CHARACTERIZATION of the INTESTINAL TRANSPORT PARAMETERS for SMALL Peptlde DRUGS*

Journal of Controlled Release, 13 (1990) 141-146 141 Elsevier Science Publishers B.V., Amsterdam

CHARACTERIZATION OF THE INTESTINAL TRANSPORT PARAMETERS FOR SMALL PEPTlDE DRUGS*

Doron I. Friedman and Gordon L. Amidon* College of Pharmacy, The University of Michigan, Ann Arbor, MI 48 109- 1065 (U.S. A .)

(Received August 30, 1989; accepted in revised form January 5, 1990)

Keywords: ACE inhibitors; j34actams; oral bioavailability of peptide drugs; carrier-mediated transport; absorption kinetics

Several laboratories have recently shown that many peptides and peptide-similar drugs are well absorbed in the mammalian intestine via the peptide transport system for nutrient smallpeptide (dipeptides and tripeptides) absorption. This facilitated non-passive transport in the intestine is saturable and concentration dependent, and further characterized by the mutual inhibition of transport among the compounds. Peptide and peptide derivative drugs are often zwitterionic compounds, generally ionized at the intestinal pH and therefore are expected to undergo only limited passive absorption. Nonpassive and passive transport are independent and may occur simultaneously. Peptide drugs that have been shown to be transported by the peptide carrier systems are various &?actams, the angiotensin converting enzyme (ACE) inhibitors captopril, enalapril, lisinopril and SQ 29,852, TRH and its analogues, and alafosfaline and some of its analogues. The carrier characterization is based upon transport parameters, K, J,,, (or V,,.J and P, (Michaelis-Menten constant, maximal flux and the carrier permeability constant respec- tively) determined over the last years for several dipeptides, f3-lactam antibiotics and ACE inhibitors.

INTRODUCTION and tripeptides. Larger peptides are hydrolyzed by the brush border enzymes before being ab- Proteins and peptides are hydrolyzed by pro- sorbed as tripeptides, dipeptides or amino acids. teolytic enzymes while traversing the G.I. sys- The peptide carrier is distinct from the amino tem to finally reach the intestinal absorptive acid transport carriers. The small peptide car- membrane either as readily absorbable amino rier transports dipeptides and tripeptides of acids or as short peptides [ 11. Uptake of the different amino composition and is versatile to- short peptides into the intestinal microvillus wards many structure modifications as will be cells is accomplished by the peptide carrier shown in this paper. Nevertheless, this peptide transport system and is restricted to dipeptides carrier has not yet been described in a way that enables structure-uptake relationship pre- *Paper presented at the Second International Symposium diction. on Disposition and Delivery of Peptide Drugs (F.I.P. Sat- Many peptide drugs which are derivatives of ellite Symposium) held at Leiden, The Netherlands, l-3 September 1989. dipeptides and tripeptides are absorbed in a fa- *To whom correspondence about this paper should be cilitated manner via the peptide carrier-me- addressed. diated transport system [2-121. Unlike large

0168-3659/90/$03.50 0 1990 - Elsevier Science Publishers B.V. 142

peptide and protein drugs, which are highly manner are thyrotrophin-releasing hormone susceptible to proteolytic enzymes and are cur- (TRH ) and its analogues which are tripeptide rently administered parenterally, small peptide and tripeptide derivatives [ 81 and alafosfalines or peptide-like drugs are more suitable for oral which comprise a unique group of antibiotics administration. Non=passive transport can undergoing hydrolysis concurrently with ab- provide an efficient route for absorption for sorption by the peptide carrier transport sys- these compounds which are generally charged tem [9]. More recently, several angiotensin or zwitterionic at the physiological pH and may converting enzyme (ACE) inhibitors which are not readily penetrate the lipophilic intestinal dipeptide and tripeptide derivatives were shown membrane. to be transported by the intestinal peptide car- Transport via the peptide carrier-mediated rier-mediated transport system and their system is a saturable and concentration depen- transport profile and Michaelis-Menten pa- dent phenomenon which is characterized by the rameters were defined [ 10-121. Michaelis-Menten constants K, and JmaX (or In this paper the carrier transport parame- V,,,). Uptake inhibition by other compounds ters L, Lax and J,,,/K, (P,) for the /3-lac- transported by the peptide carrier is typical of tam antibiotics and ACE inhibitors are re- this type of carrier-mediated transport. Al- viewed and compared in order to initiate a though apparent competitive inhibition seems structure-absorption characterization of the to occur, the inhibition mechanism for peptide intestinal small peptide carrier-mediated drugs has not yet been extensively studied. transport system. Passive transport of peptide derivative drugs may occur simultaneously with a non-passive transport and it is proportional to the relative hydrophobic character of the compound. While TRANSPORT PARAMETERS FOR SMALL passive transport may take place along the en- PEPTIDE DRUGS tire intestine, nonpassive transport via the peptide carrier may exhibit a site-specific absorption profile due to the localization of the carrier Detailed transport parameters are available to the small intestine. for several hydrolysis stable dipeptides, P-lac- The in situ jejunal perfusion or ex uivo jejunal tam antibiotics and ACE inhibitors. The trans- rings and everted intestine have been fre- port parameters are the permeability parame- quently employed to characterize carrier-me- ters PCand P, for non-passive (carrier ) and for diated transport. Varying drug concentrations passive diffusion transport pathways, respec- are used to detect Michaelis-Menten uptake tively and the maximal transport rate V,,, (or behavior and to define the transport parame- J,,,) and the Michaelis-Menten constant, K,, ters, K, J,,, and P, (J,,,/K,). Transport in- for non-passive transport. The data presented hibitors are studied in order to validate non- in Table 1 are transport parameters derived passive transport and explore its mechanism. from in situ single-pass rat perfusion applying The penicillins which are D (Ala)-D (Ala) ana- the boundary layer solution for estimation of logues were the first peptide drugs identified to the aqueous permeability [ 131. be transported non-passively in the intestine by The interaction between the peptide and the the peptide carrier-mediated transport system. carrier is a primary factor in the nonpassive or Subsequently, the absorption mechanism of facilitated transport phenomenon. Unlike the other fi-lactams was recognized to involve a non- more specific amino acid carriers identified, the passive transport [ 2-71. Other groups of bioac- peptide carrier transports a variety of di- and tive compounds absorbed in a non-passive tripeptides composed of different amino acids 143

TABLE 1

Summary of permeability data for some dipeptides and peptide drugs p-lactams and ACE inhibitors (*-dimensionless parameter, K,-Michaelis-Menten constant in mM, JL,, -maximal transport rate, Pr = J&,, /K,-the carrier transport parameter, Pz-dimensionless passive parameter)

Compound K, J&X p: pit Inhibitor Ref.

Peptides Carnosine 12.9 6.62 0.51 3 Phe-Gly 1.29 6.87 5.33 3 P-Lactams Amoxicillin 0.058 0.044 0.558 0.76 21 Cefaclor 16.1 21.3 1.32 0.00 6 Cefadroxil 5.9 8.4 1.43 0.00 Cyclacillin, Gly-Gly 6 Cefatrizine 0.58 0.73 1.25 0.20 6 Cephalexin 7.2 9.1 1.30 0.00 Cyclacillin, Gly-Gly 6 Cephradine 1.48 1.57 1.06 0.30 Cephalexin, Gly-Leu 6 Cefixime 0.031 0.016 0.184 0.52 Carnosine, Gly-Pro 21 ACE inhibitors Captopril 5.9 12.3 2.08 1.00 Gly-Gly, Cephradine 10 Enalapril 0.07 0.13 1.9 0.35 Tyr-Gly, Cephradine 11 Lisinopril 0.082 0.032 0.39 0.00 Tyr-Gly, Cephradine 12 SQ 29,852 0.080 0.160 2.00 0.25 Tyr-Gly, Cephradine 12

and peptide-like drugs which bear various sub- MACROSCOPIC CORRELATION stitution groups and which are not necessarily natural peptides by their chemical structure. The transport parameters for several acid Although some evidence for the existence of stable p-lactam antibiotics are presented in Ta- multiple peptide carriers has been recently ble 1. The carrier permeability parameter, e or published [ 141, it has not been confirmed and S,,,/K,, is relatively constant for these D-lac- so remains to be resolved. The parameters pre- tam antibiotics while Jkax and Km vary more sented in Table 1 for peptide transport dem- significantly. The Km and S,,,, as summarized onstrate different “affinities” of the peptide for in Table 1, for the different P-lactams are cor- the carrier as well as variability with respect to related in Fig. 1 with the maximal plasma lev- stability against enzymatic hydrolysis. For ex- els, C,,,, in humans after a 500 mg oral dose. ample, carnosine (P-Ala-His) is very stable The C,,, values in Fig. 1 are the mean values of against enzymatic hydrolysis and therefore was the reported maximal plasma levels [ 151. A sig- extensively studied [ 31. nificant correlation between Km and C,,, after It is difficult to compare non-passive trans- oral administration of the different /3-lactams port parameters measured by different labora- is observed. The maximal plasma concentra- tories due to differing experimental approaches tion, C,,,, after a 500 mg oral dose for several and methods for the calculation of JmaX.Since j?-lactams increases as Km increases. Since P,, J,,, and J,,,/K, may be calculated for a spe- also increases proportionally, J*,/K, will not cific amount of dry or wet tissue weight, protein change significantly between the various P-lac- or intestinal length or surface area, a simple tams studied and the permeability constant of transformation and matching of the data is im- the carrier may therefore be assumed relatively possible. We will focus on results from our lab- constant. From this correlation it can be con- oratory for consistency [5,6,10-121. cluded that the maximal flux or the carrier ca- 144

15 For the ACE inhibitors, a linear relation is / shown in Fig. 2 between S,,, (as measured in /’ rat jejunum) and fraction of dose absorbed in / humans ( % oral bioavailability ). It can be con- eluded that, for the ACE inhibitors, the J&,, or I 7 j ‘Cephulexiz~ the capacity of the peptide carrier in the intestine is the rate limiting factor for bioavailability. The carrier permeability, J~,,/K,, is 2.0 +-0.1 for captopril, enalapril and SQ 29,852 and is lower, 0.39, for lisinopril. The low capacity and carrier permeability of lisinopril results in lower non-passive permeability and poor bioavailability. A more general relationship be- ’ /’ /’ ,’ tween fraction of dose absorbed in human and

-5 ,/_ rat jejunal permeability parameters has been

0 10 20 published elsewhere [ 161. Apparently two distinct drug carrier affinity Mean C,ax. c(g /mL profiles are observed: the p-lactams transport parameters show consistently high capacity and Fig. 1. (0 ) Michaelis-Menten constant (I&, miU) and low affinity to the carrier while the ACE inhib- ( A ) dimensionless maximal flux (P,.) as derived from in situ jejunal rat perfusion versus mean reported maximal itors show low capacity and high affinity. The plasma concentration (C,,,) after 500 mg oral administra- kinetic mechanism that accounts for this dif- tion of /I-lactams in humans. ference is yet unknown.

020 I pacity for the specific p-lactam has the highest impact on the absorption process and subse- quent human plasma levels. Higher capacity low-affinity /?-lactams are better absorbed. This correlation is further supported by the fact that 2 / higher doses do not always produce a propor- E 0.10 / tional area urider curve (AUC ) for the p-lac- l9 tams with low K,, while only a very slight trend ( / / towards non-linear pharmacokinetics is re- / 005 ; /’ ported for /?-lactams with K, above 4.0 mM. As ,I the K, and S,,, increase, it is less likely for the drug to saturate the carrier. Beta-lactams with low K, and dose dependent bioavailability include ampicillin and cyclacillin and those with 0 20 40 60 80 100 relatively higher Km include cephradine, ce- Oral Bioavailability. %. phalexin and cephadroxil. The latter group, with relatively high S,,, and Km, are completely ab- Fig. 2. The dimensionless maximal flux (J&) of different sorbed in humans, while amoxicillin and cycla- ACE inhibitors in in situ rat perfusion versus human oral cillin are not. bioavailability. 145

STRUCTURAL CONSIDERATIONS acids comprising the peptide drugs is also im- portant for drug-carrier recognition and facil- The “peptide” nature of cephradine, enala- itated absorption. Numerous series of ACE in- pril and TRH is demonstrated in Fig. 3 by a hibitors with rings of varying size replacing the comparison of their chemical structures to the proline side chain or different second or third tripeptide, Phe-Ala-Pro. The /3-lactams and amino acid side chains have been synthesized ACE inhibitors which are absorbed by the pep- in order to obtain orally active ACE inhibitors tide carrier-mediated transport system are di- [ 17-201. A notable inconsistency between in or tripeptide derivatives having at least one vitro ACE inhibition and oral activity of the peptide bond and a free terminal carboxilic acid. various derivatives prevails throughout the dif- Although a second peptide bond and an amino ferent series of ACE inhibitor analogues syn- terminal or an a-amino group exist in most /?- thesized. The best intravenous or in vitro ACE lactams as in tripeptides, it is not a prerequisite inhibitors are not necessarily the most potent for facilitated absorption, as can be seen from orally [ 17-201. the non-passive absorption of the ACE inhibi- Although the inconsistency between in vitro tors which are lacking a second peptide bond or ACE inhibition and oral activity among these a terminal amino group [ 10,121. Recently, a /3- derivatives hints towards facilitated absorp- lactam antibiotic without an a-amino group, tion, no evidence for facilitated absorption has cefixime, has also been shown to be non-pas- yet been published for most of the orally active sively absorbed [ 71. Furthermore, the ACE in- ACE inhibitors. Considering their zwitterionic hibitors enalapril and lisinopril have only the nature and the cumulative data on the non-pas- amino residue of the second amino acid group sive absorption of peptide drugs, it may be hy- while SQ 29,852 is totally devoid of a second pothesized that those which are orally active are peptide bond (which is replaced with a phos- absorbed in a facilitated manner. However, oral phonic group) yet still exhibits non-passive ab- activity by passive diffusion only may be the sorption [ 11,121. case for those zwitterions have large dominant The nature of the side chains of the amino hydrophobic residues such as in fosinopril [ 111.

Phe-Ala-Pro

Enalapril Cephradine 0-l/- \ Oh &-;qcH.

f” 0 H,C HO

Fig. 3. Comparison of the chemical structures of the peptide drugs cephradine, enalapril and TRH to the tripeptide Phe- Ala-Pro. 146

Further characterization of the structural re- falin: Design, synthesis, and structure-activity rela- quirements of the peptide transporter will con- tionship, Antimicrob. Agents Chem., 18 (1980) 897- 905. siderably aid in the development of small pep- 10 M. Hu and G.L. Amidon, Intestinal absorption of cap- tides with high oral efficacy. topril in rats, J. Pharm. Sci., 77 (1988) 1007-1011. 11 D.I. Friedman and G.L. Amidon, Passive and carrier- mediated intestinal absorption components of two ACE inhibitor prodrugs in rats: enalapril and fosino- ACKNOWLEDGMENT pril, Pharm. Res., 6 (1990) 1043-1047. 12 D.I. Friedman and G.L. Amidon, Intestinal absorp This work was supported in parts by a grant tion mechanism of dipeptide ACE inhibitors of the ly- from the National Institute of General Medical syl-proline type: Lisinopril and SQ 29,852, J. Pharm. Sci., 78 (1989) 995-998. Sciences, GM 37188. 13 D.A. Johnson and G.L. Amidon, Determination of intrinsic membrane transport parameters from perfused intestine experiments: A boundary layer approach to REFERENCES estimating the aqueous and unbiased membrane permeabilities, J. Theor. Biol., 131 (1988) 93-106. D.H. Alpers, Digestion and absorption of carbohy- 14 K. Inui, T. Okano, H. Maegawa, M. Kato, M. Takano drates and proteins, in: L.R. Johnson (Ed.), Physiol- and R. Hori, H+ coupled transport of P.O. cephalo- ogy of the Gastro-Intestinal Tract, Raven Press, New sporins via dipeptide carriers in rabbit intestinal brush- York, NY, 1987, pp. 1469-1487. border membranes: Difference of transport character- A. Tsuji, E. Nakashima, I. Kagami and T. Yamana, istic between cefixime and cephradine, J. Pharmacol. Intestinal absorption mechanism of amino-p-lactam Exp. Ther., 247 (1988) 235-241. antibiotics. I: Comparative absorption and evidence 15 Martindale The Extra Pharmacopoeia, 29th edn., for saturable transport of amino-/I-lactam antibiotics J.E.F. Reynolds (Ed.), The Pharmaceutical Press, by in situ rat small intestine, J. Pharm. Sci., 70 (1981) London, 1989. 768-773. 16 G.L. Amidon, P.J. Sinko and D. Fleisher, Estimating E. Nakashima, A. Tsuji, S. Kagatani and T. Yamana, human oral fraction dose absorption: A correlation us- Intestinal absorption mechanism of amino-/I-lactam ing rat intestinal membrane permeability for passive antibiotics. III. Kinetics of carrier-mediated transport and carrier-mediated compounds, Pharm. Res., 5 across the rat small intestine in situ. J. Pharm. Dyn., (1988) 651-654. 7 (1984)452-464. 17 J.L. Stanton, J.W.H. Watthey, M.N. Desai, B.M. Finn, E. Nakashima, A. Tsuji, H. Mizuo and T. Yamana, J.E. Babiarz and H.C. Tomaselli, ACE inhibitors: Kinetics and mechanism of in vitro uptake of amino- Structure-activity profile of 1-benzazepin-2-one de- /I-lactam antibiotics by rat small intestine and rela- rivatives, J. Med. Chem., 28 (1985) 1603-1606. tion to the intact-peptide transport system, Biochem. 18 D.S. Karanewsky, M.C. Badia, D.W. Cushman, J.M. Pharmacol., 33 (1984) 3345-3352. DeForrest, T. Dejneka, M.J. Loots, M.G. Perri, E.W. M. Hu, P.J. Sinko, A.L.J. DeMeere, D.A. Johnsonand Petri110 and J.R. Powell, (Phosphinyloxy)acyl amino G.L. Amidon, Membrane permeability for some amino acid inhibitors of ACE. I. Discovery of a novel orally acids and p-lactam antibiotics: application of the active inhibitor of ACE, J. Med. Chem., 31 (1988) 204- boundary layer approach. J. Theor. Biol., 131 (1988) 212. 107-114. 19 J. Krapcho, C. Turk, D.W. Cushman, J.R. Powell, J.M. P.J. Sinko and G.L. Amidon, Characterization of oral DeForrest, E.R. Spitzmiller, D.S. Karanewsky, M. absorption of p-la&am antibiotics. I. Cephalosporins. Duggan, G. Rovnyak, J. Schwartz, S. Natarajan, J.D. Determination of intrinsic membrane absorption pa- Godfrey, D.E. Ryono, R. Neubeck, K.S. Atwal and W. rameters in the rat intestine in situ, Pharm. Res., 5 Petrillo, ACE inhibitors. Mercapto, carboxyl dipep- (1988) 645-650. tide, and phosphinic acid inhibitors incorporating 4- A. Tsuji, T. Terasaki, I. Tamai and H. Hirroka, H+ substituted prolines, J. Med. Chem., 31 (1988) 1148- gradient dependent and carrier-mediated transport of 1160. cefixime, a new cephalosporin antibiotic, across brush- 20 S. Klutchko, C.J. Blankley, R.W. Fleming, A.E. Wer- border membrane vesicles from rat small intestine, J. ner, I. Nordin, A. Holmes, M.L. Hoefle, D.M. Cohen, Pharmacol. Exp. Ther., 241 (1986) 594-601. A.D. Essenburg and H.R. Kaplan, Synthesis of novel M.J. Humphrey and P.S. Ringrose, Peptides and re- ACE inhibitor quinapril and related compounds. A di- lated drugs: A review of their absorption, metabolism, vergence of structure-activity relationship for non- and excretion. Drug Met. Rev., 11 (1986) 283-310. sulfhydryl and sulfhydryl types, J. Med. Chem., 29 F.R. Atherton, M.J. Hall, C.H. Hassall, SW. Holmes, (1986)1953-1961. R.W. Lambert, W.J. Lloyd and P.S. Ringrose, Phos- 21 D.I. Friedman and G.L. Amidon, unpublished results, phonopeptide antibacterial agents related to alafos- 1989.