The present invention relates to preparation of new corticoid compounds.

In particular it relates to steroid-structured compounds having anti-inflammatory, immunodepressive and angiostatic activities (the so-called steroid anti-inflamma¬ tory drugs) .

The compounds according to the present invention are therapeutically useful in the treatment of pathologic condi¬ tions where generally corticosteroid (corticoids) prepara¬ tions are used, but with increased benefits.

This represents an unexpected advantage over the known corticoids products. In fact, by taking into account the various defined therapeutic uses of a specific product, it is always possible, with the new products of the present invention, to find a better combination of results with respect to the known corticoids. Contrary to any expectation the products of the present invention are characterised by the fact that they show an improved therapeutic profile: high activity combined with low side-effects.

Corticoids are well known as a first-choice pharmacolo¬ gical measure in the treatment of inflammatory disease. Drugs in this category - which include, for example, hydro- cortisone, cortisone, prednisolone, prednisone, fludrocor-

tisone, desoxycorticosterone, methylprednisolone, triamcino- lone, paramethasone, betametasone, dexame asone, triamcino- lone acetonide, fluocinolone acetonide, beclomethasone, ace- toxypregnelone, etc. - have marked pharmacotoxicological ef¬ fects on various organs. Because of this, the clinical use and discontinued use thereof cause a series of side effects„ some of which are very severe. See for example Goodman & Gilman: "The Pharmaceutical Basis of Therapeutics", 9th Ed., pages 1459-1465, 1996.

These toxic effects include: those on bone which lead to changed cell metabolism and a high frequency of osteoporosis; those on the cardiovascular system which cause hyper¬ tensive reactions; those on the gastrointestinal tract which cause gastric damage. See for instance Martindale: "The Ex rapharmacopoeia", 30th Ed., pages 712-723, 1993.

According to the above mentioned art it appears to be almost impossible for therapeutic activities to be separated from side effects, see Goodman et al., as mentioned above, at page 1474.

Known in the art are non-steroid anti-inflammatory

drugs either with or without acidic ending, see patents WO 94/04484, WO 94/12463, WO 95/09831, WO 95/30641 for non-aci- dic ending and the patents therein mentioned for those with acidic ending.

However, it should be noted that steroid compounds are completely different from non-steroid compounds chemically, pharmacologically and biochemically as the pharmaco-toxico- logical mechanism of action of non-steroid products is based on inhibition of one or more cyclo-oxygenases (COX) , while steroid products have nothing to share with COX and have more complex pharmaco-toxicological mechanisms of action which have not yet been fully explained.

It is well known that these two groups of compounds are listed in completely separate categories in international pharmacopoeias.

The applicant has surprisingly and unexpectedly found corticosteroids (corticoids) which are very effective, even superior to those in the known art, and have, at the same time, a higher tolerance than the known corticoids as unexpectedly they do not cause the above side effects, or when they do, these are lower.

An object of the present invention are corticosteroids and their use as anti-inflammatory, immunosuppressive and

angiostatic agents having the general formula:

B-X ₁ -N0 ₂ or their esters or salts, where: B has the following structure:

where, in place of the hydrogens H in the CH group or two hydrogens H ₂ in the CH ₂ group shown in the general formula, there may be the following substituents: at position 1-2: there may be a double bond; at position 2-3: there may be the following substituent:

at position 2: there may be Cl, Br; at position 3: there may be CO, -0-CH ₂ -CH ₂ -Cl, OH;

at position 4-5: there may be a double bond; at position 5-6: there may be a double bond; at position 6: there may be Cl, F, CH ₃ , -CHO; at position 7: there may be Cl; at position 9: there may be Cl, F; at position 11: there may be OH, CO, Cl; at position 16, there may be CH _3/ OH, =CH ₂ ; at position 17: there may be OH, CH ₃ , 0C0(0) _^ (CH ₂ ) _va CH ₃ , or

where ua is an integer equal to 0 or 1, va is an integer from 0 to 4; at positions 16-17: there may be the following groups

R and R' are equal or different one from the other and may be hydrogen or linear or branched alkyls having from 1 to 4 carbon atoms, preferably R = R' = CH ₃ ; B being a corticosteroid residue;

R" is - ( CO-L) _c - (X) _eι -

where t and tl are integers equal or different one from the other and equal to 0 or 1, provided that they cannot be both equal to 0 when B contains no -OH groups,- the bivalent bridging group is selected from:

(CR ₄ R _S ) „(0) ^ (CR ₄ R _S ) „.. (CO) _n , _b (0) _n .. _b (CO) _n ... _b (CR ₄ R _S ) _n ... where na, n'a and n''a are equal or different one from ,the

other and are integers from 0 to 6, preferably from l to 3;

nb, nb' , n''b and n'''b are equal or different one from the

other and are integers equal to 0 or l; R ₄ and R _s are equal or different one from the other and are chosen from H, line¬ ar or branched alkyl having from 1 to 5 carbon atoms, prefe¬ rably from 1 to 3;

X is equal to X ₀ = 0, NH, NR _1C where R _1C is a linear or bran¬

ched alkyl having from 1 to 10 C atoms; or equal to X ₂ where X ₂ is equal to OH, CH ₃ , Cl, N(-CH ₂ -CH ₃ ) ₂ , SCH ₂ F, SH,

_x is a bivalent connecting bridge chosen from:

YO

where Y is a linear or whenever possible branched C^G^

alkylene, preferably having from 2 to 5 carbon atoms, or an optionally substituted cycloalkylene having from 5 to 7 carbon atoms; Y _JL selected from

where n ₃ is an integer from 0 to 3 ;

-(CH ₂ -CH-CH ₂ -0) _πf .- 0N0 ₂ where nf is an integer from l to 6, preferably from 2

to 4;

-(CH-CH ₂ -0) _nf -

where R _lf = H, CH ₃ and nf is an integer from l to 6, pre¬

ferably from 2 to 4.

The compounds which can be mentioned, and which are those preferred, are the ones listed below where B can be obtained according to the known processes of the art.

For example, the precursors and related processes de¬ scribed for example in The Merck Index, 12th Ed. of 1996, herein incorporated by reference, can be mentioned as pre¬ cursors and related processes. The precursors (according to the Merck nomenclature) include the following, where H ₂ , H, R, R' , R' ' have the meaning as defined in the compounds li¬ sted below: budesonide, hydrocortisone, alclometasone, alge- stone, beclomethasone, betamethasone, chloroprednisone, clo- betasol, clobetasone, clocortolone, cloprednol, cortisone, corticosterone, deflazacort, desonide, desoximetasone, de- xamethasone, diflorasone, diflucortolone, difluprednate, fluazacort, flucloronide, flumethasone, flunisolide, fluoci- nolone acetonide, fluocinonide, fluocortyn butyl, fluocorto- lone, fluorometholone, fluperolone acetate, fluprednidene acetate, fluprednisolone, flurandrenolide, formocortal, hal- cinonide, halobetasol propionate, halometasone, halopredone acetate, hydrocortamate, loteprednol etabonate, medrysone, meprednisone, methylprednisolone, mometasone furoate, para- metasone, prednicarbate, prednisolone, prednisolone 25-die- thylaminoacetate, prednisolone sodium phosphate, prednisone,

prednival, prednylidene, rimexolone, triamcinolone, triamci- nolone acetonide, 21-acetoxypregnenolone, cortivazol, amci- nonide, fluticasone proprionate, mazipredone, tixocortol, triamcinolone hexacetonide.

The X connecting bridges as above defined are obtaina- ^' ble by using the methods known in the art as indicated abo¬ ve or by modifing the known methods by introducing X _x bridges when these are different from the connecting bridges described in the listed patents, using processes known in the art. Generally the connection between B and Xj _. is, as seen, of an ester or amide type (NH or NR _1C/ as defined in X) . Any well known synthetic route for forming these bonds can be used to form this connection.

In the case of esters, the most direct synthethic route includes: reaction of acyl chlorides B-C0-C1 in halogen alcohols of the H0-Y _a -Cl, HO-Y _a -Br, HO-Y _a -I type, where Y _a is equal to Y or Y _x without the oxygen atom, in test conditions which are part of the known art.

The reaction products of formula B-C0-0-Y-C1(Br,I) can also be obtained by reacting the sodium or potassium salts of salts acids B-CO-OH with dihalogen derivatives of the general formula Y _a Cl ₂ ,Y _a Br _a or Y _a I ₂ , ClY _a Br, ClY _a I, BrY _a I.

The reaction products are converted into the final pro¬ ducts by reacting with AgN0 ₃ in acetonitrile according to what is known in the literature.

The general scheme is as follows:

B-CO-Cl+HO-Y _a -Br > _<>ι B-C0-0-Y _a -Br+AgN0 ₃ > where X _x = Y _a 0.

The general scheme may also be as follows:

B-C0-0Na+Br ₂ Y _a > B-C0-0-Y _a -Br+AgN0 ₃ - - - > B-X^O-,

where X _x = Y _a 0.

In this case of amide, the synthetic sequence includes reaction of the same acyl chlorides BCOCl with aminoalcohols of the general formula NH ₂ -Y _a -OH, NHR _1C -Y _a -0H to give amides of the general formula: B-C0-NH-Y _a -0H and B-C0-NR _1C -Y _a -0H according to known methods.

Reaction of these amides with halogenating agents such as, for example PC1 ₅ , PBr ₃ , SOCl ₂ , etc., gives the halogen derivatives of the general formula: B-CO-NH-Y _a -Br(Cl) and B-C0-NR _1C -Y _a -Br(Cl) .

The latter give the final products BX- TO;, by reacting with AgN0 ₃ in acetonitrile according to methods known in the literature.

The sequence may be represented as:

PC1 _S B - CO - Cl+NHR _lc -Y _a -OH > B - C0-NR _1C -Y _a -0H >

B - CO-NR _lc -Y _a -Br+AgN0 ₃ > B - C0-NR _1C -Y _a -0N0 ₂ where Y _a 0 is X _x .

An alternative route to ester formation is reaction of the sodium or potassium salts of the acids with the nitric esters of halogen alcohols of the general formula:

N0 ₂ -0-Y _a -Cl(Br,I) to directly give the products of the invention.

The reaction scheme is as follows:

B-C0-0Na+Br-Y _a -0N0 ₂ > B-C0-0-Y _a -0N0 ₂ where Y _a 0 is X _x .

Other synthetic routes similar to those described above include those in which the dihalogen derivative Br ₂ Y _a is rea¬ cted with enolates. The reaction products are then converted by reacting with AgN0 ₃ in acetonitrile according to the above reaction. The general scheme is shown for an -OH in group B, of the type -CH ₂ -0H, =CH-0H, is as follows:

AgN0 ₃ -ONa+Br ₂ -Y _a > -0-Y _a -Br > -0-Y _a -0N0 ₂

The processes to obtain these X _x connecting groups are described in patent application WO95/30641 herein incorpora¬ ted by reference.

As said above the compounds of the invention of

formula B-X ₁ -N0 ₂ or their pharmaceutical compositions, are used for the treatment of diseases in which the well known corticoids products are employed.

In particular, it can be specifically mentioned the use in respiratory disorders, e.g. antiasthmatic, the use as antiarthritic, antipruritic, antipsoriatic, antieczematic; the use in vascular disorders, e.g. as angiostatic, the use in immunology disorders, e.g. as immunosoppressive.

The compounds, or their compositions, of the present invention can be administered for example by oral, rectal (intestinal disorders) , parenteral route or by local (der¬ mal, topical, transdermal, ocular, inhalatory, etc.) appli¬ cation.

The following examples are given only for illustrative purpose as an explanation but not as a limitation of the present invention. EXAMPLE 1

CHEMICAL SYNTHESIS: preparation of hydrocortisone nitroderi- vative (HCN)

EXAMPLE 1A

Preparation of hydrocortisone (4-chloro)butanoate

4 portions of 4-chlorobutanoylchloride (0.32 ml x 4) and triethylamine (0.3 g x 4) were added in 24 hours to a solution of hydrocortisone (l g) in CHC1 ₃ dried over P ₂ 0 _s , and stirred for 3 days. The solution was treated with water, the organic phase was separated, dried (Na ₂ S0 ₄ ) and deprived of the solvent at reduced pressure. The crude residue was ground with hexane and CH ₂ C1 ₂ to give a white solid with a 53% yield by weight, which had a melting point (m.p.) of 155°C.

The product was characterised by mass spectometry: M ⁺ 493.

^X H NMR (300 MHz CDC1 ₃ ) : 0.95 (3H,S,CH ₃ ), 1.45 (3H, S, CH ₃ ) , 2.12 (2H, t, CH ₂ in 2), 2.6 (2H, t, CH ₂ C00) , 3.65 (2H, t, CH ₂ C1) , 4.45 (1H, m, CHOH) , 4.35 and 5.05 (2H, 2d, C0CH ₂ 0) , 5.70 (1H, s, olefin H) . Preparation of hydrocortisone (4-nitroxy)butanoate

AgN0 ₃ (0.2 g) was added to a solution of hydrocortisone-

4-chlorobutanoate prepared as above (0.23 g) in acetonitrile (70 ml) and refluxed for 16 hours. The solution was deprived of the solvent at reduced pressure and chromatographed on silica gel using a solution of ethyl acetate and CH ₂ C1 ₂ (3:7) ' as an eluant.

Cortisone 4-nitroxybutanoate was recovered from the head fractions.

The product was characterised by ¹ H NMR (300 MHz CDCL ₃ ) : 0,95 (3H, s, CH ₃ ) , 1.45 (3H, s, CH ₃ ) , 2.12 (2H, t, CH ₂ in 2), 2.6 (2H, t, CEjCOO) , 4.45 (1H, m, CHOH) , 4.45 (2H, t, CH ₂ 0- N0 ₂ ) , 4.35 and 5.05 (2H, 2d, COCH ₂ 0) , 5.68 (1H, s, olefin H) . EXAMPLE IB

The product from Example 1A was also prepared using another synthetic route. Preparation of hydrocortisone 4-bromobutanoate

Five portions of 4-bromobutanoylchloride (0.35 ml x 5) and potassium carbonate (0.4 g x 5) were added in 24 hours to a solution of hydrocortisone (l g) in CHC1 ₃ dried over P ₂ O _s and stirred for 5 days. The solution was treated with water,

the organic phase was separated, dried (Na ₂ S0 ₄ ) and deprived of the solvent at reduced pressure.

Preparation of hydrocortisone 4-nitroxybutanoate (HCN)

AgN0 ₃ (0.2 g) was added to a solution of hydrocortisone 4-bromobutanoate prepared as above (0.23 g) in acetonitrile (70 ml) and stirred for 48 hours at room temperature.

The solution was deprived of the solvent at reduced pressure and chromatographed on silica gel using a solution of ethyl acetate and CH ₂ C1 ₂ (3:7) as an eluant.

Cortisone 4-nitroxybutanoate was recovered from the head fractions and characterised by mass spectometry: M* 493. The spectrum was the same as that shown in Example 1A. EXAMPLE 2 Evaluation of safety and activity

The products were administered in a 2%-by-weight car- boxymethyl cellulose suspension during in vivo tests, while a 0.1%-by-weight dimethylsulphoxide suspension was used for in vitro studies.

The test groups always included 8 samples (except when differently stated in the examples) for adequate statistical evaluation, to be carried out when necessary according to common statistical procedures.

EXAMPLES 2A

Acute toxicity study

The acute toxicity of the product from Example 1A was roughly evaluated by orally administering a single dose of substance to a group of 10 mice of the Swiss strain.

Death incidence and appearance of toxic symptoms were observed during a period of 14 days after the compound administration.

The animals showed no sign of apparent toxicity even after administration of a 50 mg/kg dose. EXAMPLE 2B Study of antiarthritic activity

Adjuvant arthritis was induced in male rats of Lewis strain, weighing 170 ± 15 g by intracaudal injection of 0.6 mg of Mycobacterium butyricum (Difco) suspended in 0.1 ml of mineral oil. The animals were treated with a vehicle made up of an intraperitoneal (i.p.) 2%-by-weight suspension of car- boxymethyl cellulose in water, with intraperitoneal hydro¬ cortisone of HCN (a suspension as described above) at doses of 5 mg/kg or doses of 10 mg/kg, starting from the first day after mycobacterium inoculation.

Arthritis development was assessed 21 days later. To the arthritic lesions an arbitrary score was assigned accor-

ding to the following scale: hind limbs: 0 to 7 for each (0 for no lesions and 7 for most severe lesions) ; forelimbs: 0 to 4.5 for each (0 for no lesions and 4.5 for most severe lesions) ; tail: 0 to 5 (0 for no lesions and 5 for most severe lesions) ; ears: 0 to 2 for each (0 for no lesions and 2 for most severe lesions) ; nose and eyes: 0 to l ofr each (0 for no lesions and l for most severe lesions) .

The results were expressed as a percentage of inhibi¬ tion compared to the value obtained in the control group (animals treated withe the vehicle alone) .

The result are shown in Table 1.

TABLE 1 STUDY OF ANTIARTHRITIC ACTIVITY OF COMPOUND HCN

VERSUS HYDROCORTISONE IN RATS

As shown by the results in Table 1, the test products were capable of similarly inhibiting development of the arthritic process caused by mycobacterium. However, being the tolerability of HCN much higher than that of hydrocorti¬ sone (see ex 2C below) , the results in terms of activity are much better in the case of HCN (see for comparison 40% of antiarthritic activity obtained with 5 mg/kg hydrocortisone with respect to 62% obtained with 10 mg/kg HCN) . EXAMPLE 2C Study of crastric tolerability (safety)

Male Sprague-Dawley rats fasted for 24 hours were trea¬ ted with 5 to 10 mg/kg of intraperitoneal hydrocortisone or HCN.

Twenty-four hours later the animals were sacrified, the stomach was removed and tissue was grossly examined for the presence of lesions as described by Del Soldato et al.: "The influence of fasting and cimetidine on the relationship be¬ tween ulcerogenic and anti-inflammatory properties of cime¬ tidine", Br.J.Pharmacol. 67, 33-37, 1979. The degree of se¬ verity of the disease was evaluated according to common me¬ thods and expressed as arbitrary values. The results are shown in Table 2.

TABLE 2 STUDY OF GASTRIC TOLERABILITY OF COMPOUND HCN

VERSUS HYDROCORTISONE IN RATS

Data are expressed as arbitrary values according to the fol¬ lowing scale: 0 = absent; 1 = mild lesions; 2 = moderate lesions; 3 = punctiform ulcers; 4 = severe and numerous ul¬ cers.

* P < 0.05 (where P is probability) compared to correspon¬ ding value in group treated with hydrocortisone.

As shown in Table 2, the rats treated with hydrocorti¬ sone exhibited a marked disease in the gastrointestinal tract, varying in severity from mucosal erosion to ulcer involving the muscle layer, wall adhesions, ascites, perito¬ nitis. In the other groups treated with the vehicle alone or HCN, the damage was much lower or even absent. EXAMPLE 2D Study of nitroxysynthetase activity

The nitroxy-sinthetase inhibiting activity induced by lipopolysaccharide (LPS) was determined in rat neutrophils

and stomach after administration of one of the test com¬ pounds and compared with that obtained after treatment with the suspending vehicle only. Wistar rats fasted for 24 hours before treatment received one of the test compounds intrape- ritoneally (10 mg/kg) or LPS intravenously (caudal vein) (5 mg/kg) . Four hours later the animals were sacrificed. Blood for neutrophil isolation and stomach were removed.

Enzymatic activity was determined according to the me¬ thod described by Assreuy et al. : "Feedback inhibition of nitric oxide synthase activity by nitric oxide", Br.J.Phar¬ macol. 108, 833-7, 1993. The results are shown in Table 3.

TABLE 3 STUDY OF NITROXYSYNTHETASE ACTIVITY IN COMPOUND

HCN VERSUS HYDROCORTISONE IN RATS

^a per cent inhibition compared to group treated with vehicle alone p < 0.05 compared to corresponding value in group treated with vehicle.

As shown by Table 3 , both test products proved to be very effective in inhibiting nitroxysynthetase compared to

the group treated with the vehicle alone.

EXAMPLE 2E

Study of bone toxicity

Bone tissues (parietal bone from rat foetus) grown in vitro according to the method described by Doherty et al. ("The effect of glucocorticoids on osteoblast function. The effect of corticosterone on osteoblast, expression of beta-l integrins", Journal of Bone and Joint Surgery, Series A77/3, 396-404, 1995) was used. Hydrocortisone or HCN or the vehi¬ cle were incubated at concentrations of 100 nmol.

Ninety six hours later calcium content and bone dry weight were measured.

The results are shown in Table 4. TABLE 4 - EFFECT OF HCN AND HYDROCORTISONE ON BONE GROWTH IN RATS

° Compared to initial value (incubation time zero) * P < 0.05 compared to values obtained in control group (ve¬ hicle)

As shown in Table 4, a significant increase in tissue dry weight and increased calcium were observed after incuba-

tion with the vehicle or HCN. After incubation with hydroco¬ rtisone, the calcium content decreased and the bone dry weight did not increase. This shows that this treatment with hydrocortisone adversely affected bone growth. EXAMPLE 2F Study of some cardiovascular parameters

The effect of the test products on some cardiovascular parameters was studied in conscious Long Evans rats (350 to 450 g) which were appropriately monitored, as described by Gardiner et al.: "Influence of dexamethasone on the regional haemodynamic responses to lipopolysaccharide in conscious rats: effect of the non-selective endothelin antagonist: SB 209670", Br. . Pharmacology 117, 49P, 1996. The animals were treated with the vehicle (physiologic saline solution, 0,9% sodium chloride, s.c.) subcutaneous hydrocortisone or HCN (10 mg/kg) . Heart rate and blood pressure were recorded 4 hours after treatment.

Table 5 shows the data obtained as per-cent variation from control values .

TABLE 5 - STUDY OF COMPOUND HCN VERSUS HYDROCORTISONE IN SO¬ ME CARDIOVASCULAR PARAMETERS IN RATS

* P < 0.05 compared to group treated with vehicle ^a per-cent change compared to value recorded in group treated with vehicle alone (324+7 beats per minute) ^b per-cent change compared to value recorded in group treated with vehicle alone (101±2 mm Hg)

The results in Table 5 show that the product of the invention HCN does not affect the cardiovascular parameters measured. On the contrary, hydrocortisone used in the known art shows significant pressure as well as cardiac changes. EXAMPLE 2G Study of angiostatic activity in rats

Male Wistar rats weighing 180 to 200 g were used accor¬ ding to the procedure described by Andrade et al.: "Quanti¬ tative in vivo studies on angiogenesis in a rat sponge mo¬ del", Brit. J. Exp. Pathol. 68, 755-766, 1987. The neovascu- larisation was evaluated in relation to blood flow by im¬ planting a small sponge in the subcutaneous tissue for 14 days and determining ¹³³ Xe clearance. Briefly, an amount of ¹³³ Xe equal to lOμl was injected into the sponge using a small polyethylene cannula. The residual radioactivity from

implantation using a gamma ray detector and the ¹³³ Xe clearan¬ ce for 6 minutes was measured as a percentage of the initial value. The validity of this method for measuring neovascula¬ risation was recently demonstrated by HU et al. : "Correla¬ tion of ¹³³ Xe clearance, blood flow and histology in the rat sponge model for angionenesis. Further studies with angioge- nic modifiers", Lab. Invest. 72, 601-610, 1995.

The test compounds were administered by the subcutane¬ ous route at a dose of 10 mg/kg from day 1 to day 13 after implantation. ¹³³ Xe was measured at day 14 from subcutaneous implantation, the animals were then sacrificed and the weights of thymus and spleen were recorded.

Table 6 shows the data obtained regarding the effect of the test products on neovascularisation and on the weight of spleen and thymus.

TABLE 6 - EFFECT OF HCN AND HYDROCORTISONE ON ¹³³ Xe CLEARANCE AND WEIGHT OF SPLEEN AND THYMUS AT DAY 14

* P < 0.05 compared to values obtained in control group (ve¬ hicle)

As evident, HCN proved to be capable of exerting a

marked angiostatic effect without changing the weight of spleen or thymus, differently from the reference product.

As it is clear from the whole of the data shown in Ta¬ bles 1 to 6, the pharmacodynamic activity - anti-arthritic, immunosuppressive and antiangiogenic activities - and tole¬ rability of the nitroderivative are superior than those of the corticoid from the known art. EXAMPLE 3

Dexamethasone 21- (4-bromobutyrate) nil Dexamethasone [I] 3.5 g 8.9 mmol

4-Bromobutyryl chloride 4.06 ml 35 mmol

Potassium carbonate 4.9 g 35 mmol

Tetrahydrofuran 70ml

The solution of compound I in tetrahydrofuran is por- tionwise treated with 4-bromobutyryl chloride (0.81 ml x 5) and potassium carbonate (0.98 g x 5) during 7 hours. The mixture is stirred overnight, the solvent is evaporated un¬ der vacuum and the residue is treated with ethyl ether and water. The organic layer is separated, washed with water and dried with anhydrous sodium sulfate. After evaporation of the solvent, the residue is purified by silica gel flash column chromatography eluting with t.butyl methyl ether ~ hexane l-l to give:

less polar compound 1.0 g; derivative II 1.5 g (m.p. 184-187 °C; yield 31%) TLC: t.butyl methyl ether-hexane 2-1.

Dexamethasone 21- (4-nitrooxybutyrate) πill (compound DXN) Compound II 1.5 g 2.7 mmol

Silver nitrate 2.4 g 14.1 mmol

Acetonitrile 250 ml

The mixture of compound II and silver nitrate in aceto¬ nitrile is refluxed for 7 hours. After filtration of inorga¬ nic salts, the solvent is evaporated under vacuum and the residue is treated with ethyl ether. The organic layer is twice washed with water, dried with anhydrous sodium sulfate and evaporated under vacuum. The residue is poured into ethyl ether and filtered to give 1.27 g of pure compound III as a white solid (m.p. 183-185° C; yield 90%) TLC: t.butyl methyl ether-hexane 2-1. The following forms are enclosed: synthetic scheme;

NCX 1005 batch 1;

NCX 1005/1 analysis.

SINTHETIC SCHEME FOR THE PREPARATION DEXAMETHASONE 21- (4-NITROOXIBUTYRATE)

π

NCX 1005 (III)

EXAMPLE 4

Study of the activity on leucocyte accumulation

Male Swiss albino mice (27-33 g) maintained on a stan¬ dard chow pellet diet and tap water ad libitum were used. The experiment was done as previously described by Perretti et al. (Perretti M. , Solito E., Parente L., "Evidence that endogenous interleukin-l is involved in neutrophil migration in acute experimental inflammation in rats and mice", Agents Actions, 35,71,1992). Animals were pretreated with zymosan (lmg/0.5ml) i.p. at time 0. Two hours later DEXAMETHASONE (1 mg/kg) (Ex3-I) , DXN (Ex3-III) (lmg/kg) or phosphate buf¬ fered saline (PBS) was given intravenously. The animals were sacrificed at 4 and 24h the lavage fluids were collected and differential cell counts were performed following staining in Tur 's.

Table 7 reports results obtained on the inhibitory ef¬ fect of the tested compounds on zymosan-induced leucocyte migration in mice. As can be observed the nitroderivative steroid is much more active than dexamet asone. TABLE 7-Inhibition of neutrophil and monocyte recruitement by DEXAMETHASONE and DXN (lmg/kg) σiven 2h i.v. after zymosan (lmg/0.5ml) i.p.

EXAMPLE 5

Study of the anti-proliferative activity in human airway smooth muscle cells

Human airway smooth muscle cells were cultured by stan¬ dard explant methods. Tissues were collected into sterile pots containing PBS and penicillin and streptomycin. Under sterile tissue culture conditions, tissues were cut into small pieces (approximately l mg weight) and placed into standard medium containing 20% fetal calf serum (FCS) for several days (medium changed every 2-4 days) . ³ H-thymidine was measured in the DNA fraction of cells cultured into 48 well plates. Cells were cultured to confluence in the medium containing 10% FCS. Cells were deprived of serum for 24h before the addition of 10% FCS, together with different con¬ centration of steroids. After 24h, ³ H-thymidine was added to the cells for 4h. Cells were washed with phosphate buffered saline and ethanol. The DNA was extracted with sodium hydroxide solution and the ³ H material counted by scintilla¬ tion. The data represent observations made in triplicate

wells from smooth muscle cultured from one healthy lung do¬ nor. Table 8 reports results obtained on the inhibitory ef¬ fect of the tested compounds on human airway smooth cell proliferation. As can be observed the nitroderivative ste¬ roid is much more active than dexamethasone.

TABLE 8-Inhibition of human airway smooth cell mitogenesis by different concentrations of DEXAMETHASONE and DXN

CONCLUSIONS

As can be observed from results reported above, both activity and safety of the new nitroderivatives are better than those owned by the precursor steroids.