.0
B
The plots logkW vs. log PN of the compounds considered in the present work, superimposed to the plots of the neutral compounds, are reported in Fig 1. It is interesting to note that not only, as expected, the point relative to carbamazepine, a neutral compound, but also the points relative to the zwitterions amoxicillin, cefalexin, and piroxicam are very close to the line of the neutral compounds; this confirms that phospholipid interaction of zwitterions is related to the n-octanol lipophilicity of their neutral form despite of the fact that the latter
log Peff jejunum
IAM:MG
3.0
log PN
In our previous work (Taillardat-Bertschinger et al., 2002) we
logkW
2.0
-4.0
-4.5
cimetidine
-5.0
-5.5 -3.0
hydrochlorothiazide
-2.0
-1.0
0.0
1.0
2.0
3.0
log D6.5 bases
acids
Fig. 3. Relationships between log Peff and either log PN values (A) or log D6.5 values (B) for the 15 compounds considered.
192
L. Grumetto et al. / International Journal of Pharmaceutics 489 (2015) 186–194
that DlogkW values parameterize the excess of polar, mainly electrostatic, intermolecular interaction forces encoded in phospholipid partition as compared to n-octanol partition. However, in our previous works (Grumetto et al., 2013, 2014) IAM
A -3.0
log P effjejunum
-3.5
we observed that BBB passage data related with DlogkW values for only basic compounds whereas for acids they related with IAM
-4.0
D0 logkIAM W values; the latter are the distances from the regression
-4.5 -5.0 -5.5 -0.5
0.0
0.5 ∆ logkW
1.0
IAM.MG
(∆'log kW
IAM.MG
1.5
2.0
)
line of neutral compounds calculated taking into account their log D7.4 values, i.e., the lipophilicity actually displayed at the physiological pH of the blood. At jejunum level the compounds considered are in solution at pH 6.5. As can be seen in Fig. 2, taking into account the log D6.5 values for the acidic compounds considered in the present work, the distances of the points from the regression line of neutral compounds noticeably increase. Table 2 summarizes DlogkW
IAM:MG
bases
acids
D
B
0
, DlogkW
IAM:DD2
0
, D logkW
IAM:MG
,
IAM:DD2 logkW ,
and log Peff values for the compounds considered. On both IAM phases, a moderate inverse linear relationship was
-3.5
found between DlogkW values and log PN values (r2 = 0.589, F1,13 = 18.65 and r2 = 0.602, F1,13 = 19.68 for IAM.PC.MG and IAM.PC. DD2, respectively).
-4.0
3.3. Relationships with intestinal absorption data
-4.5
Since inter-laboratory variability of biological data is generally too high for correlation studies, we only took into account the data, reported in the literature, from a single source (Lennernäs, 2014) to obtain a consistent scale of jejunal Peff values. Obviously, we could not take into account the compounds undergoing active transport mechanisms, e.g., L-dopa and amino acids. The log Peff values, experimentally measured at pH 6.5 at jejunum level (Lennernäs, 2014), moderately related linearly with log PN values (Fig. 3A and Eq. (4)).
IAM
log P effjejunum
-3.0
-5.0 -5.5 -1.5
-1.0
-0.5
0.0
0.5
1.0
1.5
2.0
2.5
∆ logkW IAM.DD2 (∆' logkW IAM.DD 2)
bases
acids
0 IAM:MG IAM:MG ð logkW Þ (A) or IAM:DD2 ð logkW Þ (B) values for the 15 compounds considered. 0 IAM:DD2 and logkW were considered for only acids.
Fig. 4. Relationships between log Peff and either DlogkW
D D
IAM:DD2 logkW 0 IAM:MG logkW
0
D
D
D
should be remembered that at so low lipophilicity values the IAM
linearity of the relationship logkW vs. log PN is no longer observed for neutral compounds, too (Taillardat-Bertschinger et al., 2002). The behavior of poorly lipophilic compounds suggests that polar interactions between analytes and phospholipids (including the electrostatic ones in the case of ionizable compounds) become predominant when lipophilicity falls down. The other points of basic compounds are close to the regression line of neutral compounds, with the exception of desipramine that, as can be seen IAM:DD2
vs. log PN, lies below the regression in Fig 1B, reporting logkW line. Desipramine is the most lipophilic compound in the set considered with a log PN value of 4.90. Its behavior confirms that strongly lipophilic bases interact with phospholipids weaker than isolipophilic neutral compounds, as already observed on IAM.PC. DD2 phase (Grumetto et al., 2012, 2014). The distance of the points from the regression line is expressed as DlogkW
IAM:MG
D
IAM:DD2 logkW ,
and
for IAM.PC.MG and IAM.PC.DD2 phases, respectively.
Therefore, DlogkW
IAM:MG
values are the differences between the IAM:MG
experimentally measured logkW calculated from log P
N
values and the values
by Eq. (2); DlogkW
IAM:DD2
values are the IAM:DD2
differences between the experimentally measured logkW values and the values calculated from log PN by Eq. (3). Differently from n-octanol, but analogously to membrane phospholipids, phosphatidylcholine moieties of IAM stationary phases are electrically charged species. Therefore, it is reasonable to assume
logPeff ¼ 0:273ð0:062ÞlogPN 4:427ð0:177Þ
(4)
n = 15 r = 0.598 s = 0.437 F1,13 = 19.32 F1,13 a,0.001 = 17.82 2
Parabolic relationships between log Peff values and log PN values were not statistically significant (at both a levels of 0.001 and 0.01) for both the whole set and a set reduced by the exclusion of acids. Since the permeability values were measured at pH 6.5 we also verified the possible relationships with log D6.5. Both linear and parabolic relationships were not statistically significant (data not shown). However, after the exclusion of two compounds (cimetidine and hydrochlorothiazide) (Fig. 3B), a parabolic relationship was significant at a level 0.01 (r2 = 0.733, F2,10 = 13.72). Finally, we plotted log Peff values against the combination of log PN values and, for only acids, log D6.5 values. Both linear and parabolic relationships were not statistically significant (data not shown). No significant relationship was found between log Peff values IAM:MG
IAM:DD2
or logkW values (linear relationship and either logkW statistics: r2 = 0.223, F1,13 = 3.72 and r2 = 0.233, F1,13 = 3.95, for IAM. PC.MG and IAM.PC.DD2, respectively). The lack of a direct relationship between log Peff and membrane phospholipid interaction data was not surprising since it was already verified by us in previous studies on the mechanism of BBB passage (Grumetto et al., 2012, 2013, 2014). In those studies we only found significant relationships with biological data when we 0
took into account both DlogkW and D logkW values, i.e., DlogkW IAM
0
IAM
IAM
values for bases and D logkW values for acids. However, in the present study, no relationship was found between log Peff and the IAM
0
analogous combination of DlogkW and D logkW values (Fig. 4) (r2 = 0.262, F1,13 = 4.63 and r2 = 0.238, F1,13 = 4.07, for IAM.PC.MG and IAM
IAM
0
IAM.PC.DD2 respectively), i.e., taking into account D logkW values IAM
L. Grumetto et al. / International Journal of Pharmaceutics 489 (2015) 186–194
Based on a so small number of data, it is difficult to rationalize why only acids, but not bases, are affected by polar/electrostatic forces more strongly at level of BBB than at level of jejunal barrier. As a hypothesis to be verified on a larger set of data, we would suggest that the different behavior observed for acids may be related to different physical phenomena encoded in the two biological parameters considered. Indeed, Peff values account for the rate of disappearance of a drug from the jejunal content whereas log BB are parameters related to the concentrations at the steady-state observed after a given time and also reflect other processes including plasma protein binding and tissue binding (Bickel, 2005).
A -3.0
log P effjejunum
-3.5 -4.0 -4.5 -5.0 -5.5 -1.0
193
-0.5
0.0
0.5 ∆ log kW
bases
1.0
1.5
2.0
4. Conclusions
IAM.MG
In this work we found that both log PN and, even more
acids
significantly, DlogkW values, related with log Peff values, which, in turn, are non-linearly related to the drug fraction absorbed. This IAM
B -3.0
suggests that DlogkW parameter is a suitable measure of the polar/electrostatic interactions occurring in vivo at membrane level. These results are partially in accordance with our previous findings on BBB passage and, once again, are consistent with the “flip-flop” model of membrane passage. Indeed, this model suggests that the charged forms of the analytes are able to interact with the charged head groups of phospholipid bilayers but unable to migrate in their uncharged inner moieties. This latter step is operated by the neutral forms in dynamic equilibrium. Accordingly, the linear inverse IAM
log P effjejunum
-3.5 -4.0 -4.5 -5.0 -5.5 -1.5
-1.0
-0.5
0.0
0.5
1.0
1.5
2.0
2.5
∆ log kW IAM.DD 2
bases
acids
Fig. 5. Relationships between log Peff and either DlogkW (B) values for the 15 compounds considered.
IAM:MG
(A) or DlogkW
IAM:DD2
for acids. Similarly, no relationship was observed with log Peff 0
taking into account D logkW values for all the analytes (data not shown). In contrast, significant inverse linear relationships were found IAM
between log Peff and DlogkW values taking into account DlogkW values for all the compounds (Fig. 5A and Eq. (5) for IAM.PC.MG and Fig. 5B and Eq. (6) for IAM.PC.DD2). IAM
IAM
Appendix A. Supplementary data
IAM
logPeff ¼ 0:807ð0:111ÞDlogkW
IAM:MG
3:607ð0:084Þ
References
n = 15 r = 0.803 s = 0.306 F1,13 = 53.00 F1,13 a,0.001 = 17.82 logPeff ¼ 0:674ð0:098ÞDlogkW
IAM:DD2
3:545ð0:092Þ
(6)
n = 15 r = 0.784 s = 0.321 F1,13 = 47.06 F1,13 a,0.001 = 17.82 2
Therefore, differently from BBB permeation, jejunum permeability data of not only bases but also acids relate with DlogkW . As hypothesized by other authors (Lennernäs, 1997), these results suggest that jejunum absorption and BBB passage are essentially realized by similar mechanisms, as the two barriers are similar in their chemical composition. However, polar/electrostatic interactions appear as more effective in hindering BBB passage than jejunal absorption, but for only acidic compounds. As a matter IAM
of fact, these forces are quantified by DlogkW parameter, as for basic compounds, in the relationships with jejunal absorption but IAM
had to be quantified by D logkW , i.e., magnified, in the relationships with BBB passage. IAM
Supplementary data associated with this article can be found, in the online version, at http://dx.doi.org/10.1016/j. ijpharm.2015.04.062.
(5)
2
0
values relationships observed between log Peff and DlogkW indicate that polar/electrostatic interactions act as “trapping” forces at membrane level, although promoting drug partition. Furthermore, the direct linear relationship found between log Peff and log PN values may suggest that it is the lipophilicity of the neutral forms to act as a driving force for membrane passage. The method proposed can be useful for a preliminary screening of drugs and drug-candidates for the evaluation of their oral bioavailability so contributing to reduce the number of in vivo experiments on humans and animals.
Amato, M., Barbato, F., Morrica, P., Quaglia, F., La Rotonda, M.I., 2000. Interactions between amines and phospholipids: a chromatographic study on immobilized artificial membrane (IAM) stationary phase at various pH values. Helv. Chim. Acta 83, 2836–2847. Avdeef, A., 2012. Absorption and Drug Development: Solubility, Permeability, and Charge State, second ed. John Wiley & Sons, Inc., N.J, pp. 310–318. Barbato, F., Caliendo, G., La Rotonda, M.I., Morrica, P., Silipo, C., Vittoria, A., 1990. Relationships between octanol–water partition data, chromatographic indices and their dependence on pH in a set of beta-adrenoceptor blocking agents. Il Farmaco 45 (6), 647–663. Barbato, F., La Rotonda, M.I., Quaglia, F., 1996. Chromatographic indices determined on an immobilized artificial membrane (IAM) column as descriptors of lipophilic and polar interactions of 4-phenyldihydropyridine calcium-channel blockers with biomembranes. Eur. J. Med. Chem 31, 311–318. Barbato, F., La Rotonda, M.I., Quaglia, F., 1997a. Interactions of nonsteroidal antiinflammatory drugs with phospholipids: comparison between octanol/ buffer coefficients and chromatographic indexes on immobilized artificial membranes. J. Pharm. Sci. 86, 225–229. Barbato, F., La Rotonda, M.I., Quaglia, F., 1997b. Chromatographic indexes on immobilized artificial membranes for local anesthetics: relationships with activity data on closed sodium channels. Pharm. Res. 14, 1699–1705.
194
L. Grumetto et al. / International Journal of Pharmaceutics 489 (2015) 186–194
Barbato, F., Cappello, B., Miro, A., La Rotonda, M.I., Quaglia, F., 1998. Chromatographic indexes on immobilized artificial membranes for the prediction of transdermal transport of drugs. Il Farmaco 53, 655–661. Barbato, F., di Martino, G., Grumetto, L., La Rotonda, M.I., 2004. Prediction of drug– membrane interactions by IAM–HPLC: effects of different phospholipid stationary phases on the partition of bases. Eur. J. Pharm. Sci. 22, 261–269. Barbato, F., di Martino, G., Grumetto, L., La Rotonda, M.I., 2005. Can protonated beta-blockers interact with biomembranes stronger than neutral isolipophilic compounds? A chromatographic study on three different phospholipid stationary phases (IAM–HPLC). Eur. J. Pharm. Sci. 25, 379–386. Barbato, F., 2006. The use of immobilized artificial membrane (IAM) chromatography for determination of lipophilicity. Curr. Comput. Aided Drug Des. 02 (4), 341–352. Barbato, F., Cirocco, V., Grumetto, L., La Rotonda, M.I., 2007. Comparison between immobilized artificial membrane (IAM) HPLC data and lipophilicity in n-octanol for quinolone antibacterial agents. Eur. J. Pharm. Sci. 31, 288–297. Bernhard, E., Zimmermann, F., 1984. Contribution to the understanding of oxicam ionization constants. Arzneim. Forsch. 34 (6), 647–648. Bickel, U., 2005. How to measure drug transport across the blood–brain barrier. NeuroRx 2 (1), 15–26. Braumann, T., Weber, G., Gromme, L.H., 1983. Quantitative structure–activity relationships for herbicides. Reversed phase liquid chromatographic retention parameters, log kw versus liquid–liquid partition coefficients as a model of the hydrophobicity of phenylureas, S-triazines and phenoxycarbonic acid derivatives. J. Chromatogr. A 261, 329–343. Bretschneider, B., Brandsch, M., Neubert, R., 1999. Intestinal transport of b-lactam antibiotics: analysis of the affinity at the H+/peptide symporter (PEPT1), the uptake into Caco-2cell monolayers and the transepithelial flux. Pharm. Res. 16 (1), 55–61. Comer, J., Tam, K., 2001. Lipophilicity profiles: theory and measurement. In: Testa, B., van de Waterbeemd, H., Folkers, G., Guy, R.H. (Eds.), Pharmacokinetic Optimization in Drug Research. Wiley-VHCA, Zurich, pp. 275–304. Ducarme, A., Neuwels, M., Goldstein, S., Massingham, R., 1998. IAM retention and blood brain barrier penetration. Eur. J. Med. Chem. 33, 215–223. Grumetto, L., Carpentiero, C., Barbato, F., 2012. Lipophilic and electrostatic forces encoded in IAM–HPLC indexes of basic drugs: their role in membrane partition and their relationships with BBB passage data. Eur. J. Pharm. Sci. 45, 685–692. Grumetto, L., Carpentiero, C., Di Vaio, P., Frecentese, F., Barbato, F., 2013. Lipophilic and polar interaction forces between acidic drugs and membrane phospholipids encoded in IAM–HPLC indexes: their role in membrane partition and relationships with BBB permeation data. J. Pharm. Biomed. Anal. 75, 165–172. Grumetto, L., Russo, G., Barbato, F., 2014. Indexes of polar interactions between ionizable drugs and membrane phospholipids measured by IAM–HPLC: their relationships with data of blood–brain barrier passage. Eur. J. Pharm. Sci. 65, 139–146. Gurtovenko, A.A., Vattulainen, I., 2007. Molecular mechanism for lipid flip-flops. J. Phys. Chem. B 111, 13554–13559. ska, E., Bojarski, J., Haber, P., Kaliszan, R., 2000. Retention of barbituric acid Kepczyn derivatives on immobilized artificial membrane stationary phase and its correlation with biological activity. Biomed. Chromatogr. 14 (4), 256–260. Khan, S., Guha, A., Yeole, P.G., Katariya, P., 2007. Strong cation exchange resin for improving physicochemical properties and sustaining release of ranitidine hydrochloride. Indian J. Pharm. Sci. 69 (5), 626–632. Krämer, S.D., Lombardi, D., Primorac, A., Thomae, A.V., Wunderli-Allenspach, H., 2009. Lipid-bilayer permeation of drug-like compounds. Chem. Biodivers. 6, 1900–1916.
La Rotonda, M.I., Amato, G., Barbato, F., Silipo, C., Vittoria, A., 1983. Relationships between octanol–water partition data, chromatographic indices and their dependence on pH in a set of nonsteroidal anti-inflammatory drugs. Quant. Struct. -Act. Relat. 2, 168–173. Law, V., Knox, C., Djoumbou, Y., Jewison, T., Guo, A.C., Liu, Y., Maciejewski, A., Arndt, D., Wilson, M., Neveu, V., Tang, A., Gabriel, G., Ly, C., Adamjee, S., Dame, Z.T., Han, B., Zhou, Y., Wishart, D.S., 2014. DrugBank 4.0: shedding new light on drug metabolism. Nucl. Acids Res. 42 (1), D1091–D1097. Lennernäs, H., 1997. Human jejunal effective permeability and its correlation with preclinical drug absorption models. J. Pharm. Pharmacol. 49, 627–638. Lennernäs, H., 2014. Human in vivo regional intestinal permeability: importance for pharmaceutical drug development. Mol. Pharm. 11, 12–23. Leo, A.J., Hansch, C., Elkins, D., 1971. Partition coefficients and their use. Chem. Rev. 71, 525–616. Liu, X., Testa, B., Fahr, A., 2011. Lipophilicity and its relationship with passive drug permeation. Pharm. Res. 28, 962–977. Lombardo, F., Shalaeva, M.Y., Tupper, K.A., Gao, F., Abraham, M.H., 2000. ElogPoct: a tool for lipophilicity determination in drug discovery. J. Med. Chem. 43, 2922–2928. Pagliara, A., Carrupt, P.A., Caron, G., Gaillard, P., Testa, B., 1997. Lipophilicity profiles of ampholytes. Chem. Rev. 97, 3385–3400. Panigrahi, L., Pattnaik, S., Ghosal, S.K., 2005. The effect of pH and organic ester penetration enhancers on skin permeation kinetics of terbutaline sulfate from pseudolatex-type transdermal delivery systems through mouse and human cadaver skins. AAPS PharmSciTech 6 (2), E167–E173. Péhourcq, F., Matoga, M., Bannwarth, B., 2004. Diffusion of arylpropionate nonsteroidal anti-inflammatory drugs into the cerebrospinal fluid: a quantitative structure–activity relationship approach. Fundam. Clin. Pharmacol. 18 (1), 65–70. Reichel, A., Begley, D.J., 1998. Potential of immobilized artificial membranes for predicting drug penetration across the blood–brain barrier. Pharm. Res. 15 (8), 1270–1274. Salminen, T., Pulli, A., Taskinen, J., 1997. Relationship between immobilised artificial membrane chromatographic retention and the brain penetration of structurally diverse drugs. J. Pharm. Biomed. Anal. 15 (4), 469–477. Taillardat-Bertschinger, A., Barbato, F., Quercia, M.T., Carrupt, P.A., Reist, M., La Rotonda, M.I., Testa, B., 2002. Structural properties governing retention mechanisms on immobilized artificial membrane (IAM) HPLC columns. Helv. Chim. Acta 85, 519–532. Taillardat-Bertschinger, A., Carrupt, P.A., Barbato, F., Testa, B., 2003. Immobilized artificial membrane HPLC in drug research. J. Med. Chem. 46, 655–665. Tsai, R.S., Carrupt, P.-A., El Tayar, N., Giroud, Y., Andrade, P., Testa, B., Brée, F., Tillement, J.-P., 1993. Physicochemical and structural properties of non-steroidal anti-inflammatory oxicams. Helv. Chim. Acta 76, 842–854. Van de Waterbeemd, H., Testa, B., 2008. Drug Bioavailability: Estimation of Solubility, Permeability, Absorption and Bioavailability, second ed. John Wiley & Sons, Inc., N.J, pp. 185–219. Vrakas, D., Giaginis, C., Tsantili-Kakoulidou, A., 2008. Electrostatic interactions and ionization effect in immobilized artificial membrane retention. A comparative study with octanol–water partitioning. J. Chromatogr. A 1187, 67–78. Winiwarter, S., Bonham, N.M., Ax, F., Hallberg, A., Lennernäs, H., Karlén, A., 1998. Correlation of human jejunal permeability (in vivo) of drugs with experimentally and theoretically derived parameters. A multivariate data analysis approach. J. Med. Chem. 41, 4939–4949.
Contents lists available at ScienceDirect
International Journal of Pharmaceutics journal homepage: www.elsevier.com/locate/ijpharm
Pharmaceutical nanotechnology
Relationships between human intestinal absorption and polar interactions drug/phospholipids estimated by IAM–HPLC. Lucia Grumetto, Giacomo Russo, Francesco Barbato * Dipartimento di Farmacia, Università degli Studi di Napoli Federico II, Via D. Montesano, 49, I-80131 Naples, Italy
A R T I C L E I N F O
A B S T R A C T
Article history: Received 5 March 2015 Received in revised form 16 April 2015 Accepted 22 April 2015 Available online 23 April 2015
Phospholipid affinity indexes ðlogkW Þ for 15 structurally non-related basic, acidic, ampholytic, and neutral drugs were measured by HPLC on two different phospholipid stationary phases (immobilized artificial membrane – IAM). According to a method we previously proposed, polar and electrostatic forces
IAM
0
involved in drug/membrane interactions were quantified both as DlogkW and as D logkW . These values IAM
IAM
IAM logkW
Keywords: Intestinal absorption Oral bioavailability Phospholipids IAM–HPLC Membrane Lipophilicity
are the differences between the experimental and the values expected for a neutral compound having the lipophilicity value equal to either that of the neutral form of the analyte (log PN) or that of the mixture of charged and neutral forms of the analyte at jejunum pH 6.5 (log D6.5), respectively. Jejunum IAM
absorption values, log Peff, measured by the Loc-I-Gut technique, did not relate with logkW values. A moderate linear relationship was observed with log PN values for all the analytes and a weak parabolic relationship was observed with log D6.5 values, but only after the exclusion of two analytes. In contrast, a highly significant linear inverse relationship was observed with DlogkW values. Therefore, differently from the results of our recent studies on blood–brain barrier passage, the intestinal absorption data for IAM
0
not only bases and zwitterions but also for acids relate significantly with DlogkW and not with D logkW values. The results suggest that membrane passage at jejunum level can be described according to the “flipflop” model; indeed, the lipophilicity of the neutral forms (log PN) appears related to the passage through IAM
IAM
the non-polar inner moieties of phospholipids whereas DlogkW parameter appears related to the “trapping” forces at their polar surfaces. The method, easy to perform and at medium throughput, could be of use for preliminary screening of new drugs based on oral absorption potential. ã 2015 Elsevier B.V. All rights reserved. IAM
1. Introduction The extent of intestinal absorption of drugs in humans is usually determined by in vivo experiments. However, in the last years increasing ethical issues have been raised about in vivo pharmacological experiments on both animals and humans, making desirable the development of alternative in vitro methods. The methods based on the determination of physicochemical parameters, albeit ineffective when active transport mechanisms occur, are highly reproducible and at high/medium throughput; furthermore, they may be useful both to predict the oral bioavailability of drug candidates at the early stages of their development and to formulate mechanistic hypotheses useful for drug/prodrug design. Their use could avoid, or at least reduce, in vivo experiments, such as those
* Corresponding author. Tel.: +39 81 678627; fax: +39 81 678107. E-mail address: [email protected] (F. Barbato). http://dx.doi.org/10.1016/j.ijpharm.2015.04.062 0378-5173/ ã 2015 Elsevier B.V. All rights reserved.
based on the Loc-I-Gut method. The latter yields the effective intestinal permeability values, Peff, which, although actually accounting for the enterocyte apical membrane absorption, were demonstrated strictly related to the extent of human intestinal absorption (Lennernäs, 1997). Passive drug absorption of orally administered drugs is assumed to be related to drug lipophilicity (Liu et al., 2011), expressed as the logarithm of the n-octanol/water partition coefficient, log P, when referred to single species (neutral or ionized). Indeed, this parameter appears to describe adequately the partition coefficients of neutral compounds in the membrane, which, according to the Fick’s first law (as modified to take into account the existence of a barrier) is a driving force for the membrane passage. However, a large majority of drugs support at least one ionizable function (Comer and Tam, 2001) and their noctanol lipophilicity depends on their ionization degree. The values corrected for ionization, i.e., log D values, are the weighted average of log P values of the various forms, ionized and neutral,
L. Grumetto et al. / International Journal of Pharmaceutics 489 (2015) 186–194
existing in solution as a function of the pH value (Leo et al., 1971) and are always smaller than log P values of the related neutral forms. Membrane passage of ionizable compounds seems to be more realistically described by the so-called “flip-flop” model (Gurtovenko and Vattulainen, 2007; Krämer et al., 2009). According to this model, both neutral and ionized forms, in dynamic equilibrium, are involved in the passage of membrane phospholipid bilayer. Furthermore, comparisons between partition data in phospholipids and log D values demonstrated that the latter are often inadequate at describing the interactions actually occurring between ionizable analytes and membrane phospholipids (Amato et al., 2000; Barbato, 2006; Barbato et al., 1996, 1997a, b, 1998, 2004, 2005, 2007). A rapid and reproducible way to achieve partition data in phospholipids is the measurement of chromatographic retention factors, k, by HPLC (high performance liquid chromatography) on phosphatidylcholine-like stationary phases (so-called immobilized
187
artificial membrane – IAM). These values are a direct measure of phospholipid affinity (Barbato, 2006) and are usually reported as IAM
logkW , i.e., the logarithm of k values measured at, or extrapolated to, 100% aqueous mobile phase (Amato et al., 2000; Barbato, 2006; Barbato et al., 1997b, 2004, 2005; Taillardat-Bertschinger et al., 2003). In contrast with log D values, IAM–HPLC data indicate that the interactions between ionized forms and phospholipids can be even stronger than those of the corresponding neutral forms. This occurs because, differently from the interaction forces in noctanol/water partition systems, the recognition forces involved in the interactions with phospholipids include ionic bonds (Taillardat-Bertschinger et al., 2003; Vrakas et al., 2008). The latter can be estimated by interpolating lipophilicity values in n-octanol and phospholipid interaction data, according to a simple procedure we proposed in our recent works (Grumetto et al., 2012, 2013, 2014), 0
giving DlogkW and D logkW parameters. The soundness of these parameters was recently supported by studies indicating that data IAM
Scheme 1. Chemical structures of the compounds considered.
IAM
188
L. Grumetto et al. / International Journal of Pharmaceutics 489 (2015) 186–194
of penetration through the blood–brain barrier (BBB) inversely 0
related with either DlogkW (for bases) or D logkW (for acids) values (Grumetto et al., 2012, 2013, 2014), but did not with either IAM
IAM
IAM
log D or logkW values, in contrast with other authors (Ducarme ska et al., 2000; Péhourcq et al., 2004; Reichel et al., 1998; Kepczyn and Begley, 1998; Salminen et al., 1997). This suggests that, despite ionic bonds enhance partition in phospholipids, they act as “trapping” forces at the membrane level, hindering its permeation. On the basis of the results reported above (Grumetto et al., 2012, 2013, 2014), we decided to investigate about possible relationships between these parameters and intestinal permeation data. In fact, it was suggested that “pure passive membrane diffusion is universal for membranes with different physiological functions and physicochemical properties” (Lennernäs, 1997). In this work we considered fifteen structurally unrelated molecules, usually orally administered and supposed to be absorbed by mainly passive mechanism at the intestinal level. Indeed, an involvement of active transport mechanism was reported for the intestinal absorption of cefalexin, a beta lactam antibiotic, but its contribution to the total absorption would play an only minor role (Bretschneider et al., 1999). The set consisted of six bases (cimetidine, desipramine, propranolol, ranitidine, terbutaline and verapamil), five acids (fluvastatin, hydrochlorothiazide, isotretinoin, ketoprofen and naproxen), three zwitterions (amoxicillin, cefalexin and piroxicam), and one neutral compound (carbamazepine) (Scheme 1). Their phospholipid affinity data were experimentally measured by IAM–HPLC technique on two different phospholipid stationary
phases (IAM.PC.MG and IAM.PC.DD2) to reasonably exclude that the data were affected by secondary retention mechanisms. Comparisons between the scales of n-octanol lipophilicity and IAM data were preliminarily performed to highlight similarities and dissimilarities. Finally, possible relationships between jejunal absorption data and either n-octanol lipophilicity data or IAM data were investigated.
2. Materials and methods 2.1. Chromatographic conditions and equipment The analyses were performed according to a method we previously reported in the literature (Grumetto et al., 2012, 2013, 2014). The analytical HPLC columns were a IAM.PC.MG (4.6 mm 150 mm; Regis Chemical Company, Morton Grove, IL) and a IAM.PC. DD2 (4.6 mm 100 mm; Regis Chemical Company, Morton Grove, IL). Only one IAM.PC.MG and only one IAM.PC.DD2 columns were used throughout the present study. Chromatographic retention data are reported as log k (the logarithm of the retention factor), calculated by the expression: k = (tr – t0)/t0, where tr and t0 are the retention times of the drug and a non-retained compound (acetone), respectively. Direct measurements of k values in fully IAM
aqueous mobile phases ðkW Þ, i.e., 0.1 M phosphate buffer at pH 7.0, were only possible for the compounds eluting within 20 min,
Table 1 pKa values, ionization degrees at pH 6.5, and logarithms of lipophilicity values in n-octanol and of chromatographic retention factors on IAM phases for the compounds considered. Compound
pKa
[Ionized]/[unionized] ratio at pH 6.5
log PN
log D6.5
logkW
Amoxicillin Hydrochlorothiazide Ranitidine Cimetidine Cefalexin Terbutaline Carbamazepine Piroxicam Ketoprofen Propranolol Naproxen Verapamil Fluvastatin Isotretinoin Desipramine
2.44o/7.14o 7.90b 8.20d 6.80e 3.12o /6.84o 10.10f – 5.46i /1.86i 4.45b 9.42b 4.15b 8.92b 4.30n 4.76o 10.40b
p
1.22a 0.03c 0.27b 0.40b 0.65b 0.90b 2.19g 3.00j 3.12b 3.28c 3.34m 3.79b 4.17c 4.20b 4.90b
1.70a 0.05 1.44 0.08 1.00c,q 2.70 2.19g 1.20j,r 1.07 0.36l,s 0.99 1.37 1.97 2.45 1.38c,q
-0.920 0.540 1.130 1.030 0.220 0.662 1.039h 1.850k 1.120k 1.821e 1.260k 2.049 2.210 2.807 2.826h
a
0.04 50 2 p
3981 p p
112 832 224 263 158 55 7943
Winiwarter et al. (1998). Law et al. (2014). c Avdeef (2012). d Khan et al. (2007). e Grumetto et al. (2012). f Panigrahi et al. (2005). g Lombardo et al. (2000). h Grumetto et al. (2014). i Bernhard and Zimmermann (1984). j Tsai et al. (1993). k Barbato et al. (1996). l Barbato et al. (1990). m La Rotonda et al. (1983). n Van de Waterbeemd and Testa (2008). o pKa values calculated. p Not reported because either zwitterion or neutral compound. q log D7.4. r log D6.07. s log D6.7. b
IAM:MG
IAM:DD2
logkW
0.728 0.977 0.860 0.783 0.021 0.863 1.717h 1.767 1.360 2.480e 1.339 3.085 2.843 3.704 2.741h
L. Grumetto et al. / International Journal of Pharmaceutics 489 (2015) 186–194
189
2.2. Sample preparation
whereas, for the solutes requiring the addition of acetonitrile to the IAM
eluent, the kW values were calculated by an extrapolation method (Braumann et al., 1983). All reported log k values are the average of at least three measurements; for each log k value the 95% confidence interval associated with each value never exceeded 0.04. To avoid that the experimental measurements were affected by retention changes due to column aging, the retention times of five test compounds (amlodipine, p-nitroaniline, toluene, isradipine and ketoprofen) were checked weekly. No correction was done to the experimental retention values since no retention value of test compounds changed more than 4% during the study.
A
Each analyte was dissolved in the mobile phase or in methanol to ca. 104 M concentration. 2.3. Lipophilic and biological activity parameters The log PN values, i.e., partition coefficients n-octanol/aqueous phase of the neutral form of analytes, were from the literature (Avdeef, 2012; Law et al., 2014; La Rotonda et al., 1983; Lombardo et al., 2000; Tsai et al., 1993; Winiwarter et al., 1998).
3.5 3.0
desipramine
2.5
logk W IAM.MG
2.0
piroxicam
1.5 1.0 0.5 0.0
cefalexin
-0.5 amoxicillin
-1.0 -1.5 -2.0
-1.0
0.0
1.0
2.0
3.0
4.0
5.0
6.0
log PN bases
B
acids
neutral
4.0 3.5 3.0 desipramine
logk W IAM.DD2
2.5 2.0 piroxicam 1.5 1.0 0.5 cefalexin
0.0 -0.5
amoxicillin -1.0 -2.0
-1.0
0.0
1.0
2.0
3.0
4.0
5.0
6.0
log PN bases IAM:MG
IAM:DD2
Fig. 1. Relationships between either logkW (A) or logkW compounds (Taillardat-Bertschinger et al., 2002).
acids
neutral
(B) and log PN values for the 15 compounds considered in comparison to the plots of 36 neutral
190
L. Grumetto et al. / International Journal of Pharmaceutics 489 (2015) 186–194
The n-octanol/aqueous buffer at pH 6.5 partition coefficients (log D6.5) were calculated according to the following equations:
A
3.5 3.0
logD6:5 ¼ logPN logð1 þ 106:5pK a Þðfor acidsÞ
2.5
with the exception of (i) amoxicillin, whose log D6.5 value was taken from the literature (Winiwarter et al., 1998), (ii) cefalexin and desipramine, whose experimental log D7.4 (Avdeef, 2012) were assumed as a reasonable estimate of log D6.5 values, and (iii) piroxicam and propranolol, whose experimental log D6.07 and log D6.7, respectively, (Tsai et al., 1993; Barbato et al., 1990) were also assumed as a reasonable estimate of log D6.5 values. The pKa values were either calculated by the program ACD/labs (release 12.00) or taken from the literature (Bernhard and Zimmermann, 1984; Grumetto et al., 2012; Khan et al., 2007; Law et al., 2014; Panigrahi et al., 2005; Van de Waterbeemd and Testa, 2008). The log Peff values, measured at pH 6.5, were from the literature (Lennernäs, 2014).
2.0
log k W IAM.MG
logD6:5 ¼ logPN logð1 þ 10pK a 6:5 Þðfor basesÞ
1.5 1.0 0.5 0.0 -0.5 -1.0 -1.5 -2.0
-1.0
0.0
1.0
2.0 N
3.0
4.0
5.0
6.0
4.0
5.0
6.0
6.5
log P (log D ) bases
neutral
acids
B 4.0 3.5
2.4. Statistical analysis
3.0
3. Results and discussion
2.5
log k W IAM.DD2
Linear regression analysis was performed by a commercially available statistical package (Microsoft Excel 2003) for personal computer observing the requirements of significant regression analysis.
2.0 1.5 1.0 0.5 0.0
3.1. Selection of the physicochemical parameters Ionizable compounds show different lipophilicity values in noctanol at different pH of the aqueous phase (log D values), according to the abundance of the neutral and ionized forms in solution. The log D values theoretically calculated by the equations above reported in Section 2 do not take into account the contribution of the ionized forms to the partition. This implies that the theoretical values are close to the experimental ones only if the contribution of the ionized forms is negligible. Therefore, log D calculated at the pH values at which the fraction of the neutral form is 1% do not adequately reflect the actual lipophilicity values. For ten of the considered compounds that showed at pH 6.5 an appropriate ionized/neutral form ratio we took into account the calculated log D6.5 values. For the bases propranolol and desipramine we considered their experimentally determined log D6.7 and log D7.4, respectively. Since propranolol is already extensively ionized at pH 6.7 (ionized/neutral form ratio >500) and, even so more at pH 6.5, its log D6.7 is expected to be very close to its log D6.5 value. Analogously, log D7.4 value for desipramine, showing an ionized/neutral form ratio 1000 at pH 7.4, is expected to be very close to its log D6.5 value. The log D6.5 values for the zwitterionic ampholytes amoxicillin, cefalexin, and piroxicam cannot be theoretically calculated by the equations reported above. The value of log D6.5 for amoxicillin was taken from the literature (Winiwarter et al., 1998) whereas, to the best of our knowledge, no experimental log D6.5 value is reported in the literature for both cefalexin and piroxicam. Since zwitterions behave as “lipophilicity buffers” (Pagliara et al., 1997), we assumed the experimentally determined log D7.4 value for cefalexin (Avdeef, 2012) and the experimentally determined log D6.07 value for piroxicam (Tsai et al., 1993) as reasonable estimates of their log D6.5. Table 1 summarizes pKa values, the ratios between the concentrations of ionized and unionized forms at pH 6.5
-0.5 -1.0 -2.0
-1.0
0.0
1.0
2.0
3.0
log PN (log D6.5) bases
neutral
acids
IAM:MG
IAM:DD2
Fig. 2. Relationships between either logkW (A) or logkW (B) and the combination of log PN values for 10 compounds (bases, zwitterions and neutral) and log D6.5 values for five acids, in comparison to the plots of 36 neutral compounds.
(calculated by the Henderson–Hasselbalch equation), log PN, and log D6.5 values for the compounds considered, as well as their IAM:MG
IAM:DD2
IAM
and logkW values, i.e., logkW on IAM.PC.MG and logkW IAM.PC.DD2 stationary phases, respectively. As can be seen, the compounds considered span a very large range of log PN values (from 1.22 to 4.90). It is worth underlining that using eluents at pH 7.0, to maximize column stability and data reproducibility, does not have negative impact on the significance of the data as measures of membrane interactions occurring at slightly different pH values (e.g., pH 7.4 and pH 6.5, for the BBB passage and the jejunum absorption, respectively). Indeed, it was demonstrated that retention on IAM phases, even for ionizable compounds, is only negligibly affected by variations of the pH of the eluent within the range of 5.5–7.0 (Amato et al., 2000). IAM
Finally, logkW values were determined on two phospholipid stationary phases differing from each other in the end-capping of residual amino groups of the silica–propylamine core. IAM.PC.MG supports hydroxy groups (being end-capped by methyl glycolate) whereas IAM.PC.DD2 supports C10 and C3 alkyl chains (being endcapped by both decanoic and propionic anhydrides). Therefore, retention on IAM.PC.MG could be supposed to be affected by
L. Grumetto et al. / International Journal of Pharmaceutics 489 (2015) 186–194
191
Table 2 Values of the differences between observed and expected logarithms of retention factors on IAM.PC.MG and IAM.PC.DD2 stationary phases (DlogkW
IAM:MG
D
0
D
IAM:MG logkW ,
IAM:DD2 logkW
0
and D
IAM:DD2 logkW ,
and
respectively) and logarithms of the human effective jejunum permeability values for the compounds considered
(log Peff). Compound
DlogkIAM:MG W
IAM:MG D0 logkW
DlogkIAM:DD2 W
IAM:DD2 D0 logkW
log Peff
Amoxicillin Hydrochlorothiazide Ranitidine Cimetidine Cephalexin Terbutaline Carbamazepine Piroxicam Ketoprofen Propranolol Naproxen Verapamil Fluvastatin Isotretinoin Desipramine
0.778 1.296 1.648 1.445 0.003 0.681 0.038 0.206 0.619 0.045 0.653 0.221 0.361 0.216 0.319
1.158 1.309 3.001 1.823 1.304 3.532 0.038 1.632 1.008 2.268 1.209 1.697 1.384 1.600 2.469
1.294 1.888 1.491 1.292 0.297 0.905 0.555 0.152 0.671 0.299 0.898 0.428 0.169 0.664 0.953
1.743 1.904 3.087 1.737 1.838 4.268 0.555 1.529 1.247 3.027 1.299 2.690 1.889 2.297 2.335
4.50 5.06 4.57 4.58 3.81 4.52 3.37 3.18 3.06 3.54 3.07 3.17 3.62 4.00 3.35
secondary mechanisms (e.g., H bonds) involving hydroxy groups of methyl glycolate whereas retention on IAM.PC.DD2 could be supposed to be affected by secondary mechanisms, partition based, occurring at level of C10 and C3 alkyl chains. However, IAM:DD2
IAM:DD2
logkW
IAM:MG
¼ 1:106ð0:104ÞlogkW
þ 0:171ð0:170Þ
A -3.0
-3.5
(1)
n = 15 r2 = 0.896 s = 0.404 F1,13 = 112.23 F1,13 a,0.001 = 17.82 In this and the following equations, n is the number of data considered to derive the regression equation, r2 is the square of the correlation coefficient, s is the standard error of the estimate, F (the subscripts are the number of variables and the degrees of freedom) is the Fisher statistic of the regression, followed by the tabulated critical value for a = 0.001. The standard errors of the regression coefficients are reported in parentheses.
log Peff jejunum
IAM:MG
and logkW values were found strongly collinear logkW (Eq. (1)) and IAM data can be assumed as substantially reflecting the interactions between analytes and phospholipids, with secondary interaction mechanisms playing an only minor role.
does not exist as sole form at any pH of the medium (Barbato et al., 2007). The points of the less lipophilic compounds (log P < 1) are shifted upward with respect to the regression line. However, it
-4.0
-4.5
-5.0
-5.5 -2.0
-1.0
0.0
1.0
3.2. Relationships among the physicochemical parameters
IAM logkW
found that, differently from ionizable compounds, the values of structurally non-related neutral compounds relate unambiguously with n-octanol lipophilicity values in the log P range of 1.0–4.8. The relationships are expressed by Eqs. (2) and (3).
bases
¼ 0:792ð0:038ÞlogP 0:732ð0:105Þ
¼ 0:934ð0:038ÞlogP 0:883ð0:104Þ
6.0
acids
-3.0
-3.5
(3)
n = 36 r2 = 0.946 s = 0.246 F1,34 = 595.74 F1,34 a,0.001 = 12.90 IAM
5.0
(2)
n = 36 r2 = 0.926 s = 0.248 F1,34 = 422.40 F1,34 a,0.001 = 12.90 IAM:DD2 logkW
4.0
B
The plots logkW vs. log PN of the compounds considered in the present work, superimposed to the plots of the neutral compounds, are reported in Fig 1. It is interesting to note that not only, as expected, the point relative to carbamazepine, a neutral compound, but also the points relative to the zwitterions amoxicillin, cefalexin, and piroxicam are very close to the line of the neutral compounds; this confirms that phospholipid interaction of zwitterions is related to the n-octanol lipophilicity of their neutral form despite of the fact that the latter
log Peff jejunum
IAM:MG
3.0
log PN
In our previous work (Taillardat-Bertschinger et al., 2002) we
logkW
2.0
-4.0
-4.5
cimetidine
-5.0
-5.5 -3.0
hydrochlorothiazide
-2.0
-1.0
0.0
1.0
2.0
3.0
log D6.5 bases
acids
Fig. 3. Relationships between log Peff and either log PN values (A) or log D6.5 values (B) for the 15 compounds considered.
192
L. Grumetto et al. / International Journal of Pharmaceutics 489 (2015) 186–194
that DlogkW values parameterize the excess of polar, mainly electrostatic, intermolecular interaction forces encoded in phospholipid partition as compared to n-octanol partition. However, in our previous works (Grumetto et al., 2013, 2014) IAM
A -3.0
log P effjejunum
-3.5
we observed that BBB passage data related with DlogkW values for only basic compounds whereas for acids they related with IAM
-4.0
D0 logkIAM W values; the latter are the distances from the regression
-4.5 -5.0 -5.5 -0.5
0.0
0.5 ∆ logkW
1.0
IAM.MG
(∆'log kW
IAM.MG
1.5
2.0
)
line of neutral compounds calculated taking into account their log D7.4 values, i.e., the lipophilicity actually displayed at the physiological pH of the blood. At jejunum level the compounds considered are in solution at pH 6.5. As can be seen in Fig. 2, taking into account the log D6.5 values for the acidic compounds considered in the present work, the distances of the points from the regression line of neutral compounds noticeably increase. Table 2 summarizes DlogkW
IAM:MG
bases
acids
D
B
0
, DlogkW
IAM:DD2
0
, D logkW
IAM:MG
,
IAM:DD2 logkW ,
and log Peff values for the compounds considered. On both IAM phases, a moderate inverse linear relationship was
-3.5
found between DlogkW values and log PN values (r2 = 0.589, F1,13 = 18.65 and r2 = 0.602, F1,13 = 19.68 for IAM.PC.MG and IAM.PC. DD2, respectively).
-4.0
3.3. Relationships with intestinal absorption data
-4.5
Since inter-laboratory variability of biological data is generally too high for correlation studies, we only took into account the data, reported in the literature, from a single source (Lennernäs, 2014) to obtain a consistent scale of jejunal Peff values. Obviously, we could not take into account the compounds undergoing active transport mechanisms, e.g., L-dopa and amino acids. The log Peff values, experimentally measured at pH 6.5 at jejunum level (Lennernäs, 2014), moderately related linearly with log PN values (Fig. 3A and Eq. (4)).
IAM
log P effjejunum
-3.0
-5.0 -5.5 -1.5
-1.0
-0.5
0.0
0.5
1.0
1.5
2.0
2.5
∆ logkW IAM.DD2 (∆' logkW IAM.DD 2)
bases
acids
0 IAM:MG IAM:MG ð logkW Þ (A) or IAM:DD2 ð logkW Þ (B) values for the 15 compounds considered. 0 IAM:DD2 and logkW were considered for only acids.
Fig. 4. Relationships between log Peff and either DlogkW
D D
IAM:DD2 logkW 0 IAM:MG logkW
0
D
D
D
should be remembered that at so low lipophilicity values the IAM
linearity of the relationship logkW vs. log PN is no longer observed for neutral compounds, too (Taillardat-Bertschinger et al., 2002). The behavior of poorly lipophilic compounds suggests that polar interactions between analytes and phospholipids (including the electrostatic ones in the case of ionizable compounds) become predominant when lipophilicity falls down. The other points of basic compounds are close to the regression line of neutral compounds, with the exception of desipramine that, as can be seen IAM:DD2
vs. log PN, lies below the regression in Fig 1B, reporting logkW line. Desipramine is the most lipophilic compound in the set considered with a log PN value of 4.90. Its behavior confirms that strongly lipophilic bases interact with phospholipids weaker than isolipophilic neutral compounds, as already observed on IAM.PC. DD2 phase (Grumetto et al., 2012, 2014). The distance of the points from the regression line is expressed as DlogkW
IAM:MG
D
IAM:DD2 logkW ,
and
for IAM.PC.MG and IAM.PC.DD2 phases, respectively.
Therefore, DlogkW
IAM:MG
values are the differences between the IAM:MG
experimentally measured logkW calculated from log P
N
values and the values
by Eq. (2); DlogkW
IAM:DD2
values are the IAM:DD2
differences between the experimentally measured logkW values and the values calculated from log PN by Eq. (3). Differently from n-octanol, but analogously to membrane phospholipids, phosphatidylcholine moieties of IAM stationary phases are electrically charged species. Therefore, it is reasonable to assume
logPeff ¼ 0:273ð0:062ÞlogPN 4:427ð0:177Þ
(4)
n = 15 r = 0.598 s = 0.437 F1,13 = 19.32 F1,13 a,0.001 = 17.82 2
Parabolic relationships between log Peff values and log PN values were not statistically significant (at both a levels of 0.001 and 0.01) for both the whole set and a set reduced by the exclusion of acids. Since the permeability values were measured at pH 6.5 we also verified the possible relationships with log D6.5. Both linear and parabolic relationships were not statistically significant (data not shown). However, after the exclusion of two compounds (cimetidine and hydrochlorothiazide) (Fig. 3B), a parabolic relationship was significant at a level 0.01 (r2 = 0.733, F2,10 = 13.72). Finally, we plotted log Peff values against the combination of log PN values and, for only acids, log D6.5 values. Both linear and parabolic relationships were not statistically significant (data not shown). No significant relationship was found between log Peff values IAM:MG
IAM:DD2
or logkW values (linear relationship and either logkW statistics: r2 = 0.223, F1,13 = 3.72 and r2 = 0.233, F1,13 = 3.95, for IAM. PC.MG and IAM.PC.DD2, respectively). The lack of a direct relationship between log Peff and membrane phospholipid interaction data was not surprising since it was already verified by us in previous studies on the mechanism of BBB passage (Grumetto et al., 2012, 2013, 2014). In those studies we only found significant relationships with biological data when we 0
took into account both DlogkW and D logkW values, i.e., DlogkW IAM
0
IAM
IAM
values for bases and D logkW values for acids. However, in the present study, no relationship was found between log Peff and the IAM
0
analogous combination of DlogkW and D logkW values (Fig. 4) (r2 = 0.262, F1,13 = 4.63 and r2 = 0.238, F1,13 = 4.07, for IAM.PC.MG and IAM
IAM
0
IAM.PC.DD2 respectively), i.e., taking into account D logkW values IAM
L. Grumetto et al. / International Journal of Pharmaceutics 489 (2015) 186–194
Based on a so small number of data, it is difficult to rationalize why only acids, but not bases, are affected by polar/electrostatic forces more strongly at level of BBB than at level of jejunal barrier. As a hypothesis to be verified on a larger set of data, we would suggest that the different behavior observed for acids may be related to different physical phenomena encoded in the two biological parameters considered. Indeed, Peff values account for the rate of disappearance of a drug from the jejunal content whereas log BB are parameters related to the concentrations at the steady-state observed after a given time and also reflect other processes including plasma protein binding and tissue binding (Bickel, 2005).
A -3.0
log P effjejunum
-3.5 -4.0 -4.5 -5.0 -5.5 -1.0
193
-0.5
0.0
0.5 ∆ log kW
bases
1.0
1.5
2.0
4. Conclusions
IAM.MG
In this work we found that both log PN and, even more
acids
significantly, DlogkW values, related with log Peff values, which, in turn, are non-linearly related to the drug fraction absorbed. This IAM
B -3.0
suggests that DlogkW parameter is a suitable measure of the polar/electrostatic interactions occurring in vivo at membrane level. These results are partially in accordance with our previous findings on BBB passage and, once again, are consistent with the “flip-flop” model of membrane passage. Indeed, this model suggests that the charged forms of the analytes are able to interact with the charged head groups of phospholipid bilayers but unable to migrate in their uncharged inner moieties. This latter step is operated by the neutral forms in dynamic equilibrium. Accordingly, the linear inverse IAM
log P effjejunum
-3.5 -4.0 -4.5 -5.0 -5.5 -1.5
-1.0
-0.5
0.0
0.5
1.0
1.5
2.0
2.5
∆ log kW IAM.DD 2
bases
acids
Fig. 5. Relationships between log Peff and either DlogkW (B) values for the 15 compounds considered.
IAM:MG
(A) or DlogkW
IAM:DD2
for acids. Similarly, no relationship was observed with log Peff 0
taking into account D logkW values for all the analytes (data not shown). In contrast, significant inverse linear relationships were found IAM
between log Peff and DlogkW values taking into account DlogkW values for all the compounds (Fig. 5A and Eq. (5) for IAM.PC.MG and Fig. 5B and Eq. (6) for IAM.PC.DD2). IAM
IAM
Appendix A. Supplementary data
IAM
logPeff ¼ 0:807ð0:111ÞDlogkW
IAM:MG
3:607ð0:084Þ
References
n = 15 r = 0.803 s = 0.306 F1,13 = 53.00 F1,13 a,0.001 = 17.82 logPeff ¼ 0:674ð0:098ÞDlogkW
IAM:DD2
3:545ð0:092Þ
(6)
n = 15 r = 0.784 s = 0.321 F1,13 = 47.06 F1,13 a,0.001 = 17.82 2
Therefore, differently from BBB permeation, jejunum permeability data of not only bases but also acids relate with DlogkW . As hypothesized by other authors (Lennernäs, 1997), these results suggest that jejunum absorption and BBB passage are essentially realized by similar mechanisms, as the two barriers are similar in their chemical composition. However, polar/electrostatic interactions appear as more effective in hindering BBB passage than jejunal absorption, but for only acidic compounds. As a matter IAM
of fact, these forces are quantified by DlogkW parameter, as for basic compounds, in the relationships with jejunal absorption but IAM
had to be quantified by D logkW , i.e., magnified, in the relationships with BBB passage. IAM
Supplementary data associated with this article can be found, in the online version, at http://dx.doi.org/10.1016/j. ijpharm.2015.04.062.
(5)
2
0
values relationships observed between log Peff and DlogkW indicate that polar/electrostatic interactions act as “trapping” forces at membrane level, although promoting drug partition. Furthermore, the direct linear relationship found between log Peff and log PN values may suggest that it is the lipophilicity of the neutral forms to act as a driving force for membrane passage. The method proposed can be useful for a preliminary screening of drugs and drug-candidates for the evaluation of their oral bioavailability so contributing to reduce the number of in vivo experiments on humans and animals.
Amato, M., Barbato, F., Morrica, P., Quaglia, F., La Rotonda, M.I., 2000. Interactions between amines and phospholipids: a chromatographic study on immobilized artificial membrane (IAM) stationary phase at various pH values. Helv. Chim. Acta 83, 2836–2847. Avdeef, A., 2012. Absorption and Drug Development: Solubility, Permeability, and Charge State, second ed. John Wiley & Sons, Inc., N.J, pp. 310–318. Barbato, F., Caliendo, G., La Rotonda, M.I., Morrica, P., Silipo, C., Vittoria, A., 1990. Relationships between octanol–water partition data, chromatographic indices and their dependence on pH in a set of beta-adrenoceptor blocking agents. Il Farmaco 45 (6), 647–663. Barbato, F., La Rotonda, M.I., Quaglia, F., 1996. Chromatographic indices determined on an immobilized artificial membrane (IAM) column as descriptors of lipophilic and polar interactions of 4-phenyldihydropyridine calcium-channel blockers with biomembranes. Eur. J. Med. Chem 31, 311–318. Barbato, F., La Rotonda, M.I., Quaglia, F., 1997a. Interactions of nonsteroidal antiinflammatory drugs with phospholipids: comparison between octanol/ buffer coefficients and chromatographic indexes on immobilized artificial membranes. J. Pharm. Sci. 86, 225–229. Barbato, F., La Rotonda, M.I., Quaglia, F., 1997b. Chromatographic indexes on immobilized artificial membranes for local anesthetics: relationships with activity data on closed sodium channels. Pharm. Res. 14, 1699–1705.
194
L. Grumetto et al. / International Journal of Pharmaceutics 489 (2015) 186–194
Barbato, F., Cappello, B., Miro, A., La Rotonda, M.I., Quaglia, F., 1998. Chromatographic indexes on immobilized artificial membranes for the prediction of transdermal transport of drugs. Il Farmaco 53, 655–661. Barbato, F., di Martino, G., Grumetto, L., La Rotonda, M.I., 2004. Prediction of drug– membrane interactions by IAM–HPLC: effects of different phospholipid stationary phases on the partition of bases. Eur. J. Pharm. Sci. 22, 261–269. Barbato, F., di Martino, G., Grumetto, L., La Rotonda, M.I., 2005. Can protonated beta-blockers interact with biomembranes stronger than neutral isolipophilic compounds? A chromatographic study on three different phospholipid stationary phases (IAM–HPLC). Eur. J. Pharm. Sci. 25, 379–386. Barbato, F., 2006. The use of immobilized artificial membrane (IAM) chromatography for determination of lipophilicity. Curr. Comput. Aided Drug Des. 02 (4), 341–352. Barbato, F., Cirocco, V., Grumetto, L., La Rotonda, M.I., 2007. Comparison between immobilized artificial membrane (IAM) HPLC data and lipophilicity in n-octanol for quinolone antibacterial agents. Eur. J. Pharm. Sci. 31, 288–297. Bernhard, E., Zimmermann, F., 1984. Contribution to the understanding of oxicam ionization constants. Arzneim. Forsch. 34 (6), 647–648. Bickel, U., 2005. How to measure drug transport across the blood–brain barrier. NeuroRx 2 (1), 15–26. Braumann, T., Weber, G., Gromme, L.H., 1983. Quantitative structure–activity relationships for herbicides. Reversed phase liquid chromatographic retention parameters, log kw versus liquid–liquid partition coefficients as a model of the hydrophobicity of phenylureas, S-triazines and phenoxycarbonic acid derivatives. J. Chromatogr. A 261, 329–343. Bretschneider, B., Brandsch, M., Neubert, R., 1999. Intestinal transport of b-lactam antibiotics: analysis of the affinity at the H+/peptide symporter (PEPT1), the uptake into Caco-2cell monolayers and the transepithelial flux. Pharm. Res. 16 (1), 55–61. Comer, J., Tam, K., 2001. Lipophilicity profiles: theory and measurement. In: Testa, B., van de Waterbeemd, H., Folkers, G., Guy, R.H. (Eds.), Pharmacokinetic Optimization in Drug Research. Wiley-VHCA, Zurich, pp. 275–304. Ducarme, A., Neuwels, M., Goldstein, S., Massingham, R., 1998. IAM retention and blood brain barrier penetration. Eur. J. Med. Chem. 33, 215–223. Grumetto, L., Carpentiero, C., Barbato, F., 2012. Lipophilic and electrostatic forces encoded in IAM–HPLC indexes of basic drugs: their role in membrane partition and their relationships with BBB passage data. Eur. J. Pharm. Sci. 45, 685–692. Grumetto, L., Carpentiero, C., Di Vaio, P., Frecentese, F., Barbato, F., 2013. Lipophilic and polar interaction forces between acidic drugs and membrane phospholipids encoded in IAM–HPLC indexes: their role in membrane partition and relationships with BBB permeation data. J. Pharm. Biomed. Anal. 75, 165–172. Grumetto, L., Russo, G., Barbato, F., 2014. Indexes of polar interactions between ionizable drugs and membrane phospholipids measured by IAM–HPLC: their relationships with data of blood–brain barrier passage. Eur. J. Pharm. Sci. 65, 139–146. Gurtovenko, A.A., Vattulainen, I., 2007. Molecular mechanism for lipid flip-flops. J. Phys. Chem. B 111, 13554–13559. ska, E., Bojarski, J., Haber, P., Kaliszan, R., 2000. Retention of barbituric acid Kepczyn derivatives on immobilized artificial membrane stationary phase and its correlation with biological activity. Biomed. Chromatogr. 14 (4), 256–260. Khan, S., Guha, A., Yeole, P.G., Katariya, P., 2007. Strong cation exchange resin for improving physicochemical properties and sustaining release of ranitidine hydrochloride. Indian J. Pharm. Sci. 69 (5), 626–632. Krämer, S.D., Lombardi, D., Primorac, A., Thomae, A.V., Wunderli-Allenspach, H., 2009. Lipid-bilayer permeation of drug-like compounds. Chem. Biodivers. 6, 1900–1916.
La Rotonda, M.I., Amato, G., Barbato, F., Silipo, C., Vittoria, A., 1983. Relationships between octanol–water partition data, chromatographic indices and their dependence on pH in a set of nonsteroidal anti-inflammatory drugs. Quant. Struct. -Act. Relat. 2, 168–173. Law, V., Knox, C., Djoumbou, Y., Jewison, T., Guo, A.C., Liu, Y., Maciejewski, A., Arndt, D., Wilson, M., Neveu, V., Tang, A., Gabriel, G., Ly, C., Adamjee, S., Dame, Z.T., Han, B., Zhou, Y., Wishart, D.S., 2014. DrugBank 4.0: shedding new light on drug metabolism. Nucl. Acids Res. 42 (1), D1091–D1097. Lennernäs, H., 1997. Human jejunal effective permeability and its correlation with preclinical drug absorption models. J. Pharm. Pharmacol. 49, 627–638. Lennernäs, H., 2014. Human in vivo regional intestinal permeability: importance for pharmaceutical drug development. Mol. Pharm. 11, 12–23. Leo, A.J., Hansch, C., Elkins, D., 1971. Partition coefficients and their use. Chem. Rev. 71, 525–616. Liu, X., Testa, B., Fahr, A., 2011. Lipophilicity and its relationship with passive drug permeation. Pharm. Res. 28, 962–977. Lombardo, F., Shalaeva, M.Y., Tupper, K.A., Gao, F., Abraham, M.H., 2000. ElogPoct: a tool for lipophilicity determination in drug discovery. J. Med. Chem. 43, 2922–2928. Pagliara, A., Carrupt, P.A., Caron, G., Gaillard, P., Testa, B., 1997. Lipophilicity profiles of ampholytes. Chem. Rev. 97, 3385–3400. Panigrahi, L., Pattnaik, S., Ghosal, S.K., 2005. The effect of pH and organic ester penetration enhancers on skin permeation kinetics of terbutaline sulfate from pseudolatex-type transdermal delivery systems through mouse and human cadaver skins. AAPS PharmSciTech 6 (2), E167–E173. Péhourcq, F., Matoga, M., Bannwarth, B., 2004. Diffusion of arylpropionate nonsteroidal anti-inflammatory drugs into the cerebrospinal fluid: a quantitative structure–activity relationship approach. Fundam. Clin. Pharmacol. 18 (1), 65–70. Reichel, A., Begley, D.J., 1998. Potential of immobilized artificial membranes for predicting drug penetration across the blood–brain barrier. Pharm. Res. 15 (8), 1270–1274. Salminen, T., Pulli, A., Taskinen, J., 1997. Relationship between immobilised artificial membrane chromatographic retention and the brain penetration of structurally diverse drugs. J. Pharm. Biomed. Anal. 15 (4), 469–477. Taillardat-Bertschinger, A., Barbato, F., Quercia, M.T., Carrupt, P.A., Reist, M., La Rotonda, M.I., Testa, B., 2002. Structural properties governing retention mechanisms on immobilized artificial membrane (IAM) HPLC columns. Helv. Chim. Acta 85, 519–532. Taillardat-Bertschinger, A., Carrupt, P.A., Barbato, F., Testa, B., 2003. Immobilized artificial membrane HPLC in drug research. J. Med. Chem. 46, 655–665. Tsai, R.S., Carrupt, P.-A., El Tayar, N., Giroud, Y., Andrade, P., Testa, B., Brée, F., Tillement, J.-P., 1993. Physicochemical and structural properties of non-steroidal anti-inflammatory oxicams. Helv. Chim. Acta 76, 842–854. Van de Waterbeemd, H., Testa, B., 2008. Drug Bioavailability: Estimation of Solubility, Permeability, Absorption and Bioavailability, second ed. John Wiley & Sons, Inc., N.J, pp. 185–219. Vrakas, D., Giaginis, C., Tsantili-Kakoulidou, A., 2008. Electrostatic interactions and ionization effect in immobilized artificial membrane retention. A comparative study with octanol–water partitioning. J. Chromatogr. A 1187, 67–78. Winiwarter, S., Bonham, N.M., Ax, F., Hallberg, A., Lennernäs, H., Karlén, A., 1998. Correlation of human jejunal permeability (in vivo) of drugs with experimentally and theoretically derived parameters. A multivariate data analysis approach. J. Med. Chem. 41, 4939–4949.
