- Email: [email protected]
Disposition of a Polymeric Prodrug of Mitomycin C, Mitomycin C-Dextran Conjugate, in the Perfused Rat Liver
Disposition of a Polymeric Prodrug of Mitomycin C, Mitomycin C-Dextran Conjugate, in the! Perfused Rat Liver KIMIHIKO SATO, b Z U E ITAKURA, HITOSHISEZAKI'
KOYO
NISHIDA,YOSHINOBU TAKAKURA, MITSURUHASHIDA,
AND
Received April 20, 1988, from the Department of Basic Pharmaceufics, Faculty of Pharmaceutical Sciences, Kyoto University, Accepted for publication August 15. 1988. Sakyo-ku, Kyoto 606, Japan. -~ ~~~~
Abstract 0 The disposition of a polymeric prodrug of mitomycin C
(MMC), mitomycin C-dextran conjugate (MMC-D), was studied in the single-pass perfused rat liver in order to clarify the effect of physico-
chemical properties, such as molecular weight and electric charge, on the hepatic uptake of MMC-D. Six types of MMC-D were used: both cationic MMC-D (MMC-D,,,) and anionic MMC-D (MMC-Dan) conjugated with dextran with molecular weights of 10,000,70,000, and 500,000. Outflow curves were analyzed using statistical moment theory. Remarkable hepatic uptake of MMC-D,,, was observed and the uptake amount increased with an increase in molecular weight (i.e.,-80% of the dose was taken up by the liver during a single passage of the conjugate with a molecular weight of 500,000). Intrinsic clearance ( CLint,i)and apparent distribution volume (VJ also increased as the molecular weight increased. On the other hand, almost 100% of applied MMC-Dan was recovered in the outflow regardless of molecular weight, with almost the same moment parameters as those of the vascular reference substance (VRS), '311-labeledhuman serum albumin (HSA).In a repeated application, the uptake of MMC-D,., decreased in a stepwise manner, suggesting a saturation in the hepatic uptake of MMC-D,,, while the uptake of MMC-D,, was unchanged. The MMC-D,,, pretreatment also affected the uptake of Evans blue (EB) bound to bovine serum albumin (BSA). These results demonstrate that molecular weight and electric charge determine the hepatic disposition of macromolecular prodrugs.
One of the major limitations of cancer chemotherapy is that most antitumor drugs have indiscriminate action on both tumor and normal cells. Therefore, a n approach to modify their pharmacokinetic properties to concentrate their action a t the target cells would be promising.' Mitomycin C (MMC) is a highly active antibiotic in a variety of human neoplastic disease, but its use is limited by side effects such as severe myelosuppression and gastrointestinal damage. To overcome these disadvantages and enhance therapeutic efficacy, the utilization of analogues3 and lipophilic4 or polymeric5 prodrugs of MMC has been reported. In our series of investigations, we have developed polymeric prodrugs of MMC [MMC-dextran conjugates having cationic (MMC-D,,J and anionic (MMC-D,,) charges], and have examined their physicochemical, pharmacodynamic, and pharmacokinetic characteristics.621 These studies revealed that MMC-D both acts as a reservoir of MMC that behaves characteristically as a macromolecule, supplying active MMC in the body,'0J6 and has higher activity than MMC against various murine tumors7 and human malignant tumor~.~~ The in vivo disposition study demonstrated that MMC-D,,, was rapidly cleared from plasma and accumulated in the liver and spleen to a great extent, while MMC-Dan showed slow plasma clearance and accumulated gradually in the liver, spleen, and lymph nodes following intravenous inject i ~ n . ~ The ~ Jin~ vitro study showed that MMC-DCatwas highly adsorbed on the surface of hepatocytes; however, the 0022-3549/89/0 100-001 1 $0 1.OO/O 0 1989, American Pharmaceutical Association
interaction between MMC-D, and hepatocytes was negligible.zO In the present study, hepatic disposition of MMC-D was studied using the isolated perfused rat liver in order to clarify the effect of physicochemical properties of the conjugates, such as electric charge and molecular weight, on the clearance by the reticuloendothelial organs. The MMC-D conjugate was introduced as a pulse function from the portal vein and venous outflow patterns were analyzed by statistical moment analysis to derive the disposition parameters.22 Furthermore, the effect of MMC-D treatment on the disposition of a model anionic drug, Evans blue (EB), bound to albumin was studied.
Experimental Section Chemicals-Mitomycin C (MMC) was kindly supplied by Kyowa Hakko Kogyo (Tokyo). Dextrans with various molecular weights were purchased from Pharmacia (Uppsala, Sweden), and had average molecular weights of -10,000 (T-10),70,000(T-701,and 500,000 (T-500). I3'I-Labeled human serum albumin (HSA; 100 pCi/mg) was purchased from Daiichi Radioisotopes (Tokyo). Bovine serum albumin (BSA; fraction V) was from A m o u r Pharmaceutical (U.K.), Evans blue (EB) was from Merck (Darmstadt, Germany), and Na,["Cr]04 (536 pCi/mg) was obtained from New England Nuclear (Boston, MA). Red blood cells (RBC) were labeled by incubation with Na2[51Cr]04a t 37 "C for 20 min and suspended in saline. All other chemicals were commercial, reagent-grade products. Preparation of MMC-D,., and MMC-Dan-The MMC-DCatand MMC-D,, conjugates were synthesized as reported p r e v i ~ u s l y . ~ J ~ For the synthesis of MMC-D,,,, dextran was activated by cyanogen bromide, a spacer (E-aminocaproic acid) was introduced, and then MMC was coupled by a carbodiimide-catalyzed reaction. Three structures were proposed for the linkage between the spacer and the dextran,TJ and two of them have the possibility to give a cationic charge in MMC-D,,,. For the synthesis of MMC-Dan,6-bromohexanoic acid was introduced to dextran as a spacer and then MMC was coupled as above. Since not all spacer arms were used for the coupling of MMC, the remaining free carboxyl groups give anionic charges in MMC-Dan.All conjugates were estimated to contain MMC to almost equal extents of -10% for MMC-D,,, and 8% (wr'w) for MMC-Dan.The MMC-D conjugates were prepared in saline solution at a concentration of 5 mg equivalent MMC/mL for injection. Analytical Method-The concentration of MMC-D and EB were determined spectrophotometrically a t 364 and 620 nm, respectively. In the case of free MMC, the concentration was determined a t 364 nm. The I3'I and 51Crradioactivities were counted with a well NaI scintillation counter. Single-Pass Perfusion of Rat Liver-Male Wistar albino rats (170-200 g) were obtained from Shizuoka Agricultural Co-operative Association for Laboratory Animals (Shizuoka, Japan). Perfusion of the rat liver was carried out using the method of Mortimore et al.24.25 with slight modifications. The rats were anesthetized by intraperitoneal injection of sodium pentobarbital (60 mgkg), the abdomen and chest were opened, and the portal vein and the inferior vena cava were cannulated with a polyethylene tubing (PE-160). Krebs Ringer Journal of Pharmaceutical Sciences 111 Vol. 78, No. 1, January 1989
bicarbonate buffer (pH 7.4) was aerated with 95% 02:5% C 0 2 and pumped at 37 "C into the cannulate portal vein with the aid of a peristaltic pump a t a rate of 12 mlimin. The bile duct was cannulated with polyethylene tubing (PE-10) and bile was collected every 10 min during experiments to check the viability of the liver. Bile production was almost constant in the range 3-8 pL/min. After a n equilibration period of 30 min, test compounds dissolved in saline (0.1328 mL) were injected into the line of perfusion flow as a pulse function using a six-position rotary valve injector (type 50 teflon rotary valves, Rheodyne, Cotati, CA). The outflow perfusate was collected into weighed test tubes at appropriate time intervals ( 1 3 s). The sample volume was calculated from the gain in weight of the tube, assuming a density of the outflow perfusate of 1.0. The sample was subjected to assay after an appropriate dilution with the perfusion buffer. In the case of repeated application, the interval between injections was 15 min. Data Analysis-Moment analysis was used for analyzing the outflow pattern. A detailed theoretical description of this analysis was given previously.21.22The first three (zero to second) statistical moment parameters (moments) are defined as follows:
LL
0 $.p Y
z
0 + a
Dc
+ z
w
0
z
0
0
0
15
45
30
60
T I M E( S E C ) AUCi
=
Cdt
(1)
-ti = [tCdtiAUCi
(t - ?i)2Cdt/AUCi
17: =
Figure 1-Typical outflow patterns of three types of MMC-D,,,, ['3'uHSA in the liver perfusion experiment. Key: (0)MMC-D,,, (T-10); (U) MMC-Dca, (T-70); (A) MMC-Dcat (T-500);( 0 )['3'IlHSA.
64
(3)
r
h
where t is the time, C is the concentration of compounds normalized by the injection dose and has a dimension of 5% of dose/mL, and AUCi, ?,, and r2i,are the area under the concentration-time curve, mean transit time of the drug through the liver, and variance of transit time, respectively. These moments are calculated by numerical integration using a linear trapezoidal formula from the outflow concentration-time curve. Disposition parameters are calculated from these moments as follows:
z --. 40 J
0
0 I&
0
be
z
32
0 L I
(4)
k Dc
AUCiQii100
(5)
E 16
te,,i = t i / ( l - FJ
(6)
0
CLjnt,i =
(7)
Vi Fi
= ti/AUCi100 =
-
vjfie*,i
0
z
where Vi is the apparent distribution volume, Fi is the recovery ratio, Q, is the flow rate, tel,i is the mean elimination time, and CL,,,, is the intrinsic clearance. These parameters can be divided into two groups [i.e.,those representing distribution (V,) and those representCLint,J]. ing elimination (Fi,
ResuIts Perfusion of MMC-D,,,, MMC-D,., [l3lI1HSA, [SICrlRBC, and Free MMC-Figures 1 and 2 show the representative outflow patterns of three types of MMC-D,,, and MMC-Dan together with that of [1311]HSA in a single-pass liver perfusion. "Cr-Labeled RBC was the earliest to appear in the hepatic venous outflow, and its concentration rose to the highest and earliest peak and decayed most rapidly (data not shown). The appearance of [1311]HSA was a little slower and the peak value was a little lower than that of [51CrlRBC. Every MMC-D,,, conjugate had a lower peak than [l3lI1HSA, and the peak value decreased with an increase of molecular weight (Figure 1).On the contrary, MMC-Danshowed almost the same outflow patterns as that of ['311]HSA, regardless of molecular weight (Figure 2). The moments, AUCi, ti, and u2i were calculated from these outflow patterns and the disposition parameters were derived from the moments. Table I summarizes the moments and disposition parameters for [51Cr]RBC,[1311]HSA,MMC-IIfat, MMC-Dan,and MMC. The recoveries of [51CrlRBCand [ IIHSA were almost 100% and 12 / Journal of Pharmaceutical Sciences Vol. 78, No. I , January 7989
+
0
0
15
30 TIME
45
60
(SEC)
Figure 2-Typical outflow patterns of three types of MMC-Dan, ['3'l]HSA in the liver perfusion experiment. Key: (0)MMC-Dan (T-10); ( 0 )MMC-Dan (T-70); (A)MMC-0," (T-500); ( 0 )['"I]HSA.
the V, were 0.209 and 0.252 m u g liver, respectively,As the molecular weight of MMC-D,,, increased, F, and tel,l decreased and CLlnt,, and V, increased. On the other hand, parameters of MMC-Dan were similar to those of [13111HSA, irrespective of molecular weight. In the case of free MMC, -35% of the dose was extracted by the liver (Table I). Repeated Application of MMC-D-Figure 3 shows the outflow patterns of MMC-D,,, (T-500) applied repeatedly three times to the perfused rat liver. The peak value was increased in a stepwise manner. Moments and disposition parameters for MMC-DCat(T-500) in these runs, together with those of MMC-DCat(T-10) and MMC-Dan (T-701, are summarized in Table 11. For MMC-D,,, (T-5001, F, was increased and CLlnt,,and V, were decreased successively, and the same tendency was observed in the case of MMC-D,,, (T10). However, there was no significant difference between the parameters of MMC-Dan (T-70) in the first and the second applications.
Table I-Moments Experiment
and Disposition Parameters for [51Cr]RBC,['311]HSA, MMC-D,,,, AUC,, % of dosesiml
Compound
471.3 485.9 396.1 285.5 89.3 508.0 505.4 485.3 325.9
RBC HSA MMC-Dcat(T-l0) MMC-Dca,(T-70) MMC-D-t(T-500) MMC-Dan(T-l0) MMC-Dan(T-70) MMC-Dan(T-500) MMC aMean 2 SD (n
28
=
f
f, s
24.5a
f 14.3 2 9.5
t 24.4 2 16.2 2 21.6 f 13.7 t 22.4 t 10.2
MMC-Dan,and MMC in the Liver Perfusion
s2
UZI,
6 1 .o 1.o 0.81 1 0.571 0.180 1.o 1.o 0.979 0.654
42.1 +- 14.3 36.8 f 17.5 59.8 2 9.6 92.7 f 7.3 70.7 2 33.0 55.6 ? 23.7 53.3 t 5.3 54.5 t 7.1 9.9 47.1
8.89 t 0.83 9.33 2 1.09 12.01 ? 1.35 115 4 & 2.71 8.22 -t 2.59 9.51 ? 1.80 9.75 t 2.15 8.68 f 1.28 8.88 2 0.49
*
CL",,,,
tel,bv
mllminig
min
-
-
0.313 1.06 6.56 0 0 0.031 0.682
1.06 0.445 0.167
-
413 0.428
v,
mLig 0.209 0.252 0.331 0.478 1.08 0.277 0.236 0.229 0.292
3-9).
r L
0
z
0,
14
z
0
I-
<
I-
I K -
<
i
L
K
W
z0
E 11 0
7
z
0
0
0
15
0
30 TIME
45
60
Effect of Preperfusion of MMC-D on t h e Disposition of EB/BSA-Figure 4 illustrates the outflow patterns of EB injected with BSA (4.7%, w/v) after treatment with MMCDcat,and the calculated moments and disposition parameters are shown in Table 111. In the case of a single injection of EB/BSA, the recovery of EB was -82%. After MMC-DCat treatment, the recovery of EB was decreased depending on the uptake amount of MMC-DCat(i.e.,the larger the molecuand Disposition Parameters for MMC-D,,, AUC,, Yo of doseslrnl
Compound MMC-Dca,(T.,500) 1st 2nd 3rd MMC-Dcat(T-10) 1st 2nd MMC-Dan(T-70) 1st 2nd 'Mean t SD (n
=
45
60
TIME (SEC)
(SEC)
Figure 3- Typicd outflow patterns of MMC-D,,, (T-500) applied repeatedly three times in the liver perfusion experiment. Key: firsf perfusion (O ),second perfusion (0), third perfusion (A).
Table Il-Moments
30
15
0
Figure 4-Typicai outf/ow patterns of EB injected with BSA after preperfusion of three types of MMC-Dcaiin the liver perfusion experiment. Key: (a)no freatment; (0)MMC-Dca,(T-10); (0) MMC-Dca,(T70); (A)MMC-Dcai (T-500).
lar weight of MMC-D,,,, the less the recovery of EB in the outflow). On the contrary, MMC-D,, treatment did not affect the disposition of EB/BSA. Effect of EB/BSA on the Disposition of MMC-D-Figure 5 shows the outflow patterns of MMC-D,,, (T-70) before and after EBiBSA treatment in the liver perfusion experiment. Moments and disposition parameters are summarized in
and MMC-Dan in the Liver Perfusion Experiment UZ,,
v,,
s2
*
mLig
89.3 t 16.2a 181.6 f 8.7 236.8 ? 5.3
8.22 ? 2.59 8.74 2 3.77 11.03 t 2.31
70.7 33.0 73.0 f 10.2 72.2 2 12.8
0.180 0.366 0.474
6.56 2.29 1.16
0.167 0.232 0.380
1.08 0.505 0.442
396.1 f 9.5 429.1 2 10.1
12.01 5 1.35 15.09 f 0.75
59.8 t 9.6 90.7 t 3.5
0.81 1 0.873
0.313 0.199
1.06 1.99
0.331 0.395
505.4 f 13.7 506.8 k 15.0
9.75 t 2.15 8.35 k 1.14
53.3 t 5.3 53.1 2 8.4
1.o 1 .o
0 0
-
0.236 0.192
-
3-9).
Journal of Pharmaceutical Sciences / 13 Vol. 78, No. 7, January 1989
Table ill-Moments and Disposition Parameters for Evans Blue (EB) Injected with Bovine Serum Albumin (BSA) after Preperfusion of MMC-D,., and MMC-D,, in the Llver Perfusion Experiment Treatment
AUCi, % Of dosesiml
None MMC-Dcat(T-l0) MMC-D-,(T-70) MMC-D-t(T-500) MMC-Dan(T-l0) MMC-Dan(T-70) MMC-Dan(T-500)
410.2 ? 6.2a 393.2 5 9.5 299.3 2 21.7 206.5 -t 25.2 419.1 5 13.7 398.6 ? 1097 389.1 f 6.74
aMean 2 SD (n
=
10.12 2 1.05 10.48 -+ 2.08 9.52 2 2.06 8.78 k 1.41 9.20 ? 0.79 11.30 f 0.11 10.07 -t 0.81
up,, s2
F,
58.6 ? 11.9 61.7? 4.7 55.5 ? 7.2 53.1 2 4.50 62.7 ? 14.2 70.1 ? 10.6 60.2 2 13.0
0.824 0.788 0.598 0.418 0.839 0.795 0.792
0.960 0.825 0.392 0.253 0.952 0.919 0.809
0.297 0.313 0.371 0.519 0.260 0.31 1 0.297
3-6).
surface of both parenchymal and nonparenchymal cells with an electrostatic force. On the other hand, MMC-D,, showed a long circulating life and was cleared slowly by the liver a t almost the same rate as that of poly(vinylpyrrolidone), which is known to be taken up by fluid-phase endocytosis.26In the present study, the disposition characteristics of MMC-D in the liver were studied at organ level using a single-pass perfusion technique. It was reported that MMC was inactivated rapidly by an acid-catalyzed27 or enzymatic28 degradation, while MMC-D was shown to be more stable than MMC.g The MMC-D conjugate was converted slowly to MMC at a physiological condition (pH 7.4,37 “C), with a half-life of 24 h, by chemical hydrolysis; enzyme did not accelerate this reaction.9 Therefore, MMC-D was considered to be stable in its form during liver perfusion experiments, and the behavior of intact MMC-D can be followed by spectrophotometrical measurements. We have established a n analytical method for drug disposition in an organ perfusion system by applying the statistical moment theoryz2and have examined the disposition of lipophilic and polymeric prodrugs of MMC in normal and tumorbearing muscles of The present liver perfusion system for MMC-D seemed to obey linear pharmacokinetics in the tested dose range because almost the same AUCi values were obtained a t different doses (data not shown). In this report, therefore, disposition parameters were used as the indices to evaluate the hepatic disposition characteristics of MMC-D. The dispersion ratio (di),which gives information about mixing within the system,ZZ was not derived since the second moment (2Jin the present experiments included more computational errors than the zero and first moments. In contrast to most other organs in which the capillary presents a substantial barrier between the vascular and interstitial spaces, the liver has the discontinuous endothelia1 capillaries which allow free contact to the surface of the hepatocytes of substances circulating in plasma. This anatomical characteristic explains the interaction of MMC-D with the surface of liver cells in vivo and in this perfusion study.
tL
0
w
v
30
15
0
TIME
60
45
(SEC)
Figure 5-Typical oufflow patterns of three types of MMC-D,,, (T-70) before and after applicationof EB/BSA in the liver perfusion experiment. Key: before fB/BSA (m), after €B/BSA ( O ) ,EB recovered with MMCoCal (T-70) (A).
Table IV. The disposition of MMC-DCat (T-70) was not changed by the pretreatment with EB/BSA. However, EB was recovered in the MMC-DCat(T-70) fractions (Figure 51, and the recovery, -4% of the dose, corresponded to -19% of the amount taken up by the liver during the first perfusion.
Discussion Our previous in vivol~JsJgand in vitroZ0studies revealed that MMC-DCatwas cleared rapidly from plasma and accumulated mostly in the liver due to the adsorption on the
Table IV-Moments and Disposition Parameters for MMC-D,,,(TdO) Albumin (BSA) In the Liver Perfusion Experiment
285.5 2 24.4a 290.9 40.4
Before EBIBSA After EBiBSA aMean t SD (n
0.310 0.380 0.933 2.07 0.275 0.338 0.368
*
=
3,4).
14 / Journal of Pharmaceutical Sciences
Vol. 78, No. 1, January 1989
11.54 f 2.71 12.28 1.13
*
before and after Application of Evans Blue (EB) with Bovine Serum
92.7 92.6
? 5
7.3 9.7
0.571 0.582
1.06 1.11
0.445 0.499
0.478 0.535
tained in a clinical trial involving intratumoral administraIt was demonstrated by indicator dilution analysis that the tion.13 Rapid clearance by the liver will be favorable to avoid labeled RBC space (sinusoidal blood volume) was 0.25 mL/g of liver and the extravascular space for albumin, which systemic side effects of MMC-D,,, when it is absorbed by the corresponds to the volume of the space of Disse, was 0.057 vascular system from the administration site. On the other mL/g in dogs.29 The total space of distribution for albumin hand, MMC-D,,,, with a long circulating life, will be suitable for systemic targeting to the tumor tissue with capillaries of was also reported to be 0.30 mWg of liver in rats.30These values agree well with our results (Table I). The recovery and enhanced permeability to macromolecules.~~ the distribution volume of MMC-Dansuggest that its bchavThus, the study of the mechanism of hepatic uptake of ior in the perfused liver is similar to that of 11311]HSA.FluidMMC-D at the organ level has demonstrated that the methphase endocytosis of MMC-Dan could not be detected in this odology in this study would be useful for evaluating the perfusion experiment since the rate of endocytosis is very hepatic disposition of not only macromolecular prodrugs, but low. On the contrary, MMC-D,,, was taken up by the liver also other types of drug delivery systems such as liposome giving larger CLi,t,i and V ivalues than MMC-Dan,and these and microspheres. values varied with molecular weight. More than 80% of injected MMC-D,,, (T-500) was trapped by the liver during single passage and the apparent distribution volume was extraordinarily large (1.08 mWg of liver). References and Notes In the repeated application, the hepatic uptake of MMC1. Poznansky, M. J.; Cleland, L. G. In Drug Delivery Systems; DCatdecreased in a stepwise manner (Figure 3, Table 11), but Juliano, R. L., Ed.; Oxford University: New York, 1980; pp 253the hepatic uptake of EWBSA increased in proportion to the 315. uptake amount of pretreated MMC-D,,, (Figure 4, Table 111). 2. Sezaki, H.; Hashida, M. CRC Crit. Rev. Ther. Drug Carrier Syst. Furthermore, EB taken up by the liver was recovered in the 1984,1, 1 3 8 . outflow when MMC-D,, was perfused afterward (Figure 5). 3. Keyes, S. R.; Rockwell, S.;Sartorelli, A. C. CancerRes. 1987,47, 5654-5657. These results suggest that perfused MMC-DaL adsorbed on 4. Sasaki, H.; Kakutani, T.; Hashida, M.; Sezaki, H. J . Pharm. Sci. the surface of hepatocytes based on the electrostatic force, as 1986, 75, 1166-1170. was shown in isolated hepatocytes.17 Evans blue (EB) with 5. Sezaki, H.; Hashida, M. In Pharmacokinetics: A Modern View: an anionic charge might interact with MMC-D,,, on the Benet, L. Z.; Levy, G.; Ferraiolo, B. L., Eds.; Plenum: New York, surface of hepatocytes. This interaction of anionic drugs with 1984; pp 345-358. 6. Kojima, T.; Hashida, M.; Muranishi, S.; Sezaki, H. J . Pharm. cationic polymer would offer a novel way for controlling the Pharmcol. 1980, 32, 30-34. in vivo disposition of the former. 7. Hashida, M.; Kato, A.; Ko'ima, T.; Muranishi, S.; Sezaki, H.; In our previous study,21muscle showed little permeability Tanigawa, N.; Satomura, Hikasa, Y. Gann 1981, 72, 226to MMC-D, regardless of molecular weight and electric 234. charge, since this organ has a continuous capillary endotheli8. Kato, A,; Takakura, Y.; Hashida, M.; Kimura, T.; Sezaki, H. Chem. Phurm. Bull. 1982,30, 2951-2957. um. In the tumor, increased permeability was observed, but 9. Hashida, M.; Takakura, Y.; Matsumoto, S.; Sasaki, H.; Kato, A,; the uptake amount was much less than that by the liver in Kojima, T.; Muranishi, S.; Sezaki, H. Chem. Pharm. BuLZ. 1983, a single passage. Finck et aL31reported that the single-pass 31,2055-2363. hepatic uptake (extraction) of lZ5I-labeledpolymeric immu10. Hashida, M.; Kato, A.; Takakura, Y.; Sezaki, H. Drug Metab. noglobulin A and [12511asialoorosomucoid,which are known Dispos. 1984.12, 492-499. 11. Takakura, Y.; Matsumoto, S.; Hashida, M.; Sezaki, H. Cancer to be trapped by a receptor-mediated mechanism, averaged Res. 1984,44, 2505-2510. 18.0 and 71.8%, respectively. Hepatic uptake of MMC-DCat 12. Matsumoto, S.; Arase, Y . ;Takakura, Y.; Hashida, M.; Sezaki, H. (T-5001, based on an electrostatic adsorption, is comparable Chem. Pharm. Bull. 1985,33, 2941-2947. to the latter. 13. Honda. K.: Satomura. K.: Hashida. M.: Sezaki. H. J D ~J. . Cancer Positively charged macromolecules, such as p r ~ t a m i n e , ~ ~ Chemother. (Tokyo) 1985, 12, 3111317. cationized ferriti11,3~ and p ~ l y l y s i n e , showed ~ ~ . ~ ~ high cellu14. Takakura, Y.; Kato, A,; Hashida, M.; Honda, K.; Arimoto, A,; Satomura, K.; Sezaki, H. J . Pharmucobio-Dyn. 1985, 8, 357lar interaction when cells were continuously exposed to the 364. medium containing them in vitro. It is of interest to examine 15. Matsumoto, S.; Yamamoto, A.; Takakura, Y.; Hashida, M.; their interaction with liver and explore the general feature of Tanieawa. N.: Sezaki. H. Cancer Res. 1986.46. 4463-4468. the in vivo fate of the polycationic compound. In our series of 16. Takaukura; Y.:Mori, K.; Hashida, M.; Sezaki, H. Chem. Phurm. investigations, diethylaminoethyl-dextran, which is a dexBull. 1986,34, 1775-1783. tran derivative with a positive charge, and cationized albu17. Takakura, Y.; Kitajima, M.; Matsumoto, S.; Hashida, M.; Sezaki, H. Znt. J . Pharm. 1987,37, 135-143. min showed remarkable hepatic uptake.36. 18. Takakura, Y.; Atsumi, R.; Hashida, M.; Sezaki, H. Znt. J . Pharm. The fate of MMC-D,,, on the cell surface of hepatocytes is 1987,37,145-154. not clear, but little seems to be internalized by the cells. The 19. Takakura, Y.; Takagi, A.; Hashida, M.; Sezaki, H. Pharm. Res. in vitro study demonstrated that a large part of radiolabeled 1987,4, 293-300. MMC-D,,, remained a t the membrane fraction of the tumor 20. Nakane, S.; Matsumoto, S.; Takakura, Y.; Hashida, M.; Sezaki, H. J . Pharm. Pharmacol. 1988,40, 1-6. cells even after 3 h of incubation.12 The fact that only 2% of 21. Atsumi, R.; Endo, K.; Kakutani, T.; Takakura, Y.; Hashida, M.; the dose was excreted in bile, in spite of a large accumulation Sezaki, H. Cancer Res. 1987,47, 5546-5551. in parenchymal cells in the in vivo experiment (unpublished 22. Kakutani, T.; Yamaoka, K.; Hashida, M.; Sezaki, H. J . P h a r m data), also supports this speculation. The MMC-D,,, conjucokinet. Biophurm. 1985,13, 609-631. gate would be retained on the surface of hepatocytes for a 23. Schaar, R. L.; Sparks, T. F.; Roseman, S. Anal. Biochem. 1977, considerably long time and release of MMC which would be 79, 513425. rapidly inactivated due to metabolic degradation. 24. Mortimore, G. E.; Tietze, F. Ann. N.Y. Acad. Sci. 1959,82, 329. 25. Mortimore, G. E.; Tietze, F.; Steffen, D., Jr. Diabetes 1959, 8, Considering tumor targeting by systemic administration, 307-314. MMC-D,,, seems to be disadvantageous because rapid hepat26. Munniksma, J.; Noteborn, M.; Kooistra, T.; Steinstra, S.; ic uptake will result in low comprehensive availability. In Bouma, J. M. W.; Gruber, M.; Brouwer, A,; Dalen, D. P. V.; local injection, however, MMC-D,,, showed sustained retenKnook, D. L. Biochem. J . 1980, 292, 613-621. tion a t the injection site and enhanced lymphatic deliv27. Underberg, W. J. M.; Lingeman, H. J. Pharm. Sci. 1983, 72, 549-553. ery,11J6 and remarkable reduction in tumor size was ob-
k.;
I
.
Journal of Pharmaceutical Sciences / 15 Vol. 78, No. 1, January 1989
28. Schwartz, H. S. J. Pharmacol. Exp. Ther. 1962,136, 250-258. 29. Goresky, C. A. Fed. Proc. 1982,41, 3033-3039. 30. Stollman, Y. R.; Gartner, G.; Theilmann, L.; Ohmi, N.;Wolkoff, A. W. J. Clin. Invest. 1983, 72, 118-723. 3 1. Finck, M.H.; Reichen, J.; Vierling, J. M.; Kloppel, T. M.; Brown, W. R. A m . J . Physiol. 1985,248, G450-455. 32. Becker, F. F.; Green, H. Exp. Cell Res. 1960,19, 361375. 33. Farquhar, M. G. J . Cell B i d . 1978, 77, R35-R42. 34. Korgnguth, S. E.; Stahman, M. A.; Anderson, J. W. Exp. Cell Res. 1961,24, 484-494.
16 / Journal of Pharmaceutical Sciences Vol. 78, No. 1, January 1989
35. Nevo, A.; Vries, A.; Katchalsky, A. Biochim. Biophys. Actu 1955, 17,536-547. 36. Takakura, Y.; Nishida, K.; Hashida, M.; Sezaki, H., unpublished results.
Acknowledgments A part of this work was supported by Grant-in-Aid from the Mochida Memorial Foundation for Medical and Pharmaceutical Research.
KOYO
NISHIDA,YOSHINOBU TAKAKURA, MITSURUHASHIDA,
AND
Received April 20, 1988, from the Department of Basic Pharmaceufics, Faculty of Pharmaceutical Sciences, Kyoto University, Accepted for publication August 15. 1988. Sakyo-ku, Kyoto 606, Japan. -~ ~~~~
Abstract 0 The disposition of a polymeric prodrug of mitomycin C
(MMC), mitomycin C-dextran conjugate (MMC-D), was studied in the single-pass perfused rat liver in order to clarify the effect of physico-
chemical properties, such as molecular weight and electric charge, on the hepatic uptake of MMC-D. Six types of MMC-D were used: both cationic MMC-D (MMC-D,,,) and anionic MMC-D (MMC-Dan) conjugated with dextran with molecular weights of 10,000,70,000, and 500,000. Outflow curves were analyzed using statistical moment theory. Remarkable hepatic uptake of MMC-D,,, was observed and the uptake amount increased with an increase in molecular weight (i.e.,-80% of the dose was taken up by the liver during a single passage of the conjugate with a molecular weight of 500,000). Intrinsic clearance ( CLint,i)and apparent distribution volume (VJ also increased as the molecular weight increased. On the other hand, almost 100% of applied MMC-Dan was recovered in the outflow regardless of molecular weight, with almost the same moment parameters as those of the vascular reference substance (VRS), '311-labeledhuman serum albumin (HSA).In a repeated application, the uptake of MMC-D,., decreased in a stepwise manner, suggesting a saturation in the hepatic uptake of MMC-D,,, while the uptake of MMC-D,, was unchanged. The MMC-D,,, pretreatment also affected the uptake of Evans blue (EB) bound to bovine serum albumin (BSA). These results demonstrate that molecular weight and electric charge determine the hepatic disposition of macromolecular prodrugs.
One of the major limitations of cancer chemotherapy is that most antitumor drugs have indiscriminate action on both tumor and normal cells. Therefore, a n approach to modify their pharmacokinetic properties to concentrate their action a t the target cells would be promising.' Mitomycin C (MMC) is a highly active antibiotic in a variety of human neoplastic disease, but its use is limited by side effects such as severe myelosuppression and gastrointestinal damage. To overcome these disadvantages and enhance therapeutic efficacy, the utilization of analogues3 and lipophilic4 or polymeric5 prodrugs of MMC has been reported. In our series of investigations, we have developed polymeric prodrugs of MMC [MMC-dextran conjugates having cationic (MMC-D,,J and anionic (MMC-D,,) charges], and have examined their physicochemical, pharmacodynamic, and pharmacokinetic characteristics.621 These studies revealed that MMC-D both acts as a reservoir of MMC that behaves characteristically as a macromolecule, supplying active MMC in the body,'0J6 and has higher activity than MMC against various murine tumors7 and human malignant tumor~.~~ The in vivo disposition study demonstrated that MMC-D,,, was rapidly cleared from plasma and accumulated in the liver and spleen to a great extent, while MMC-Dan showed slow plasma clearance and accumulated gradually in the liver, spleen, and lymph nodes following intravenous inject i ~ n . ~ The ~ Jin~ vitro study showed that MMC-DCatwas highly adsorbed on the surface of hepatocytes; however, the 0022-3549/89/0 100-001 1 $0 1.OO/O 0 1989, American Pharmaceutical Association
interaction between MMC-D, and hepatocytes was negligible.zO In the present study, hepatic disposition of MMC-D was studied using the isolated perfused rat liver in order to clarify the effect of physicochemical properties of the conjugates, such as electric charge and molecular weight, on the clearance by the reticuloendothelial organs. The MMC-D conjugate was introduced as a pulse function from the portal vein and venous outflow patterns were analyzed by statistical moment analysis to derive the disposition parameters.22 Furthermore, the effect of MMC-D treatment on the disposition of a model anionic drug, Evans blue (EB), bound to albumin was studied.
Experimental Section Chemicals-Mitomycin C (MMC) was kindly supplied by Kyowa Hakko Kogyo (Tokyo). Dextrans with various molecular weights were purchased from Pharmacia (Uppsala, Sweden), and had average molecular weights of -10,000 (T-10),70,000(T-701,and 500,000 (T-500). I3'I-Labeled human serum albumin (HSA; 100 pCi/mg) was purchased from Daiichi Radioisotopes (Tokyo). Bovine serum albumin (BSA; fraction V) was from A m o u r Pharmaceutical (U.K.), Evans blue (EB) was from Merck (Darmstadt, Germany), and Na,["Cr]04 (536 pCi/mg) was obtained from New England Nuclear (Boston, MA). Red blood cells (RBC) were labeled by incubation with Na2[51Cr]04a t 37 "C for 20 min and suspended in saline. All other chemicals were commercial, reagent-grade products. Preparation of MMC-D,., and MMC-Dan-The MMC-DCatand MMC-D,, conjugates were synthesized as reported p r e v i ~ u s l y . ~ J ~ For the synthesis of MMC-D,,,, dextran was activated by cyanogen bromide, a spacer (E-aminocaproic acid) was introduced, and then MMC was coupled by a carbodiimide-catalyzed reaction. Three structures were proposed for the linkage between the spacer and the dextran,TJ and two of them have the possibility to give a cationic charge in MMC-D,,,. For the synthesis of MMC-Dan,6-bromohexanoic acid was introduced to dextran as a spacer and then MMC was coupled as above. Since not all spacer arms were used for the coupling of MMC, the remaining free carboxyl groups give anionic charges in MMC-Dan.All conjugates were estimated to contain MMC to almost equal extents of -10% for MMC-D,,, and 8% (wr'w) for MMC-Dan.The MMC-D conjugates were prepared in saline solution at a concentration of 5 mg equivalent MMC/mL for injection. Analytical Method-The concentration of MMC-D and EB were determined spectrophotometrically a t 364 and 620 nm, respectively. In the case of free MMC, the concentration was determined a t 364 nm. The I3'I and 51Crradioactivities were counted with a well NaI scintillation counter. Single-Pass Perfusion of Rat Liver-Male Wistar albino rats (170-200 g) were obtained from Shizuoka Agricultural Co-operative Association for Laboratory Animals (Shizuoka, Japan). Perfusion of the rat liver was carried out using the method of Mortimore et al.24.25 with slight modifications. The rats were anesthetized by intraperitoneal injection of sodium pentobarbital (60 mgkg), the abdomen and chest were opened, and the portal vein and the inferior vena cava were cannulated with a polyethylene tubing (PE-160). Krebs Ringer Journal of Pharmaceutical Sciences 111 Vol. 78, No. 1, January 1989
bicarbonate buffer (pH 7.4) was aerated with 95% 02:5% C 0 2 and pumped at 37 "C into the cannulate portal vein with the aid of a peristaltic pump a t a rate of 12 mlimin. The bile duct was cannulated with polyethylene tubing (PE-10) and bile was collected every 10 min during experiments to check the viability of the liver. Bile production was almost constant in the range 3-8 pL/min. After a n equilibration period of 30 min, test compounds dissolved in saline (0.1328 mL) were injected into the line of perfusion flow as a pulse function using a six-position rotary valve injector (type 50 teflon rotary valves, Rheodyne, Cotati, CA). The outflow perfusate was collected into weighed test tubes at appropriate time intervals ( 1 3 s). The sample volume was calculated from the gain in weight of the tube, assuming a density of the outflow perfusate of 1.0. The sample was subjected to assay after an appropriate dilution with the perfusion buffer. In the case of repeated application, the interval between injections was 15 min. Data Analysis-Moment analysis was used for analyzing the outflow pattern. A detailed theoretical description of this analysis was given previously.21.22The first three (zero to second) statistical moment parameters (moments) are defined as follows:
LL
0 $.p Y
z
0 + a
Dc
+ z
w
0
z
0
0
0
15
45
30
60
T I M E( S E C ) AUCi
=
Cdt
(1)
-ti = [tCdtiAUCi
(t - ?i)2Cdt/AUCi
17: =
Figure 1-Typical outflow patterns of three types of MMC-D,,,, ['3'uHSA in the liver perfusion experiment. Key: (0)MMC-D,,, (T-10); (U) MMC-Dca, (T-70); (A) MMC-Dcat (T-500);( 0 )['3'IlHSA.
64
(3)
r
h
where t is the time, C is the concentration of compounds normalized by the injection dose and has a dimension of 5% of dose/mL, and AUCi, ?,, and r2i,are the area under the concentration-time curve, mean transit time of the drug through the liver, and variance of transit time, respectively. These moments are calculated by numerical integration using a linear trapezoidal formula from the outflow concentration-time curve. Disposition parameters are calculated from these moments as follows:
z --. 40 J
0
0 I&
0
be
z
32
0 L I
(4)
k Dc
AUCiQii100
(5)
E 16
te,,i = t i / ( l - FJ
(6)
0
CLjnt,i =
(7)
Vi Fi
= ti/AUCi100 =
-
vjfie*,i
0
z
where Vi is the apparent distribution volume, Fi is the recovery ratio, Q, is the flow rate, tel,i is the mean elimination time, and CL,,,, is the intrinsic clearance. These parameters can be divided into two groups [i.e.,those representing distribution (V,) and those representCLint,J]. ing elimination (Fi,
ResuIts Perfusion of MMC-D,,,, MMC-D,., [l3lI1HSA, [SICrlRBC, and Free MMC-Figures 1 and 2 show the representative outflow patterns of three types of MMC-D,,, and MMC-Dan together with that of [1311]HSA in a single-pass liver perfusion. "Cr-Labeled RBC was the earliest to appear in the hepatic venous outflow, and its concentration rose to the highest and earliest peak and decayed most rapidly (data not shown). The appearance of [1311]HSA was a little slower and the peak value was a little lower than that of [51CrlRBC. Every MMC-D,,, conjugate had a lower peak than [l3lI1HSA, and the peak value decreased with an increase of molecular weight (Figure 1).On the contrary, MMC-Danshowed almost the same outflow patterns as that of ['311]HSA, regardless of molecular weight (Figure 2). The moments, AUCi, ti, and u2i were calculated from these outflow patterns and the disposition parameters were derived from the moments. Table I summarizes the moments and disposition parameters for [51Cr]RBC,[1311]HSA,MMC-IIfat, MMC-Dan,and MMC. The recoveries of [51CrlRBCand [ IIHSA were almost 100% and 12 / Journal of Pharmaceutical Sciences Vol. 78, No. I , January 7989
+
0
0
15
30 TIME
45
60
(SEC)
Figure 2-Typical outflow patterns of three types of MMC-Dan, ['3'l]HSA in the liver perfusion experiment. Key: (0)MMC-Dan (T-10); ( 0 )MMC-Dan (T-70); (A)MMC-0," (T-500); ( 0 )['"I]HSA.
the V, were 0.209 and 0.252 m u g liver, respectively,As the molecular weight of MMC-D,,, increased, F, and tel,l decreased and CLlnt,, and V, increased. On the other hand, parameters of MMC-Dan were similar to those of [13111HSA, irrespective of molecular weight. In the case of free MMC, -35% of the dose was extracted by the liver (Table I). Repeated Application of MMC-D-Figure 3 shows the outflow patterns of MMC-D,,, (T-500) applied repeatedly three times to the perfused rat liver. The peak value was increased in a stepwise manner. Moments and disposition parameters for MMC-DCat(T-500) in these runs, together with those of MMC-DCat(T-10) and MMC-Dan (T-701, are summarized in Table 11. For MMC-D,,, (T-5001, F, was increased and CLlnt,,and V, were decreased successively, and the same tendency was observed in the case of MMC-D,,, (T10). However, there was no significant difference between the parameters of MMC-Dan (T-70) in the first and the second applications.
Table I-Moments Experiment
and Disposition Parameters for [51Cr]RBC,['311]HSA, MMC-D,,,, AUC,, % of dosesiml
Compound
471.3 485.9 396.1 285.5 89.3 508.0 505.4 485.3 325.9
RBC HSA MMC-Dcat(T-l0) MMC-Dca,(T-70) MMC-D-t(T-500) MMC-Dan(T-l0) MMC-Dan(T-70) MMC-Dan(T-500) MMC aMean 2 SD (n
28
=
f
f, s
24.5a
f 14.3 2 9.5
t 24.4 2 16.2 2 21.6 f 13.7 t 22.4 t 10.2
MMC-Dan,and MMC in the Liver Perfusion
s2
UZI,
6 1 .o 1.o 0.81 1 0.571 0.180 1.o 1.o 0.979 0.654
42.1 +- 14.3 36.8 f 17.5 59.8 2 9.6 92.7 f 7.3 70.7 2 33.0 55.6 ? 23.7 53.3 t 5.3 54.5 t 7.1 9.9 47.1
8.89 t 0.83 9.33 2 1.09 12.01 ? 1.35 115 4 & 2.71 8.22 -t 2.59 9.51 ? 1.80 9.75 t 2.15 8.68 f 1.28 8.88 2 0.49
*
CL",,,,
tel,bv
mllminig
min
-
-
0.313 1.06 6.56 0 0 0.031 0.682
1.06 0.445 0.167
-
413 0.428
v,
mLig 0.209 0.252 0.331 0.478 1.08 0.277 0.236 0.229 0.292
3-9).
r L
0
z
0,
14
z
0
I-
<
I-
I K -
<
i
L
K
W
z0
E 11 0
7
z
0
0
0
15
0
30 TIME
45
60
Effect of Preperfusion of MMC-D on t h e Disposition of EB/BSA-Figure 4 illustrates the outflow patterns of EB injected with BSA (4.7%, w/v) after treatment with MMCDcat,and the calculated moments and disposition parameters are shown in Table 111. In the case of a single injection of EB/BSA, the recovery of EB was -82%. After MMC-DCat treatment, the recovery of EB was decreased depending on the uptake amount of MMC-DCat(i.e.,the larger the molecuand Disposition Parameters for MMC-D,,, AUC,, Yo of doseslrnl
Compound MMC-Dca,(T.,500) 1st 2nd 3rd MMC-Dcat(T-10) 1st 2nd MMC-Dan(T-70) 1st 2nd 'Mean t SD (n
=
45
60
TIME (SEC)
(SEC)
Figure 3- Typicd outflow patterns of MMC-D,,, (T-500) applied repeatedly three times in the liver perfusion experiment. Key: firsf perfusion (O ),second perfusion (0), third perfusion (A).
Table Il-Moments
30
15
0
Figure 4-Typicai outf/ow patterns of EB injected with BSA after preperfusion of three types of MMC-Dcaiin the liver perfusion experiment. Key: (a)no freatment; (0)MMC-Dca,(T-10); (0) MMC-Dca,(T70); (A)MMC-Dcai (T-500).
lar weight of MMC-D,,,, the less the recovery of EB in the outflow). On the contrary, MMC-D,, treatment did not affect the disposition of EB/BSA. Effect of EB/BSA on the Disposition of MMC-D-Figure 5 shows the outflow patterns of MMC-D,,, (T-70) before and after EBiBSA treatment in the liver perfusion experiment. Moments and disposition parameters are summarized in
and MMC-Dan in the Liver Perfusion Experiment UZ,,
v,,
s2
*
mLig
89.3 t 16.2a 181.6 f 8.7 236.8 ? 5.3
8.22 ? 2.59 8.74 2 3.77 11.03 t 2.31
70.7 33.0 73.0 f 10.2 72.2 2 12.8
0.180 0.366 0.474
6.56 2.29 1.16
0.167 0.232 0.380
1.08 0.505 0.442
396.1 f 9.5 429.1 2 10.1
12.01 5 1.35 15.09 f 0.75
59.8 t 9.6 90.7 t 3.5
0.81 1 0.873
0.313 0.199
1.06 1.99
0.331 0.395
505.4 f 13.7 506.8 k 15.0
9.75 t 2.15 8.35 k 1.14
53.3 t 5.3 53.1 2 8.4
1.o 1 .o
0 0
-
0.236 0.192
-
3-9).
Journal of Pharmaceutical Sciences / 13 Vol. 78, No. 7, January 1989
Table ill-Moments and Disposition Parameters for Evans Blue (EB) Injected with Bovine Serum Albumin (BSA) after Preperfusion of MMC-D,., and MMC-D,, in the Llver Perfusion Experiment Treatment
AUCi, % Of dosesiml
None MMC-Dcat(T-l0) MMC-D-,(T-70) MMC-D-t(T-500) MMC-Dan(T-l0) MMC-Dan(T-70) MMC-Dan(T-500)
410.2 ? 6.2a 393.2 5 9.5 299.3 2 21.7 206.5 -t 25.2 419.1 5 13.7 398.6 ? 1097 389.1 f 6.74
aMean 2 SD (n
=
10.12 2 1.05 10.48 -+ 2.08 9.52 2 2.06 8.78 k 1.41 9.20 ? 0.79 11.30 f 0.11 10.07 -t 0.81
up,, s2
F,
58.6 ? 11.9 61.7? 4.7 55.5 ? 7.2 53.1 2 4.50 62.7 ? 14.2 70.1 ? 10.6 60.2 2 13.0
0.824 0.788 0.598 0.418 0.839 0.795 0.792
0.960 0.825 0.392 0.253 0.952 0.919 0.809
0.297 0.313 0.371 0.519 0.260 0.31 1 0.297
3-6).
surface of both parenchymal and nonparenchymal cells with an electrostatic force. On the other hand, MMC-D,, showed a long circulating life and was cleared slowly by the liver a t almost the same rate as that of poly(vinylpyrrolidone), which is known to be taken up by fluid-phase endocytosis.26In the present study, the disposition characteristics of MMC-D in the liver were studied at organ level using a single-pass perfusion technique. It was reported that MMC was inactivated rapidly by an acid-catalyzed27 or enzymatic28 degradation, while MMC-D was shown to be more stable than MMC.g The MMC-D conjugate was converted slowly to MMC at a physiological condition (pH 7.4,37 “C), with a half-life of 24 h, by chemical hydrolysis; enzyme did not accelerate this reaction.9 Therefore, MMC-D was considered to be stable in its form during liver perfusion experiments, and the behavior of intact MMC-D can be followed by spectrophotometrical measurements. We have established a n analytical method for drug disposition in an organ perfusion system by applying the statistical moment theoryz2and have examined the disposition of lipophilic and polymeric prodrugs of MMC in normal and tumorbearing muscles of The present liver perfusion system for MMC-D seemed to obey linear pharmacokinetics in the tested dose range because almost the same AUCi values were obtained a t different doses (data not shown). In this report, therefore, disposition parameters were used as the indices to evaluate the hepatic disposition characteristics of MMC-D. The dispersion ratio (di),which gives information about mixing within the system,ZZ was not derived since the second moment (2Jin the present experiments included more computational errors than the zero and first moments. In contrast to most other organs in which the capillary presents a substantial barrier between the vascular and interstitial spaces, the liver has the discontinuous endothelia1 capillaries which allow free contact to the surface of the hepatocytes of substances circulating in plasma. This anatomical characteristic explains the interaction of MMC-D with the surface of liver cells in vivo and in this perfusion study.
tL
0
w
v
30
15
0
TIME
60
45
(SEC)
Figure 5-Typical oufflow patterns of three types of MMC-D,,, (T-70) before and after applicationof EB/BSA in the liver perfusion experiment. Key: before fB/BSA (m), after €B/BSA ( O ) ,EB recovered with MMCoCal (T-70) (A).
Table IV. The disposition of MMC-DCat (T-70) was not changed by the pretreatment with EB/BSA. However, EB was recovered in the MMC-DCat(T-70) fractions (Figure 51, and the recovery, -4% of the dose, corresponded to -19% of the amount taken up by the liver during the first perfusion.
Discussion Our previous in vivol~JsJgand in vitroZ0studies revealed that MMC-DCatwas cleared rapidly from plasma and accumulated mostly in the liver due to the adsorption on the
Table IV-Moments and Disposition Parameters for MMC-D,,,(TdO) Albumin (BSA) In the Liver Perfusion Experiment
285.5 2 24.4a 290.9 40.4
Before EBIBSA After EBiBSA aMean t SD (n
0.310 0.380 0.933 2.07 0.275 0.338 0.368
*
=
3,4).
14 / Journal of Pharmaceutical Sciences
Vol. 78, No. 1, January 1989
11.54 f 2.71 12.28 1.13
*
before and after Application of Evans Blue (EB) with Bovine Serum
92.7 92.6
? 5
7.3 9.7
0.571 0.582
1.06 1.11
0.445 0.499
0.478 0.535
tained in a clinical trial involving intratumoral administraIt was demonstrated by indicator dilution analysis that the tion.13 Rapid clearance by the liver will be favorable to avoid labeled RBC space (sinusoidal blood volume) was 0.25 mL/g of liver and the extravascular space for albumin, which systemic side effects of MMC-D,,, when it is absorbed by the corresponds to the volume of the space of Disse, was 0.057 vascular system from the administration site. On the other mL/g in dogs.29 The total space of distribution for albumin hand, MMC-D,,,, with a long circulating life, will be suitable for systemic targeting to the tumor tissue with capillaries of was also reported to be 0.30 mWg of liver in rats.30These values agree well with our results (Table I). The recovery and enhanced permeability to macromolecules.~~ the distribution volume of MMC-Dansuggest that its bchavThus, the study of the mechanism of hepatic uptake of ior in the perfused liver is similar to that of 11311]HSA.FluidMMC-D at the organ level has demonstrated that the methphase endocytosis of MMC-Dan could not be detected in this odology in this study would be useful for evaluating the perfusion experiment since the rate of endocytosis is very hepatic disposition of not only macromolecular prodrugs, but low. On the contrary, MMC-D,,, was taken up by the liver also other types of drug delivery systems such as liposome giving larger CLi,t,i and V ivalues than MMC-Dan,and these and microspheres. values varied with molecular weight. More than 80% of injected MMC-D,,, (T-500) was trapped by the liver during single passage and the apparent distribution volume was extraordinarily large (1.08 mWg of liver). References and Notes In the repeated application, the hepatic uptake of MMC1. Poznansky, M. J.; Cleland, L. G. In Drug Delivery Systems; DCatdecreased in a stepwise manner (Figure 3, Table 11), but Juliano, R. L., Ed.; Oxford University: New York, 1980; pp 253the hepatic uptake of EWBSA increased in proportion to the 315. uptake amount of pretreated MMC-D,,, (Figure 4, Table 111). 2. Sezaki, H.; Hashida, M. CRC Crit. Rev. Ther. Drug Carrier Syst. Furthermore, EB taken up by the liver was recovered in the 1984,1, 1 3 8 . outflow when MMC-D,, was perfused afterward (Figure 5). 3. Keyes, S. R.; Rockwell, S.;Sartorelli, A. C. CancerRes. 1987,47, 5654-5657. These results suggest that perfused MMC-DaL adsorbed on 4. Sasaki, H.; Kakutani, T.; Hashida, M.; Sezaki, H. J . Pharm. Sci. the surface of hepatocytes based on the electrostatic force, as 1986, 75, 1166-1170. was shown in isolated hepatocytes.17 Evans blue (EB) with 5. Sezaki, H.; Hashida, M. In Pharmacokinetics: A Modern View: an anionic charge might interact with MMC-D,,, on the Benet, L. Z.; Levy, G.; Ferraiolo, B. L., Eds.; Plenum: New York, surface of hepatocytes. This interaction of anionic drugs with 1984; pp 345-358. 6. Kojima, T.; Hashida, M.; Muranishi, S.; Sezaki, H. J . Pharm. cationic polymer would offer a novel way for controlling the Pharmcol. 1980, 32, 30-34. in vivo disposition of the former. 7. Hashida, M.; Kato, A.; Ko'ima, T.; Muranishi, S.; Sezaki, H.; In our previous study,21muscle showed little permeability Tanigawa, N.; Satomura, Hikasa, Y. Gann 1981, 72, 226to MMC-D, regardless of molecular weight and electric 234. charge, since this organ has a continuous capillary endotheli8. Kato, A,; Takakura, Y.; Hashida, M.; Kimura, T.; Sezaki, H. Chem. Phurm. Bull. 1982,30, 2951-2957. um. In the tumor, increased permeability was observed, but 9. Hashida, M.; Takakura, Y.; Matsumoto, S.; Sasaki, H.; Kato, A,; the uptake amount was much less than that by the liver in Kojima, T.; Muranishi, S.; Sezaki, H. Chem. Pharm. BuLZ. 1983, a single passage. Finck et aL31reported that the single-pass 31,2055-2363. hepatic uptake (extraction) of lZ5I-labeledpolymeric immu10. Hashida, M.; Kato, A.; Takakura, Y.; Sezaki, H. Drug Metab. noglobulin A and [12511asialoorosomucoid,which are known Dispos. 1984.12, 492-499. 11. Takakura, Y.; Matsumoto, S.; Hashida, M.; Sezaki, H. Cancer to be trapped by a receptor-mediated mechanism, averaged Res. 1984,44, 2505-2510. 18.0 and 71.8%, respectively. Hepatic uptake of MMC-DCat 12. Matsumoto, S.; Arase, Y . ;Takakura, Y.; Hashida, M.; Sezaki, H. (T-5001, based on an electrostatic adsorption, is comparable Chem. Pharm. Bull. 1985,33, 2941-2947. to the latter. 13. Honda. K.: Satomura. K.: Hashida. M.: Sezaki. H. J D ~J. . Cancer Positively charged macromolecules, such as p r ~ t a m i n e , ~ ~ Chemother. (Tokyo) 1985, 12, 3111317. cationized ferriti11,3~ and p ~ l y l y s i n e , showed ~ ~ . ~ ~ high cellu14. Takakura, Y.; Kato, A,; Hashida, M.; Honda, K.; Arimoto, A,; Satomura, K.; Sezaki, H. J . Pharmucobio-Dyn. 1985, 8, 357lar interaction when cells were continuously exposed to the 364. medium containing them in vitro. It is of interest to examine 15. Matsumoto, S.; Yamamoto, A.; Takakura, Y.; Hashida, M.; their interaction with liver and explore the general feature of Tanieawa. N.: Sezaki. H. Cancer Res. 1986.46. 4463-4468. the in vivo fate of the polycationic compound. In our series of 16. Takaukura; Y.:Mori, K.; Hashida, M.; Sezaki, H. Chem. Phurm. investigations, diethylaminoethyl-dextran, which is a dexBull. 1986,34, 1775-1783. tran derivative with a positive charge, and cationized albu17. Takakura, Y.; Kitajima, M.; Matsumoto, S.; Hashida, M.; Sezaki, H. Znt. J . Pharm. 1987,37, 135-143. min showed remarkable hepatic uptake.36. 18. Takakura, Y.; Atsumi, R.; Hashida, M.; Sezaki, H. Znt. J . Pharm. The fate of MMC-D,,, on the cell surface of hepatocytes is 1987,37,145-154. not clear, but little seems to be internalized by the cells. The 19. Takakura, Y.; Takagi, A.; Hashida, M.; Sezaki, H. Pharm. Res. in vitro study demonstrated that a large part of radiolabeled 1987,4, 293-300. MMC-D,,, remained a t the membrane fraction of the tumor 20. Nakane, S.; Matsumoto, S.; Takakura, Y.; Hashida, M.; Sezaki, H. J . Pharm. Pharmacol. 1988,40, 1-6. cells even after 3 h of incubation.12 The fact that only 2% of 21. Atsumi, R.; Endo, K.; Kakutani, T.; Takakura, Y.; Hashida, M.; the dose was excreted in bile, in spite of a large accumulation Sezaki, H. Cancer Res. 1987,47, 5546-5551. in parenchymal cells in the in vivo experiment (unpublished 22. Kakutani, T.; Yamaoka, K.; Hashida, M.; Sezaki, H. J . P h a r m data), also supports this speculation. The MMC-D,,, conjucokinet. Biophurm. 1985,13, 609-631. gate would be retained on the surface of hepatocytes for a 23. Schaar, R. L.; Sparks, T. F.; Roseman, S. Anal. Biochem. 1977, considerably long time and release of MMC which would be 79, 513425. rapidly inactivated due to metabolic degradation. 24. Mortimore, G. E.; Tietze, F. Ann. N.Y. Acad. Sci. 1959,82, 329. 25. Mortimore, G. E.; Tietze, F.; Steffen, D., Jr. Diabetes 1959, 8, Considering tumor targeting by systemic administration, 307-314. MMC-D,,, seems to be disadvantageous because rapid hepat26. Munniksma, J.; Noteborn, M.; Kooistra, T.; Steinstra, S.; ic uptake will result in low comprehensive availability. In Bouma, J. M. W.; Gruber, M.; Brouwer, A,; Dalen, D. P. V.; local injection, however, MMC-D,,, showed sustained retenKnook, D. L. Biochem. J . 1980, 292, 613-621. tion a t the injection site and enhanced lymphatic deliv27. Underberg, W. J. M.; Lingeman, H. J. Pharm. Sci. 1983, 72, 549-553. ery,11J6 and remarkable reduction in tumor size was ob-
k.;
I
.
Journal of Pharmaceutical Sciences / 15 Vol. 78, No. 1, January 1989
28. Schwartz, H. S. J. Pharmacol. Exp. Ther. 1962,136, 250-258. 29. Goresky, C. A. Fed. Proc. 1982,41, 3033-3039. 30. Stollman, Y. R.; Gartner, G.; Theilmann, L.; Ohmi, N.;Wolkoff, A. W. J. Clin. Invest. 1983, 72, 118-723. 3 1. Finck, M.H.; Reichen, J.; Vierling, J. M.; Kloppel, T. M.; Brown, W. R. A m . J . Physiol. 1985,248, G450-455. 32. Becker, F. F.; Green, H. Exp. Cell Res. 1960,19, 361375. 33. Farquhar, M. G. J . Cell B i d . 1978, 77, R35-R42. 34. Korgnguth, S. E.; Stahman, M. A.; Anderson, J. W. Exp. Cell Res. 1961,24, 484-494.
16 / Journal of Pharmaceutical Sciences Vol. 78, No. 1, January 1989
35. Nevo, A.; Vries, A.; Katchalsky, A. Biochim. Biophys. Actu 1955, 17,536-547. 36. Takakura, Y.; Nishida, K.; Hashida, M.; Sezaki, H., unpublished results.
Acknowledgments A part of this work was supported by Grant-in-Aid from the Mochida Memorial Foundation for Medical and Pharmaceutical Research.