AU622890B2 - Formulation and use of retinoids in treatment of cancer and other diseases - Google Patents

Description

8f 1; .k i~ r ;,rr ~~l.~IIt -r r- OPI DATE 25/08/89 wo AOJP DATE 28/09/89 APPLN. ID 30475 89 PCT PCT NUMBER PCT/US89/00435 INTERNATIONAL APPLICATION PUBLISHED UNDER THE PATENT COOPERATION TREATY (PCT) (51) International Patent Classification 4 (11) International Publication Number: WO 89/ 06977 ,A6 1K 47/00, 31/595 Al (43) international Publication Date: 10 August 1989 (10.08.89) (21) International Application Number: PCT/US89/00435 t DNt: lDGINS, Daniel, S' Arnold, White Dur- S Bo,433, tpuston, TX 77210 (US).

(22) International Filing Date: 3 February 1989 (03.02.89) (81) Designated States: A T opean patent), AU, (31) Priority Application Number: 152,183 BE (European patent), B GBJ (OAPI patent), BR, CF (OAPI patent), CG (OAPI patent), CH, CH (Eu- (32) Priority Date: 4 February 1988 (04,02.88) ropean patent), CM (OAPI patent), DE, DE (European patent), DK, FI, FR (European patent), GA (33) Priority Country: US (OAPI patent), GB, GB (European patent), HU, IT (European patent), JP, KP, KR, LK, LU, LU (European patent), MC, MG, ML (OAPI patent), MR (OA- (71) Applicant: BOARD OF REGENTS, THE UNIVERSI- PI patent), MW, NL, NL (European patent), NO, TY OF TEXAS SYSTEM [US/US]; 201 West Seventh RO, SD, SE, SE (European patent), SN (OAPI pa- Street, Austin, TX 78701 tent), SU, TD (OAPI patent), TG (OAPI patent).

(72) Inventors: MEHTA, Kapil 8711 Ilona, Houston, TX 77025 PEREZ-SOLER, Roman 3023 South Published Braeswood, Houston, TX 77025 LOPEZ-BER- With international search report.

ESTEIN, Gabriel 5630 Rutherglen, Houston, TX 77096 (US).

(54) Title; FORMIULATION AND USE OF RETINOIDS IN TREATMENT OF CANCER AND OTHER DISEASES (57) Abstract The present invention, involves a method for therapeutic administration of retlnoid to an animal. This method, in a preferred embodiment, comprises the basic steps of- preparing liposomes comprising phospholipid and retinoid; and parenterally adminis.

tering a quantity of the resultant liposomes to the animal, said quantity containing a therapeutically effective amount of the retinoid. The animel being administered the liposomes may bear a tumor impeded by retinolds and the administering step serve to impede growth of said tumor, The most preferred retinoid is all-trans retinoic aC' iiough other retinoids may prove useful. In certain cases, the retinoid may be retinol, particularly all trans-retinol.

The phospholipids of the present invention may be one or more of phosphatidylcholine, phosphatidylserine, phosphatidylglycerol, sphingomyelin and phosphatidic acid, These phospholipids, their derivatives and those of analogous structure and hydropathic properties may be used to prepare the liposome-encapsulated retinoids of the present invention as Would be apparent to one skilled in the relevant arts upon examination of the present descriptions.

The phospholipids of these retinold-containing liposomes, in a preferred embodiment, comprise dimyristoylphosphatidylcholine and dimyristoylphosphatidylglycerol, still more preferably in about a 73 ratio.

(U-937)

I

0 1 RETINOIC ACID CONCENTRATION (ng/mi c 1- n rn:n-- S WO 89/06977 PCT/US89/00435 FORMULATION AND USE OF RETINOIDS IN TREATMENT OF CANCER AND OTHER DISEASES The present invention relates to therapeutic usage of retinoids encapsulated in liposomes.

It has been known for more than 50 years that retinoids, the family of molecules comprising both the natural and synthetic analogues of retinol (vitamin A), are potent agents for control of both cellular :.ifferentiation and cellular proliferation (Wolbach et al., J.

Exp. Med., 42:753-777, 1925). Several studies have shown that retinoids can suppress the process of carcinogenesis in vivo in experimental animals (for reviews, see e.g., Bollag, Cancer Chemother. Pharmacol., 3:207-215, 1979, and Sporn et al., In Zedeck et al. Inhibition of Tumor induction and development, pp. 71-100. New York: Plenum Publishing Corp., 1981). These results are now the basis of current attempts to use retinoids for cancer prevention in humans. furhermore, there is an extensive evidence which suggests that retinoids can suppress the development of malignant pheotype in vitro (for review, see e.g., Bertram et al., In: M.S. Arnott et al., Molecular .i WO 89/06977 PCT/US89/0043 5 t -2interactions of nutrition and cancer, pp 315-335. New York, Raven Press, 1982; Lotan et al., The modulation and mediation of cancer by vitamins, pp 211-223. Basel: S.

Karger AG, 1983) thus suggesting a potential use of retinoids in cancer prevention. Also, recently it has been shown that retinoids can exert effects on certain fully transformed, invasive, neoplastic cells leading in certain instances to a suppressior. of proliferation (Lotan, Biochim. Biophys. Acta, 605:33-91, 1980) and in other instances to termital differentiation of these cells, resulting in a more benign, non-neoplastic phenotype (see Brietman et al., Proc. Natl. Acad. Sci.

77:2936-2940, 1980).

Retinoids have also been shown to be effective in the treatment of cystic acne (see Peck, et al., New Engl. J. Med., 300:329-333, 1979). In addition to cystic acne, retinoid therapy has been shown to be effective in gram-negative folliculitis, acne fulminans, acne conglobata, hidradenitis suppuritiva, dissecting cellulitis of the scalp, and acne rosacea (see e.g., Plewig et al., J. Am. Acad. Dermatol., 6:766-785, 1982).

However, due to highly toxic side effects of naturally occurring forms of vitamin A (hypervitaminosis A) at therapeutic dose level, clinical use of retinoids has been limited (Kamm et al., In: The Retinoids. Sporn et al., Academic Press, pp 228-326, 1984; Lippman et al., Cancer Treatment Reports, 71:493-515, 1987). In free form, the retinoids may have access to the surroundin normal tissues which might be the basis of their profound toxicity to liver, centrai. nervous system and skeletal tissue.

Therefore, one potential method to reduce the toxicity associated with retinoid administr;ation would be WO 89/06977 PCT/US89/00435 01 -3the use of a drug delivery0 system. The liposomal format is a useful one for controlling the topography of drug distribution in ,ivo. This, in essence, involves attaining a high concentration and/or long duration of drug action at a target a tumor) site where beneficial effects may occur, while maintaining a.low concentration and/or reduced duration at other sites where adverse side effects may occur (Juliano, et al., In: Drug Delivery Systems, Juliano ed., Oxford Press, pp 189-230, 1980). Liposome-encapsulation of drug may be expected to impact upon all the problems of controlled drug delivery since encapsulation riTGically alters the pharmacokinetics, distribution and metabolism of drugs, The present invention involves a method for therapeutic administration of retinoid to an animal. This method, in'a preferred embodiment, comprises the basic steps of: preparing ,liposomes comprising phospholipid and retinoid; and administering a quantity of the resultant liposomes to the ariimal, said quantity containing a therapeutically effective amount of the retinoid. The retinoids may be administered parenterally, topically, orally or intraperitoneally. The animal being administered the liposomes may bear a tumor impeded by retinoids and the administering step serve to impede growth of said tumor or the animal may have a dermatological disorder, opthalmic disSase, rheumatic disease or vitamin deficiency resportive td retinoids wherein the administering step results in clinical improvement. The most preferred retinoid is all-trans retinoic acid although other retinoic acids may prove useful, In certain cases, the retinoid may be retinol, particularly all trans-retinol.

The phospholipids of the present invention may be one or. more of phosphatidy'holini, phosphatidylserine,

F-

WO 89/06977 PCT/US89/00435 -4phosphatidylglycerol, sphingomyelin and phosphatidic acid.

These phospholipids, their derivatives, and those of analogous structure and hydropathic properties may be used to prepare the liposome-encapsulated retinoids of the present invention as would be apparent to one skilled in the relevant arts upon examination of the present descriptions. In addition, the liposomes may also comprise a sterol component, for example, cholesterol.

The phospholipids of these retinoid-containing liposomes, in a preferred embodiment, comprise dimyristoylphosphatidylcholine and dimyristoylphosphatidylglycerol, till more preferably in about a 7:3 ratio.

The processes of the present invention are particularly useful as a method for therapy or prophylaxis of an animal afflicted with cancer. Such a method may comprise: identifying an animal afflicted with cancer; preparing liposomes comprising phospholipid and a retinoid; and parenterally administering a quantity of said liposomes to the animal, said quantity containing a therapeutically effective amount of the retinoid.

From another view, the present invention may also comprise a method of inducing cellular differentiation.

Such induced differentiation may be useful to impede proliferatidn of undifferentiated neoplastic cells or to promote the differentiation of normal cells having the potential differentiated capacity to attack neoplastic cells. More particularly as to the latter use, the liposome-encapsulated retinoids of the present invention may, tor example, be used for inducing the in vivo differentiation of peripheral blood monocytes. This particular method comprises the steps of: preparing liposomes comprising phospholipid and retinoid; and parenterally administering a quantity of said liposomes to an animal, said quantity containing an amount of the i) WO 89/06977 PCT/US89/00435 retinoid effectively inducing peripheral blood monocyte differentiation.

The present invention further includes a process for producing a powder which forms liposomes comprising a retinoid upon suspension of the powder in an aqueous solution. This process comprises the steps of: dissolving retinoid in t-butanol to form a first solution, mixing the first solution and a dry phospholipid to form a second solution, and lyophilizing the second solution to produce a powder. In a preferred embodiment, the phospholipids are defined as a phospholipid film. The solution or powder preferably has a ratio of retinoid to phospholipid between about 1:5 and about 1:50, more preferably between about 1:10 and about 1:15. A composition of matter produced essentially by this process is also an object of this invention. A reconstituted liposomal retin.oid preparation may be produced by simple agitation of the above powder in an aqueous solution.

Such a reconstituted liposomal retinoid preparation may be used for therapy or prevention by parenteral administration.

The term "liposomes" as used herein means man-made lipid vesicles which may include plurilamellar lipid vesicles, stable plurilamellar vesicles, small sonicated multilamellar vesicles, reverse phase evaporation vesicles, large multilaAellar lipid vesicles, and so forth.

Figure 1 shows a time profile of liposomal retinoic acid (L-RA) stability in the presence and absence of serum.

Figure 2 shows human red blood cell (RBC) lysis as a un4ction of time with RA and L-RA WO 89/06977 PCT/US89/00435 -6- Figure 3 shows RBC lysis as a function of retinoic acid (RA) concentration and L-RA concentration Figure 4 shows tne inhibition of THP-1 cell growth as a function of RA concentration L-RA concentration or empty liposome concentration Figure 5 shows the induction of transglutaminase (TGase) in human monocytic THP-1 cells as a function of ,0 treatment with RA or L-RA Figure 6 shows the inhibition of human histiocytic U-937 cell growth as a function of RA concentration L-RA concentration and empty liposome concentration Figure 7 shows the time course of accumulation of tissue TGase activity in cultured human peripheral blood monocytes (HPBM). HPBM were fractionated into small (0) and large subpopulations by centrifugal elutriation, and they were cultured in 35-mm-well tissue culture plates as described in Materials and Methods. At the indicated time points the cells were washed, sonicated, and assayed for TGase activity. Values are the means of six determinations from two dishes.

Figure 8 shows dose-dependent effects of recombinant interferon-gamma (rIFN-g) on induction of tissue TGase activity in HPBM subpopulations. Small and large monocytes were cultured in serum containing medium alone or medium containing increasing concentrations of rIFN-g. After 72 hr, the cells were harvested and the cell lysates assayed for tissue TGase activity. The results shown represent mean SD of three determinations from)an individual donor.

il 1 WO 89/06977 PCT/US89/00435 -7- Figure 9 shows effects of retinol (ROH) and RA on induction of tissue TGase activity in cultured HPBM.

Cells were cultured in the presence of 5% human AB serum and the absence or presence of 500 nM ROH or RA for varying periods of time. At the end of each time point, the cells were harvested and assayed for enzyme activity. Values shown arx the means SD of six determinations from two independent experiments. Inset, doseresponse curve for tissue TGase induction by ROH and RA (0 in HPBM after 72-hr culture.

Figure 10 shows effects of free- and liposomeencapsulated RA on induction of tissue TGase in HPBM.

A: The cells were cultured in tissue culture dishes in presence of serum-containing medium alone 500 nm liposomal RA or medium containing 500 nM free-RA or "empty liposomes" for indicated periods of time. Both the liposomal RA and "empty liposomes" contained 200 ug/ml lipid, At tlhe end of each time point, the cultures were washed and cell lysates assayed for TGase activity. Values shown are the mean SD of six determinations from two independent experiments.

B: Western-blot analysis of the levels of tissue TGase in freshly isolated HPBM (lane 1) and in HPBM cultured for 72 hr in the presence of serum-containing medium alone (lane in medium containing 500 nM free RA (lane 500 nM liposomal RA (lane or "empty liposomes" (lane Cell lysates containing 25 ug of protein were subjected to Western-blot analysis as described in Materials and Methods.

Figure 11 shows effect of free and liposomeencapsulated ROH on induction of tissue TGase in HPBM.

A: HPBM monolayers were cultured in serum-containing medium alone (A or medium containing 1 uM of free- or liposomal-ROH for 72 hr. Then the cultures were

(I

II 1

I

WO 89/06977 PCT/US89/0043i -8washed and the cell lysates assayed for enzyme activity as described in Materials and Methods. B: Western-blot analysis of tissue TGase levels in freshly isolated EPBM (lane 1) and in HPBM cultured for 72 hr in the presence of ser,,m-containing medium alone (lane in medium containing 1 uM of free ROE (lane 3),,or liposorneencapsulated ROH (lane 4) as descri~bed in Materials and Methods, Twenty-five micrograms of cell protein was loaded onto each lane.

In a broad and general sense, the present invention relates to liposome-encapsulated retinoids. As generally known, the lipid membranes of liposornes are formed of a bimolecular layerL of one or more naturally occurring and/or synthetic lipid compounds having polar heads and nonpolar tails. The present inventors have sh;own that the encapsulation of retinQoc acid in liposomes decreases the toxicity observed with use of the free drug.

Represontatlvlee sl~itable compounds for forming liposomes useful in the present invention are phosphatldylcholine, both naturally occurring and synthetically preparea, phosphatidic acid, phosphatidylserine, phosphatidylglycerol, sphingoliplds, sphingotnyelin, cardiolipin, glycolipids, gangliosides, cerebrosides, phosphatides, sterols, and the like.

More particularly useful phospholJipids include dimyristoylphosphatidylcho3.ine and dimyristoylphosphatidylglycerol. In addition, the following compounds may be suitable:, egg phosphatidy.choineo. dilauryloylphosphatidyicholinef dipalmitoylphosphatidylcholine, disteatoyJlphosphatidylchollne# J-mytistoyl-2-palmitoylphosphatidylcholine, I-palmitoyl-2-myristoyl phosphatidyicholine, 1-palmitoyl-2-stearoy.

phosphatidylcholine, 2-tearoy1-2-palznitoyl I e WO 89/06977 WO 8906977PCT/VS89/00435 -9phosphatidyicholive, dioleoylphosphatidylcholile, dila~tryloylphosphatidylglycerol, dipalmitoylphosphatidylglycerol, distearoylphosphatidylglycerolr dioleoylphosphatidylglycero-, dirnyristpylphosphatidic acid, dipalmitoyl phosphatidic acid, dimyristoyl phosphati(dylethanolamine, dipairnitoyl phosphatidylethanolamine, dimyristoyl phosphatidylseririe, dipalmi tzy'L phosphatidylserine, brain phosphatidylserine, brain sphingomyelin, dipalmitoyl sphingomyelin, and distearoyl sphingomyelin.

In addition, other lipids such as steroids and cholesterol may be intearmixed with the phosphoQlipid components to confer certain desired and known properties on the resultant liposotnes. Further, synthetic phospholipids containing either altered aliphatic portions, such as hydroxyl qroupst branched carbon chains, cycloderivatives, aromatic derivatives, ethers, amides, polyunsaturated derivatives, halogenated derivatives, or altered hydr~jphi)Ac portions containing carbohydrate, clycol, phosphate, phosphonate, quaternary amine, sulfate, sulfonate, carboxy, amine, sQlfhydryll imidazole groups and combinations of such groups, can be either substituted or intermixed With, the phospholipids, Suitable therapeutic agents for encapsulation may include various rehinoids. Although retinolic acid, more particularly trans-retinoic acid and~ retinot mote particularl~y$ all-trans-retinol, are preferred, it also believed that the- following compounds, may be successfully encapsulated: all-trans-ret4.noic. acid, retinoic acid methyl ester, retinoic acid ethyl. ester, phenyl analog of retinoic acid, etretinate, retinol, retinyl acetate, retinaldehyde, and l3-qia-retinoic acid. Thrs, compounds an~d their pharmacologio activities are det,.ribed in

;Y

WO 89/06977 PCr/US89/0043 N Roberts and Goodman, Biological Methods for Analysis and Assay of Retinoids: Relationship Between Structure and Activity In: Retinoids, Sporn and Roberts, Eds., Academy Press, p. 235, 1984.

The discoveries presented herein involve manifold uses of retinoids encapsulated in liposomes. For ixample, the encapsulation of retinoic acid in liposomes results into at least 15-fold decrease in the toxicity as compared to the free drug.

The encapsulation of certain retinoids, such as retinol, within liposomes permits their direct delivery to the intracellular sites and thus circumvents the requirement for cell surface receptors. This may be of particular significance, for example, in therapy of tumors which lack the cell surface receptors for serum retinol binding protein but possess intracellular receptors for retinoic acid.

Encapsulation of retinoids within liposomes allows an intravenous administration of the drug. There has been no acceptable vehicle available to permit intravenous administration of retinoids. Due to the highly lipohilic character of retinoids the use of liposomes to deliver retinoids is an attractive approach to reduce their toxicity. Using multilameller liposomes, retinoids can be efficiently encapsulated without losing their activity while reducing their toxicity by at least The use of vitamin A and its analogues (Retinoids) in the prevention and treatment of human cancer represents a relatively new direction in oncologic therapeutics.

Recent laboratory investigations have documented that the retinoids can block phenotypic expression of cancer, whether initiated by chemical, viral, physical, or SWO 89/06977 PCT/US89/00435 -11biologic carcinogens. In humans also, the retinoids have been shown to cause regression of pcemalignant lesions and leukoplakia. However, the use of retiioids has been associated with both short te-m toxicity such as central nervous system alterations, as well as chronic intoxication such as skin and mucous membrane dryness and liver impairment, which eventually become irreversible. A considerable amount of effort has been devoted to develop vitamin A derivatives with an improved therapeutic/toxicity ratio. The effort has met with very little success.

Human peripheral blood monocytes (HPBM) from normal donors, isolated by counercurrent centrifugal elutriation into two subpopulations, showed no significant difference in their ability in vitro to differentiate into macrophages, as determined by induction of the protein cross-linking enzyme, tissue transglutaminase (TGase).

The two subpopulations were equally responsive to the augmenting effect of recombinant interferon-gamma (rIFN-g on expression of tissue TGase, In vitro maturation and treatment with rIFN-g of HPBM were associated with increased binding of tritiated retinol. Intracellular delivery of retinol rendered this hormone active in inducing the differentiation of HPBM. The retinoldinduced expression of tissue TGase was the result of increased accumulation of the enzyme peptide and not activation of preexisting enzyme. Maturation of HPBM, induced by in vitro culture or treatment with rFN-g, appeared to be associated with acquisition of cell surface receptors for serum retinol-bndinng protein.

In addition, it iL believed that the liposomal retinoids may be used As anti-inflammatory reagents.

Recent studies by the present inventors have shown that serum retinoids (vitamin A and aralogs) exert a strong 2. The method of claim 1, wherein the therapeutically .effective amo;flt is effective to inhibit the growth of P reticulosa1-coma cells.

WO 89/06977 PCT/US89/00435 -12inhibitory effect on gamla-interferon-lipopolysaccharide (LPS) triggered, -ytostatic activation of murine macrophages Mehta, et al., J. Inununol., June 1987, and incorporated herein by reference). Retinoid induced ztippression of macrophage activation is associated with induction and accumulation of a protein cross-linking eflzym~e, tissue transglutaminase. Recent studies suggest that the inhibition of cystostatic activation is mediated through inh.bition of secretion of tumor necrosis factor (TNF-Alpha). In tli .ese studies, mouse peritoneal znacrophages were activated in the presence of either intact ser~um or de-lipi ized serum supplemented with retinoids, The cells were metabolically labeled with 35 S-methonine. Imniunoprecipitation with anti--TNF antibodies followed by Sr(S polyacrylaniide gel Oleptrophoztesis showed, in addition to a majo- band at, 17,5 Kd for TNFf the peesence of high molecular weight banA(0dad6 dMrcvr n nyai assay,.

can serv4- as an endogeftous substrate, for tissue TGa-.

re -inoid-induced expression of tissue TGase may cause inter- or intra-molecular cross-linking of TNF, thereby inactivating it or tnhibiting its secretion into the eYhracellular environment. Since factors such as tumor' necrosls factor antd laterleukin-1 (both are released by activated macrophages) are the 'main mediators of inflammation (Nawroth, et J. Exp. Med., 163:1363- 1375r 1986)f by inhibiting the ielease of such mediators from macrophaqei myb p(~ssible to inhibit the whole cascadle leading to inflammtion.

SW 89/06977 PCT/US89/00435 -13- Indeed, retinoids have been shown to be effective as anti-inflammatory agents (Hensby, Agents and Actions, 21:238, 1987). Also, TNF has been shown to induce the release of IL-1 by endothelial cells (Dinarello, et al., J. Exp. Med., 163:1433-1439, 1986). In addition, certain retinoids (Etretinate and Isotretinoin) have been reported to inhibit neutrophil and monocyte migration in patients with dermatological disorders. Retinoids have also been shown to inhibit other discrete polymorphonuclear leukocyte functions in vitro. It has been suggested that retinoids may exert their anti-inflammatory effects by interacting with neutrophil membranes to inhibit a variety of responses, such as lysosomal enzyme release and superoxide generation, In addition, retinoids have proven effective in treating a wide variety of dermatological diseases, including all types of acne, and recent studies have shown that topical adl,,inistration of retinoids may be effective in reversing UV-induced aging of the skin. Retinoids have also been used to treat rheumatic diseases, as immunomodulators (against cancer, infectious diseases, and parasitic diseases), as eye drops or ointments for preventing certain dy:e diseases, for treatment of vitamin A deficiency disorder, and as dietary supplements. It is contemplated that the liposome encapsulated retinoids of the present invention will prove as effective in treating these diseases as the free retinoids, but will not have the associated toxicity. Therefore, use of liposomal encapsulated retinoids should prove advantageous for treating these disorders and is considered to be within the scope of thM present invention.

These examples are presented to describe preferred embodiments and utilities of the present invention and are about aJL 7:3 ratio. 01 -K- 0 1 5 10 RETINOIC ACID CONCENTRATION (ngAmI)/ WO 89/06977 PCT/US89/00435 -14'rnot meant to limit the present invention unless otherwise stated in the claims appended hereto.

EXAMPLE 1 Preparation of liposomal-all trans-retinoic acid (L-RA) Preparation of lyophilized powder containing all trans-retinoic acid and phospholipids was carried out as follows.

A solution of :.,,tinoic acid in t-butanol (1-5 mg/ml) was added to a dry lipid film containing dimyristolyphosphatidyl choline (DMPC) and dimyristoylphosphatidyl glyjcerol (DMPG) at a 7:3 molar ratio. The phospholipids wo~re solubilized in the t-butaaol containing the all-trans retinoic acid and the solution was freeze-dried overnight.

A powder containing diryristoylphosphatidylcholine (DMPC), dimyristoylphosphatidylglyrarol (D)MPG)f and all-trans retinoi,c acid was obtainedi. The lipid:drug ratio uised was erom 3,0:1 to 15:1.

Reconstitution of liposomal retinoic acid from the lyophilized powder was done as follows. The lyophilized powder was mixed with aormal saline at room temperature to form multilamellar liposomes containing all trans-retitnoic acid. This reconstitution method required mild handshaking for 1 min to obtain a preparation devoid of any aggregates or clumps. By light microscopy, the reconstituted preparation contained muiltilamellar liposomes of a close'size range. N~o aggregates or drug clumps were identified in the liposomal preparation in three diI~ferent experiments.

I WO 89/06977 PCT/US89/00435 0 Encapsulation efficiency and size distribution of the liposomal all-trans retinoic acid preparation were determined as follows.

The liposomal all-trans retinoic acid preparation was centrifuged at 30,000 x g for 45 minutes. A yellowish pellet containing the retinoic acid and the lipids was obtained. By light microscopy, the pellet was composed of liposomes with no crystals or drug aggregates. The encapsulation efficiency was calculated to be greater than by measuring the amount of free retinoic acid in the supernatant by UV spectrophotometry. Liposomes were sized in a Coulter-Counter and Channelizer. The size distribution was as follows: 27% of liposomes less than 2 micrometers 65% between 2 um and 3 um, 14% between 3 um and 5 um, 1% more than 5 um. The method used for encapsulation of retinoids was simple, reproducible and could be used for large scale production, for example, for clinical trials.

-I

Further experiments were performed by the same procedure but with different lipids, ratios of lipids and the use of H-all-trans retinoic acid. Additional lipids utilized were dipalmitoylphosphatidylcholine

(DPPC)

stearylamine (SA) and cholesterol. After sedimentation of the liposomes, residual 3H was determined and encapsulation efficiency calculated. Table 1 shows encapsulation efficiencies determined by this method for various L-RA preparations.

Therefore, one potential method to reduce the toxicity associated with retinoid administjation would be i 1 i 'I -1vI hi IW- WO 89/O 77 PCT/US89/00435 i TABLE 1 Encapsulation Efficiency of Retinoic Acid in LiDosomes LIPOSOME COMPOSITION DMPC:Cholesterol 9:1 DMPC:Cholesterol 9:3

DPPC

DMPC;SA:Cholesterol 8:1:1

DMPC:DMPG

7:3

DMPC:DMPG

9:1

ENCAPSULATION

EFFICIENCY (%I 69.3 64.5 69.1 56.7 90.7 Of the lipid compositions studies, DMPC:DMPG at ratios between 7:3 and 9:1 gave superior encapsulation efficiencies.

Liposomal all-trans retinol (L-ROH) was prepared by the methods described above for L-RA with DMPC:DMPG, 7:3.

EXAMPLE 2 Stability of Liposomal Retinoic Acid 3 3 Liposomal 3 H-retinoic acid H-RA) was prepared with DMPC:DMPG, 7:3 as described in Example 1. Samples of the L- 3 H-RA were incubated with either phosphate-buffered saline (PBS) or PBS with 20% (by volume) fetal calf serum (FCS). After various periods of incubation at about 37'C, L, _1 I The phospholipids of the prespnt invention may be K one or more of phosphatidyliholine- phosphatidylserine, :i

I

e eas of e _w d h-R a WO 89/06977 PCT/US89/00435 -17liposomes. The tritium in the supernatant solution measured to determine 3 H-RA release. Figure 1 shows the release of 3H-RA over a two day period. The L- H-RA was over about 80% stable over the period of the experiment, even in the presence of 20% FCS.

3 When H-all-trans retinol was used to label L-ROH and stability in PBS measured, only about 5% of the H-ROH was released after a 24 hr incubation at 37'C.

EXAMPLE 3 In Vitro Lysis of Human Erythrocytes (RBCs) by Retinoic Acid or Liposomal Retinoic Acid Lysis of human red blood cells (RBCs) was quantitated by measuring the release of hemoglobin in the supernatants by observation of increases in optical density at 550 naiometers as described previously (Mehta, et al., Biochem. Biophys, Acta., Vol. 770-, pp 230-234 (1984).

Free-RA dissolved in dimethyl formamide (DMFA), was added to the RBCs. Results with appropriate solvent controls, empty liposomes, and empty liposomes plus free-drug were also noted. Release of hemog.',obin by hypotonic lysis of the same number of human RBCs Jy water was taken as a 100% positive control, while cells treated with PBS were taken as negative controls.

Preparations of L-RA comprising various lipids were incubated at a concentration of 20 microgram (ug) RA per ml with RBCs in PBS for 4 hr at 37'C. The toxicity of the L-RA preparations on the basis of percent RBC lysis is shown in Table 2.

1.;1UJKL5"9PnQspnOJipia ana retinoid; anid parenterally administering a quantity of said liposomes to an animal, said quantity containing an amnount of the WO 8941977PCT/US89/00435 -TABLE 2 in Vitro Toxicity Of-L-RA Preparations To RBCs LIPOSOME COMPOSITION RBC LYSIS DMPC:Cholesterol 9:1 DMPC:CholesL~rol 90.2 9:3 DPPC 6.7 DMPC;SA:Cholesterol 70.4 8:1:1 DMPC:DMPG 8 7:3 DMPC:DMPG 8.3 9:1 As may be seen from the data of Table 2, L-RA of DMPC:cholesterol, DPPC, DMPC:DMPG and DMPC:DMPG exhibited low RBC toxicity under these conditions.

It is of interest to note that the latter two L-RA compositions exhibited superior encaipsulation efficiencies (Table 1).

A further experiment concerning the toxicity over time of free RA and L-RA (DMPC:DMPG-7:3) toward RBC was conducted. Human erythrocytes wert-incubated at 37'C in PBS with 10 ug/mi free RA or 120 ug/m. L-RA, and RBC lysis monitored over a period of 5 hr. Figure 2 shows time courses of REC lysis. At between about 1 hr and about 3 hr, the free RA extensively lysed a large majority of the erythrocytes. When a similar manipulation was performed with 1 L-RA (DMPC:DMPG(7:3) at a RA-,concentration of 120 ug/mi, little RBC lysis occurred less than after 6 hr)

NPAM

Figure 2 shows human red blood cell (RBC) lysis as a i, function of time with RA and L-RA 2 i S WO 89/06977 PCT/US89/00435 -19- A study was also conducted concerning the effects upon RBC lysis in 2 hr of free RA and L-RA (DMPC:DMPG(7:3) at various concentrations. Figure 3 shows the results of this study. Free RA showed linearly increasing RBC lysis between about 5 ug RA/ml and about 30 ug RA/ml. Liposomal RA caused RBC lysis of only about 5% at a concentration of 160 ug RA/ml.

EXAMPLE 4 Acute Toxicity Of Free And Lipsomal Retinoic Acid The acute toxicity of free and liposomal all-trans retinoic acid was studied in CD1 mice. Free all-trans retinoic acid was prepared as an emulsion in normal saline containing 10% DMSO and 2% Tween 80 at a concentration of 3 to 5 mg/ml. Liposomal all-trans retinoic acid was prepared using a lipid:drug ratio of 15:1. The final concentration of all-trans retinoic acid in the liposomal preparation was 3 mg/m1. Empty liposomes of the same lipid composition (DMPC;DMPG 7:3) were also tested at doses equivalent to 80 mg/kg, 100 mg/kg, and 120 mg/kg of liposomal-all trans retinoic acid. Normal saline containing 10% DMSO and 2% Tween 80 was also tested as a control at a dose equivalent to 50 mg/kg of free all-trans retinoic acid. All drugs tested were injected intravenously via tail vein as a single bolus. The injected volumes of free and liposomal-all-trans retinoic acid were the same for each dose.

Table 3 shows data obtained from these acute toxici y experiments.

reeni cdwspeae s neuso nnrasln results shown represent mean SD of three determinations from/ian individual donor.

ii f -i 'Ni~ aaap: ilTi~l~ I WO 89/06977 PCT/US89/00435 TABLE 3 Acute Toxicity of Free and Lioosomal All-Trans retinoic Acid Drug Dose mg/kg Number Animals with s mi7lrP.

Number Animals alivp (72 hrl Free RA

L-RA

10 40 100 120 80 100 120 50 0/6 6/6 6/6 3/3 3/3 0/6 0/6 0/6 0/6 0/6 0/6 0/6 0/6 0/6 /6 5/6 4/6 0/3 0/3 6/6 6/6 6/6 6/6 6/6 6/6 5/6 6/6 6/6 Empty Liposomes Normal saline

DMSO

2% Tween 80 The maximum non-toxic dose of free all-trans retinoic acid was 10 mg/kg. Higher doses caused seizures immediately after injection. The acute LD 5 (deaths occurring up to 72 hours after injection) of free alltrans retinoic acid was 32 mg/kg. The cause of death was cardioulmonary arrest after seizures for 1-2 minutes in all animals. No seizures or deaths were observed in the animals treated with liposomal all-trans retinoic acid at a dose of 120 mg/kg (maximum non-toxic dose and LD 50 greater than 120 mg/kg). Higher doses were not tested.

No seizures were observed in the animals treated with empty liposomes or normal saline with 10% DMSO and 2% Tween

N'

WO89/06977 PCT/US89/00435, -21- EXAMPLE In Vitro Inhibition of Tumor Cell Growth Liposomal all-trans retinoic acid (L-RA) was prepared as described in Example 1.

Cells of the human monocytic cell line THP-1 were inoculated into samples of eucaryotic cell culture medium in the presence or absence of L-RA, at a final RA concentration of 1 micromolar After 24 hr at 37'C, 3 H-thymidine was added to each culture and incorporation thereof into cellular polynucleotides measured. Table 4 shows the percentage of tumor growth inhibition as reflected by decreases in 3H-thymidine incorporation induced by L-RA of differing lipid compositions.

phosphatidyicholifle, 1-palmitoyl7 I2-myristoyl phosphatidyiCholile, i-palmitoyl-2-stearoyl phosphatidylcholile, 1-s tearoyl-2-palmitoyl

(L

WO 89/06977 PCT/US89/00435 -22- TABLE 4 L-RA Inhibition of Tumor Cell Growth LIPOSOME COMPOSITION TUMOR CELL (TiIP-l) INHIBITION M% DMPC:Cholesterol 9:1 DMPC :Cholesterol 9:3

DPPC

DMPC; SA:Cholesterol DMPC :DMPG 7:3

DMPC:DMPG

9:1 From Table 4, it should be noted that L-RA (DMPC:DMPG-7:3), whi,ch, as previously shoWn herein, gave a superior encapsulation efficiency and shoWed a low RBC toxicity (Tables 1 and also effectively iqhibited the tumor cell growth.

Cells of the human monocytic cell line THP-l and of the human, histiocytic cell line U-937 were inoculated at about 2,000O cells per cell in aliquots of eucarlyotic cell culture edium contained in wells of a 96 well microtiter plate. The medium in various wells contained different amount of free RA or L-RA (DMPC:DMPG The cells were incubated for 72' hr at 37"C and cell growth determined and compared to that of controls without any form of retinoic acid. Figure 4 shows the inhibition of THP-l cell),,growth by increasing concentrations of free RA or L-RA (DMPC:DMPG ,WO 89/06977 PCT/US89/00435 -23- At concentrations of less than 1 ug RA/ml, both preparations inhibited cell growth by over The human monocytic leukemia THP-1 cells, after a 72 hr incubation with either free RA or L-RA at a concentration of 0.3 ug RA/ml, were observed to have lost their generally ovate form and to have a more flattened and spread morphological appearance often associated with cellular differentiation. The generally ovate form was retained when the cells were cultured in the absence of any free or liposomal retinoic acid.

After incubation for 24 hr with 0.3 ug/ml or 0.6 ug/ml RA or L-RA in another experiment, THP-1 cells had increased levels of tissue transglutaminase enzymic activity, or marker for monocytic cell differentiation.

As shown in Figure 5, THP-1 cells, at 4 x 106 cells/ml, showed about 56% greater transglutaminase activity when incubated with L-RA as compared to free RA at equivalent retinoic acid concentrations.

Cells of the human histiocytic cell line U-937 were distributed and cultured under the same conditions as the THP-1 cells in the prior experiment. Figure 6 shows the effects upon cell growth of increasing concentrations of free all-trans retinoic acid liposomal (DMPC:DMPG 7:3) all-trans retinoic acid (L-RA) and empty liposomes (which were devoid of retinoic acid). It should be noted that the U-937 cells were almost completely growthinhibited by L-RA at a retinoic acid concentration of about 10 ug/ml while this amount of free RA inhibited growth less than W8906,77 PCT[US89/06435 V -24- EXAMPLE 6 Antitumnor Activity of Liposornal All-Trans Retinoic Acid in vivo The antitumor activity of liposornal-all trans retinoic acid (DMPC:DMPG 7:3) was tested in -vivo against liver metastases of M5076 reticulosarcoma. C57BL/6 mice were inoculated with 20,000 M5076 cells on day 0. Intravenously treatment with 60 mg/kg liposomal all-trans retinoic acid was given on day 4. The mean survival of control animals (non-treated3) was 21,8 1.6 days. The mean survival of treated animal 3 was 27.0 1.6 days, Liposoma. all-trans retinoic acid was shown, therefore, to have antitumor activity at a dose w~ell below the maximum non-toxic dose, agist a, cell ],ink (M5076) which was rel' 1 stant to free tetinoic acid in iu-m.itro studies.

THP-l cells treated~i in vitro with R~A {l 'MM) for 72 hours when Injected subcutaneously into male mice, failed to develop into tumors, whereas untreated cells formed a huge mass of tumors in such mice.

EXAMPE 7 Induction of Tissue Tcansglutaminase in Human Peripheral E3lood Monocytes by intracellular Delivery of Retiitoids Circulating blood monocytes are the precursors of macrophages whichi accumulate at the sites of tumor rejection delaye~d hypersensitivity (251f chronic inflammation W6] and at the site of damage, tizoue as a par1 of the healing processes (see reterence citations in section At these site" t peripheral blood monocytes acquire new functional '"and bi~)chemical characteristics that are associated with the maturationi or retinoids may be used -s anti-inflammatory reagents.

Recent studies by the present inventors have shown that serum retinoids (vitamin A and analogs) exert a strong I r- WO 89/06977 PCT/US89/00435 differentiation process. To understand clearly the mechanisms involved in differentiation, it is necessary to manipulate the extracellular environment and assess precisely a variety of cellular functions and biochemical activities.

Vitamin A and its analogues (retinoids) have been shown to exert a profound effect on the differentiation of monocytic cells. Both normal [19] and leukemic (7,17,28) monocytic cells differentiate in response to retinoids which might suggest that retinoids play a role in regulating the differentiation of these cells. According to recent reports, the cellular activity of transglutaminase (TGase), an enzyme that catalyzes the covalent cross-linking of proteins, may be directly linked to the retinoid's action [4,15,21,23,35,39.39]. Recently, the present inventors found that in vitro maturation of human peripheral blood monocytes (HPBM) to macrophage-like cells was associated with the induction and accumulation of a specific intracellular TGase, tissue TGase (19,22]. Gamma (g)-interferon, which promotes the tumoricidal properties in HPBM, also augmented the expression of tissue TGase [191. Similarly, the activation of guinea pig and mouse macrophages in vivo was associated with a marked increase in tissue TGase activity [10,24,34]. Terminal differentiation of human monocytic leukemia cells (THP-1) induced by phorbol ester and retinoic acid was associated with induction and accumulation of tissue TGase (17], suggesting that the induction of tissue TGase was a marker of monocytic cell differentiation. The present invention involves further definition of the role of retinoids in differentiation and maturation of HPBM and comprises studies of culture conditions that inhibit or facilitate the internalization of retinoids by HPBM on expression of tissue TGase. The studies herein demonstrate that HPBM, isolated into two subpopulationst show no significant WO 89/06977 PCT/US89/ 0435 -26difference in their ability to express!!tissue TGase actiluity induced by either in vitro cul'ture or exposure'to recombinant interferon gaLtma (r2 FN-g), ind that the expression of tissuie TGase i'r coltured HPBM may be indkiced by e direct delivpry of retinhoids to intracell,4ar sites, A. Materials ard5141ethocdh 101. MaterLals: RPM4'r-1640 modium supplemented with L-gliltamine and human serum~ were from Gibco Laboratories (Grand Island, NY); Es.-herichia coi-derived human recombinant gint~irfearon (rIFN-g) was kindly supplied by Genentech Yiic, (South San Francisco, CA); and all-trans retinol (ROFI) and all-tzans retinoic acid (RA) were purchased fromn Sigma Chemtical Co. Louis, MO). The~chromatp'7r4phically pure lipids dimyristoy. phosphatidylcholine (DMPC) and dimyristoyl ph'osphatidylglycere2. (DMPG) were from Avanti Polar Lipids, Birmingham, AtL); tritiated putrescinte (sp.

act. 28.8 Ci/nunol)p from New England Nuclear (Boston, MA); and tritiated ROH (sp. act. 15 mci/mmol.), trom Aznersharn (Arlington Vieights, IL). '4.lpids, cultu're medium, and serum Were screened for endotoxin witb the Limulus amebocyte lysate assay (YLA Bioproducl-s, Walkersville,, MD), and theyr were used otnly when endotoxin contaminatiori was less than 0.25 ng/ml.

2. HPBM Iola~lii, Purification, and culture Pure populatic~ns of HPLE2 were obtained by countercut-rent cent-rifuga,'. elutriation of miononuclear launkocytre-r;.oh fractitons obtaintoa from normal donors,, who were undergoing routin(,t platfletapheresis HPP 4 M were isolated into two SUhpolations according to si4ze with a Coulter ZB1 coanter anid C-1.000 channelizer (Coul(;er W'O 89/06977 PCT/I1S8910043S 1 WO 89/06977 PCr/US89/00435 -27- Electronics, Hialeah, FL). The median volume of small monocytes waa 255 mm3, and that of the large monocytes was

S+

280 m.!C The small monocytes were 95% t 3% nonspecific esterase-positive and the large monocytes were 98% 2% positive. Detailed procedures for isolation anp characteristics of these subpopulations have been published elsewhere (36,371. Small, large, or mixed (obtained by mixing equal parts of small and large HPBM) HPBM subpopulations were washed once with medium (RPMI-1640 supplemented with L-glutamine, 20 mM HEPES buffer, ug/ml gentamicin, and 5% human AB Serum) and resuspended to 0.5 million/mi density in the same medium. The cells were dispensed in 4-ml samples into 35-mm-well plates and cultupred under appropriate conditions.

3. Enzyme Assay Tissue TGase adtivity in cell extracts was measured as a Ca2+ dependent incorporation of 3 H] putrescine into dirvthycasein. In brief, 'zultuved HPBM were %shed three times in Tris-buffered saline (20 mM Tris-HC1, 0.15 M NaCl, pH 7,6) and scraped from the dish in a minimal volume of the same buffer containing 1 mM EDTA and 15 mM Bta-mercaptoethanol. The cells were lysed by sonication, and TGase activity in the lysates was determined as described previously (13,201. The protein content ir cell lysates was determined by Lowry's method (141 with bovine gamma globulin as standard. The enzyme activity was expressed as nanomoles of putrescinrr incorporated into dimethyl-casein pet riour per milligram of cell protein.

4. Immunochemical Detection of Tissue TGase 9Q detect tissue TGase in cell extracts, the cell lysates wete solubilized in 20 mt 'ris-HC1 (pH 6.8) contfiinin4 1% sodium dodeey). sulfate (SD8), O.75 M B6ta-

JI

WO 89/06977 PCTIUS800435 -28mercaptoethanol, 2.5% sucrose and 0.001% bromophenol blue.

Solubilized extracts were fractionated by eleetrophoresis on a 6.5% discontinuous polyacrylamide gel and electroblotted onto nitrocellulose paper. The paper was neutralized with 5% bovine serum albumin and treated with iodinated anti-tissue TGase antibody; the preparation, characterization and properties of this antibody h ve been described elsewhere The unbound antibody was removed by washing the paper in Tris-HCL buffer (50 mM,-pH 7.5) containing 200 mM NaCI, 5 mM EDTA, 0.5% Triton X-100, 0.1% SDS, and 0.25% gelatin, and the paper was dried and autoradiographed as described earlier [20,24].

Preparation of Liposomes Multilamellar vesicles (liposomes) containing DMPC and DMPG at a molar ratio of 7:3 were prepared as described [16,181. All-trans ROH or RA were encapsulated by adding the required amount of the drug (predissolved in ethanol) in lipid-containing organic solvents before vacuum (:ying. The dried lipid-drug film was dispersed by agitation in sterile saline solution. Retinoids up to a 1:10 drug:lipid ratio could be completely encapsulated within the liposomes and were highly stable. The stability and encapsulation efficiency of the liposome preparations were studied by using radiolabelled retinol and showed that only 5% 2% of the incorporated radioactyvity leaked out in the supernatant after 24-hr incubation at 37'C, 6. Binding Assay for 3

H]ROH

Freshly isolated HPBM were cit tured in serum containing medium alone or medium plus 50 units rIFN-g for varying periods of time. At theiend of indicated time periods, HPBM monolayers were washed twice WO 89/06977 PCT/US89/00435 -29in ice cold medium and resuspended in 0.5 ml of prechilled reaction mixture containing 5.0 microcuries (uCi)/ml [11,12(n) H] vitamin A (free ROH) in RPMI medium suppleminted with 5% delipidized human AB serum (serum deipidization was done by organic solvent extraction as described earlier Binding assays were carried out for 1 hr in an ice bath. After a 1-hr incubation, the monocyte monolayers were washed six times with ice-cold medium and the cells were lysed in 200 ul of Triton X-100.

Fifty-microliter aliquchs of cell lysates, in triplicate, were counted for the cell-associated radioactivity.

Background counts, obtained by adding the reaction mixture toward the end of the 1-hr incubation before harvesting, were subtracted from the experimental values.

B. Results 1. Tissue TGase Induction During In Vitro Culture of HPBM The culture of HPBM in the presence of serumcontaining medium for up to 10 days was associated with a marked induction of tissue TGase activity in both small and large HPBM (Fig. the increase in enzyme activity being more rapid after about 4 days of culture. After days in culture, small monocytes showed a 93-fold increase in enzyme activity (from 0.44 to 41.1 nmol/hr/mg), whereas large HPBM accumulated about 103-fold increase in the enzyme activity (from 0.36 to 37.4 nmol/hr/mg). Small and large HPBM mixed together and cultured under similar conditions showed no significant difference in the rate and amount of accumulation of tissue TGase activity compared with that of individual HPBM fractions (data not shown). Induction of enzyme activity was associated with a change in the morphology of cultured monocytes. Freshly isolated HPBM looked rounded, but after 6-8 days in (FCS). After various periods of incubation at about 37'C, WO 89/06977 PCT/US89/00435 culture both the large and small HPBM became firmly adherent to the plastic surface, were more spread and flattened, and had the appearance typical of mature macrophages. By day 10, when the cells had accumulated maximal levels of enzyme activity, these levels then either plateaued or started declining.

2. Effect of rIFN-g on Tissue TGase Expression The effect of continuous exposure to rIFN-g on induction of tissue TGase activity in HPBM is shown in Figure 8. Small and large monocytes were cultured in serum-containing medium for 72 hr in the presence of increasing concentrations of rIFN-g. Enzyme activity in the HPBM populations increased significantly after their continuous exposure to rIFN-g compared with that of cells cultured in the presence of medium along. However, rIFN-g dose size produced no significant difference in enzyme activity between the two HPBM populations. As previously noted a 100-U/ml dose of rIFN-g seemed to be optimal for augmenting TGase activity; higher rIFN-gconcentrations were less effective. The inductive effect of rIFN-g on tissue TGase activity was evidence at 5 U/ml and pretreatment of HPBM cultures with rIFN-g (100 U/ml) followed by washing and subsequent culture in medium alone did not enhance the expression of tissue TGase. The rIFN-g-induced augmentation of tissue TGase was associated with morphologic changes in HPBM so that the rIFN-gtreated cells ware more spreadout and flattened than the untreated control cells after three days in culture.

3. Effect of Retinoids on Tissue TGase Induction Since the two HPBM populations showed no hetero- 33 geneity in terms of induced tissue TGase levels, our subsequent studies were done with whole HPBM fraction I }i WO 89/06977 PC US89/00435 -31without separation into subsets. HPBM cultured in the presence of 500 nM RA for 24 hr accumulated at least three-fold higher enzyme activity than did the control cells cultured in medium along (Fig. Continuous exposure to RA caused a rapid and linear increase in the enzyme activity, whereas in the control cells no significant change in the level of tissue TGase activity was observed for up to 2 days of culture. By day 3, the control cells accumulated about six-fold higher enzyme activity (3.4 nmol/hr/mg) than did freshly isolated HPBM (0.6 nmol/hr/mg), but they still had significantly less enzyme activity than the RA-treated cells (9.8 nmol/hr/mg). Retinoic acid-induced expression of tissue TGase was dose dependent (Fig 9 inset). ROH, the physiologic analogue of RA, did not induce the expression of tissue TGase in HPBM even at a dose of 1 uM. Thus, HPBM cultured in the presence of ROH for up to 3 days showed no significant difference in accumulation of tissue TGase activity when compared with that of control cells cultured in medium along (Fig. 9).

4. Effect of Liposome-Encapsulated Retinoids on Tissue TGase Induction Liposome-encapsulated RA was more effective in inducing tissue TGase expression than was free RA at an equimolar concentration. After 24-hr culture, the amount of tissue TGase activity in HPBM induced by free or liposomal RA at an equimolar concentration of 500 nM was not significantly different (3.4 and 3.7 nmol/hr/mg, respectively); after 48 and 72 hr, however, liposomal RAtreated cells accumulated at least 50% more enzyme activity than did free RA-treated cells (Fig. 10A). That increase in enzyme activity by liposome-encapsulated RA was a specific effect of RA and not of lipids was demonstrated by the fact that a culture of HPBM in the after 6 hr). i WO 89/06977 PCT/U$89/00435 -32presence of "empty liposomes," and containing Ecuivalent amount of lipids did not induce enzyme activity throughout the incubation period. "Empty liposomes," as reported earlier inhibited serum-induced expression of ti sue TGase after 72 hr of culture (Fig 10A). The free or liposomal RA-induced increase in enzyme activity was caused by an increased amount of the enzyme peptide, as revealed by Western-blot analysis of cell lysates using a iodinated antibody to tissue TGase (Fig. 10B). The increase in enzyme activity was proportional to the increase in enzyme peptide and not caused by activation of preexisting enzyme.

Retinol, which in its free form was unable to enhance the expression of tissue TGase in HPBM, became active when presented in liposomal form. Liposome-encapsulated

ROH

caused a rapid and linear increase in tissue TGase activity with time in culture (Fig 11A). After 72 hr of culture, liposomal-ROH caused a nine-fold increase in enzyme activity (7.1 nmol/hr/mg) when compared to that of control cells exposed to free ROH under similar conditions nmol/hr/mg). Liposomal ROH-induced expression of tissue TGase resulted from increased accumulation of the enzyme peptide as demonstrated by Western-blot analysis (Fig. 11B).

Tissue TGase induction is Related to HPBM Uptake of Retinoids The effect of in vitro maturation and rIFN-g treatment on the binding of tritiated-ROH by HPBM was examined.

After 4 days of control culture (medium dose), tritiated- ROH binding by HPBM increased 50% compared to this binding by freshly isolated cells. After 9 days the control culture binding value increased to 350%. The increases in

K*'

f1-

K_

i WO 89/06977 PCT/US89/00435 1S -33- ROH binding were associated with parallel increases in tissue TGase activity (Table TABLE Effect of In V tro Culture and rIFN-g Treatment on I HIROH Binding by HPBM C a Tissue TGase Culture Days in H]ROH bound activity conditions culture (cpm/10 ug protein) (nmol/hr/mg) Medium alone 0 684 25 0.25 0.13 4 994 115 2.96 0.75 9 2,220 144 32,60 8 Medium alone 3 626 37 2.9 0,23 Medium rIFN-g 3 1,782 130 7.6 0.7 20 aHPBM were cultured in serum-containing medium alone or medium containing 50 U/ml rIFN-g for indicated periods of time.

bBinding of tritiated ROH during different periods of culture was determined as described in Materials and Methods.

Cparallel cultures of HPBM maintained under similar conditions were used for assaying enzyme activity as described in Materials and Methods.

Exposure of HPBM to rIFN-g augmented the ROH binding and the expression of enzyme activity. The rIFN-g-treated cells showed a threefold higher [3 H]ROH binding than did control cells incubated in the presence of serumcontaining medium alone for the same period of time. The presence of delipidized serum in the reaction mixture was essential; only 10% of the total counts were cellassociated when delipidized serum was omitted from the reaction mixture.

WO 8906977 PCT/US89/00435 *-34- C. Discussion The results reported in this Example suggested that HPBM, isolated into two populations base .A their size and density, have equal potential to differentiate into mature macrophages. The in vitro maturation of HPBM to macrophages was associated with enhanced binding and uptake of retinol, presumably as a result of the acquisition of cell surface receptors for serum retinolbinding protein. Exposure of HPBM to rIFN-g for 72 hr led to enhanced binding of [3H]ROH that was comparable to the binding activity of control HPBM cultured in vitro for 9 days. HPBM maturation induced by in vitro culture or by exposure to rIFN-g was accompanied by similar morphologic and enzymatic changes. The requirement of cell surface receptor for serum retinol-binding protein could be circumvented by direct intracellular delivery of ROH.

Recently, several reports have suggested an association between monocytic cell differentiation and induction of tissue TGase [10,17,19,21-24,34]. Freshly isolated HPBM that have very low levels of tissue TGase accumulate large amounts of this enzyme after their in vitro maturation [19,22]. Just as the two subpopulations of HPBM showed no significant difference in their ability to induce and accumulate tissue TGase activity during in vitro differentiation to macrophages, both fractions were equally responsive to the effect of rIFN-g in terms of augmented enzyme expression (Fig 8).

Functional heterogeneity among HPBM subpopulations isolated by similar criteria has beenI reported earlier.

Thus, the subsets of HPBM isolated in)o small anc large populations have been reported to produce different amounts of reactive oxygen species prostaglandins antibody dependent cell-medicated cytotoxicity and tumor-cell killing This functional WO 89/0i,977 PCT/US89/00435 heterogeneity among HPBM subpopulations has been 4ttributed to either iaturat;ional oriclonal differences.

The data presented hereqin, however, suggest no heterogeneity among HPBM subpopulations in induction of tissue TGase, a marker for monocytic cell differentiation, and equal potential for differentiating into mature macrophages. The ability of rIFN-g to enhance tissue TGase expression igl both HPBM subpopulations suggests that this endogenous cytokine may play an important role in the maturationo differentiation, and expression of differentioted functions in monocytic cells.

The factors in serum responsible for induction and accumulation of tissue TGase in cultured HPBM and macrophages have been shown to be endogenous retinoids and serum retinol-binding protein Extraction of retinoi~ds by' delipidization or depletion of retinolbinding protein from the serum completely abolished its enzyme-inducing abillty (19,21]. serum retinol-binding protein is believed t~o be responsible for intravascular transport and delivery of retinol to specific target tissues [8,9,29,31). Receptors for serum retzinol-bindirng protein present on the surface of target cells are responsible for the specificity of the delivery process The binding of ROH-retinol-binding protein complexc to cell surface receptors apparently facilitates the delivery of ROH into the interior of the cell (9,31].

At supraphysiologic doses (greater than 10 nM) on the other hand, RA can enter the cells directly by simple diffusion without the participation of surface receptors *for retinol-binding protein This suggested that freshly isolated HPBM probably lack the cell surface *receptors for serum retinol-binding protein and -therefore cannot internalize the endogenous or exogenous retinoids.

Indeed, the addition of exogenous RA to HPBM culturesa at doses (eg. greater than 10 nM) at which the receptor- WO 89/06977 PCT/US89/00435 -36mediated delivery becomes irrelevant resulted in a marked induction of tissue TGase activity (Fig. The enzymeinducing ability of RA was augmented further by encapsulating RA within the liposomes and allowing its internalization via phagocytosis (Fig. Of particular interest was the effect of ROH, which, in its free form did not induce the expression of tissue TGase in freshly isolated HPBM. When ROH was encapsulated within liposomes, however, the requirement for a cell surface receptor for serum retinol-binding protein was obviated. Thus liposomal ROH induced a significant level of tissue TGase activity in HPBM (Fig. 11), This suggested an effective approach for targeting retinol or its inactive analogues to the monocytic cells with no or minimal toxic effects. Because HPBM lack cell surface receptors for serum retinol-binding protein makes administered ROH subject to nonspecific internalization by other cell types. The present studies suggested, furthermore, that interaction of ROH-retinol binding-protein complex with the cell surface receptor is requ;red only for the intracellular delivery of retinol and that, unlike in the case of other hormones ligand-receptor interaction may not require a second messenger for expression of the final event. The increase in TGase enzyme activity induced by free RA or liposome-encapsulated RA or ROH, was the result of the accumulation of enzyme protein rather than the activation of preexisting enzyme, as revealed by immunoblots of the cell lysates using an iodinated antibody to tissue TGase (Figs. 10,11).

Preliminary data on tritiated ROH-binding (Table further suppirted the concept that in vitro differentiation of HPBM to mature macrophages was associated with acquisition of cell surface receptors for retinol-binding protein and that treatment with rIFN-g augmented the t- ;i i WO 89/0977 PCT/US89/00435 -37expression of these receptors. Once the HPBM acquire these receptors, they could internalize the endogenous retinoids and induce the expression of tissue TGase.

Indeed, retinoids have been shown specifically to trigger the gene for tissue TGase in myelocytic cells (23].

Impairment of macrophage function in retinoiddeficient animals has been well documented to lead to increased incidence of infections and decreased tumor-cell killing In cultures of guinea pig peritoneal macrophages, RA has been reported to increase the intracellular levels for the tumoricidal enzyme arginase [32].

The present findings that retinoids play an important role in the differentiation process of HPBM support the idea that retinoids are the important regulators of monocyte/macrophage functions.

D. REFERENCES 1. Bankhurst, Hastein, Goodwin, JiS., and Peake, J.E. The nature of the prostaglandinproducing mononuclear cells in human peripheral bleed. J. Lab. Clin. Med. 97.179, 1981, 2. Baum, and Fisher, S. Macrophage production by bone marrow of tumor bearing mice. Cancer Res, 32e2813, 1972.

3. Cuatrecasas, P. Hormone receptors, membrane phspholipids and protein kinases. Harvey Lect. 80,89, 1986, 4. Davies, Murtaugh, Moore, W.T., Johnson, and Lucas, D.S. Retinoic acid-induced expression of tissue transglutaminase in human WO 89/06977 j/C/S8/03 PCr/US89/00435 -38promyelocytic leuk~gmia (HL-60) cells, J. Biol. Chem.

260, 5116, 1985.

Dennert, G. Retinoids and the immune system; Inuunostimulation by vitamin A. in the Retinoids, Vol. 2 (Sporn, Robc *tst an6, Goodman, Eds.) New York: Academic Press, P. 373? 1984.

6. Dizon, and Southam, C.M. Abnormal cellular IQ response to skin abrasions in cancer patients.

Cancer 16f 1288, 1963.

7. Fetrero, Pressano, Pagliardi, and Robera. Induction of differentiation of human myeloid leukemias: surface changes probed with mnonoclonal antibodies, Blood 61.171, 1983.

8. GanguJly, Roa, Murthy, S.K.j and Scirada, K. systemic mode of action of vitamin A. Vitan.

HO~rm. 38.1, 1980.

9. Hell~er J. Interaction of plasma retinol-binding protein with its receptors; specit'.c binding of bovit."e and human retinol-binding p roteins to pigment epithelium cells from bovine eyez. Biol. Chem.

248t3613, 4975,~ and Birckbichler, P.J. Enhanced transglutaminase activity associatRd with macrolphage activiation, Exp. Ce2L. Res. 141,191 1982.

Liebovichr S4J., and Ross, R, The role of the macrophage in wound repair. A study with hydrocortisone and antimacrophage serumu. Am, J. Pathol.

78,71, 1975.

WO 89/06977 PCI/US89/0043.5 -39- 12. Lopez -Bereste inl Reuben, Hersh Kilbourn, Hester, J.P.,j Bielski, M Ta2,.apaz and Mavligit, G4M. Comparative functional analysis of lymnphocytes and monocytes from platoletapheresis,.

Transfusionl 23,201, 1983.

13. Lorand, Rule, Ong, Furlanetto,

R.,

Jacobsen, Downey, Over, and Lorand, J.

Amiine specificity in transpetidation. inhibition of fibrin cross-linking. Biochemistry 7,1214f 1980.

14. Lowry, Rosebroigh, Farr, and Randall, R.J. Protein measurement with folin-phenol reagent. J. Biol. Chem. 193,265, 1951, Mehta, Claringbold, and Lopez-Beresterl,

G,

Anphotericin a. inhitbits the serum-induced expression of tissue transglitaminase in murine peritoneal macrophages, J. Immnunol. 136t4306$ 1986, 16. Mehtar JuliAnof Rot.,f and Lopez-Beresteil,

G.

stimulation of m~acrophage protease secretion via liposomal delivery of rnuraniyl dipephide de:1vatives to int ,*acellular sites, Im'munology 51.,517f 1984, 17. Mehta, and Jopez-,Beresteii4, Expression of tissue transg)-utaminase in ct.1u1~ced monocytic leukemia cells during differentiation, Cancer Rest 46,1388f 1986.

18. Mehta, R. Lopez-Berestein, Hiersh$ E.M.0 and Juli1ano, Uptake of liposoiues and liposomeencapsulated muramyl dipeptide by *human peripheral.

blood nl~nocytes. J. Reticu3.oeodothelial Soc. 321.55, 1982.

-w c---u-L1 ujii o~ si.ze witth a Coulter ZBI counter-and C-10OO channelizer (Coultzer WO~- 89097PC[V';'/03 \V-0 19. Mehta, Lopez-Berestein, Moore, and Davies; P.J.A. Interferon-g requires serum retinoids to promote the expression of tissue transglutaminase in cultured human blood monocytes. J. Immnunol.

1985.

Mehta, Murtaugh, a1iano, FR.L., and Davies, P,,,14A. Phiagocytosis inhibits the expression of tissue tranisglutamvnase in mouse peritorieal1 imacrophage8 J. Immunol. 132,2552, 1984.

21. Moore, Murtaugh, and Davies, P.J.A.

Retirlnc acid induced expression of Lissue transglutaminase in mouse peritoneal macrophages. J.

Biol. Chein. 259,12794, 1984.

22. Murtaugh, Arend, and Davies, induction of tissue transglutaminaso in human peripheral blood monocytes. J. E4xp. Med. 159,114, 1984.

23, Murtaugh, Dennison, Stein, and Davies, P.J.A. Rptinoic acid induced gene expression in normal acid leukemic rnyeloid cells. J. Exp, Med.

163,1325, 1986.

24s. Murtaughi, idehta, Johnson, Meyers, M., Juliano, and Davies, P4J.A. Induction of tissue transglutar.ase in miouse peritoneal. macrophages. J. Biol. Chemi. 258,11074, 1983.

Nelson, D.S. Macrophage5 as effectors of cellmediated immnunity. CR. Crit. Rev. Microbiol 1, 45,197 2.

WO 89/06977 PC7/US89/00435 -41- 26. Normann, and Weiner, R. Cytotoxicity of human peripheral blood monocytes. Cell. Irnmunol. 81,413, 1983.

27. Norris, Morris Sanders, in, and Kohler, P.F. Isolation of functional subsets of human peripheral blood monocytes. J. Immunol.

123,166, 1979.

28. Olsson, and Brietman, T.R. Induction of differentiation of the human histiocytic lymphoma cell line U-937 by retinoic acid and cyclic adenosine 5'-monophosphate inducing agents.

Cancer Res., 42,3924, 1982.

29. Peterson, P.A. Characteristics of a vitamin A transporting' protein complex occurring in human serum. J. Biol. Chem. 246,34, 1971.

30. Picker, Raff, Goldyne, and Stobo, JiB. Metabolic Heterogeneity among .human monocytes and its modulation of PGE 2 4 J. Tmmunol. 124,2557, 1980.

31. Rask, and Peterson, P.A. In vitro uptake of vitamin A from the retinol-binding plasma protein to muco sal. epithq1ial cells from the monkey's small intestine. J. Biol. Chem. 251,6360, 1976, 31. Roberts, A.B.r and Sporn, M.B. Cellular biology and biochemistry of the retinoids, in The Retinoids, Vol. 2 (Sporn, Roberts and Goodman, Eds New York: Academic press, p. 209, 1984.

33. Rothblat, Arborgast, Ouellette and Howard, BV. Preparation of delipidized serum rIFN-g for varying periods of time. At the end of indicated time periods, HPBM monolayers were washed twice Ii'I m WO 89/06977 PCT/US89/00435 -42proteins for use in cell culture systems. In Vitro 12,554, 1976.

34. Schroff, Neumann, and Sorg, C.

Transglutaminase as a marker for subsets of murine macrophages. Eur. J. Immuno. 11,637, 1981.

Scott, K. Meyskens, Jr., and Russel, D.H.

Retinoids increase transglutaminase activity and inhibit ornithine decarboxylase activity in Chinese ovary hamster cells and in melanoma cells stimulate to differentiate. Proc. Natl. Acad. Sci. USA, 79,4053, 1982.

36. Turpin, Hester, Hersh, EM., and Lopez- Beresteir. G. Centrifugal elutriation as a method for isolation of large number of functionally intact hkman peripheral blood monocytes. J. Clin.

Apheresis. 3,111, 1986.

37. Turpin, Hersh, and Lopez-Berestein, G.

Characterization of small and large human peripheral blood monocytes; effects of in vitro maturation on hydrogen peroxide release and on the response to macrophage activators. J. Immunol. 136,4194, 1986.

38. Yuspa, S. Ben, and Litchir, U. Regulation of epidermal transglutaminase activity and terminal differentiation by retinoids and phorbol esters.

Cancer Res. 43,5707, 1983.

39. Yuspa, Ben, and Steinert, 1. Retinoic acid induces transglutaminase activity but inhibits cornification of cultured epidermal cells. J. Biol.

Chem. 257,9906, 1982.

isolated HPBM looked rounded, but after 6-8 days in i :i i SWO 89/06977 PCT/US89/00435 -43- Changes may be made in the const: .ction, operation and arrangement of the various steps nd procedures and components such as various retinoids -r lipids described herein without departing from the co, ept and scope of the invention as defined in the followin claims.

Claims (18)

1. A therapeutic method for administration of a retinoid to a mammal, comprising the steps of: administering to a mammal a therapeutically effective amount of a retinoid encapsulated in liposomes, wherein: the retinoid is selected from the group consisting of all-trans-retinoic acid and all- trans-retinol; the liposomes comprise phospholipids selected from the group consisting of dimyristoyl phosphatidyl choline and dimyristoyl phosphatidyl glycerol in an approximately 7:3 molar ratio, and dimyristoyl phosphatidyl choline and dimyristoyl phosphatidyl glycerol in an approximately 9:1 molar ratio; and S. the molar ratio of retinoid to phospholipids is between about 1:5 and about 1:50.

2. The method of claim 1, wherein the therapeutically effective amount is effective to inhibit the growth of reticulosarcoma cells.

3. The method of claim 1, wherein the retinoid is administered to the mammal in a dosage greater than about 0 iag retinoid per kg of body weight of the mammal.

4. The method of claim 1, wherein the retinoid is administered by a method selected from parenteral, topical, oral, and intraperitoneal administration.

A therapeutic method for administration of a retinoid to a mammal, comprising the steps of: administering to a mammal a therapeutically effective amoilnt of all-trans-retinoic acid encapsulated in liposomes, where the liposomes comprise the pho-pholipids dimyristoyl 9202053j res.021,30475/89,44 A-0 <n I phosphatidyl choline and dimyristoyl phosphatidyl glycerol in an approximately 7:3 molar ratio, and where the molar ratio of retinoic acid to phospholipids is between about 1:5 and about 1:50.

6. The method of claim 5, wherein the molar ratio of retinoic acid to phospholipid is between about 1:10 and about 1:15.

7. The method of claim 5, wherein the therapeutically effective amount is effective to inhibit the growth of reticulosarcoma cells.

8. A therapeutic composition comprising a retinoid encapsulated in liposomes, wherein: the retinoid is selected from the group consisting of all-trans-retinoic acid and all- 0 trans-retinol; :the liposomes comprise phospholipids selected from the group consisting of dimyristoyl phosphatidyl choline and dimyristoyl phosphattiyl glycerol in an approximately 7:3 molar ratio, and dimyristoyl phosphatidyl choline and dimyristoyl phosphatidyl glycerol in an approximately 9:1 molar ratio; and the molar ratio of retinoid to phospholipids is between about 1:5 and about 1:50.

9. The therapeutic composition of claim 8, which comprises an aqueous suspension of said retinoid encapsulated in liposomes.

The method of claim 1, wherein said mammal being administered said composition bears a tumor impeded by retinoids and said administering step serves to impede growth of said tumor. 920205jnires.021,30475/89,45 Cr o j' ,4 46

11. The method of claim 1, wherein said mammal being administered said composition has a dermatological disorder responsive to retinoids and said administering step results in clinical improvement of said disorder.

12. The method of claim 1, wherein said mammal being administered said composition has an ophthalmic disease responsive to retinoids and said administering step results in clinical improvement of said disease.

13. The method of claim 1, wherein said mammal being administered said composition has a rheumatic disease .1 responsive to retinoids and said administering step results in clinical improvement of said disease.

14. The method of claim 1, wherein said mammal being administered said composition has a vitamin A deficiency responsive to retinoids and said administering step results in clinical improvement of said deficiency.

The method of claim 1 for inducing the in vivo differentiation of peripheral blood monocytes and other cells, wherein said method comprises parenterally administering a quantity of said composition to said mammal, said quantity containing an amount of the retinoid effectively inducing peripheral blood monocyte differentiation.

16. A process for producing a powder which forms a composition as claimed in claim 8 upon suspension in an aqueous solution; said process comprising the steps of: dissolving said retinoid in t-butanol to form a first solution; mixing the first solution and said phospholipids to dissolve the phospholipids and form a second solution; and lyophilizing the second solution to produce a powder. 92O2O5Jm&02t30475/89,46 antibody dependent cell-medicated cytotoxicity and tumor-cell killing This functional 'i .i 47

17. A composition as claimed in claim 8 produced by a process comprising the steps of: dissolving said retinoid in t-butanol to form a first solution; mixing the first solution and said phospholipids to dissolve the phospholipids and form a second solution; and lyophilizing the second solution to produce a powder.

18. A process for producing an aqueous suspension of a composition as claimed in claim 8, said process comprising the steps of: dissolving said retinoid in t-butanol to form a first solution; mixing the first solution and said phospholipids to dissolve the phospholipids and form a second solution; and lyophilizing the second solution to produce a powder; and agitating a sample of the powder in an A aqueous solution. 92026$54l.- s.21.30475/89,47 I i i