FIELD OF THE INVENTION This invention concerns the use of isotype-specific retinoic acid receptor (RAR) agonists and antagonists for the modulation of alveolar destruction and/or to promote the formation of alveoli in mammalian lung tissue deficient in adequate numbers of functional alveoli.

BACKGROUND OF THE INVENTION Among aerobic animals, the lung functions to provide an interface for the exchange of gases between blood and the atmosphere. The agents of this exchange are numerous small sacs termed alveoli (in adult humans about 300,000,000 per lung) that provide a gas permeable-liquid impermeable barrier between the gas and liquid phases. Between the alveoli are numerous capillaries carrying deoxygenated blood to the lung from the tissues and oxygenated blood from the alveoli to the tissues. The partial pressure of oxygen in the lungs is approximately 100 mm Hg at sea level; at this pressure the binding of oxygen by hemoglobin in the erythrocytes is favored. The alveoli thus provide a means for presenting the oxygen to hemoglobin to permit the conversion of deoxyhemoglobin to hemoglobin.

Because the exchange occurs at the surface of the gas/blood barrier, alveoli have evolved as a means for providing extremely high surface area in a compact overall area, thus maximizing possible gas exchange. Lack of adequate gas exchange would lead to disability, which could progress to death.

Diseases that result in fewer alveoli can be quite serious, and are common causes of inadequate oxygenation and resultant disability and/or death. Among such diseases are bronchiopulmonary dysplasia (BPD) and emphysema. BPD is a disease of prematurely born infants, and is characterized mainly by a failure of the infant to form a sufficient number of appropriately-sized alveoli. Emphysema, a disease of middle and advanced age, appears to be due to progressive proteinase-induced alveolar destruction.

The process of alveoli formation has been studied from a gross developmental standpoint, and seems to be similar in rat, mouse, and human. The process includes the subdivision (septation) of the saccules that constitute the gas-exchange region of the immature lung. Septation results in the formation of smaller, more numerous gas exchange structures (alveoli). The timing of the onset and cessation of septation vary among species, but both onset and cessation are critical to the formation of alveoli of the size and number needed for adequate oxygenation.

The molecular basis of the initiation and cessation of alveoli formation have not been as well understood as the structural events and timing accompanying alveolar development.

Knowledge of the molecular signals that initiate and end septation, and that govern the spacing of septa relative to the O2-demand, are virtually unknown; however, several lines of evidence suggest that certain retinoids (retinoic acid and its derivatives) may play a key signaling role. In Massaro et al., Nature Medicine 30 3: 675 (1997), hereby incorporated by reference herein, rats were treated with elastase, causing destruction of alveolar walls in a manner similar to that seen in emphysema. Treatment of the rats with all-trans-retinoic acid (ATRA), an antagonist of all RAR isotypes, appeared to reverse this destruction. Similarly, treatment of newborn rats (which are bom with immature lungs lacking an adult complement of alveoli) with ATRA induced the formation of an increased number of alveoli in rats without enlarging the lung. See Massaro et al., Am. J. Physiol. 270: L305 (1996) incorporated by reference herein.

However, ATRA can have a multiplicity of physiological effects. The retinoid receptors, when bound by an appropriate ligand, are mediators of various life processes, including reproduction, metabolism, differentiation, hematopoiesis, and embryogenesis.

Thus, the therapeutic use of the pan-agonist ATRA may not be desirable in all instances.

Additional means for inhibiting alveolar destruction and/or promoting the formation of alveoli in a postnatal aerobic animal, particularly a mammal such as a human are desirable.

Additionally, it would be useful to have more specific therapeutic methods treat such a condition.

SUMMARY OF THE INVENTION The present invention is directed to methods and compositions for modulating the formation of alveoli in mammalian lung tissue. In one embodiment the invention comprises a therapeutic method for inducing the formation of alveoli in mammalian lung tissue by administration of a composition comprising a therapeutically effective amount of a ligand that is an RARa antagonist and that does not have RARP agonist activity. Preferably it is also an PAR ? antagonist. In a preferred aspect, the RARa receptor antagonist has specific RAR modulating activity at the RARa receptor and is not an antagonist of the RARy receptor. In an alternative aspect of this embodiment, the RARa receptor antagonist has specific RAR modulating activity at the RARa receptor and is not specific to either the RARP or the RARy receptor. Even more preferably, the ligand is specific to RARa isotypes. Preferably such a ligand is a retinoid.

By"specific RAR modulating activity"is meant that such a compound has a disassociation constant (Kd) (the ligand concentration at which 50% of the target RAR receptors are complexed with the ligand) at an RAR receptor at least 10 times, preferably at least 25 times, even more preferably at least 50 times, and most preferably at least 100 times greater than the Kd for the binding of the same ligand to an RXR receptor. Determination of the Kd at an RAR or RXR receptor for a given ligand is a routine matter. Membrane preparations of host cells expressing a cloned RAR or RXR receptor, and the amino acid and nucleotide sequences of such receptors, have been described in various publications available, and within the knowledge of, the person of ordinary skill in the art. For example, U. S. Patent No. 5,776,699, to Klein et al., describes assays employing RAR and RXR, and PCT Publication No. W093/11755 discloses ligand binding assays. These references are now incorporated by reference herein in their entirety.

By"agonist"is meant a retinoid receptor ligand that will cause the activation of transcription at a gene having an appropriate retinoid receptor response element.

By"antagonist"is meant a retinoid receptor ligand that will inhibit the activation of transcription by the retinoid receptor at a gene having an appropriate retinoid receptor response element in the presence of an antagonist of the retinoid receptor.

By"inverse agonist"is meant a retinoid receptor ligand that will inhibit the expression of transcription at a gene having an appropriate retinoid receptor response element beyond a basal expression level existing in the absence of an agonist of the retinoid receptor.

In one embodiment one uses a compound that is an RARy agonist. Preferably, it is an antagonist of at least one of RARß or RARa. In one embodiment one can use a cocktail of compounds that collectively act as an RARa antagonist, RAR/3 antagonist and an RARE agonist.

In another embodiment, the invention is directed to therapeutic compositions for the treatment of an emphysemic mammal, or of a mammal suffering from bronchopulmonary dysplasia, comprising a therapeutically effective amount of an ligand that is an RARa antagonist and that does not have RAR antagonist activity at all RAR isotypes. In a preferred aspect of this embodiment, the RARa antagonist has specific RAR modulating activity at the RARa receptor, and is also an antagonist of the RAP, receptor. In another preferred aspect, the RARa receptor antagonist has specific RAR modulating activity at the RARa receptor and is not an antagonist of the RARy receptor. In an alternative aspect of this embodiment, the RARa receptor antagonist has specific RAR modulating activity at the RARa receptor and is not specific to either the RAR (3 or the RARy receptor. In one embodiment, the ligand is specific to RARa isotypes. Preferably such a ligand is a retinoid.

The RARa antagonist may be used either as the only active ingredient in the composition, or in combination with one or more additional therapeutically active ingredients. In one aspect, the additional therapeutically active ingredient is a retinoid; in a preferred aspect, an additional therapeutically active ingredient is another RAR-active ligand, for example, all-trans-retinoic acid.

By"RAR-active retinoid"is meant that the retinoid has antagonist, agonist, or inverse agonist (negative hormone) activity at an RAR receptor.

By"therapeutically effective amount"is meant that the amount of the RAR-specific therapeutic agent is sufficient, either as the result of a single dose, or as the result of multiple doses over the term of therapy, to decrease the rate of alveolar destruction in an emphysemic mammal, or to promote the growth of alveolar septa in said mammal.

DETAILED DESCRIPTION OF THE INVENTION The present invention is directed to compositions and methods for the treatment or prevention of alveolar destruction and/or to promote the formation of alveoli in mammalian lungs deficient in adequate numbers of functional alveoli. Such methods and compositions involve RAR-active retinoids, therapeutic compositions containing such agents, and methods for their use.

The retinoid receptors are part of the steroid/thyroid/vitamin D superfamily of nuclear receptors. The retinoid receptors include the retinoic acid receptors (RAR) and the retinoid X receptor (RXR). The RAR and RXR receptors are single chain polypeptides containing a number of structural domains in common: a ligand binding domain, a sequence-specific DNA binding domain, and a leucine zipper motif. In the presence of ligand, the single RAR or RXR chains can, by virtue of the leucine zipper, form dimers. The RAR chain is believed to exist in vivo exclusively as an RAR/RXR heterodimer. RXR may form heterodimers with RAR or other members of the superfamily, such as the vitamin D receptor and the thyroid receptor.

Retinoid receptor dimers are effective transcription factors regulating gene transcription by binding to retinoic acid response elements (RAREs) or retinoid X response elements (RXREs) present in (or near) the promoters of retinoid responsive genes, or by negatively regulating the enhancer functions of other transcription factors.

Described RAR isotypes include RARß, RARa, and RARy, and described RXR isotypes include RARß, RARa, and RARy. Within each receptor class, these isotypes have sequence homology, but are encoded by different genes. Within each isotype several isoforms have been described; these isoforms differ in their N terminals and are generated by alternative splicing and/or differential usage of more than one promoter. See e. g., Nagpal & Chandraratna, Current Pharm. Design 2: 295-316 (1996) and Mangelsdorf et al., The Retinoid Receptors in The Retinoids : Biology, Chemistry and Medicine Ch. 8 (2d ed. Sporn et al. eds. 1994), both of which are hereby incorporated by reference herein.

Depending both upon the ligand and the nature (e. g., RAR or RXR; isotype ; isoform) of the monomer chains contained in the dimers, an enormous variety of biological responses regulated by the retinoid receptors are possible.

Ligands specific to RAR (e. g., ATRA) or RXR (e. g., TTNB), and to specific RAR isotypes have been described. See e. g., Nagpal & Chandraratna, Current Pharm. Design 2: 295-316 (1996), incorporated by reference herein. Therefore, the design and/or selection of RAR-specific ligands, and of RAR isotype-specific ligands is well within the ability of the person of ordinary skill in the art.

By"specific"to a given retinoid receptor is meant that the disassociation constant (Kd) for the binding of the ligand to a given target receptor or receptor isotype or isoform is at least 10 times lower than the Kd value for the ligand and a non-target receptor or receptor isotype or isoform. Kd is defined as the concentration of ligand at which 50% of the receptors are ligand bound. Even more preferably, Kd is at least 25 times lower for the target receptor than for untargeted receptors. Most preferably, Kd is at least 50, or at least 100, times lower for the target receptor than for untargeted receptors.

An aspect of the present invention comprises compositions for the treatment or prevention of alveolar destruction and/or the promotion of alveolar formation in a mammal.

Such compositions comprise a ligand that is an RARa antagonist and that does not have RAR antagonist activity for the RARy isotype. In a preferred aspect of this embodiment, the RARa antagonist has specific RAR modulating activity at the RARa receptor, and is not an RAR (3 agonist and preferably is an antagonist of the RARP receptor. In another preferred aspect, the RARa receptor antagonist has specific RAR modulating activity at the RARa receptor and is not an antagonist of the RARy receptor. In an alternative embodiment of this embodiment, the RARa receptor antagonist has specific RAR modulating activity at the RARa receptor and is not specific to either the RARP or the RARy receptor. In a different embodiment, the ligand is specific to RARa isotypes. Preferably such a ligand is a retinoid.

Alveolar destruction may be the result of a pathological condition such as emphysema. Treatment to promote alveolar formation may be in response to a condition such as bronchopulmonary dysplasia (BPD).

Another aspect of the invention provides methods for the treatment of or prevention of alveolar destruction and/or the promotion of alveolar formation in a mammal, comprising administering a therapeutic amount of a composition comprising a retinoid receptor ligand that is an RARa antagonist. In a preferred aspect of this embodiment, the RARa antagonist has specific RAR modulating activity at the RARa receptor, and is an antagonist of the PAR ? receptor. In another preferred aspect, the RARa receptor antagonist has specific RAR modulating activity at the RARa receptor and is not an antagonist of the RARE receptor. In an alternative embodiment of this embodiment, the RARa receptor antagonist has specific RAR modulating activity at the RARa receptor and is not specific to either the RARR or the RARy receptor. In one embodiment, the ligand is specific to RARa isotypes. Preferably such a ligand is a retinoid.

The above-mentioned embodiments and aspects are clearly useful for the treatment of medical conditions, such as emphysema and BPD, which have been heretofore difficult or impossible to treat without a lung transplant. As indicated above, use of the RARa antagonist in combination with another agent with activity in treatment of alveolar deficiencies, such as a RAR (3 antagonist or all-trans-retinoic acid, is useful as well. Such methods and compositions may provide a synergistic therapeutic effect and/or permit the use of lower effective doses of ATRA or another therapeutic agent (the RAR (3 antagonist) (and the RARa antagonist), thus reducing the prospect of undesired side effects resulting from the use higher concentrations of either agent when used alone.

The compositions of the present invention can be administered in any therapeutically effective manner or form, and in conjunction with any pharmacologically effective vehicle.

For example, in a particularly preferred aspect, the compositions of the invention may be administered in the form of an inhalant as a powdered or liquid aerosol. Such a formulation may comprise the active agent solubilized in a micronized hydrophobic/hydrophilic emulsion.

Such compositions are well known to those of skill in the art.

Alternatively, the compositions may be administered systemically, such as intravenously by infusion, or by intraperitoneal injection. For intravenous administration, the necessary composition of pharmacologically effective infusion solutions, such as the proper electrolyte balance and tonicity are well known to those of skill in the art, and therefore formulation of the compositions of the present invention with such solutions would be well within the ability of such a person with the disclosure of this application. Similarly, administration of drugs by intraperitoneal injection is well known, and pharmacological vehicles are well known.

Synthesis of candidate compounds having specific RAR modulating activity is well- known in the art. For example, among other references, the synthesis of RAR ligands is disclosed in U. S. Patents No. 5,739,338; 5,728,846; 5,760,276; 5,877,207; the disclosures of all of which is incorporated by reference herein. The synthesis of RARa-specific antagonists is disclosed in U. S. Patents No. 5,663,357; 5,675,024; and 5,856,490, also incorporated by reference herein. Synthesis of RAR-y specific agonists is dislosed in Swann et al., European Patent Application No. EP 0 747 347 and in Reczek et al., Skin Pharmacol. 8: 292-299 (1995), also incorporated herein by reference. One group of specific RARP antagonists have the following formula, where X is CR2, O, S, or Nr; R'and R"can be either H or a lower alkyl ; Ar and Ar'are each a single ring aryl moiety; and B is either CR'CH, CHCR', COO, and OOC, COHN, NHOC, CSHN, or NHSC.

Additionally, the construction of combinatorial libraries of compounds suitable for screening as RAR-selective ligands is now commonplace and well known to those of skill in the art.

Likewise, methods of screening candidate compounds for Specific RAR modulating activity is routine and well-known in the art. For example, U. S. Patent 5,455,265, incorporated by reference herein, describes a chimeric receptor transactivation assay which tests for RAR-agonist and antagonist activity in the RAR-a, RAR-ß, RAR-y, RXRe receptor subtypes. Briefly, this assay employs chimeric proteins containing an RAR ligand-binding domain and a heterologous polypeptide segment having the ability to bind to a response element (RE), in turn facilitating transcription of a specific, measurable target gene, such as chloramphenicol acetyltransferase (CAT). Only an agonist of the given RAR isotype whose ligand binding domain comprises part of the chimeric protein will permit the activation of CAT transcription and expression. These results can be compared with those obtained using chimeric proteins having non-target ligand binding moieties. For example, those ligands able to stimulate CAT synthesis when used in combination with chimeric proteins having RAR binding domains, but not those having RXR ligand binding domains, will be RAR specific agonists.

Variation of this transactivation assay permits testing ligands as RAR antagonists, or antagonists of a given target isotype. A competitive assay involving the use of a stimulatory concentration of a known agonist of a given receptor (for example, ATRA is known to be an agonist of all RAR isotypes), and measuring the reduction in a reporter gene expression (e. g., CAT expression) as a function of the added candidate compound.

Additionally, straightforward receptor binding studies can be performed as competition, rather than transactivation assays. For example, and without limitation, membrane preparations containing cloned retinoid receptors can be used, and the receptors loaded with a known radiolabeled ligand. The release of radioactivity from these preparations as a function of test compound concentration can be determined.-The lower the Kd for a given ligand, the more likely the ligand will be effective as a modulator of receptor activity. Further disclosure is available in, e. g., U. S. Patent No. 5,776,699, previously incorporated by reference herein.

Example 1: Involvement of RAR Receptors in Alveolus Formation In most mammals new alveoli are formed postnatally; thus the lung of many newborn mammals is immature, and not merely a smaller version of the adult lung. For example, in humans alveolus formation can continue up to the age of 20.

The study of alveolus development in newborn mammals, in this case newborn rats, therefore provides an opportunity to study the effect of various agents on alveolus formation.

The rate of increase in specific lung volume (expressed as cm3/100 g body weight) in rats is greatest within the first 10 days following birth, after which it increases at a less steep rate.

However, the lung's efficiency is determined not merely by the volume of air that can be contained in the lung, but by the alveolar surface area, which is a function of tissue growth within the lung. Burri et al., Anat. Rec. 178: 711-730 (1973) and Burri, Anat. Rec. 180: 77-98 (1974) performed extensive studies of the postnatal rat lung and demonstrated that three developmental phases can be shown. In the first stage (day 1-4) there is a massive expansion of the lung's volume. In the second stage, there is an increase in lung tissue proliferation due to septation, including alveolar and capillary surface areas. In the final stage (day 21 to adult) there is a period of concomitant tissue redistribution, lengthening of the septa and alveolar growth.

In this example, healthy 3-day old Sprague-Dawley rat pups were injected intraperitoneally with either cottonseed oil 5 (carrier) or with 5 ul/gram cottonseed oil containing 1.7 mM BMS 195614, a retinoid with RARa antagonist activity. Therefore, with regard to RARa, this drug has an activity opposite to that of ATRA.

Following the initial injection, the rat pups were subjected to identical daily injections until day 13. A set of rat pups were sacrificed at day 4 and at day 21 by anesthesia with phenobarbitol sodium and scission of the abdominal aorta. Following sacrifice, rat lungs were fixed and histological sections prepared essentially as described in Massaro et al., Am.

K Physiol. 270: L305 (1996), incorporated by reference herein. The histological sections were viewed and photographs taken under light microscopy.

Histological sections show that at age 14 days alveoli were smaller, and the degree of septation greater, in rats that had been treated with a RARa antagonist daily from age 3 through 13 days (the separation stage of lung development), than in rats treated with cottonseed oil alone.

The differences in alveolar dimensions between treated and untreated rat pups were so pronounced, they were quantitated by measuring the distance between alveolar walls (Lm) and the surface-to-volume ratio (S/V) of the gas-exchange structures (septa). The Lm in cottonseed oil-treated rats was 86 um i 5. 5 um (N=3) compared to 74.0 u. m 1.0 {im (N=3) in RARa antagonist-treated rats (N=3; P=0. 05 by Mann Whitney nonparametric analysis).

The SIV was 470 yri'32 ni' (N=3) in cottonseed oil-treated rats compared to 541 pm~} 47 l (N=3) in RARa antagonist-treated rats (P=0. 05). Lung volume was the same in both rat groups but surface area was smaller (714 cm2 i69 cm2) in cottonseed oil-treated rats than in RARa antagonist-treated rats (820 cm2+ 82 cm2, P=0. 12).

These data demonstrate that treatment of immature rats with a RARa antagonist promotes septation of the saccules of the gas-exchange region, resulting in smaller and more numerous alveoli and a higher surface to volume ratio, and clearly suggest that antagonism of the RARa receptor results in the up-regulation of naturally occurring, spontaneous alveolus formation. A clear implication of this finding is that the inverse is also true: that agonist activity at the RARa receptor is inhibitory for septation and alveolar development. Thus, therapeutic treatment of mammals with a RARa antagonist will promote the growth of septa.

It is quite surprising, in light of the previous finding that ATRA (an RAR-specific agonist of RARa, RARp, and RARy) stimulates septation, that an antagonist of RARa is able to promote septation and alveolar development.

These data therefore provide evidence that alveolus formation is promoted in vivo by inhibition of RARa, and that stimulation of the RARa receptor is not required.

The data also indicate that the increase in lung air volume seen in the first few days is an event independent from the presence or absence of RAR receptors or the presence or absence of an RAR modulating ligand. The septation stage of lung development can therefore be uncoupled from the increase in lung volume seen immediately post birth. Thus, modulating of RAR receptors can be used to specifically affect septation without modulating other events in alveolar formation.

Thus, signal transduction through RAR receptors appears to be an endogenous modulator of alveolus formation. The generation of agents having specific RAR modulating activity (e. g., RARa antagonist activity) permits more specific targeting of alveolar formation. In addition, treatment with a RARa antagonist plus all-trans retinoic acid may well augment the effect of either agent alone, allowing the use of lower doses of these drugs.

It will be recognized that the foregoing examples and preferred embodiments are exemplary only, and that the invention is defined solely by the claims that conclude this specification.