Polymeric Drugs and Drug Administration - American Chemical Society

Chapter 17

Use of α-1,4-Polygalactosamine as a Carrier of Macromolecular Prodrug of 5-Fluorouracil Tatsuro Ouchi, Keigo Inosaka, and Yuichi Ohya

Downloaded by FUDAN UNIV on March 21, 2017 | http://pubs.acs.org Publication Date: May 5, 1994 | doi: 10.1021/bk-1994-0545.ch017

Department of Applied Chemistry, Faculty of Engineering, Kansai University, Suita, Osaka 564, Japan

α-1,4-Polygalactosamine (PGA) and N-acetyl-α-1,4-polygalactosamine ( N A P G A ) have been reported to be low toxic and biodegradable polymers and to act as growth inhibitors of certain tumor cells through stimulation of the host immune system. Moreover, these polysaccharides can be expected to have affinities for hepatocytes. A water soluble macromolecular prodrug of 5-fluorouracil (5FU) with affinity for tumor cells and without side-effects was obtained as CM-NAPGA/amide/pentamethylene/ester/monomethylene/5FU conjugate. The release behavior of 5 F U from this conjugate and its antitumor activity were investigated in vitro at 37°C in physiological saline. The release of free 5 F U was observed, while no 5 F U derivative was detected. The cytotoxicity of C M - N A P G A / a m i d e /pentamethylene/ester/monomethylene/5FU conjugate was lower than that of free 5 F U against p388D1 lymphocytic leukemia, while the conjugate was more potent than the free 5 F U against HLE hepatoma cells in vitro. The administration of CM-NAPGA/amide/pentamethylene/ester/monomethylene/5FU conjugate via intraperitoneal (i.p.) transplantation/i.p. injection in mice resulted in a significant survival effect against p388 lymphocytic leukemia. The conjugate in the high dose range caused neither rapid decrease of body weight of the treated mice, nor acute toxicity in the treated mice.

5-Fluorouracil (5FU) has a remarkable antitumor activity which is accompanied, however, by undesirable side-effects. a-l,4-Poly-galactosamine ( P G A ) is a novel basic polysaccharide purified from the culture fluid of Paecilomyces sp. 1-1 and free from acute toxicity (1). P G A and N-acetyl-a-l,4-polygalactosamine ( N A P G A ) are non-toxic, non-immunogenetic biodegradable polymers. These polymers have been reported to act as growth inhibitors of certain tumors through stimulation of the host immune system (2). Similar to galactose, galactosamine has an affinity for hepatocytes (3) and can be used as a targeting moiety for hepatoma cells (4).

0097-6156/94/0545-0204$08.00/0 © 1994 American Chemical Society

Ottenbrite; Polymeric Drugs and Drug Administration ACS Symposium Series; American Chemical Society: Washington, DC, 1994.


17.

OUCHI ET AL.

PGA as a Carrier of Macromolecular Prodrug of 5FU

205

Thus, P G A and N A P G A are potential drug carriers with targetability to hepatocytes. Previously, we synthesized P G A , its oligomer (GOS), and N A P G A linking 5 F U via urea bonds. We prepared PGA/urea/hexamethylene/urea/5FU, GOS/urea/hexamethylene/urea/5FU, and NAPGA/urea/hexamethylene/ urea/5FU conjugates which significantly increased the survival effects against p388 lymphocytic leukemia in mice by intraperitoneal (i.p.) transplantation/i.p. injection in vivo and exhibited strong growth inhibition effects against Meth-A fibrosarcoma in mice by subcutaneous implantation/intravenous injection in vivo. However, the poor solubility of these compounds in water (5) limited their therapeutic effect. In order to obtain a water-soluble macromolecular prodrug of 5 F U , we prepared N-carboxymethyl-N-acetyl-a-l,4-polygalactosamine ( N - C M - N A P G A ; M n = 3000-10000) and 0-carboxymethyl-N-acetyl-a-l,4-polygalactosamine ( O - C M N A P G A ; M n = 3000-10000) as drug carriers. The present paper is concerned with the syntheses and antitumor activities of N-CM-NAPGA/amide/pentamethylene/ ester/methylene/5FU conjugate 5 and O-CM-NAPGA/amide/pentamethylene/ ester/methylene/5FU conjugate 6. Experimental Materials. a-l,4-Polygalactosamine (PGA) from supernatant fluid of Paecilomyces sp. 1-1 strain (6,7) was supplied by Higeta Shoyu Co. Ltd. The supplied P G A was treated with polygalactosaminidase isolated from Pseudomonas sp. 881 to give low molecular weight P G A . The low molecular weight P G A obtained was purified by ultrafiltration (cut off Mw 3000 and 10000) and the degree of polymerization of the P G A was estimated to be 20 to 60. N - C M - N A P G A and O - C M - N A P G A were prepared according to the method reported previously (8). The molecular weight of N - C M - N A P G A and O - C M N A P G A was determined by G P C (column: Tosoh TSKgel G M P W ; eluent: 0.1 m m o l / L acetic acid buffer; detector: RI; standard: pulluran) and found to be M w = 16000, M n = 10900, and M w = 16300, M n = 6200, respectively. The degree of substitution of carboxyl group per galactosamine unit ( D C M ) was estimated from the ratio of absorbance at 1737 cm-1 (COO) to that at 1647 cm-1 ( C O N H ) . It was 40 mol% for N - C M - N A P G A , and 40 and 70 mol% for O - C M - N A P G A , respectively (9). Ν,Ν'-Dimethylformamide ( D M F ) and other organic solvents were purified by distillation. 5-Fluorouracil (5FU), l-ethyl-3-(3-dimethylaminopropyl)carbodiimide (EDC-HC1), and other materials were commercial grade reagents and used without further purification. Tumor Cell Line. The p388Dl lymphocytic leukemia cell line was maintained in RPMI-1640 medium containing 10% heat-inactivated fetal calf serum (FCS), 2 m m o l / L L-glutamine, 18 mmol/L sodium hydrogen carbonate, and 60 m g / L kanamycin. The H L E human hepatoma cell line was maintained in Dulbecco's modified Eagle minimum essential medium containing 10% FCS, 2 m m o l / L L glutamine, 18 m m o l / L sodium hydrogen carbonate, and 60 m g / L kanamycin.



206

POLYMERIC DRUGS AND DRUG ADMINISTRATION

Synthesis of l-[(Amino-n-pentyl)-ester]-methylene-5FU Hydrochloride 4. l-[(Amino- n-pentyl)-ester]-methylene-5FU hydrochloride 4 was prepared through the reaction steps shown in scheme I (10). Formalin (14.4 m L of 37 vol%, or 100 mmol) was added to 17.8 g (220 mmol) of 5 F U . The reaction mixture was stirred at 60°C for 45 min. The solvent was then evaporated to give the oily 1,3-dimethylol5 F U 2. D C C (12.3 g, 120 mmol), t-butoxycarbonyl-amino-n-capronic acid (15.6 g, 120 mmol), and 0.7 g of 4-dimethylaminopyridine ( D M A P ) were added to the solution of 2 in 300 m L of acetonitrile and then stirred for 4 h at room tempera ture. The D C U formed was filtered off and washed with dichloromethane. The resulting solution was washed three times with 1 Ν HC1 and aqueous saturated solution of NaCl. The organic layer was dried over anhydrous sodium sulfate. The solvent was evaporated to give an oily residue. The crude product was purified by recrystallization from diethyl ether to afford a white solid of l-[(t-butoxycarbonylamino-n-pentyl)-ester]-methylene-5FU 3. Compound 3 was dissolved in dioxane containing 4 Ν HC1, stirred for 1 h at room temperature. The solution was evaporated to give the white powder of l-[(amino-n-pentyl)-ester]- methylene-5FU hydrochloride 4. M.p. 188-189°C.; yield 30%. IR (KBr): 3050 ( Ν Ή ) , 3000, 2950 (CH.), 1720 (C = 0 of 5FU), 1740, (COO), 1270 (C-F), 1200 (COO), 1110 c m ^ C N). H - N M R ( ( C D ) S O ) , 61.34 (m,2H,CH ), 1.6 (m,4H,CH ), 2.35 (t,2H,CH ), 2.73 (m,2H,CH ), 5,58 (s,2H,CH ), 8.05 (s,3H,N H ), 8.25 (d,lH,H-6). C-NMR ((CD ) SO)), δ 23.9, 25.4, 26.8, 33.2, 40.3, 70.8 (CH ), 129.7, 138.6, 140.9, 157.7 (5FU), 172.8 (C = 0 ) . U V (MeOH): X 262nm, e 8800. 3

f

3

2

2

2

2

+

2

3

1 3

2

3

2

2

m a x

max

Bonding of 5 F U to N - C M - N A P G A and O-CM-NAPGA. The bonding of 5 F U to N C M - N A P G A was carried out according to Scheme II. N - C M - N A P G A (0.5 g) and 0.37 g of E D C ^ H C l were dissolved in 1 m L of water. The ice-cooled aqueous

H N(CH ) COOH 2

2

+

5

6-amino hexanoic auia acid

(CH^COÇS-^""^ δ N-K D oc-our* Boc-SDP

TEA ^ »» ' 2°' "Room Temp.

^ D

< c u

M

F

H

1

5

Boc-NHtCH^COOH

h

CH3 H

0

Q

A

H O H 2 C

F

Η

Boc-NH(CH ) CO-CH -N\:0 2

3

5

2

Ο.

2

4

N

Η

HCl-Dioxane^" H C I - H N ( C H ) C O - C H - N ^ > 0

F

Scheme I. Preparation route for hydrochloride 4.

2

2

5

4

2

>

F

l-[(amino-n-pentyl)-ester]-methylene-5FU



17. OUCHI ET AL.


207

Scheme II. Synthesis of N - C M - N A P G A / a m i d e / Q / e s t e r / Q / S F U conjugate 5. solution was added to 1.19 g of 4 dissolved in 5 m L of D M F containing equivalent amount of triethylamine (TEA). The reaction mixture was stirred at 0°C for 1 h, and then at room temperature for 24 h. After evaporation of D M F , the crude product was dissolved in water and subjected to gel-filtration chromatography (Sephadex G-15, eluent: water). The high M w fraction with U V absorbance characteristic of 5 F U was collected and freeze dried in vacuo to give the resulting N - C M - N A P G A / a m i d e / Q / e s t e r / Q / S F U conjugate 5. The bonding of 5 F U to OC M - N A P G A was carried out according to Scheme ΠΙ using O - C M - N A P G A ( D C M = 40, 70 mol%) to give 0 - C M - N A P G A / a m i d e / C 5 / e s t e r / C l / 5 F U conjugate 6. The amount of 5 F U (mol%) per galactosamine unit (D5FU) of the conjugates was determined by G P C of the 5 F U released after the conjugate hydrolysis with 3N N a O H (column: Shodex OHpak KB-803; eluent: water; detector: U V 254.5 nm). The 5 F U bonding to N - C M - N A P G A and O - C M - N A P G A was confirmed by U V absorbance due to the presence of 5 F U residue in the high M w fraction (Àmax = 262 nm, in H 0 ) , and by the enhanced absorbance of amide bands (1646, 1556 cm" ) in the IR spectra. 2

1

Determination of the Extent Release of 5 F U from Conjugates. The release behavior of 5 F U from the conjugates 5, 6, and P G A / u r e a / C 6 / u r e a / 5 F U was investigated in 1/15 m o l / L K H P 0 - N a P 0 buffer solution (pH = 7.4) at 37°C in vitro. The amount of the 5 F U released from the conjugates was estimated by G P C (column: Shodex OHpak KB-803; eluent: water; detector : U V 265 nm). 2

4

2

4

Measurement of Cytotoxic Activity in vitro. The cytotoxic activity of the conjugates was measured against p388Dl lymphocytic leukemia cells and H L E human hepatoma cells in vitro. The tumor cell suspension (100 M L ) containing 1 χ 10 p388Dl lymphocytic leukemia cells in culture medium with 10% FCS was placed in a 96-wells multi-plate (Corning 25860MP) and incubated with conjugate 5, 6, or 4


208


4, TEA, WSC DMF/Water

Ο

Η

H OCH CONH(CH2)5Cg-CH -N \=0 —Q Ô =< 2

2

2


Nc

NCOCH Η

N C O C H 3 Η

NCOCH Η

3

3

Scheme III. Synthesis of O - C M - N A P G A / a m i d e / Q / e s t e r / Q / S F U conjugate 6. free 5 F U in a humidified atmosphere (5% C 0 ) at 37°C for 48 h. The number of viable cells was determined by M T T [3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyl tetrazolium bromide] assay (11), using a microplate reader (MTP-120, Corona Electric Co.). O n the other hand, aliquots of tumor cell suspension (100 μ ί ) containing 8.0 χ 10 H L E human hepatoma cells in culture medium with 10% F C S were placed in wells of 96-well multi-plate (Corning 25860MP) and incubated in a humidified atmosphere (5% C 0 ) at 37°C for 48 h. The cells, after washing with the culture medium, were added to 100 m L of the fresh culture medium containing 20 μ ι of PBS solution of conjugates 5, 6, or free 5 F U , and incubated under the same condition for 48 h. The number of viable cells was determined by the method described above. The cytotoxic activity was calculated by the following equation: 2

4

2

Cytotoxic activity (%) = ( N - N ) / N χ 100 c

T

C

where N is the number of control cells; N is the number of treated cells. c

T

Measurement of Antitumor Activity in vivo. The survival effects for N - C M N A P G A / 5 F U conjugate 5 and 5 F U derivative 4 were tested against p388 lymphocytic leukemia in female C D F 1 mice (30 untreated mice/group and 6 treated mice/group) in vivo i.p./i.p. according to the protocol of the Japanese Foundation for Cancer Research (JFCR). 1 χ 10 of leukemia cells were injected to mice i.p. on day 0. The conjugates were dissolved in a sterile normal saline solution and administrated i.p. The mice received two doses of 200-800 mg/kg of the conjugate on day 1 and 5. The ratio of life prolongation, T / C (%), which is the ratio of median survival for the treated mice (T) to that for the control mice (C), was evaluated as a survival effect. The average C value was generally 10 days. The T / C (%) values over 120 were estimated as active. The in vivo screening was performed at the Cancer Chemotherapy Center of the Japanese Foundation for Cancer Research. 6


17. OUCHIETAL.

PGA as a Carrier of Macromoleadar Prodrug of 5FU

209

Results and Discussion Synthesis of N-CM-NAPGA/5FU and O-CM-NAPGA/SFU Conjugates. The N - C M N A P G A and O - C M - N A P G A reacted with 5 F U derivative 4 in DMF/water solvent by using E D O H Q as a condensation reagent to give N - C M - N A P G A amide/C / e s t e r / C y 5 F U conjugate 5, and 0 - C M - N A P G A / a m i d e / C / ester/C75FU conjugate 6, respectively. The reaction conditions, the degree of 5 F U introduction per galactosamine unit (D5FU) for conjugates, and the water solubility data are summarized in Table I. These conjugates were easily separated from nonimmobilized 4 by gel filtration chromatography. The formation of amide bond was observed in IR spectra for conjugates 5 and 6. These conjugates were watersoluble. Downloaded by FUDAN UNIV on March 21, 2017 | http://pubs.acs.org Publication Date: May 5, 1994 | doi: 10.1021/bk-1994-0545.ch017

5

5

Release Behavior of 5FU from the Conjugate in vitro. The release behaviors of 5 F U from N - C M - N A P G A / a i n i d e / Q / e s t e r / Q / S F U conjugate 5, O - C M - N A P G A / a m i d e / C / e s t e r / C / 5 F U conjugate 6, and P G A / u r e a / C 6 / u r e a / 5 F U conjugate were investigated in 1/15 m o l / L Κ Η Ρ 0 - ^ Ρ 0 buffer solution (pH = 7.4) at 37°C in vitro. In these release tests, only the formation of free 5 F U was observed, while no 5 F U derivative was detected. The release rates of 5 F U from these conjugates are shown in Figure 1. The release rates of 5 F U from conjugates 5 and 6 were lower than that from P G A / u r e a / C 6 / u r e a / 5 F U conjugate. These release results may be due to the fact that the hydrolysis rates for the ester and amide groups in conjugates 5 and 6 are lower than that for urea bonds in P G A / u r e a / C 6 / u r e a / 5 F U conjugate. The slow release of 5 F U from the conjugates was achieved by bonding 5 F U to N - C M - N A P G A or O - C M - N A P G A through pentamethylene/monomethylene spacer groups via amide/ester bonds. These conjugate can be expected to release free 5 F U slowly after intravenous injection in vivo. 5

1

2

4

4

Cytotoxic Activity against Tumor Cells in vitro. The cytotoxic activity of the conjugates 5, 6 was investigated against p388Dl lymphocytic leukemia cells and H L E human hepatoma cells in vitro. The conjugates dose effects on the cytotoxic activity against p388Dl lymphocytic leukemia cells and against H L E human hepatoma cells are shown in Figure 2 and Figure 3, respectively. The I C values (concentration at which the cytotoxic activity is 50%) determined from the data in Figures 2 and 3, and the ratio of I C values of the conjugates to that of the free 5 F U are summarized in Table II. While these two conjugates did not show a higher cytotoxic activity against p388Dl lymphocytic leukemia cells, they showed a slightly higher cytotoxic activity against H L E human hepatoma cells in vitro. These results suggested affinities to H L E human hepatoma cells for N - C M - N A P G A and O - C M - N A P G A as polymer carriers. The receptor-mediated uptake of these conjugates into hepatoma cells through endocytosis was suggested. 50

50

Survival Effect. The survival effect of the N - C M - N A P G A / 5 F U conjugate 5 was tested against p388 lymphocytic leukemia in female C D F 1 mice i.p./i.p. The survival effect results for conjugate 5, 5 F U derivative 4, and the free 5 F U are presented in Figure 4. In the case of free 5 F U , the life prolongation significantly decreased due to its toxicity in high dose ranges. O n the other hand, the life



Ο

Ο

Ο

Ο

q

^

vq


22 S od 2

ON

q

H

ΙΛ

en H

es

T-h

en m

es es m ON es en

0\

1-H

»-H

τι- T t ο es oo ^ oo es en es* es

es

ο es oo oo es en es' es' S-X

ON

^

VO

VO

ON NO

ο

m ο

d

ο

ON

I 1

NO ON

S S en oo m es d d

NO

d d o es vn en d d

ιι


17. OUCHI ET AL.


211

m

Φ DC

10

20

30

Tlme(h)

Figure 1. Release rate of 5 F U from N - C M - N A P G A / a m i d e / C / e s t e r / Q / 5 F U conjugate 5, O - C M - N A P G A / a m i d e / Q / e s t e r / Q / S F U conjugate 6, and P G A / u r e a / C / u r e a conjugate in 1/15 m o l / L K H P 0 - N a 2 P 0 buffer solution (pH = 7.4). ( • ) 5 ( D 5 F U = 18.1 mol%); ( A ) 6 ( D 5 F U = 10.6 mol%); ( · ) P G A / u r e a / C / 5 F U conjugate ( D 5 F U = 43 mol%).


5

6

2

4

4

6

10"

9

10"

8

10*

7

10

6

10"

5

10"

4

10

3

10*

2

Concentration of 5FU (mol/l)

Figure 2. Cytotoxic activity of N - C M - N A P G A / 5 F U conjugate 5 , 0 - C M - N A P G A / 5 F U conjugate 6, and free 5 F U against p388Dl lymphocytic leukemia cells in vitro. ( · ) 5 ( D 5 F U = 18.1 mol%); ( • ) 6 ( D 5 F U = 10.6 mol%); ( A ) free 5 F U .

10"

Y

10"

E

10"'

10°

10°

ΗΓ

10

3

Concentration of 5FU (mol/l)

Figure 3. Cytotoxic activity of N - C M - N A P G A / 5 F U conjugate 5 , 0 - C M - N A P G A / 5 F U conjugate 6, and free 5 F U against H L E human hepatoma cells in vitro. ( · ) 5 ( D 5 F U = 18.1 mol%); ( • ) 6 ( D 5 F U = 10.6 mol%); ( A ) free 5 F U .


212


Table II. I C values of N - C M - N A P G A / 5 F U conjugate 5 and O - C M - N A P G A / 5 F U conjugate 6 against p388Dl lymphocytic leukemia cells and H L E human hepatoma cells 50

HLE

p388Di sample

ICso'

Free 5 F U

1.0 xlO'

5

1.1 X l O

4

11.0

5

1.0

5


6

1.0 XlO'

ratio"

ICso

ratio

-

9.0 x 10^ 1.8 XlO"

4

0.20

6.9 x 10" 0.76 4

"iC^: concentration at which the cytotoxic activity shows 50%. ratio = ( I C of conjugate ) / ( I C of free 5FU ).

b

50

50

180

160

ϋ

140

120

1

2

3

4

5

6

7

4

Dose(x10" unit mol/kg) Figure 4. Survival effect per unit mol of 5 F U for 4 and N - C M - N A P G A / 5 F U conjugate 5 against p388 lymphocytic leukemia in mice i.p./i.p. ( # ) free 5 F U ; ( A ) 4; ( • ) 5 ( D 5 F U = 10.1 mol%).

prolongation due to conjugate 5 tended to increase with an increase in dose. While a T / C value for 5 F U derivative 4 was lower than that for the free 5 F U , the T / C value for conjugate 5 was as high as that for the free 5 F U . Moreover, conjugate 5 did not cause a rapid decrease of body weight and did not display an acute toxicity in the treated mice even in high dose ranges, as shown in Figure 4. These results indicate that the covalent bonding of 5 F U to N - C M - N A P G A effectively suppresses the side-effects of 5 F U . Therefore, a novel macromolecular prodrug of 5 F U was prepared by using N - C M - N A P G A as a drug carrier. The prodrug is water soluble, exhibits high antitumor activity, and reduces the side-effects typical of 5 F U . Since the conjugates 5 and 6 are water soluble, they can be easily tested for antitumor activity by intravenous injection. From the standpoint of targeting, the pharmacokinetics of the conjugates with N - C M - N A P G A or O - C M - N A P G A is of


17. OUCHI ET AL.


interest. We will report the antitumor activity of these conjugates hepatoma in vivo by intravenous injection in next paper.

213

against

Acknowledgement. The authors wish to express their sincere appreciation to Dr. Tazuko Tashiro of Cancer Chemotherapy Center of the Japanese Foundation for Cancer Research for the screening test of the survival effect against p388 lymphocytic leukemia in mice i.p./i.p. The authors wish to give their thanks to Higeta Shoyu Co. Ltd. for providing P G A .


Literature Cited 1. Takagi, H . , Kadowaki, K., Agric. Biol. Chem. 1985, 49, 3159. 2. Ishitani, K., Suzuki, S., Suzuki, M . J. Pharmacobiol. Dyn. 1988, 11, 58. 3. Ashwell, G., Hardford, J. Ann. Rev. Biochem. 1982, 51, 531. 4. Duncan, R., Hume, I. C., Kopeckova, P., Ulbrich, K., Strihalm, J., Kopecek, J. J. Control. Rel. 1989, 10, 51. 5. Ohya, Y., Huang, T. Z., Ouchi, T., Hasegawa, K., Tamura, J., Kadowaki, K., Matsumoto, T., Suzuki, S., Suzuki, M . J. Control. Rel. 1991, 17, 259. 6. Takagi, H . , Kadowaki, K. Agric. Biol. Chem. 1985, 49, 3151 7. Takagi, H . , Kadowaki, K. Chitin in Nature and Technology, R. Muzzorelli, C. Jeuniaux, Ed. Plenum Publishing Corporation, New York, 1986, pp. 121-128. 8. Ohya, Y., Inosaka, K., Ouchi, T., Hasegawa, K., Arai, Y., Kadowaki, K., Matsumoto, T., Suzuki, S., Suzuki, M . Carbohydr. Polym.,submitted. 9. Nishimura, S., Ikeuchi, Y., Tokura, S. Carbohydr. Res. 1984, 134, 305. 10. Ohya, Y., Kobayashi, H . , Ouchi, T. React. Polym. 1991, 15, 153. 11. Mosmann, T. J. Immunol. Methods 1983, 65, 55. RECEIVED

May

21,

1993


Polymeric Drugs and Drug Administration - American Chemical Society

Recommend Documents