2-Hydroxypropyl-b-cyclodextrin increases aqueoussolubility and photostability of all-trans-retinoic acid

H. S. Lin BSc, C. S. Chean BSc, Y. Y. Ng BSc, S. Y. Chan PhD and P. C. Ho PhD

Department of Pharmacy, National University of Singapore, 10 Kent Ridge Crescent, Singapore 119260

SUMMARY

Background: All-trans-retinoic acid (ATRA, vita-

min A acid or tretinoin) is effective in the treat-

ment of acute promyelocytic leukaemia (APL).

Unfortunately, the oral absorption of ATRA is

highly variable. Its poor aqueous solubility also

makes it dif®cult to be formulated into parenteral

formulation. To date, there is no parenteral for-

mulation of ATRA available commercially.

Objective: To undertake the preformulation work

necessary for developing such a product.

Method: We investigated the solubility and

stability pro®le of ATRA in various formulations.

Results: The aqueous solubility of ATRA could be

greatly increased by the inclusion of ATRA

in 2-hydroxypropyl-b-cyclodextrin (HP-b-CD).

Adjusting the pH value further improved the water

solubility of ATRA. The photostability of HP-b-

CD-based formulation of ATRA was evaluated

and it was found that inclusion ATRA into HP-b-

CD did improve the photostability of ATRA.

Conclusion: These results showed that it is possi-

ble to develop a parenteral formulation and/or an

aqueous oral formulation of all-trans-retinoic acid

by using 2-hydroxypropyl-b-cyclodextrin. How-

ever, the biopharmaceutical properties of such a

formulation would be necessary before its use.

Keywords: 2-hydroxypropyl-b-cyclodextrin, all-

trans-retinoic acid, aqueous solubility, photo-

stability1

INTRODUCTION

All-trans-retinoic acid (ATRA, vitamin A acid or

tretinoin), an endogenous metabolite of vitamin A,

plays an important role in many physiological

processes (1). It has been used in acne treatment for

many years (2). During the past 20 years, its anti-

cancer and differential induction effects evoked

interest for its use in cancer chemotherapy (1). In

patients with acute promyelocytic leukaemia (APL),

ATRA has demonstrated its dramatic anti-leukae-

mic activity. A high proportion of newly diagnosed

APL patients goes into complete remission at the

beginning of chemotherapy (1, 3, 4). ATRA also has

high potential as an anticancer drug for other

tumours (5±8). As ATRA is a hydrophobic lipidic

acid, which is almost insoluble in aqueous solution,

the oral absorption of ATRA is affected by the pH

and the fatty acid composition of the intraluminal

bile (4). Therefore, plasma concentration following

an oral dose of ATRA is highly variable (9). In certain

cases, such as in comatose patients, oral adminis-

tration of ATRA is not suitable (10). Due to its poor

water solubility, no aqueous parenteral formulation

of ATRA is available for clinical use and only a lip-

osomal parenteral formulation is being tested (11).

Cyclodextrins (CDs), cyclic oligosaccharides

derived from starch, are well known for their

ability to increase aqueous solubility of hydro-

phobic drugs (12±15). Previous studies showed that

inclusion of certain drugs into cyclodextrin could

change their physical properties (12, 16). Among

the b-cyclodextrins that have been studied, only

SBE-b-CD and HP-b-CD are suitable for intrave-

nous administration (13). Recently, a HP-b-CD-

based parenteral formulation of itraconazole which

contained 40% (w/v) HP-b-CD has been devel-

oped and tested in AIDS patients (17). Recently, a

study has been carried out to improve the ATRA

solubility with cyclodextrins for dermatological

applications (18). The aqueous solubility of ATRA

was greatly improved. However, the reported

physical stability of the HP-b-CD-based ATRA

solution was low and rapid re-precipitation of

ATRA occurred within 25 min of dissolution of the

complex (18). It is known that the stability of

cyclodextrin complex could be affected by theCorrespondence: Dr Paul C. Ho. Tel: +65 874 2651;

fax: +65 779 1554; e-mail: phahocl@nus.edu.sg

Journal of Clinical Pharmacy and Therapeutics (2000) 25, 265±269

Ó 2000 Blackwell Science Ltd 265

method of preparation. In this study, the

ATRA:HP-b-CD inclusion complex was prepared

through an alternative method with the aim of

developing a parenteral formulation or an aqueous

oral formulation suitable for clinical use. The

photostability and solubility of the inclusion com-

plex were also assessed.

MATERIALS AND METHODS

Special precautions

To avoid the isomerization of ATRA, all laboratory

procedures involving the manipulation of ATRA

were executed in a dimly-lit environment.

Materials

ATRA was purchased from Tokyo Chemical

Industry Co., Ltd. (Tokyo, Japan). Acitretin (inter-

nal standard) and 2-hydroxypropyl-b-cyclodextrin

(HP-b-CD) (mw � 1396 and degree of substitu-

tion � 4á5) were obtained from F. Hoffmann±

La Roche Ltd. (CH-4002 Basel, Switzerland) and

Nihon Shokuhin Kako Co. Ltd. (Tokyo, Japan),

respectively. All other reagents used were of

reagent grade and all solvents were of HPLC grade.

Milli-Q water was used to prepare buffer solutions

and other aqueous solutions.

Apparatus

A Waters (Milford, MA, U.S.A.) HPLC system,

which consisted of a M-45 model solvent delivery

pump, and2 a Model 440 ultraviolet detector with a

340 nm wavelength ®lter was used. Separation of

compounds was made on a reversed-phase column

(ODS Hypersil, 5 lm, 250 ´ 4 mm, Hewlett Pack-

ard, Germany). The mobile phase consisted of

acetonitrile-methanol-2% w/v ammonium acetate

(55:25:20 v/v) and the ¯ow rate was 1á5 ml/min.

Acitretin was used as an internal standard. The

calibration curve of ATRA was linear (r > 0á99)

over the examined range of 1±16 lg/ml. The

in-between run coef®cients of variation were less

than 5% at all concentration levels.

Phase solubility studies

HP-b-CD aqueous solutions were prepared by

dissolving HP-b-CD into 0á067 M phosphate buffer

(pH � 5á9, 6á5 or 7á4). ATRA:HP-b-CD complex-

ation was obtained through a routine method pre-

viously reviewed by Loftsson and Brewster (12): an

excess of ATRA (10 mg/ml) was added to HP-b-

CD solutions of different concentrations (0á01, 0á02,

0á03, 0á05, 0á10, 0á15, 0á20, 0á30 M) and the suspen-

sion was then shaken on a horizontal rotary shaker

at a speed of 200 r.p.m. under ambient temperature

(24 °C) for 8 days; ®nally the suspension was

®ltered through a 0á45 lm disc ®lter (Millipore Co.)

to collect a clear solution. All samples were

prepared in triplicate. The concentration of ATRA

in ATRA:HP-b-CD inclusion complex solutions

was determined by HPLC assay. The inclusion

constants (K1:1 and K1:2) were calculated by the

method described by Higuchi and Connors (19):

K1:1 � slope1/S0 á (1 ± slope1); K1:2 � slope1/S¢0 á(1 ± slope2).

Photostability studies

The photostability of ATRA in methanol and in

phosphate buffer (pH � 7á4) containing 0á10 M

HP-b-CD was tested. All the solutions tested were

of the same ATRA concentration (0á8 mg/mL).

These solutions were kept at a distance of 40 cm

away from a light source (six ¯uorescent tubes of

6 W each) in an incubator with temperature set at

40 °C. The solutions were sampled at speci®ed

time intervals (0, 0á25, 1, 2, 4, 6 and 8 h) and then

analysed for the quantities of ATRA. The results

were expressed as percentages of ATRA remaining.

The results were compared with the controls,

which consisted of the same samples, but were

protected from light by wrapping with aluminium

foil and placed in the same incubator. Each test was

carried out in triplicate.

Dilution and re-dissolution study

of ATRA-HP-b-CD complex

HP-b-CD-based ATRA solution (2á76 mg/mL) was

diluted with 0á3 M phosphate buffer (pH � 7á4) to

produce 5, 10, 50, 100 and 500 dilutions. The diluted

solutions were centrifuged at 5000 r.p.m. for 5 min

and carefully observed to detect any precipitation.

Two millilitres of HP-b-CD-based ATRA solution

(2á76 mg/mL) in 0á3 M phosphate buffer (pH � 7á4)

was freeze-dried overnight. The collected freeze-

dried product was then re-dissolved in Milli-Q

Ó 2000 Blackwell Science Ltd, Journal of Clinical Pharmacy and Therapeutics, 25, 265±269

266 H. S. Lin et al.

water and then made up to 2á0 mL. The re-dissolved

solutions were checked for the presence of re-

precipitation. Both tests were carried out in triplicate.

RESULTS

Phase solubility studies

The solubility pro®le of ATRA:HP-b-CD inclusion

complex is shown in Fig. 1. Inclusion of ATRA into

HP-b-CD dramatically improved its aqueous solu-

bility. Adjusting the pH value was found to further

improve the aqueous solubility of ATRA. The

aqueous solubility of ATRA, which is almost nil

(»8 ´ 10)5 mg/ml) (18), was increased to 2á76 mg/

ml in 0á3 M HP-b-CD (pH � 7á4). The phase solu-

bility curves of ATRA in HP-b-CD solutions dis-

played a typical AP-type with two linear segments

(Fig. 1), which indicated the formation of 1:1 and

1:2 inclusion complexes. These characteristic phase

solubility curves were similar to that of the

anandamide in HP-b-CD observed by P. Jarho et al.(15). The correlation coef®cients of the linear

regression of these phase solubility curves were

greater than 0á99, except that of the 1:2 inclusion

complex in pH 5á9 buffer, which was equal to 0á989,

indicating a good ®t for all regressions. The inclu-

sion constants (K1:1 and K1:2) calculated by the

Higuchi and Connors method are listed in Table 1.

The binding af®nity of the inclusion complex and

the solubility of ATRA in the HP-b-CD solution

were found to be increased with pH (Table 1).

Photostability studies

ATRA is unstable to light (4). In this study, we

found that inclusion of ATRA into HP-b-CD

improved its photostability. The rate of photo-

degradation of ATRA in aqueous solution of

HP-b-CD was slower than that in methanol (Fig. 2).

When the solutions of ATRA in HP-b-CD and

methanol were protected from light, more than

99% of ATRA remained in these solutions after

28 h incubation at 40 °C (data not shown).

Dilution and re-dissolution of ATRA-HP-b-CD

complex

No precipitation was observed in all samples after

5, 10 50, 100 and 500 times dilutions with 0á3 M

phosphate buffer (pH � 7á4). The freeze-dried

product prepared from the solution of ATRA in

0á30 M HP-b-CD (pH 7á4) was redissolved in Milli-

Q water and then made up to 2á0 ml. A clear yellow

solution of ATRA was formed, and no re-precipi-

tation was observed. The ATRA solution in 0á30 M

Fig. 1. Phase solubility study. The symbols represent

mean and the error bars represent SD.

Table 1. Inclusion constants of ATRA:HP-b-CD com-

plexes

pH K1:1 (M)1) K1:2 (M)1)

5á9 48á2 26á86á5 125á9 30á47á4 266á6 59á1

Fig. 2. Photostability study. The symbols represent

mean and the error bars represent SD.

Ó 2000 Blackwell Science Ltd, Journal of Clinical Pharmacy and Therapeutics, 25, 265±269

2-Hydroxypropyl-b-cyclodextrin 267

HP-b-CD (pH 7á4) was found to be very stable and

no precipitation was observed even after storage at

room temperature (24 °C) for 3 months.

DISCUSSION

Cyclodextrins are well known for improving the

water solubility of hydrophobic drugs (12±15).

Through the formation of inclusion complex with

HP-b-CD, the aqueous solubility of ATRA was

increased by more than 10 000 times to 2á76 mg/ml.

In this study, we found that higher pH enhanced

the solubilizing effect of HP-b-CD on ATRA. ATRA

is a hydrophobic acid (4), which could be weakly

ionized to ATRA) in aqueous solution at non-

acidic pH (e.g. at pH 7á4). The complexation of a

drug molecule with cyclodextrin depends on the

equilibrium reaction between the free drug, free

cyclodextrin and the drug-cyclodextrin complex

(12). The increase in pH results in the formation of

ATRA). The binding of ATRA to HP-b-CD could

be enhanced through the ionic interaction between

the COO) group on the ATRA) and the hydroxyl

groups on the HP-b-CD. As a result, the aqueous

solubility of ATRA is increased substantially by an

increase in pH. Solubilization of hydrophobic drug

with cyclodextrin usually increases its oral bio-

availability (12). The oral bioavailability of ATRA is

highly variable (4, 9), which could be affected by

the extent of solubilization and self-aggregation in

the intestinal ¯uids (20). Administration of an

aqueous cyclodextrin-based formulation of ATRA

may overcome this problem. However, to prevent

the dissociation of ATRA from the cyclodextrin

complex in stomach acid, it may be necessary to

enteric-coat the HP-b-CD-based ATRA complex.

In this study, ATRA solubilized by HP-b-CD

was found to be more photo-stable than ATRA in

methanol. ATRA is light sensitive. When exposed

to light, it will undergo isomerization (4). The cyclic

oligosaccharide ring of cyclodextrin may sterically

hinder such isomerization. Such a photo-protective

effect has also been observed with the inclusion

complex formed by butyl-ethoxydibenzoylmethane

and HP-b-CD (16).

The 1:1 inclusion constant (K1:1) observed in this

study was of the same magnitude as the K1:1 in the

inclusion complex of ATRA and DM-b-CD (18).

The 1:2 inclusion constant (K1:2) observed in this

study was also similar to the K1:2 of the inclusion

complex between ATRA and DM-b-CD deter-

mined by the phase solubility method (18).

Although the aqueous solubility of ATRA was

greatly improved, the binding af®nity between

ATRA and HP-b-CD was fairly weak. This ®nding

was in agreement with the previous study by

Montassier et al. (18).

However, unlike the ATRA:HP-b-CD solution

reported by Montassier et al. (18), the solution in

this study was found to be very stable and no

precipitation was observed, even after storage at

room temperature (24 °C) for 3 months. The

improved physical stability may be due to

the alternative preparation method employed in

this study. In the study by Montassier et al., an

ATRA:HP-b-CD complex was prepared by shaking

ATRA, HP-b-CD in ethanol. After shaking for

8 days, the solution was evaporated to dryness

(18). Using ethanol, an organic solvent, to prepare

the ATRA:HP-b-CD inclusion complex, the solu-

tion could be over saturated. When the complex is

re-dissolved in aqueous solution, re-precipitation

would be anticipated. By contrast, when the

inclusion complex was prepared solely in aqueous

solution, re-precipitation was not encountered in

the present study.

It is not surprising that no precipitation of ATRA

was observed after dilution, previous study has

shown the ability of HP-b-CD to improve the

physical property of host drugs (12, 21).

In this study, we have demonstrated that

the aqueous solubility of ATRA could be grea-

tly increased by complexation of ATRA with

2-hydroxypropyl-b-cyclodextrin (HP-b-CD). The

ATRA: HP-b-CD complex solution was stable and

no precipitation was observed on storage. Also the

photostability of ATRA was improved by the for-

mation of inclusion complex with HP-b-CD. This

preliminary study showed that it is possible to

develop a parenteral formulation of all-trans-retinoic

acid by using 2-hydroxypropyl-b-cyclodextrin.

However, knowledge of3 the biopharmaceutical

properties of such a formulation would be necessary

before its use.

ACKNOWLEDGMENTS

The authors would like to thank Ms Q. Y. Zhou and

Mr W. J. Sam for their help during the experiment

and the preparation of manuscript.

Ó 2000 Blackwell Science Ltd, Journal of Clinical Pharmacy and Therapeutics, 25, 265±269

268 H. S. Lin et al.

REFERENCES

1. Sporn MB, Roberts AB, Goodman DS, eds. (1994)

The Retinoids: Biology, Chemistry, and Medicine. 2nd

edn. New York: Raven Press.

2. Freedberg IM, Eisen AZ, Wolff K, Goldsmith LA,

Katz SI, Fitzpatrick TB, eds. (1999) Fitzpatrick's Der-

matology in General Medicine. 5th edn. New York:

McGraw-Hill, Health Professions Division.

3. Tallman MS, Andersen JW, Schiffer CA, et al, (1997)

All-trans-retinoic acid in acute promyelocytic leuke-

mia. New England Journal of Medicine, 337, 1021±1028.

4. Regazzi MB, Iacona I, Gervasutti C, Lazzarino M,

Toma S. (1997) Clinical Pharmacokinetics of Tretin-

oin. Clinical Pharmacokinetics, 32, 382±402.

5. Gill PS, Espina BM, Moudgil T, et al. (1994) All-trans

retinoic acid for the treatment of AIDS-related

Kaposi's sarcoma: results of a pilot phase II study.

Leukemia, 8(Suppl.), S26±32.

6. Oridate N, Lotan D, Xu XC, Hong WK, Lotan R.

(1996) Differential induction of apoptosis by all-trans-

retinoic acid and n-(4-hydroxyphenyl) retinamide in

human head and neck squamous cell carcinoma cell

lines. Clinical Cancer Research, 2, 855±863.

7. Jozan S, Courtade M, Mathieu-Boue A, Lochon I,

Bugat R. (1998) Cytotoxic effect of interferon-alpha

2a in combination with all-trans retinoic acid or

cisplatin in human ovarian carcinoma cell lines. Anti-

Cancer Drugs, 9, 229±238.

8. Laaksovirta S, Rajala P, Nurmi M, Tammela TLJ,

Laato M. (1999) The cytostatic effect of 9-cis-retinoic

acid, tretinoin, and isotretinoin on three different

human bladder cancer cell lines in vitro. Urological

Research, 27, 17±22.

9. Adamson PC, Pitot HC, Balis FM, Rubin J, Murphy

RF, Poplack DG. (1993) Variability in the oral bio-

availability of all-trans-retinoic acid. Journal of the

National Cancer Institute, 85, 993±996.

10. Yueh YK, Liew PPL, Ho PC, Wu TS. (1999) Sublin-

gual administration of all-trans-retinoic acid to a

comatose patient with acute promyelocytic leukemia.

Annals of Pharmacotherapy, 33, 503±505.

11. Estey E, Thall PF, Mehta K, et al. (1996) Alterations in

tretinoin pharmacokinetics following administration

of liposomal all-trans retinoic acid. Blood, 87, 3650±

3654.

12. Loftsson T, Brewster ME. (1996) Pharmaceutical

applications of cyclodextrins. 1. Drug solubilization

and stabilization. Journal of Pharmaceutical Sciences,

85, 1017±1025.

13. Rajewski RA, Stella VJ. (1996) Pharmaceutical

applications of cyclodextrins. 2. In vivo drug

delivery. Journal of Pharmaceutical Sciences, 85,

1142±1169.

14. Jarho P, Pate DW, Brenneisen R, Jarvinen T. (1998)

Hydroxypropyl-beta-cyclodextrin and its combina-

tion with hydroxypropyl-methylcellulose increases

aqueous solubility of Delta (9)-tetrahydrocannabinol.

Life Sciences, 63, PL381-PL384.

15. Jarho P, Urtti A, Jarvinen K, Pate DW, Jarvinen T.

(1996) Hydroxypropyl-beta-cyclodextrin increases

aqueous solubility and stability of anandamide. Life

Sciences, 58, PL181-PL185.

16. Scalia S, Villani S, Scatturin A, Vandelli MA, Forni

F. (1998) Complexation of the sunscreen agent,

butyl-methoxydibenzoylmethane, with hydroxy-

propyl-beta-cyclodextrin. International Journal of

Pharmaceutics, 175, 205±213.

17. Zhou HH, Goldman M, Wu J, et al. (1998) A phar-

macokinetic study of intravenous itraconazole fol-

lowed by oral administration of itraconazole

capsules in patients with advanced human im-

munode®ciency virus infection. Journal of Clinical

Pharmacology, 38, 593±602.

18. Montassier P, Duchene D, Poelman MC. (1997)

Inclusion complexes of tretinoin with cyclodextrins.

International Journal of Pharmaceutics, 153, 199±209.

19. Higuchi T, Connors KA. (1965)4 Phase-solubility

techniques. Advances in Analytical Chemical Instru-

ments, 4, 117±212.

20. Han CH, Zimmerman CL, Wiedmann TS. (1998)

Spectral properties and ionization behavior of reti-

noids, II. Bile salt solutions. International Journal of

Pharmaceutics, 172, 229±240.

21. McDonald C, Muzumdar PP. (1998) Prevention of

precipitation of phenytoin in an infusion ¯uid by

hydroxypropyl beta-cyclodextrin. Journal of Clinical

Pharmacy and Therapeutics, 23, 235±239.

Ó 2000 Blackwell Science Ltd, Journal of Clinical Pharmacy and Therapeutics, 25, 265±269

2-Hydroxypropyl-b-cyclodextrin 269