AU661278B2 - Dynemicin analogs: syntheses, methods of preparation and use - Google Patents

Description

DYNEMICIN ANALOGS:

SYNTHESES, METHODS OF PREPARATION AND USE

Description

Cross-Reference to Related Application This is a continuation-in-part of application Serial No. 673,199, filed March 21, 1991, which is a continuation-in-part of application Serial No. 562,269, filed on August 1, 1990, both of whose disclosures are incorporated by reference.

Technical Field

The present invention relates to novel DNA-cleaving, cytotoxic and anti-tumor compounds, and particularly to fused ring systems that contain an enediyne macrocyclic ring and also an epoxide ring, as well as chimeras that contain such a fused ring system.

Background Art

Dynemicin A (Compound 1 shown below),

where Me is methyl, is a potent antibacterial and anticancer agent recently isolated from Micromonospora chersina [(a) Konishi et al, J. Am. Chem. Soc..

112:3715-3716 (1990); (b) Konishi et al., J. Antibiot.. 42: 1449-1452 (1989)]. Its striking molecular structure combines characteristics of both the enediyne [Golik et al., J. Am. Chem. Soc.. 109:3461-3462 (1987); Golik et al., J. Am. Chem. Soc.. 109:3462-3464 (1987); Lee et al., J. Am. Chem. Soc. 109:3464-3466 (1987); Ellestad et al., J. Am. Chem. Soc.. 109:3466-3468 (1987)] and the anthracycline ["Anthracycline Antibiotics", H.S. E1 Khadem, ed., Academic Press, New York (1982) and "Recent Aspects in Anthracyclinone Chemistry", Tetrahedron

Symposia-in-Print No. 17, T.R. Kelly, ed., Tetrahedron; 40): 4537-4794 (1984)] classes of antibiotics, and

presents a considerable challenge to organic synthesis as well as a unique opportunity for the development of new synthetic technology and therapeutic agents.

The calicheamicin and esperamicin derivatives are perhaps the best known of the enediyne compounds. For a key paper describing the first synthesis of calicheamicinone, see: (a) Cabal et al., J. Am. Chem. Soc.. 112:3253 (1990). For other selected studies of model systems in the area of calicheamicinsesperamicins, see: (b) Nicolaou et al, J. Am. Chem.

Soc.. 110:4866-4868 (1988); (c) Nicolaou et al., J. Am. Chem. Soc.. 110:7247-7248 (1988); (d) Schoenen et al., Tetrahedron Lett., 30: 3765-3768 (1989); (e) Magnus et al., J. Am. Chem. Soc.. 110:6921-6923 (1988; (f) Kende et al., Tetrahedron Lett.. 29: 4217-4220 (1988).

Brief Summary of the Invention

The present invention relates to novel fused ring systems that contain an epoxide ring and an

enediyne macrocyclic ring, and thus have structural features similar to dynemicin A. The compounds have DNA-cleaving, antibiotic and antitumor activities.

Compositions and methods of making and using the

compounds are disclosed. A fused ring compound of the invention has a structure that corresponds to the formula

wherein A is a double or single bond;

R¹ is selected from the group consisting of H, C₁-C₆ alkyl, phenoxycarbonyl, benzyloxycarbonyl, C₁-C₆ alkoxycarbonyl, substituted C₁-C₆ alkoxycarbonyl (particularly substituted ethoxycarbonyl), and

9-fluorenylmethyloxycarbonyl;

R² is selected from the group consisting of H, carboxyl, hydroxylmethyl and carbonyloxy C₁-C₆ alkyl;

R³ is selected from the group consisting of H and C₁-C₆ alkoxy;

R⁴ is selected from the group consisting of H, hydroxyl, C₁-C₆ alkoxy, oxyacetic acid, oxyacetic C₁-C₆ hydrocarbyl or benzyl ester, oxyacetic amide, oxyimidazilthiocarbonyl and C₁-C₆ acyloxy;

R⁶ and R⁷ are each H or together with the unsaturated carbon atoms of the intervening vinylene group form a one, two or three fused aromatic six-membered ring system;

W together with the carbon atoms of the depicted, intervening vinylene group forms a substituted aromatic hydrocarbyl ring system containing 1, 2 or 3 six-membered rings such that said fused ring compound contains 3, 4 or 5 fused rings, all but two of which are aromatic, and in which that aromatic hydrocarbyl ring system, W, is joined [a, b] to the structure shown

(i.e., W is joined [a,b] to the nitrogen-containing rings of the structure shown); and

R⁸ is hydrogen or methyl, with the proviso that R⁸ is hydrogen when W, together with the carbon atoms of the intervening vinylene group is

9,10-dioxoanthra.

In preferred practice, W together with the intervening vinylidene group forms a benzo ring so that a compound has the structural formula shown below.

wherein R⁵ is selected from the group consisting of hydrogen, C₁-C₆ alkoxy, hydroxyl, C₁-C₆ acyloxy, oxyethanol, oxyacetic acid, o-nitrobenzyloxy and halo, and A and the remaining R groups are as before described.

More particularly, in one embodiment, R², R³, R⁵, R⁷ and R⁸ are hydrogen so that a compound of the invention corresponds to the structural formula shown below, where R¹ and R⁴ are as previously defined.

More preferably, R is C₁-C₆ alkoxy, hydroxyl,

C₁-C₆ acyloxy, oxyethanol, or oxyacetic acid, and R⁴ is hydrogen (H) or hydroxyl so that a fused ring compound has the structural formula shown below.

Also contemplated is a chimeric compound (also referred to as a chimer or chimera) that is comprised of a before-described fused ring compound as an aglycone portion bonded to (i) an oligosaccharide portion or (ii) a monoclonal antibody or antibody combining site portion thereof that immunoreacts with target tumor cells. The oligosaccharide portion comprises a sugar moiety selected from the group consisting of ribosyl, deoxyribosyl, fucosyl, glucosyl, galactosyl,

N-acetylglucosaminyl, N-acetylgalactasaminyl, a

saccharide whose structure is shown below, wherein a wavy line adjacent a bond indicates the position of linkage

A monoclonal antibody or binding site portion thereof is bonded to the fused ring compound aglycone portion through an R⁴ oxyacetic acid amide or ester bond, or an oxyacetic acid amide or ester bond from W. An oligosaccharide portion is glycosidically bonded to the aglycone portion through the hydroxyl of an R⁴ oxyethanol group or the hydroxyl of an oxyethanol-substituent of W.

A pharmaceutical composition is also contemplated. That pharmaceutical composition contains a DNA cleaving, antibiotic or tumor cell growth-inhibiting amount of a before-defined compound or chimera as active agent dissolved or dispersed in a physiologically tolerable diluent.

A compound, chimera or a pharmaceutical composition of either is also useful in a method for cleaving DNA, for inhibiting tumor growth and as an antimicrobial. In accordance with such a method, the DNA to be cleaved, target tumor cells whose growth is to be inhibited or target microbial cells is (are)

contacted with a composition of the invention. That contact is maintained for a time period sufficient for the desired result to occur. Multiple administrations of a pharmaceutical composition can be made to provide the desired contact.

Brief Description of the Drawings

In the drawings forming a portion of this disclosure,

Figure 1 is a photograph of an ethidium bromide stained 1 percent agarose gel that illustrates the cleavage of øX174 form I DNA by Compound 40 after 24 hours in phosphate buffers (50mM) containing 20 volume percent THF at pH 7.4. Lane 1 is the DNA alone as control, lanes 2-6 show the results obtained with 5000, 2000, 1000, 500 and lOOμM Compound 40, respectively. The designations I, II and III outside the gel indicate forms I, II and III of the DNA, respectively.

Figure 2 is a photograph of an ethidium bromide stained 1 percent agarose gel that illustrates the cleavage of 0X174 from I DNA by Compounds 40, 47, 42, 54, 55, 58 and 62 after 24 hours in pH 8.0 50mM Tris-HCl buffer. Lane 1 is the DNA alone as control, lanes 2, 3, 4, 5, 6, 7 and 8 show the results obtained with 5mM of each of compounds 40, 47, 42, 54, 55, 58 and 62, respectively. The designations Form I, II and III are as in Figure 1.

Figure 3 is a graph showing results from two studies of the percent growth inhibition of MIA PaCa-2 human pancreatic carcinoma cells over a four-day time period by various concentrations of Compound 2 (DY-1).

Figure 4 is a graph showing the results from four studies of the percent growth inhibition over a four-day time period of MB49 murine bladder carcinoma cells by various concentrations of Compound 2 (DY-1) .

IC₅₀ values for two of the studies were 43 nM and 91 nM.

Figure 5 is a graph showing the results from two studies of the percent growth inhibition over a four-day time period of MB49 murine bladder carcinoma cells by various concentrations of Compound 21 (DY-2).

Detailed Description of the Invention

I. The Compounds

A compound of the invention contains an enediyne macrocycle linked to a fused ring that

corresponds to structural Formula I

wherein A is a double or single bond;

R¹ is selected from the group consisting of H, C₁-C₆ alkyl, phenoxycarbonyl, benzyloxycarbonyl, C₁-C₆ alkoxycarbonyl, substituted C₁-C₆ alkoxycarbonyl

(particularly a substituted ethoxycarbonyl) and

9-fluorenylmethyloxycarbonyl;

R² is selected from the group consisting of H, carboxyl, hydroxylmethyl and carbonyloxy-C₁-C₆ alkyl;

R³ is selected from the group consisting of H and C₁-C₆ alkoxy;

R⁴ is selected from the group consisting of H, hydroxyl, C₁-C₆ alkoxy, oxyacetic acid (-OCH₂CO₂H) , C₁-C₆ hydrocarbyl or benzyl oxyacetic acid ester, oxyacetic amide, oxyethanol (-OCH₂CH₂OH) , oxyimidazylthiocarbonyl and C₁-C₆ acyloxy;

R⁶ and R⁷ are each H or together with the intervening vinylene group form a one, two or three fused aromatic six-membered ring system;

W together with the bonded, intervening, vinylene group (i.e., the unsaturated carbon atoms bonded to W) forms a substituted aromatic hydrocarbyl ring system containing 1, 2 or 3 six-membered rings such that said fused ring compound contains 3, 4 or 5 fused 6-membered rings all but two of which rings are aromatic, and in which that aromatic hydrocarbyl ring system, W, is joined [a, b] to the structure shown; and

R⁸ is hydrogen or methyl with the proviso that R⁸ is hydrogen when W together with the intervening vinylidene group is 9,10-dioxoanthra.

Exemplary R⁶ and R⁷ groups are shown in Scheme III and are discussed in relation thereto, and

thereafter.

As noted above, the bond, A, between the R² and R³ substituents can be a double or single bond. The bond A is preferably a single bond.

A C₁-C₆ alkyl group, as can be present in R¹ is exemplified by methyl, ethyl, propyl, isopropyl, butyl, sec-butyl, pentyl, 2-methylpentyl, hexyl, cyclohexyl, cyclopentyl and the like. A substituted C₁-C₆ alkyl group is also contemplated as an R¹ group. Such

substituted alkyl groups include hydroxyalkyl groups such as 2-hydroxyethyl, 4-hydroxyhexyl and

3-hydroxypropyl, haloalkyl groups such as 2-chlorobutyl, 3-halopentyl such as 3-fluoropentyl, and the like. The above C₁-C₆ alkyl and substituted C₁-C₆ alkyl groups are further contemplated as the C₁-C₆ alkyl portion of a carbonyloxy C₁-C₆ alkyl group of R²; i.e., a C₁-C₆ alkyl ester of a R² carboxyl group, and of a R¹ urethane group. Those same alkyl groups can constitute the alkyl portion of a C₁-C₆ alkoxy group of R³ or R⁴. A C₁-C₆ acyloxy group as is present in R⁴ or R⁵ (discussed hereinafter) is a carboxylic acid derivative of an appropriate alkyl group, above, except for, for example, cyclohexyl and iso-propyl, and is limited to a

cyclopentylcarboxyl group for the cyclopentane

derivatives. Examples of such C₁-C₆ acyloxy groups include formyloxy, acetoxy, propionoxy, butyryloxy, iso-butyryloxy, pentanoyloxy, 2-methylbutyryloxy,

pivaloyloxy, hexanoyloxy, and the like. The alcohol-carbonyl portion of a urethane R¹ is typically formed by the reaction of a corresponding halo formate derivative, such as a chloroformate like phenylchloroformate, with the secondary amine nitrogen atom that is formed by addition of an acetyleniσ group-containing moiety to the 6-position or a correspondingly numbered position of a fused ring system such as that shown in Scheme II hereinafter. Such groups can also be prepared by base-catalyzed exchange from a formed carbamate using the substituted ethyl alcohol as is illustrated hereinafter.

Exemplary C₁-C₆ alkoxycarbonyl groups and substituted C₁-C₆ alkoxycarbonyl groups contain a before-described C₁-C₆ alkoxy group or substituted C₁-C₆ alkoxy group linked to the carbonyl group and can be formed by reaction of a C₁-C₆ alkylchloroformate.

Exemplary substituted ethoxycarbonyl groups that are a particularly preferred group of substituted C₁-C₆ alkoxycarbonyl group have a substituent other than hydrogen at the 2-position of the ethoxy group, and include 2-trimethylsilylethoxycarbonyl,

2-phenylsulfonylethoxycarbonyl, α- or β- 2-naphthylsulfonylethoxycarbonyl, α- or β- 2-anthracylsulfonylethoxycarbonyl, 2-propenoxycarbonyl, 2-hydroxyethoxycarbonyl,

2-triphenylphosphoniumethoxycarbonyl halide (e.g., chloride, bromide or iodide) and

2-trimethylammoniumethoxycarbonyl halide (as before).

It is particularly preferred that R¹ be a group that can be enzymatically or otherwise removed intracellularly to provide the resulting secondary amine free of a substituent group. A compound where R¹ contains a 2-substituted-ethoxycarbonyl group such as a 2-phenylsulfonyl-, 2-naphthylsulfonyl- and

2-anthracylsulfonyl- as are shown in Scheme III (shown as R₁ therein) can form the free secondary amine

compound via a β-elimination under relatively mild conditions. Phenylsulfonylethoxycarbonyl, α-naphthyl- and β-naphthylsulfonylethoxycarbonyl (collectively referred to as naphthylsulfonylethoxycarbonyl) are particularly preferred R¹ groups, with phenoxycarboxyl being a preferred R¹ group.

An R⁸ group can be methyl or hydrogen with the proviso that R⁸ is hydrogen when W along with the intervening vinylene group carbon atoms forms a 9,10-dioxoanthra ring. It is particularly preferred that R⁸ be methyl when W forms a benzo ring.

R⁴ groups that are hydrogen, hydroxyl,

oxyethanol (-OCH₂CH₂OH) , oxyacetic acid (-OCH₂CO₂H) , oxyacetic C₁-C₆ hydrocarbyl esters such as the before-discussed C₁-C₆ alkyl groups such as ethyl oxyacetate (-OCH₂CO₂CH₂CH₃) , as well as C₁-C₆ unsaturated esters such as the allyl, propargyl, 2-butenyl and the like, as well as the benzyl ester and oxyacetic amides constitute particularly preferred embodiments of the invention. Exemplary C₁-C₆ and benzyl esters that have been

prepared; i.e., Compounds 24a-g exhibited activity against MIA PaCa-2 tumor cells.

A pharmaceutically acceptable non-toxic salt of the oxyacetic acid such as sodium, potassium, ammonium, calcium and magnesium is also contemplated. An oxyacetic acid amide corresponds to the chemical formula -OCH₂C0NR¹³R¹⁴ wherein R¹³ is hydrogen (H) or C₁-C₆ alkyl (as before) and R¹⁴ is independently hydrogen, C₁-C₆ alkyl, phenyl, 1- or 2-napthyl, 1- or 2-anthryl, or a peptide having 1 to about six amino acid residues; or R¹³ and R¹⁴ together with the amido nitrogen atom form a 5- or 6-membered ring as is present in pyrrolidine, piperidine or morpholine. A particularly contemplated peptide is distamycin, or a derivative thereof as discussed in Taylor et al., Tetrahedron. 40:457 (1984) and Baker et al., J. Am. Chem. Soc.. 111:2700 (1989). Distamycin derivatives are themselves known DNA-cleaving agents. Indeed, a N-bromoacetyldistamycin adduct of Compound 2 has been prepared. Another particularly preferred peptide is -Ala-Ala-Ala-, [(-Ala-)₃].

An R⁴ group that contains a derivatized oxyacetic acid amide or ester can also include a

peptidyl spacer containing zero to about 6 residues such as (-Ala-)₃ that links the compound to a monoclonal antibody or an antibody binding site portion thereof, collectively referred to herein as a "Mab" , as is illustrated in relation to Scheme III hereinafter (R or R₃). The Mab utilized immunoreacts substantially only with target tumor cells; i.e., is tumor cell specific, and thereby provides further specificity to the drug molecules. Such a Mab-linked fused ring enediyne is one type of chimeric molecule of the invention.

The spacer portion of the compound-Mab

construct serves to link the two portions of the

molecule together. When there are zero peptide residues present, a lysine epsilon-amino group of the Mab forms the amido bond shown in Scheme III. The spacer peptide chain, when present, is typically comprised of amino acid residues having small side chains such as glycine or alanine, or relatively hydrophilic side chains such as serine, glutamine and aspartic acid. A peptide spacer is typically free of cysteine residues, but otherwise can have substantially any structure that does not interfere with bonding between the two portions of the chimeric compound. A peptide can be prepared by an one of several synthetic methods as are well known. The Mab portion of the above chimeric

construct can constitute an intact antibody molecule of IgG or IgM isotype, in which case, a plurality of compounds can be present per antibody molecule. The binding site portions of an antibody can also be

utilized, in which case, at least one compound is linked to the proteinaceous antibody binding site portion.

An antibody binding site portion is that part of an antibody molecule that immunoreacts with an antigen, and is also sometimes referred to as a

paratope. Exemplary antibody binding site portions include F(ab), F(ab'), F(ab')₂ and F_v portions of an intact antibody molecule, and can be prepared by well known methods. An intact monoclonal antibody and a portion that includes its antibody combining site portion can be collectively referred to as a paratope-containing molecule.

Exemplary anti-tumor Mabs are noted in the table below, listed by the name utilized in a

publication, along with its deposit accession number at the American Type Culture Collection (ATCC No.), 12301 Parklawn Drive, Rockville, Maryland 20852 U.S.A., and the tumor antigen with which the Mab paratope is

reported to react. A c'tation to a discussion of each Mab and its immunoreactivity is provided by the footnote under the antigen listing.

Exemplary Anti-Tumor Mabs

Mab ATCC No. Antigen

B 3.6 HB 8890 GD3¹

14.8 HB 9118 GD2²

11C64 - - GD3³

9.2.27 Condritin sulfate

proteoglycan⁴

R24 - - GD3⁵

HT29/26 HB 8247 colon cancer

glycoprotein gp 29⁶

HT29/36 HB 8248 colon cancer

glycoprotein gp29⁶

CLT85 HB 8240 colon cancer⁶

F64.5 - - mammary carcinoma⁷

R38.1 pan carcinoma

- - 70Kd protein⁷

F36/22 HB 8215 human breast

carcinoma⁸

T16 HB 8279 human bladder tumor, glycoprotein gp48⁹

T43 HB 8275 human bladder tumor⁹

T101 HB 8273 human bladder tumor⁹

116-NS-19-1 HB 8059 colorectal carcinoma monosialoganglioside¹⁰

126 HB 8568 GD2¹¹

CLH 6 HB 8532 colon cancer¹²

CLG 479 HB 8241 colon cancer¹²

19.9 CRL 8019 CEA¹³

CLNH5 - - lung carcinoma14

16-88 - - colon carcinoma^{15 1} Cheresch et al.. Proc. Natl. Acad. Sci., USA.

82:5155-5159 (1985): Ibid. 81:5767-5771 (1984) 2 Cheresch et al. Cancer Res. 44: 5112-5118 (1986) 3 Cheresch et al., J. Cell. Biol.. 102: 688 (1986) 4 Bumol et al., Proc. Natl. Acad. Sci., USA, 79:1245 (1982); Harper et al., J. Immunol., 132:2096 (1984) 5 U.S. Patent No. 4,507,391

6 U.S. Patent No. 4,579,827

7 U.S. Patent No. 4,522,918

8 European Patent Application No. 84400420.0,

publication No. 0 118 365, published September 12, 1984

9 European Patent Application No. 84102517.4,

publication No. 0 118 891, published September 19, 1984

10 U.S. Patent No. 4,471,057

11 Cheresch et al., J. Cell. Biol.. 102:688 (1986)

U.S. Patent No. 4,675,287

12 U.S. Patent No. 4,579,827

13 U.S. Patent No. 4,349,528

14 Patent Application PCT/US83/00781, WO 83/04313

15 European Patent Application No. 85300610.4,

publication No. 0 151 030, published August 7 , 1985 A fused ring enediyne compound of the invention can also be glycosidically linked to a sugar moiety to form a second type chimeric molecule. In such a chimer, the fused ring enediyne compound takes the place of the aglycone as in an antibi ic molecule such as doxorubicin, calicheamicin or esperamicin, with the sugar moiety taking the place of the oligosaccharide portion. Bonding between the used ring enediyne compound aglycone and oligosaccharide is typically via a hydroxyl group of a spacer group that is itself linked to the fused ring enediyne through a reacted hydroxyl group. A preferred spacer group is an oxyethanol group that can be an R⁴ group or can be a substituent of W as is discussed and illustrated hereinafter.

The oligosaccharide portion of the molecule is typically added after the synthesis of the fused ring enediyne compound (aglycone) portion is complete, except for any blocking groups on otherwise reactive

functionalities of the aglycone that are typically removed after addition of the oligosaccharide portion. A sugar moiety is added by standard techniques as are discussed hereinafter.

A glycosidically-linked sugar moiety can be a monosaccharide such as a ribosyl, deoxyribosyl, fucosyl, glucosyl, galactosyl, N-acetylglucosaminyl,

N-acetylgalactosaminyl moiety or the more preferred saccharides whose structures are shown below, wherein a wavy line adjacent a bond indicates the position of linkage

The position of the glycosyl bond to be formed in the sugar moiety used for forming a chimeric compound is typically activated prior to linkage to the fused ring enediyne compound. For example, the 1-position hydroxyl group of an otherwise protected sugar (as with ^tBuMe₂Si or Et₃Si groups) is reacted with

diethylaminosulfur trifluoride (DAST) in THF and in the presence of 4A molecular sieves at -78°C to form the 1-fluoroderivative. The enediyne having a free hydroxyl group is then reacted with the 1-fluro-protected

saccharide in the presence of silver perchlorate and stannous chloride to provide a protected desired, typically blocked, chimer molecule.

Similarly, treatment of 1-position hydroxyl of an otherwise protected saccharide with sodium hydride and trichloracetonitrile [Grandler et al., Carbohydr. res., 135:203 (1985); Schmidt, Angew. Chem. Int. Ed. , Engl., 25:212 (1986)] in methylene chloride at about room temperature provides a 1-α-trichloroacetimidate group to activate the saccharide for coupling with the fused ring enediyne (aglycon) hydroxyl. Coupling is then carried out in boron trifluoride-etherate in methylene chloride to provide the protected desired chimer compound.

Once the aglycone and oligosaccharide are coupled, the protecting groups that are present are removed to provide the desired compound, which is then recovered using standard techniques. Exemplary

syntheses are discussed hereinafter.

The 1, 2 or 3 six-membered ring fused rings that along with the depicted vinylene group constitute the structure W are aromatic hydrocarbyl rings. Such rings can thus be benzo, naphtho and anthra rings, using fused ring nomenclature. The anthra (anthracene) derivative rings contemplated here contain 9,10-dioxo groups (are derivatives of anthraquinone) and are therefore referred to as 9, 10-dioxoanthra rings.

Where a benzo, naphtho or 9,10-dioxoanthra ring forms part of the fused ring system, those fused rings are bonded to the remaining fused ring system through the carbon atoms of the 1- and 2-positions or are (a, b). A benzo, naphtho or 9 , 10-dioxoanthra fused ring portion can also contain one or more substituents at the ring positions remaining for substitution. Those substituent groups are selected from the group

consisting of hydroxyl, C₁-C₆ alkoxy, oxo, C₁-C₆ acyloxy and halo (chloro, bromo or iodo).

For a benzo ring, one or two substituents can be present at one or two of the remaining positions of the radical. Symmetrical substitution by the same substituent is preferred because of the lessened

possibility for isomer formation. When a single

substituent is present on a benzo ring, that substituent is referred to as R⁵, which designation for convenience includes hydrogen. R⁵ is thus selected from the group consisting of hydrogen (no substituent), C₁-C₆ alkoxy, benzyloxy, o-nitrobenzyloxy, hydroxyl, C₁-C₆ acyloxy, oxyethanol, oxyacetic acid, oxyacetic acid C₁-C₆

hydrocarbyl ester and halo. It is preferred that a hydroxyl group or a group that can form a hydroxyl group intracellularly be present, such that a hydroxyl group be present intracellularly at a position meta to the nitrogen in the adjacent ring. When two substituents are present on a benzo ring, they are referred to as R ¹⁰ and R¹¹ and are selected from the group consisting of

C₁-C₆ alkoxy, benzyloxy, oxo, C₁-C₆ acyloxy, hydroxyl and halo.

W is more preferably a benzo group that contains a single substituent R⁵. In one particularly preferred embodiment, R⁵ is situated in the benzo ring meta or para to the nitrogen atom bonded to R¹. That R⁵ group is more preferably selected from the group consisting of hydroxyl, C₁-C₆ alkoxy, benzyloxy,

o-nitrobenzyloxy, C₁-C₆ acyloxy, oxyethanol, oxyacetic acid and an oxycacetic C₁-C₆ hydrocarbyl ester.

When R⁵ is meta to the above nitrogen atom, it is preferred that the R⁵ group be an electron releasing group such as hydroxyl or a C₁-C₆ acyloxy group that can provide a hydroxyl group intracellularly. A C₁-C₆ acyloxy group is believed to be a pro-drug form of the hydroxyl group that is cleaved intracellularly by an endogenus esterase or the like to provide the hydroxyl group. The presence of such an electron releasing group appears to assist in enhancing the potency of the compound against target tumor cells. It is believed that the enhanced potency is due to enhanced triggering of the epoxide opening and cyclization reactions.

When R⁵ is para to the above nitrogen atom, it is preferred that the R⁵ group be an o-nitrobenzyloxy group, oxyethanol, oxyacetic acid or oxyacetic acid

C₁-C₆ hydrocarbyl ester. Those groups are particularly useful for the preparation of chimeras.

A particularly preferred compound has a structure corresponding to Formula Xlb, hereinafter.

A naphtho ring can have three substituents.

This ring can have a 4-position radical, R⁵, selected from the group consisting of hydroxyl, C₁-C₆ alkoxy, benzyloxy, C₁-C₆ acyloxy and halo, and substituents at the 5- (R¹⁰) and 8-positions (R¹¹) that are selected from the group consisting of hydroxyl, C₁-C₆ alkoxy,

benzyloxy, C₁-C₆ acyloxy, oxo and halo radicals. A 9 , 10-dioxoanthra ring can have three substituents at the 4- (R⁵) , 5- (R⁹) and 8-positions (R¹²) that are

independently selected from the group consisting of hydroxyl, C₁-C₆ alkoxy, benzyloxy, C₁-C₆ acyloxy and halo. Thus, R⁵, R⁹ and R¹² can define the same groups, and all three groups can be written as either R⁵, R⁹ or R¹², but they are shown separately herein.

Exemplary structural formulas for such fused ring compounds are illustrated below by structural Formulas II-IX, wherein each of the R groups is as discussed before.

In addition to the before-stated preference regarding R⁸ and that bond A be a single bond, several other structural features and substituents are

preferred.

Thus, it is preferred that R² and R³ be hydrogen, and that R⁶ and R⁷ be hydrogen. It is also preferred that the fused ring system W together with the depicted vinylene group be substituted benzo, or an unsubstituted benzo, naphtho or 9,10-dioxoanthra ring. It is further preferred that the fused ring compound contain a total of 3-fused six-membered rings so that W together with the depicted vinylene group forms a benzo ring.

One particularly preferred group of compounds of the invention in which W is an R⁵-substituted benzo ring corresponds to structural Formula X.

wherein A is a double or single bond; R¹ is selected from the group consisting of H,

C₁-C₆ alkyl, phenoxycarbonyl, benzoxycarbonyl, C₁-C₆ alkoxycarbonyl and 9-fluorenylmethyloxycarbonyl;

R² is selected from the group consisting of H, carboxyl, hydroxylmethyl and carbonyloxy C₁-C₆ alkyl; R³ is selected from the group consisting of H and C₁-C₆ alkoxy;

R⁴ is selected from the group consisting of H, hydroxyl, oxyacetic acid (-OCH₂CO₂H) , oxyacetic C₁-C₆ hydrocarbyl or benzyl ester, oxyacetic amide,

oxyethanol, oxyimidazylthiocarbonyl and C₁-C₆ acyloxy;

R⁵ is selected from the group consisting of hydrogen, C₁-C₆ alkoxy, benzyloxy, o-nitrobenzyloxy, hydroxyl, C₁-C₆ acyloxy, oxyethanol, oxyacetic acid, oxyacetic acid C₁-C₆ hydrocarbyl ester and halo; and

R⁶ and R⁷ are each H or together form with the intervening vinylidine group form a one, two or three fused aromatic ring system, and R⁸ is methyl or

hydrogen.

A still more preferred group of compounds of the invention correspond to structural Formulas XI, XIa and Xlb.

wherein R¹, R⁴, R⁵ and R⁸ are as previously defined.

Of the compounds corresponding to structural Formula XI, there are further preferences for R¹, R⁴ and R⁵. These preferences also relate to the previously discussed compounds. Thus, R¹ is most preferably phenoxycarbonyl phenylsulfonylethoxycarbonyl,

naphthylsulfonylethoxycarbonyl or hydrogen. R⁸ is most preferably hydrogen (H) to provide a compound of

Formulas XIa or Xlb. R⁴ is most preferably H, hydroxyl, imidazylthiocarbonyloxy, benzyl oxyacetate and C₁-C₆ hydrocarbyl oxyacetate such as ethyl oxyacetate. R⁵ in Formulas XI and XIa is H, but is more preferably

hydroxyl, C₁-C₆ alkoxy, benzyloxy, C₁-C₆ acyloxy,

oxyethanol, oxyacetic acid, oxyacetic acid C₁-C₆

hydrocarbyl or benzyl ester and o-nitrogenzyloxy as in Formula Xlb. It is noted that an R⁵ o-nitrobenzyloxy group is not usually used in a pharmaceutical

composition discussed hereinafter.

The structural formulas of particularly preferred compounds are shown below, along with compound numbers as utilized herein. In the formulas below and elsewhere herein, Ph = phenyl. Me = methyl, NBnO = fi-nitrobenzyloxy, Naph = naphthyl and ^tBuCO₂ = pivaloyl.

II. Pharmaceutical Compositions

A compound or chimera of the invention is useful as a DNA cleaving agent, and also as an

antimicrobial and a cytoxic (antitumor) agent, as are dynemicin A, calicheamicin, esperamicin and

neocarzinostatin. A compound of the invention can also therefore be referred to as an "active agent" or "active ingredient".

DNA cleavage can be assayed using the techniques described hereinafter as well as those described by et al., J. Pro. Chem.. 54:2781 (1989); Nicolaou et al., J. Am. Chem. Soc.. 110:7147 (1989); Nicolaou et al., J. Am. Chem. Soc.. 110:7247 (1988) or Zein et al., Science. 240:1198 (1988) and the citations therein. A compound or chimer of the invention is useful against Gram-positive bacteria such as S. aureus and epidermis. Micrococcus luteus and Bacillus subtillis as is dynemicin A. Such a compound or chimer also exhibits antimicrobial activity against E. coli.

Pseudomonas aeruginos. Candida albucans and Aspergillis fumigatus. Activity of a compound of the invention against the above microorganisms can be determined using various well known techniques. See, for example,

Konishi et al., J. Antibiotics. XLII:1449 (1989).

Antimicrobial and antitumor assays can also be carried out by techniques described in U.S. Patent No.

4,837,206, whose disclosures are incorporated by

reference, as well as by the procedures described hereinafter.

A before-described compound can also be shown to undergo a Bergman cycloaromatization reaction in the presence of benzyl mercaptan, triethylamine and 1,4-cyclohexadiene as discussed in Haseltine et al., J. Am. Chem. Soc.. 111:7638 (1989). This reaction forms a tetracyclic reaction as is formed during DNA cleavage, and can be used as a co-screen to select more active compounds.

A pharmaceutical composition is thus contemplated that contains a before-described compound or chimer of the invention as active agent. A

pharmaceutical composition is prepared by any of the methods well known in the art of pharmacy all of which involve bringing into association the active compound and the carrier therefor. For therapeutic use, a compound or chimer of the present invention can be administered in the form of conventional pharmaceutical compositions. Such compositions can be formulated so as to be suitable for oral or parenteral administration, or as suppositories. In these compositions, the agent is typically dissolved or dispersed in a physiologically tolerable carrier.

A carrier or diluent is a material useful for administering the active compound and must be

"pharmaceutically acceptable" in the sense of being compatible with the other ingredients of the composition and not deleterious to the recipient thereof. As used herein, the phrases "physiologically tolerable" and "pharmaceutically acceptable" are used interchangeably and refer to molecular entities and compositions that do not produce an allergic or similar untoward reaction, such as gastric upset, dizziness and the like, when administered to a mammal. The physiologically tolerable carrier can take a wide variety of forms depending upon the preparation desired for administration and the intended route of administration.

As an example of a useful composition, a compound or chimer of the invention (active agent) can be utilized, dissolved or dispersed in a liquid

composition such as a sterile suspension or solution, or as isotonic preparation containing suitable

preservatives. Particularly well-suited for the present purposes are injectable media constituted by aqueous injectable buffered or unbuffered isotonic and sterile saline or glucose solutions, as well as water alone, or an aqueous ethanol solution. Additional liquid forms in which these compounds or chimers can be incorporated for administration include flavored emulsions with edible oils such as cottonseed oil, sesame oil, coconut oil, peanut oil, and the like, as well as elixirs and similar pharmaceutical vehicles. Exemplary further liquid diluents can be found in Remmington's Pharmaceutical Sciences. Mack Publishing Co., Easton, PA (1980).

An active agent can also be administered in the form of liposomes. As is known in the art. liposomes are generally derived from phospholipids or other lipid substances. Liposomes are formed by mono- or multi-lamellar hydrated liquid crystals that are dispersed in an aqueous medium. Any non-toxic,

physiologically acceptable and metabolizable lipid capable of forming liposomes can be used. The present compositions in liposome form can contain stabilizers, preservatives, excipients, and the like in addition to the agent. The preferred lipids are the phospholipids and the phosphatidyl cholines (lecithins), both natural and synthetic.

Methods of forming liposomes are known in the art. See, for example, Prescott, Ed., Methods in cell Biology. Vol. XIV, Academic press, New York, N.Y.

(1976), p.33 et seq.

An active agent can also be used in compositions such as tablets or pills, preferably containing a unit dose of the compound or chimer. To this end, the agent (active ingredient) is mixed with conventional tableting ingredients such as corn starch, lactose, sucrose, sorbitol, talc, stearic acid,

magnesium stearate, dicalcium phosphate, gums, or similar materials as non-toxic, physiologically

tolerable carriers. The tablets or pills can be

laminated or otherwise compounded to provide unit dosage forms affording prolonged or delayed action.

It should be understood that in addition to the aforementioned carrier ingredients the

pharmaceutical formulation described herein can include, as appropriate, one or more additional carrier

ingredients such as diluents, buffers, flavoring agents, binders, surface active agents, thickeners, lubricants, preservatives (including antioxidants) and the like, and substances included for the purpose of rendering the formulation isotonic with the blood of the intended recipient.

The tablets or pills can also be provided with an enteric layer in the form of an envelope that serves to resist disintegration in the stomach and permits the active ingredient to pass intact into the duodenum or to be delayed in release. A variety of materials can be used for such enteric layers or coatings, including polymeric acids or mixtures of such acids with such materials a& shellac, shellac and cetyl alcohol, cellulose acetate phthalate, and the like. A

particularly suitable enteric coating comprises a styrene-maleic acid copolymer together with known materials that contribute to the enteric properties of the coating. Methods for producing enteric coated tablets are described in U.S. Patent 4,079,125 to Sipos, which is herein incorporated by reference.

The term "unit dose" , as used herein, refers to physically discrete units suitable as unitary dosages for administration to warm blooded animals, each such unit containing a predetermined quantity of the agent calculated to produce the desired therapeutic effect in association with the pharmaceutically acceptable

diluent. Examples of suitable unit dosage forms in accord with this invention are tablets, capsules, pills, powder packets, granules, wafers, cachets, teaspoonfuls, dropperfuls, ampules, vials, segregated multiples of any of the foregoing, and the like.

A previously noted preferred or particularly preferred compound or chimer is preferred or

particularly preferred for use in a pharmaceutical composition.

A compound or chimer of the invention is present in such a pharmaceutical composition in an amount effective to achieve the desired result. For example, where in vitro DNA cleavage is the desired result, a compound or chimer of the invention can be utilized in an amount sufficient to provide a

concentration of about 1.0 to about 5000 micromolar (μM) with a DNA concentration of about 0.02 μg/μL. As a cytoxic (antitumor) agent, an effective amount of a compound or chimer of the invention is about 0.1 to about 15 mg per kilogram of body weight or an amount sufficient to provide a concentration of about 0.01 to about 50 μg/mL to the bloodstream. A compound or chimer of the invention exhibits antimicrobial activity in a concentration range of about 0.01 mg to about 50 μg/mL. The above concentrations and dosages vary with the particular compound of the invention utilized as well as with the target, e.g., DNA, tumor, microbe, as is well known.

III. Methods

A compound or chimer of the invention is useful in cleaving DNA, as a cytotoxic agent and also in inhibiting the growth of neoplastic cells, and is utilized in a method for effecting such a result. A compound or chimer of the invention is typically

utilized in a before-described composition.

In accordance with such a method, DNA or target cells to be killed or whose growth is to be inhibited are contacted with a composition that contains a compound or chimer of the invention (active

ingredient) present in an amount effective or sufficient for such a purpose, as discussed before, dissolved or dispersed in a physiologically tolerable

(pharmaceutically acceptable) diluent. That contact is maintained for a time sufficient for the desired result to be obtained; i.e., DNA cleaved, cells killed or neoplastic cell growth inhibited. Where the desired result is carried out in vitro. contact is maintained by simply admixing the DNA or target cells with the composition and maintaining them together under the appropriate conditions of temperature and for cell growth to occur, as for

control, untreated cells. Thus, a single admixing and contacting is typically sufficient for in vitro

purposes.

The above method is also useful in vivo, as where a mammal such as a rodent like a rat, mouse, or rabbit, a farm animal like a horse, cow or goat, or a primate like a monkey, ape or human is treated. Here, contact of a composition and the cells to be killed or whose growth is to be inhibited is achieved by

administration of the composition to the mammal by oral, nasal or anal administration or by introduction

intravenously, subcutaneously or intraperitoneally.

Thus, contact in vivo is achieved via the blood or lymph systems.

Although a single administration (admixture) and its resulting contact is usually sufficient to maintain the required contact and obtain a desired result in vitro, multiple administrations are typically utilized in vivo. Thus, because of a body's breakdown and excreting pathways, contact between an active ingredient of a composition and the target cells is typically maintained by repeated administration of a compound of the invention over a period of time such as days, weeks or months, or more, depending upon the target cells.

Exemplary methods oi die invention for DNA cleavage and inhibition of MIA PaCa-2 human Pancreatic carcinoma (ATCC CRL 1420) and MB49 murine bladder carcinoma target cells (obtained from Dr. Lan Bo Chen of the Dana Farber Cancer Institute, Boston, MA) as well as ten other neoplastic cell lines are discussed

hereinafter.

IV. Compound Syntheses

A compound of the invention can be prepared by a number of routes, several of which are illustrated in the schemes hereinafter. The retrosynthetic plan for these syntheses is illustrated below in Scheme I, with the general forward synthesis shown in Scheme II, thereafter.

Briefly, the basic fused 3-, 4- or 5-fused six-membered ring system is first formed.

Following Scheme II for a general synthesis, using the tetrahydro-phenanthridine fused ring system of Compound 4 (W is benzo) as exemplary, an oxygen-containing substituent R⁴ having an oxygen atom of that group bonding to the 10-position ring carbon atom is formed as Compound 5. Introduction of that oxygen-containing substituent can be accomplished by oxidation as with m-chloroperbenzoic acid (mCPBA) , followed by acylation and reflux to rearrange the formed acylated N-oxide to the 10-position (steps a and b) of Compound 5. The acetyl group is removed to form the alcohol (Compound 6, step c), which is then blocked with a t-butyldimethylsilyl group (Si^tBuMe₂; Compound 7, step d).

An appropriate acetylenic group-containing compound is added adjacent to the nitrogen atom (at the 6-position) as by reaction of an ethynyl Grignard reagent in the presence of an activating moiety that also functions to block the secondary amine so

generated, such as phenyl chloroformate to form an R¹ substituent of Compound 8 (step e) .

The epoxide ring is added next between the 6- and 10-positions by oxidation as with (mCPBA) as in

Compound 9 (step f) . The formed epoxide ring is on the opposite side of the ring plane from the 6-position acetylenic group, and preferably, the epoxide is in an α-configuration, whereas the acetylenic group is β.

The oxygen-linked R⁴ group Si^tBuMe₂ is replaced with a hydrogen (step g) and the alcohol so formed is next converted to a ketone. Compound 11 (see also step h, Scheme III). The vinyl acetylene portion of the enediyne-containing ring is added as in Compound 13, when necessary. This step can be carried out by reacting (Z) 1-chloro-4-trimethylsilyl-but-1-en-3-yne (Compound 12) with the ketone Compound 11 in the

presence of butylamine, triphenylphosphine and

palladium¹¹ acetate (step i).

Scheme III illustrates that the entire

enediyne carbon skeleton can be bonded to the 6-position in a single step, as is discussed hereinafter. Scheme III also illustrates formation of a benzo ring W by reaction of an aniline compound with ethyl

cyclohexanone-2-carboxylate. Compound 41 was prepared starting with m-anisidine (R⁵ = m-methoxy), whereas a meta-pivaloyl ester (R⁵ = meta-C₅ acyloxy) was prepared from a meta-benzyl ether in preparing Compounds 59a and 59b. Where a R⁸ methyl group is desired, ethyl

3-methylcyclohexanone-2-carboxylate is utilized.

A derivative of compound such as Compounds 2 or 21 having a hydroxyl group (R⁵, or R₄ of Scheme III) can be prepared following the synthetic route

illustrated in Scheme III (R₂=H,H) through step j and then Scheme VII, by utilizing meta-(2-nitrobenzyloxy) aniline for preparation of a tri-cyclic compound analogous to that formed in step a of Scheme III. The meta-2-nitrobenzyloxy (NBnO) group can be removed after step j of Scheme III to form the meta-hydroxyl-substituted derivative of a compound such as

Compounds 2 or 21 by irradiation with ultraviolet light. A meta-hydroxyl-substituted derivative of Compound 2, Compound 42, has been so prepared via Compound 43, irradiated, and shown to cause cleavage of double stranded DNA at a 2mM concentration.

After removal of the trimethylsilyl group from the otherwise free (unlinked) acetylenic group (Scheme II, step j), the acetylenic group is inserted into the carbonyl group at position-10 by reaction with a base such as lithium diisopropylamide (LDA) to form a fused ring compound of the invention where R⁴ is hydroxyl (OH) (step k). The R⁴ group can later be replaced with a hydrogen or derivatized as discussed hereinafter.

As noted before, R⁶ and R⁷ are preferably hydrogen (H) . However, R⁶ and R⁷ along with the

intervening vinylidene group can together form an aromatic mono-, di- or tri-cyclic ring system that can be hydrocarbyl or heterocyclic. When such a ring system is present, the ethylenic bond of the enediyne-containing ring is also one of the unsaturated carbon-to-carbon bonds of aromatic ring system, and the entire enediyne carbon skeleton is typically bonded at the 6- position (see Scheme III, R₂) as a single unit, as is shown in Scheme XI.

Turning more specifically to Scheme III, 2-ethoxycarbonylcyclohexanone is reacted with the aniline (R₄ as shown hereinafter), and the reaction product cyclized by treatment with H₂SO₄. The cyclized material is reduced with lithium aluminum hydride and then air oxidized (step a) to form the tricyclic compound. That compound is oxidized, acetylated and rearranged (steps b and c) to form the acetoxy compound (R=OAc) whose acetyl group is removed to form the alcohol (step d, R=OH) that is then reacted with t-butyldimethylsilyl trifluoromethanesulfonate in the presence of 2,6-lutidine in step e (R=OSi^tBuMe₂).

The product of step e is reacted first with a chloroformate whose R₁ group is shown hereinafter, and then with diacetylide or an aromatic diacetylide ring system compound (R₂ as shown hereinafter) blocked with a trimethylsilyl group (TMS) and containing a mono-Grignard reagent in step f to form the partially linked macrocyclic ring precursor.

The epoxide ring is formed in step g by reaction with mCPBA. The Si^tBuMe₂ group is removed in step h by reaction with ⁿBuNF, and the resulting alcohol is oxidized with pyridinium chlorochromate in the presence of molecular sieves in step i to form the ketone.

The macrocyclic ring is closed in step j by reaction with lithium diisopropylamide (LDA). The hydroxyl group formed in step j is reacted in step k with an appropriate 2-haloacetic acid derivative (R₃ as shown hereinbelow) to form the final product.

R₃ = CH₂COOH

CH₂CONH-peptide

CO(CH₂)₂COOMe

CH₂CONJHPh

CH₂CONH-Naphthyl

CH₂CONH-Anthracyl

CH₂CO₂-spacer-Mab

spacer-Saccharide

R₄ = OMe

OH

OCOMe

OCO'Bu

ONBn

H

Subscripted R groups are used in this scheme to distinguish R groups therein from the superscripted R groups defined elsewhere herein.

The "spacer" noted for R₃ is a peptide that typically contains zero to about six amino acid residues that links a monoclonal antibody, "Mab", to the

compound. The R³ "spacer" linked to an oligosaccharide is typically an oxyethanol or oxyacetic acid group used to form the glycosidic bond to the saccharide, or bond to a paratope-containing molecule. A vicinal diyne aromatic compound suitable for introduction at the 6-position (based on Compound 4) of a 3-, 4- or 5-fused six-membered ring system can be prepared by alkylation of a vicinal dihalide with trimethylsilylacetylene in the presence of

diisopropylamine, dicyanophenylpalladium chloride, triphenylphosphene and cuprous iodide to form the vicinal bis-trimethylsilylacetylene derivative.

Exemplary vicinal dihalo aromatic compounds commercially available from Aldrich Chemical Co. of Milwaukee, WI include 1,2-diiodobenzene,

1,2-dibromobenzene and 1,2-dichlorobenzene.

2,3-Dibromonaphthalene, 2,3-dibromoanthracene and

6,7-dichloroquinoxaline have the following Chemical Abstracts Registry Numbers (R.N.) 13214-70-5,

117820-97-0, and 19853-64-6.

After preparation of the vicinal bis-trimethylsilylacetylene derivative, one of the

trimethylsilyl groups (TMS) is replaced with a hydrogen atom by reaction with silver nitrate and potassium cyanide. The resulting aromatic diacetylide is then reacted with the fused ring system as described

previously.

An appropriate vicinal diacetylide can also be prepared via a vicinal, dihydroxymethyl compound. In one exemplary synthesis, 1,4-dimethoxy-6,7-dimethylnaphthalene (R.N. 73661-12-2) is reacted with N-bromosuccinimide (NBS) and azobisisobutyronitrile (AIBN) in a halogenated solvent such as carbon

tetrachloride to form the corresponding vicinal

dibromomethyl derivative. Reaction of that dibromo compound with hydroxide ion forms the vicinal

dihydroxymethyl derivative. Mild oxidation of the dihydroxymethyl compound with pyridinium chlorochromate (PCC) provides the corresponding vicinal dialdehyde. Reaction of the dialdehyde with triphenylphosphine and carbon tetrabromide, followed by reaction with butyl lithium and then with trimethylsilyl chloride provides the vicinal di-TMS acetylene derivative. Following replacement of one TMS group with hydrogen as discussed before, the aromatic diacetylide is reacted with the fused ring system as was also discussed before.

The above method of synthesis can also be applied to unsubstituted aromatic compounds, such as vicinal dicarboxylic acids, anhydrides or esters. For example, phthalic acid and naphthalene-2,3-dicarboxylic acid are both available from Aldrich Chemical Co.

Either or both can be used to form the corresponding dimethyl esters by reaction with diazomethane.

Reduction of the diesters to vicinal dihydroxymethyl derivatives can be accomplished by reduction using diisobutylaluminum hydride (DIBAL). The resulting dihydroxymethyl compounds are thereafter reacted as described above to form a desired compound.

It is preferred that an aromatic diacetylide contain its two vicinal acetylenic groups symmetrically bonded to the ring system so as to minimize isomer formation. Thus, 2,3-disubstituted-naphthalene, anthracene or quinoxaline compounds are utilized, or a ϋ,7-disubstituted-quinoxaline, or the like.

An exemplary synthesis for compounds corresponding to structural Formula V is illustrated in Scheme IV, shown below. Thus, chloronapthazarin. Compound 25,

[Banville et al., Can J. Chem., 52:80 (1974); Savard et al., Tetrahedron. 40:3455 (1984); and Echavarren et al., J. Chem. Res. (3 ), 364 (1986)] is reacted in step i with diene Compound 26 [Schmidt et al., Synthesis, 958

(1982)] in a Diels-Alder reaction to form the

aminoanthraquinone, Compound 27. Reduction of the quinone as with DIBAL in step ii, followed by alkylation with methylene bromide in the presence of cesium

fluoride in DMF as step iii provides the O-blocked

Compound 28.

Removal of the urethane group with base in step iv, annellation and subsequent reaction as

illustrated in Scheme III provides Compounds 29 (steps v-viii) and 30 (steps ix and xii). Compound 30 is then reacted with lithium iodide in pyridine, oxidized with 2,3-dichloro-5,6-dicyanobenzoguinone (DDQ) , and reacted with pivaloyl chloride (Piv) to form Compound 31 (steps xiii-xv).

Ethylmagnesium bromide is reacted with

Compound 31 in the presence of phenyl chloroformate (step xvi). The resulting compound is reacted with mCPBA to form the epoxide (step xvii). The TBS group is removed with ⁿBu₄NF to form the alcohol (step xviii) that is then oxidized with Jones reagent (step xix) to form the corresponding ketone. That ketone is reacted with (Z)-1-chloro-4-trimethylsilylbut-1-en-3-yne, palladiumº, Cul and butylamine to form Compound 32 (step xx). Compound 32 is reacted with silver nitrate and potassium cyanide (step xxi), IDA (step xxii),

thiocarbonyldiimidazole (step xxiii), tri-n-butylstannane and AIBN (step xxiv), and then hydroxide ion (step xxv) to form Compound 33.

For preparation of a compound corresponding to structural Formula IX, 1-aminoanthraquinone (Aldrich Chemical Co.) can be used as a starting material. Here, the amino group is blocked as with a t-Boc group, the quinone function is reduced as with DIBAL and the resulting phenolic hydroxyl groups are blocked with pivaloyl chloride.

The t-Boc group is then removed as by reaction with trifluoroacetic acid (TFA) and the aromatic ring is annelated with ethyl cyclohexanone-2-carboxylate, followed by cyclization with sulfuric acid, reduction with lithium aluminum hydride and then air oxidation to reform the quinoid structure as the pivaloyl groups are lost during the prior reactions.

The above reactions form the fused ring system. The enediyne macrocycle portion is thereafter added as discussed before.

Where a double bond is desired in the otherwise saturated six-membered ring as is shown in structure Formulas VI, VII, VIII and IX, that

functionality can be added as follows. The fused ring system and epoxide are formed, one of the acetylenic linkages is made, the amine blocked and the ketone is formed, all as discussed previously. Exemplary

compounds of the required structure are Compounds 13 and 32 shown in Scheme II and IV. The compound is then oxidized to introduce hydroxyl group α (position 9 of Compound 13) to the ketone at position 10 of Compound 13. That hydroxyl group is then blocked as with

triethylsilyl (TES or Et₃Si) chloride and the

macrocyclic enediyne ring is closed. The hydroxyl group resulting from the ring closure is either removed as discussed before or blocked with a group that is not removed by removal of the TES group.

The TES group is removed and the resulting hydroxyl group is oxidized to a ketone as by Swern oxidation. That ketone is then reacted with LDA and methyl chloroformate to form a carboxy enol whose hydroxyl group can be methylated with diazomethane.

Subsequent reaction with hydroxide ion and

neutralization provides the unsaturated methoxy

carboxylic shown in structural Formulas VI, VII, VIII and IX.

For a compound of structural Formula VI, where the R⁸ methyl group is present, that methyl group is in the β-configuration. Here, the ketone formed by the above Swern oxidation is reacted with LDA,

phenylselenylbromide and hydrogen peroxide to form an enone. That enone is reduced with copper hydride, which attacks from the α-face to provide the β-stereochemistry for the R⁸ methyl group. The reduced enone is then reacted as above to provide the double bond and R² and

R³ groups

A compound of the general formula of Compound

24

can be prepared by reaction of an appropriate

2-haloacetic acid (Compound 24a) or 2-haloacetate ester (Compounds 24b-g) with Compound 2 in the presence of a base such as cesium carbonate (Cs₂CO₃) and a crown ether such as 18-crown-6. An exemplary synthesis of Compound 24c is provided hereinafter. A compound wherein R⁴ is a C₁-C₆ alkoxy group can be similarly prepared from

Compound 2 by alkylation with a C₁-C₆ halo derivative such as iodo or a triflate derivative in the presence of Cs₂CO₃.

Removal of the carbamate group from the nitrogen atom of a compound such as Compound 2 to form a free amine-containing compound such as Compound 40 is achieved by reaction with lithium aluminum hydride reduction. Variants in the alcohol portion of a

carbamate derivative of a dynemicin analog can be prepared from a phenoxycarbonyl derivative such as

Compound 2 by reaction with the replacing alcohol in the presence of the sodium salt of the alcohol as is shown schematically in Figure 15. V. Results:

The key retrosynthetic step that led to the present synthetic strategy is shown in Scheme I, as noted previously. Scheme II outlined the construction of Compound 2 starting from the quinoline derivative Compound 4 [(a) Masamune et al., J. Org. Chem.. 29:681-685 (1964); (b) Curran et al., J. Org. Chem.. 49:2063-2065 (1984); (c) Hollingsworth et al., J. Org. Chem.. 1537-1541 (1948)]. Thus, treatment of Compound 4 with m-chloroperbenzoic acid (mCPBA) in dichloromethane gave the corresponding N-oxide (step a) which underwent regiospecific rearrangement [Boekelheide et al., J. Am. Chem. Soc.. 76:1286-1291 (1954)] upon heating in acetic anhydride (step b) to give the acetoxy derivative

Compound 5 (62 percent overall yield). Compound 5 was converted to the corresponding silyl ether Compound 7 in 92 percent overall yield by standard methods via hydroxy Compound 6 (steps σ and d). Addition of phenyl chloroformate [Comins et al., J. Org. Chem., 55:292-298 (1990)] to a mixture of Compound 7 and ethynylmagnesium bromide at -78°C led to the formation of Compound 8 in 92 percent yield (step e).

Treatment of Compound 8 with mCPBA led to epoxide Compound 9 (85 percent) (step f), which was converted to ketone Compound 11 via alcohol Compound 10 by desilylation (step g) followed by oxidation (step h). Coupling Compound 11 with the vinyl chloride derivative Compound 12 via Pd(OAc₂)-CuI catalysis (step i) followed by AgNO₃/KCN treatment (step j) resulted in the

formation of the requisite precursor Compound 3 via coupling product Compound 13 (79 percent overall yield). Finally, treatment of Compound 3 with LDA in toluene at -78°C (step k) gave the targeted dynemicin A model

Compound 2 (80 percent based on 25 percent recovery of starting material). Compound 2 is also referred to as DY-l in Figures 10 and 11.

Compound 2 crystallized from ether in colorless prisms, mp 232-235°C dec. X-ray

crystallographic analysis (this X-ray crystallographic analysis was carried out by Dr. Raj Chandha, Department of Chemistry, University rf California, San Diego) confirmed its structure (ORTEP drawing Oakridge Thermal Ellipsoid Plotter) and revealed some interesting

structural features.

The acetylenic moieties are bent from linearity with the following angles: C14, 160.4°; C15, 170.8°; C18, 171.6° and C19, 162.0°. The distance between carbons C14 and C19 (cd distance) [Nicolaou et al., J. Am. Chem. Soc.. 110:4866-4868 (1988)] was found to be 3.63A, a value that agrees well with the calculated one for the MMX minimized structure of

Compound 2 (3.63A) and that of the experimentally derived [(a) Konishi et al, J. Am. Chem. Soc.. 112:3715-3716 (1990); (b) Konishi et al., J. Antibiot.. 42:1449-1452 (1989)] distance in dynemicin A (3.54A). The calculated distance for these acetylenic carbons was found to be 3.40A. See: Semmelhack et al., Tetrahedron Lett. , 31: 1521-1522 ( 1990) . [The MMX87 force field in computer programs MMX and PCMODEL from Serena Software, P.O. Box 3076, Bloomington, IN 47402-3076 was used.]

Scheme V outlines a cascade of novel transformations of model system Compound 2.

Thus, upon treatment with p-toluenesulfonic acid (TsOH·H₂O) in benzene-1,4-cyclohexadiene (3:1, 0.05 M) at 25°C for 24 hours. Compound 2 was converted to Compound 17 in 82 percent yield, presumably via

intermediates 14 and 15. This transformation (2→17) was also carried out using 1.0 equivalent of TsOH·H₂O and 50 equivalents of Et₃SiH in benzene at 25°C (24 hours, 85 percent).

For a study using Et₃SiH and a number of other hydrogen donors to trap C-centered radicals, see:

Newcomb et al., J. Am. Chem. Soc.. 108. 4132-4134

(1986). Compound 17 derived from this reaction with Et₃SiH was contaminated with about 5 percent of an, as yet, unidentified product (detected by ¹H NMR

spectroscopy).

Thus, protonation of the epoxide group in Compound 2 initiates formation of triol Compound 14 which undergoes spontaneous Bergman cyclization [ (a) Bergman, Ace. Chem. Res.. 6:25-31 (1973); Jones et al., J. Am. Chem. Soc., 94:660-661 (1972); Lockhart et al., J. Am. Chem. Soc.. 103:4091-4096; (b) Darby et al.,

J. Chem. Soc. Chem. Commun.. 1516-1517 (1971); (c) Wong et al., Tetrahedron Lett.. 21:217-220 (1980)] to form a benzenoid diradical which is, in turn, rapidly trapped by the hydrogen donor present to furnish cyclized product Compounds 15, 15a, or 15b. Under the reaction conditions, triol Compound 15 apparently undergoes a pinacol-type rearrangement leading to the observed final product Compound 17. The structure of Compound 17 was supported by its spectroscopic data and was confirmed by X-ray crystallographic analysis.

Furthermore, it was found that trimethylsilyl trifluoromethylsulfonate (TMSOTf) in the presence of Et₃SiH induces the same transformation (2→17), (Scheme V) at -78°C in less than 5 minutes (78 percent yield), suggesting a very low energy of activation for the cyclization process. The rather dramatic shortening of the cd distance in going from epoxide Compound 2 (cd= 3.63A, X-ray and MMX) to triol Compound 14a (cd= 3.19A, MMX) is noteworthy. For comparison with other similar enediyne cyclizations and calculations, see: (1)

Nicolaou et al, J. Am. Chem. Soc., 110:4866-4868 (1988); (b) Nicolaou et al., J. Am. Chem. Soc., 110:7247-7248 (1988); (c) Hazeltine et al., J. Am. Chem. Soc..

111:7638-7640 (1989); (d) for similar stabilization of enediynes via cobalt complexation, see: Magnus et al., J. Am. Chem. Soc., 110:1626-1628 (1988) and Magnus et al., J. Am. Chem. Soc., 110:6921-6923 (1988); (e)

Semmelhack et al., Tetrahedron Lett.. 32:1521-1522

(1990); (f) Snyder et al J. Am. Chem. Soc., 111:7630- 7632 (1989); (g) Magnus et al., Tetrahedron Lett.,

30:1905-1906 (1989).

In an attempt to prevent the pinacol rearrangement of triol Compound 15, the acetate

derivative Compound 2a was prepared from Compound 2

[acetic anhydride, 4-dimethylaminopyridine (DMAP) , 84 percent], and was subjected to the epoxide opening and cyclization reaction conditions as described above.

Indeed, the acetate diol Compound 15a was obtained (84 percent yield) as the final product of this cascade starting with Compound 2a and using TsOH-H₂O as the initiator (presumably via intermediate Compound 14a).

The use of anhydrous HCl in CH₂Cl₂ in the presence of Et₃SiH (Scheme V hereinbefore) also resulted in triggering of the cyclization cascade leading from

Compound 2 to Compound 15b (85 percent yield) presumably via the intermediacy of Compound 14b (cd= 3.145 A, MMX). The same conversion (2→15b) was also effected by the use of 3.0 equivalents of MgCl₂ and 50 equivalents of Et₃SiH in CH₂Cl₂ at 25°C (12 hours, 87 percent) or 1.2 equilalents of TiCl₄ and 50 equivalents of Et₃SiH in CH₂Cl₂ at -78°C (0.5 hour, 60 percent).

Thus, only Compound 15 underwent the further pinacol-type rearrangement to form Compound 17.

These cyclizations are analogous to those observed for dynemicin A. [(a) Konishi et al., J.

Am.Chem. Soc. 112:3715-3716 (1990); (b) Konishi et al., J. Antibiot.. 42:1449-1452 (1989); (c) Sugiura et al., Proc. Natl. Acad. Sci. USA. 87:3831-3835 (1990)].

An alternate mode of triggering the cyclization of Compound 2 based on cobalt complexation of the acetylenes was devised. This triggering

cyclization is illustrated in Scheme VI, below. This pathway was designed so as to prevent the acetylenes from spontaneously cyclizing upon epoxide opening and thus allow the isolation of the postulated intermediate cis-diol.

Thus, reaction of Compound 2 with Co₂(CO)₈ (2.2 equivalents) resulted in the formation of the dicobalt complex Compound 18 (step a) in 96 percent yield. Use of one equivalent of Co₂(CO)₈ resulted in the formation of a monocobalt complex in addition to dicobalt derivative Compound 18 and starting material Compound 2. [For similar stabilization of enediynes via cobalt complexation, see: Magnus et al., J. Am. Chem. Soc., 110:1626-1628 (1988); and Magnux et al., J. Am. Chem. Soc., 110:6921-6923 (1988)].

Treatment of Compound 18 with trifluoroacetic acid in CH₂Cl₂ (zero degrees C) followed by aqueous workup led to the formation of the stable cis opening product Compound 19 (step b; 92 percent yield). Upon exposure of Compound 19 to ferric nitrate, or

trimethylamine N-oxide in CH₂Cl₂ in the presence, or absence, of Et₃SiH at 25°C, the cyclized product

Compound 17 was obtained in 85-92 percent yield via liberation of the acetylenic groups to afford Compound 14 followed by spontaneous and sequential generation of Compounds 15 and 17 as shown in Schemes V and VI. The same study [reaction with Fe(NO₃)₃] carried out in

CD₂Cl₂ resulted in the incorporation of two deuterium atoms in Compound 17 confirming methylene chloride as an effective hydrogen atom donor in these aromatization studies.

To obtain a closer model to dynemicin A, the tertiary hydroxy group in Compound 2 was removed to form Compound 21 as shown in Scheme VII below. Thus, reaction of Compound 2 with

thiocarbonyldiimidazole in the presence of DMAP resulted in the formation of Compound 20 in 95 percent yield (step a). Compound 20, upon treatment with ⁿBu₃SnH in the presence of a catalytic amount of AIBN (toluene, 75°C), led to the desired Compound 21 (step b; 72 percent yield). This model system Compound 21 also underwent smooth cyclization to polycycles Compounds 23 (86 percent) and 23a (82 percent) upon suitable

triggering, (as summarized in Scheme VII).

An ORTEP drawing of dynemicin A model Compound 21 (mp 251.2 52°C dec, from ether - petroleum ether) as determined by X-ray crystallographic analysis was prepared. The following angles revealed considerable deviation of the acetylenic groupings from their

preferred linear arrangement: C17-C18-C19 = 170.2°;

C9-C19-C18 = 162.0°; C14-C15-C16 = 170.1°; and

C13-C14-C15 = 163.7°. The distance between carbons C14 and C19 (cd distance) was found to be 3.59A, which agrees well with the values derived for the MMX

minimized structure of Compound 21 (3.59A) and from the X-ray crystallographic analysis of dynemicin A (3.54A). [Konishi et al., J. Am. Chem. Soc.. 112:3715-3716

(1990)]. Again, the considerable shortening of the cd distance is noted in going from 9 to the cis diols 10 (cd= 3.21A, MMX) and 10a (cd= 3.19A, MMX).

In order to allow for the generation of the parent enediyne system Compound 40, Compound 45 was designed and synthesized from Compound 21 as shown in Scheme VIII, below.

Thus, Compounds 21 and 46 were separately reacted with ten equivalents of 2-phenylthioethanol

[PhS(CH₂)₂OH] in acetonitrile for 12 hours in the presence of three equivalents of Cs₂CO₃ and 0.5

equivalents 18-crown-6 in step a to form the

corresponding phenylthioethoxy Compounds 21a and 46a, respectively. The products of that reaction were then individually reacted with 2.5 equivalents of mCBPA in CH₂Cl₂ at 0-25°C to provide Compounds 45 (85 percent overall) and 47 (82 percent overall) as step b. When step b is carried out with one equivalent of mCBPA, the corresponding sulfoxides, Compounds 21b and 46b are prepared.

To prepare Compound 40 that was too labile for isolation, Compound 45 was reacted with an excess of Cs₂CO₃ and 0.5 equivalents of 18-crown-6 in a

dioxane: 1,4-cyclohexadiene (4:1) solution for one hour at 25°C as step c. Compound 47 was reacted with 1.2 equivalents of 1,8-diazabicyclo[5.4.0]undec-7-ene (DBU) in benzene at 5°C for one hour to provide a 97 percent yield of Compound 48 as step c.

Freshly prepared Compound 40 was reacted with either phenol (i, PhOH) or thiophenol (ii, PhSH) using two equivalents of either nucleophile at 25°C for two hours to provide Compounds 49 (25 percent yield) or 50 (33 percent yield, respectively, as step d. The

reactivity of Compound 40 and its ability to cleave DNA add credibility to the notion of a pathway including epoxide ring opening and the intermediacy of an

iminoquinone methide species in the mode of action of dynemicin A, at leasf as a partial mechanism.

Interestingly, the methoxy derivative.

Compound 48 proved to be quite stable under basic or neutral conditions. Treatment of Compound 48 with 0.5 equivalents of p-toluenesulfonic acid-H₂O (step d iii. TSOH) in dioxane:war.er:1,4-cyclohexadiene (4:1:1) at 60°C for ten minutes provided Compound 51 (20 percent yield).

An X-ray crystallographic analysis of Compound 48 confirmed its molecular structure and revealed a number of interesting parameters including a cd distance of 3.63A [Calcd. cd=3.6lA, MMX] and the non-linearity of the acetylenic groupings [angles at acetylenic carbons: C-14, 163.3°; C-15, 173.0°; C-18, 169.3°; C-19, 160.5°].

Enediyne model system Compound 43 (Scheme IX) was designed for its potential to generate species 42 under photolytic, neutral conditions. Following the strategy developed earlier for the synthesis of Compound 21, Compound 43 was constructed in good overall yield. Reactions of Compound 43 are shown in Scheme IX.

Irradiation in step a of Compound 43 in THF-H₂O (10:1) for 40 minutes [Hanover mercury lamp, pyrex filter] at ice-cooling temperature resulted in clean conversion to Compound 42 as observed by TLC and ¹H NMR spectroscopy (THF-d₈: D₂O 10:1, 300 MHz). Attempted isolation of Compound 42 led to decomposition, whereas treatment of crude Compound 42 in a pH 8.0 buffer-THF solution containing either ethanol (EtOH),

mercaptoenthanol (EtSH) or n-propylamine (ⁿPrNH₂) as nucleophiles (Nu) at 25°C under argon atmosphere for 1.5 hours provided a mixture of Compounds 44a and 54 (31 percent yield), 44b (34 percent yield), or 44c (46 percent yield), respectively, as step b.

Noteworthy was the isolation of Compound 54 (Scheme IX) in about 5-10 percent yield from the

reaction of 42 with ethanol, in air, along with Compound 44a (33 percent yield), as a sensitive but stable molecule under neutral conditions. The isolation of Compound 54 provides strong evidence for the

intermediacy of quinone methide Compound 52, [a) Dyall et al., J. Am. Chem. Soc.. 94-2196 (1972); b) Zanarotti, Tetrahedron Lett.. 23:3815 (1982); c) Zanarotti, J. Org. Chem.. 50:941 (1985); d) Angle et al., J. Am. Chem.

Soc.. 111:1136 (1989); e) Angle et al., Tetrahedron Lett.. 301193 (1989); f) Crescenzi et al., ibid 31:6095 (1990); g) Barton et al., ibid 11:8043 (1990); h) Angle et al., ibid 111:4524 (1990)], trapping of which with molecular oxygen would provide Compound 54 as is shown in Scheme X, below.

Remarkably, treatment of Compound 55 under similar conditions (THF-pH 9.0 buffer in air) resulted in the isolation of Compounds 58 and 62 in 32 percent and 25 percent yields, respectively, as stable

molecules. Quinone methide intermediates have been implicated in the mode of action of anthracycline antibiotics [for recent postulated quinone methide intermediates in the area of anthracycline antibiotics, see: a) Boldt et al., J. Org. Chem.. 52:2146 (1987); b) Gaudiano et al., J. Org. Chem., 54:5090 (1989);

c) Egholm et al., J. Am. Chem. Soc.. 111:8291 (1989); d) Ramakrishnan et al., J. Med Chem., 29:1215 (1986); e) Boldt et al., J. Am. Chem. Soc., 111:2283 (1989); f) Gaudiano et al., J. Am. Chem. Soc., 112:9423 (1990); g) Karabelas et al., J. Am. Chem. Soc., 112:5372 (1990); h) Gaudiano et al., J. Am. Chem. Soc., 112:6704 (1990)] and dynemicin A.

The pivaloate ester Compounds 59a and 59b were also synthesized according to the previously discussed general strategy for their potential to liberate

phenolic species of type of Compound 42 (Scheme IX), thus presenting yet another triggering mechanism to initiate the dynemicin A-type cascade. Indeed, upon exposure to four equivalents of LiOH in ethanol:water (3:1) at 25°C for 4-6 hours. Compounds 59a and 59b were smoothly converted to products 60a (56 percent) and 60b (42 percent), respectively, presumably via a sequence involving concomitant carbamate exchange [EtO- for

PhO-].

The methoxy Compound 41 was also synthesized and exhibited reasonable stability under neutral and basic conditions. As expected, however, this compound cyclized rapidly under acidic conditions. For example, upon treatment with TsOH-H₂O in benzene: 1,4-cyclohexadiene (1:1) at 25°C for one hour, Compound 41 afforded the aromatized product Compound 61 (32 percent yield). Treatment of Compound 41 as per steps a and b of Scheme V yields the corresponding methoxyphenyl derivative Compound 41a having a hydrogen in place of the hydroxyl of Compound 41. Treatment of Compound 41a with 2-phenylthioethanol, 2-α-naphthylthioethanol or 2-β-naphthylthioethanol under basic conditions as discussed in regard to Scheme VIII, followed by

oxidation led to Compounds 41b, 41c and 4Id,

respectively.

The synthesis of Compounds 70 and 80 proceeded as summarized in Scheme XI, shown below.

Thus, coupling of the readily available

Compounds 11 [Nicolaou et al., J. Am. Chem. Soc..

112:7416 (1990); Nicolaou et al., J. Am. Chem. Soc.. 113:3106 (1991)] and 71 using palladium (O)-copper(l) catalysis afforded product Compound 72 (55 percent yield) as step a. Desilylation of Compound 72 with lithium hydroxide in step b followed by base-induced (lithium diisopropylamide; LDA) ring closure led to Compound 74 via 73 (75 percent overall yield) in step c. Conversion of Compound 74 to the thionoimidazolide

Compound 75 (84 percent based on 67 percent conversion of staring material) as discussed elsewhere herein, followed by deoxygenation with tri-n-butylstannane

(ⁿBu₃SnH) in step e resulted in the formation of 76 (94 percent). Exchange of the phenoxy (PhO) group of

Compound 76 with 2-phenylthioethoxy (PhSCH₂CH2O) took place smoothly under basic conditions (two equivalents of PhSCH₂CH₂ONa, THF) leading to Compound 77 (92 percent yield) in step f, from which the sulfone Compound 70 was generated by oxidation using 2.5 equivalents mCPBA (81 percent yield) in step g.

Similar chemistry employing the naphthalene ditriflate Compound 81 led from Compound 11 to

arenediyne Compounds 85-80 via intermediate Compounds 82-84 in comparable yields as depicted in Scheme XI.

Step h utilized acetonitrile as a solvent and a one hour reaction time as compared with step a of the scheme that used benzene as solvent and 3.5 hours of reaction to obtain yields of 55 and 56 percent, respectively. Step i utilized five equivalents of trimethylsilylacetylene, 0.05 equivalents of Pd(PPh₃)₄, 0.2 equivalents of Cul and two equivalents of triethylamine in acetonitrile (as were present in steps a and b) with a reaction time of 20 hours at 25°C to provide Compound 82 in 76 percent yield. Steps c-g of both reactions were identical. Reaction of Compounds 77 or 88 with one equivalent each of mCPBA leads to preparation of the corresponding sulfoxides. Compounds 77a and 88a.

The free amine Compounds, 76a and 87a, shown below, corresponding to Compounds 76 and 87 were

prepared from their corresponding phenylsulfonylethyl carbamates 70 and 80 under basic conditions. Those compounds were both stable at 25°C, whereas Compound 40 decomposed at that temperature.

Treatment of Compounds 76a and 87a with silica gel in benzene initiated epoxide opening to form the corresponding diols 76b and 87b, respectively. Diol Compound 76b cyclized spontaneously at 25°C in the presence of 1,4-cyclohexadiene to form the

cycloaromatized diol. Compound 76c. In contrast to Compound 76b, diol Compound 87b was stable in the presence of 1,4-cyclohexadiene, but cycloaromatized at elevated temperature (65°C, two hours) to form

Compound 87c.

The cycloaromatization of enediyne Compounds 76, 70, 87 and 80 was then studied in order to determine the precise structural and spectroscopic changes taking place. Thus, whereas cycloaromatization of Compound 76 (about 0.02M solution) under acid conditions [1.2 equivalents of TsOH.H₂O, benzene-cyclohexdiene (4:1), 25°C, four hours] produced smoothly the corresponding naphthalene derivative (78 percent) no dramatic changes in the UV and fluorescent spectra were observed for the starting arenediyne and cycloaromatized product.

In contrast, however, similar acid-induced Bergman cycloaromatization of the naphthalene enediyne Compound 87 furnished the epoxy-containing anthracene derivative Compound 89 (49 percent) that exhibited, as expected, strong and characteristic UV and fluorescence profiles that were distinct from those of the starting arenediyne Compound 87 [UV (EtOH) , 87:λ_max (log e) 304 (3.47, 294 (4.01), 284 (4.26), 267-240 (4.53-4.55), 214 (4.50) nm; B9 : λ_max (log e) 390 (3.74), 369 (3.78), 351 (3.66), 333 (3.45), 318 (3.20), 267-244 (4.43-4.46), 214

(4.43) nm. Fluorescence (EtOH, 1 μM, excitation at 260 nm. Compound 87:λ_max 435, 412, 393, 374, 357 nm; 89: λ_max 520, 466, 442, 413, 392 nm].

Attempted cycloaromatization of Compounds 70 and 80 under a variety of basic conditions led to decomposition, presumably via the in situ generated free amines and diradical species, whereas acid treatment resulted in the formation of epoxy-opened (dihydroxy) Compounds 78 and 89a albeit in low yields (20 percent and 15 percent, respectively). Interestingly, both Compounds 70 and 80 exhibited DNA cleaving properties under basic conditions (supercoiled $174 DNA, pH 9.0) and potent anticancer activity against a variety of cell lines as discussed hereinafter.

Saccharide-Containing Chimeras

Chimeric compounds that include both a fused ring enediyne as the aglycone and a before-discussed mono- or oligosaccharide as the oligosaccharide portion is also contemplated, as noted earlier. The previously depicted saccharides are related to the calicheamicin oligosaccharide.

The before-depicted saccharides correspond to the calicheamicin oligosaccharide (Structures F and G), the oxime precursor thereto (Structures D and E), and fragments thereof (Structures A-C). More specifically, the disaccharide Structure A corresponds to the

calicheamicin A and E rings with the hydroxylamine link to a B ring analog. The monosaccharide Structure B corresponds to the A ring alone with the hydroxylamine-linked B ring analog. The trisaccharide thiobenzoate Structure C corresponds to rings A, E and B, and a C ring analog. The 5-ring Structure D corresponds to the FMOC-blocked oxime precursor to the complete

calicheamicin oligosaccharide, whereas 5-ring Structure E is the FMOC-deblocked version thereof. Structures F and G are the complete calicheamicin oligosaccharides that are epimeric about the 4-position of the A ring hydroxylamine linkage, with Structure F having the native calicheamicin oligosaccharide stereochemistry. These saccharides are discussed in more detail

hereinafter.

Inasmuch as the previously depicted saccharides of the calicheamicin oligosaccharide are derivatives of known compounds, as are their suitably protected precursors, their complete syntheses need not be discussed in complete detail herein. Those syntheses are described in Nicolaou et al., J. Am. Chem. Soc..

112:4085-4086 (1990); Nicolaou et al.. Ibid., 112:8193-8195 (1990); Nicolaou et al., J. Chem. Soc.. Chem.

Commun.. 1275-1277 (1990) as well as in U.S. Patent Application Serial No. 07/520,245 filed May 7, 1990 and Serial No. 07/695,251 filed May 3, 1991, all of whose disclosures are incorporated by reference herein.

Nevertheless, the saccharides αsed herein differ

somewhat from the reported saccharides, so their

syntheses will be discussed, at least in pertinent part herein below.

The disaccharide-linked hydroxylamine compound is prepared in a manner analogous to that of Compound 12 of Nicolaou et al., J. Am. Chem. Soc.. 112:8193-8195 (1990), except that an o-nitrobenzyl (shown as NBnO or ONBn in the schemes) glycoside is utilized instead of the methyl glycoside precursor. Compound 9 of that paper. The synthesis for the disaccharide is

illustrated in Schemes XII and XIII, and is discussed below.

Thus, D-fucose Compound 90 was peracetylated in step a to form tetraacetate Compound 91 which was converted to the anomeric bromide Compound 92 in step b, and glycosylated with o-nitrobenzyl alcohol to afford Compound 93 in step c (63 percent overall yield).

Deacetylation of Compound 93 in step d led to Compound 94, which reacted selectively with carbonyl diimidazole in step e to afford the requisite ring A intermediate 95 in 86 percent overall yield. Turning to Scheme XIII, intermediate Compound 95 was then coupled [Mukaiyama et al., Chem. Lett.. 431 (1981); Nicolaou et al., J. Am. Chem. Soc. 106:4159 (1984)] to glycosyl fluoride Compound 96 (Compound 8 of the above paper; Me=methyl) with AgClO₄-SnCl₂ catalyst in step a, leading stereoselectively to disaccharide Compound 97 as the major anomer (80 percent yield, about 5:1 ratio of anomers). Chromatographic purification of Compound 97 with removal of the carbonate protecting group (NaH-HOCH₂CH₂OH, 90 percent) in step b to form Compound 98, and treatment in step σ with ⁿBu₂SnO-Br₂ [David et al., J. Chem. Soc. Perkin Trans 1. 1568

(1979) ] led to hydroxyketone Compound 99 (65 percent yield plus 17 percent Compound 98) via intermediate Compound 98.

Oxime formation in step d with O-benzyl hydroxylamine under acid conditions led to Compound 100 (90 percent, single geometrical isomer of unassigned stereochemistry; Ph=phenyl) which was silylated in step e under standard conditions to furnish Compound 101 (90 percent). Photolytic cleavage [Zenhavi et al., J. Ore. Chem.. 37:2281 (1972); Zenhavi et al., ibid. 37:2285 (1972); Ohtsuka et al., J. Am. Chem. Soc. 100:8210 ( 978); Pillai, Synthesis, 1 (1980)] of the

o-nitrobenzyl group from Compound 101 (THF-H₂O, 15 minutes) produced lactol Compound 102 in 95 percent yield in step f. Treatment of Compound 102 with

NaH-Cl₃CC≡N [Grandler et al., Carbohydr. Res., 135:203 (1985); Schmidt, Angew Chem. Int. Ed., Engl., 25:212 (1986)] in CH₂Cl₂ for two 25°C in step g

resulted in the formation the α-trichloroacetimidate Compound 103 in 98 percent yield. Reaction of benzyl alcohol (2.0 equivalents) with trichloroacetimidate Compound 103 under the Schmidt conditions [Grandler et al., Carbohydr. Res., 135:203 (1985); Schmidt, Anoew Chem. Int. Ed., Engl.. 25:212 (1986)] [BF₃-Et₂O,

CH₂Cl₂, -60 → -30°C] resulted in stereoselective

formation of the β-glycoside Compound 104 (79 percent yield) together with its anomer (16 percent, separated chromatographically) [¹H NMR, 500MHz, C₆D₆, 104: J_1,2=6.5 Hz, epi-Compound 104: J_1,2= 2.4 Hz].

on the other hand, treatment of lactol

Compound 102 with DAST in step g' led to the glycosyl fluoride Compound 102a in 90 percent yield (about 1:1 anomeric mixture). Reaction of Compound 102a with benzyl alcohol in step h' in the presence of silver silicate [Paulsen et al., Chem. Ber.. 114:3102 (1981)] - SnCl₂ resulted in the formation of the 0-glycoside Compound 104 and its anomer in 85 percent (about 1:1 anomeric mixture).

Generation of intermediate Compound 106 via Compound 105 proceeded smoothly under standard

deprotection conditions in steps i and j. Finally, exposure of Compound 106 to Ph₂SiH₂ in the presence of Ti (OⁱPr) ₄ in step k resulted in the formation of the desired target Compound 107 as the only detectable product (92 percent yield). Interestingly, reduction of Compound 106 with NaCNBH₃-H^⊕ led predominantly to the 4-epimer of Compound 107 (90 percent yield). The stereochemical assignments of Compound 107 and epi-107 at C-4 were based on % NMR coupling constants [¹H NMR, 500MHz, C₆D₆, 107: J_3,4=9.5, J_4,5=9.5 Hz; epi-107 : J_3,4=1.9 Hz, J_4,5=1.5 HZ].

The hydroxylamine linked A ring derivative (Structure B) can be prepared starting with Compound 9 of Nicolaou et al., J. Am. Chem. Soc. 112:8193-8195 (1990). There, the 2-hydroxyl is blocked with a

t-butyldimethylsilyl (^tBuMe₂Si) group as before, and the carbonate group removed by reaction of sodium hydride in ethylene glycol-THF at room temperature. The keto group can be prepared by oxidation with dibutylstannic oxide (ⁿBu₂SnO) in methanol at 65°. The 3-position hydroxyl is similarly blocked with a ^lBuMe₂Si group, and the oxime formed as above.

The trisaccharide plus C ring analog

(Structure C) can be readily prepared from Compound 19 of Nicolaou et al., J. Am. Chem. Soc. , 112:8193-8195 (1990). Thus, that Compound 19 is reacted with benzoyl chloride in the presence of triethylamine and a

catalytic amount of DMAP in methylene chloride to provide the blocked oxime-containing trisaccharide.

The use of Compounds 2 and 103 in preparing an exemplary chimera is illustrated in Scheme XIV, below, and discussed hereinafter.

Thus, as shown in Scheme XIV, coupling of Compound 2 with ethyl bromoacetate under basic

conditions led to derivative Compound 24c (60 percent yield), which was converted to primary alcohol Compound 111 (80 percent overall yield) by: (i) ester hydrolysis; (ii) 2-pyridyl thiolester formation (collectively step b) and (iii) reduction in step c Coupling of Compound 111 (1.2 equivalent) with trichloracetimidate Compound 103 under the influence of BF₃·Et₂O as described before led to the formation of two major products (70 percent, about 1:1 ratio) and two minor products (14 percent, about 1:1 ratio), which were chromatographically

separated.

The major isomers were shown to be the

diastereomeric β-glycoside Compounds 112a (R_f-=0.12, silica, 20 percent ethyl acetate in petroleum ether) and 112b (R_f-0.10, silica, 20 percent ethyl acetate in petroleum ether) [¹H NMR, 500MHz, C₆D₆, 112a; J_1,2=6.5 Hz; 112b; J, ₂=6.5 Hz], whereas the minor isomers were shown to be the α-anomers of Compounds 112a and 112b at C-1 [¹H NMR, 500MHz, C₆D₆, epi-112a, J_1,2=2.4 Hz: epi-112b, J_1,2=2.4 Hz]. Sequential deprotection of Compounds 112a and 112b as described above for Compound 104 led to oxime Compounds 114a and 114b via intermediate Compounds 113a and 113b, respectively.

Finally, reduction of Compounds 114a and 114b under the Ph₂SiH₂-Ti(OⁱPr)₄ conditions led exclusively to the targeted Compounds 115a and 115b respectively (90 percent yield). The C-4 stereochemistry of Compounds 115a and 115b was again based on the coupling constants J₄₃=9.5 and J₄₅=9.5 Hz for the newly installed H-4.

Structures 112a-ll5a and 112b-115b are interchangeable, since the absolute stereochemistry of the aglycons has not been determined. Physical data for Compounds 115a and 115b are provided hereinafter. Still further chimeras have been prepared that contain a fused-ring enediyne glycosidically-linked to the complete calicheamicin oligosaccharide or an analog thereof. Exemplary synthetic steps are outlined in Schemes XV and XVI, below.

Thus, a compound such as Compound 111 of Scheme XIV was reacted with phenylthioethanol in the presence of a base as discussed elsewhere to exchange out the phenoxide moiety from the corresponding

carbamate. That compound was then oxidized with

m-chloroperbenzoic acid to form the corresponding

2-(phenylsulfonyl) ethyl carbamate. Compound 120.

Racemic compound 120 was reacted with Compound 123 as shown in Scheme XV. Compound 123 was prepared from Compound 121 via compound 122, and shown in the scheme and discussed in regard to Scheme XIII as to the preparation of Compound 103. Compound 121 was itself prepared in a manner analogous to that discussed in Nicolaou et al., J. Am. Chem. Soc., 112:8193-8195

(1990), except that Compound 99 herein was utilized instead of Compound 12 of the published paper, and Et₃Si blocking groups were used instead of ^tBuMe₂Si blocking groups that were used in the published paper.

Thus, Compound 121 was irradiated in THF-H₂0 (9:1, v/v) at zero degrees C in step a to provide a 95 percent yield of Compound 122. That compound was reacted with a catalytic amount of NaH and

trichloroacetonitrile in methylene chloride (1:12, v/v) at 25°C to provide a 98 percent yield of Compound 123 in step b. Compounds 120 and 123 were coupled in step c in benzyl alcohol, BF₃·Et₂O in methylene chloride at -60° → -30°C to provide Compound 124 as a mixture of

diastereomeric anomers 124a and 124b present as a β- to α-anomer ratio of about 5:1, the β-anomer being shown as Compound 124a, and the α-anomer, Compound 124b, not being shown.

Reaction of Compounds 124a and 124b with ⁿBuNF using standard conditions removed the triethylsilyl groups. The oxime- and FMOC-containing, hydroxyl deblocked compound that resulted, Compounds 125a and 125b, thus contained the oligosaccharide portion

discussed previously as oligosaccharide Structure D. Subsequent reaction with diethylamine under usual conditions removed the FMOC group to form Compounds 126a and 126b. The completely deblocked oxime-containing oligosaccharide portion of Compounds 126a and 126b is the oligosaccharide discussed previously as

oligosaccharide Structure E.

Scheme XVI shows only the oxime-containing ring of Compound 126a, with the remaining portions indicated by the wavy lines. As is shown in Scheme XVI, reduction of Compound 126a with sodium cyanoborohydride in BF₃·Et₂O at -50°C provided a 65 percent yield of epimeric Compounds 127a and 127b, that were present in about equal amounts. Those compounds were epimeric at the 4-position of the A ring as indicated by the number 4 and the arrow. The epimeric portions of Compounds 127a and 127b provide the oligosaccharide portions previously identified as oligosaccharide Structures F and G, with only the epimer present in calicheamicin (Structure F) being shown in Scheme XVI.

As prepared, Compounds 125-127 were mixtures of diastereomers. Those diastereomers have been

separated, but the absolute configurations of the fusedring enediyne portions are presently unknown, and only one is illutrated, whereas both were shown in Scheme XIV.

Preparation of other chimeras using other appropriate fused ring enediyne aglycons and

oligosaccharide portions.

As noted earlier, an R⁵ group such as that shown in Formula X is preferably a hydroxyl group or a group convertible thereto intracellualarly. The

presence of an R⁵ hydroxyl group also provides an atom (oxygen) that can be used to link a hydroxyl-containing spacer group that itself can be glycosidically linked to a before-described saccharide. A generalized synthesis is illustrated in Scheme VI. A more specific partial synthesis is illustrated in Scheme XVII, below, that was used to prepare Compound 59. Thus, in accordance with Scheme XVII, ethyl cyclohexanone-2-carboxylate. Compound 130 was reacted in step a with 1.2 equivalents of ethylene glycol and 0.1 equivalents of TsOH·H₂O in benzene at reflux for ten hours to form Compound 131. Compound 131 was reacted in step b with sodium methoxide-methanol at reflux for eight hours to form Compound 132 in 78 percent yield from Compound 130.

Compound 132 was then reacted in step c with Compound 133 in the presence of 1.2 equivalents of DCC and 0.1 equivalents of DMAP in methylene chloride at 25°C for 14 hours to form Compound 134 in 96 percent yield. Compound 134 was reacted with m-aminophenol in THF at reflux for 96 hours to provide Compound 135 in 87 percent yield in step d. Compound 135 was reacted with benzylbromide, 1.05 equivalents of NaH and 0.1

equivalents of ⁿBu₄NI in THF at 25°C in step e to form Compound 136 in 72 percent yield.

Compound 136 was cyclized in step f in 37 percent HCl-THF (1:2.7) at reflux for three hours to provide a mixture of the cyclized product Compounds 137a and 137b in 100 percent yield. Compounds 137a and 137b were formed in about a 4:1 ratio in the order named. Compounds 137a and 137b as a mixture were treated with DIBAL and two equivalents of LiAlH₄ in THF at reflux for three hours and then with oxygen and SiO₂ at 25°C for 24 hours to form Compound 138 in 50 percent yield.

A similar reaction sequence can be used with p_-anisidine in place of the m-aminophenol to form the corresponding methoxy compound. That compound when treated with two equivalents of sodium ethylthiolate in DMF at 160°C, followed by acylation with acetic

anhydride and then reaction with sodium methoxide provides the substituted phenol para- to the nitrogen atom. That phenol, when treated as in step e forms the benzyl ether. Compound 139.

Either of Compounds 138 or 139, when treated as in steps a and b of Scheme II, or c-e of Scheme III provides the corresponding benzyloxy Compounds 140a or

140b. Hydrogenation of either of Compounds 140a or 140b using 10 percent Pd/C in ethanol at 25°C provides the corresponding phenol derivatives Compounds 141a and 141b.

Reaction of either phenol with 1.05 equivalents of NaH and mercaptothiazolyl pivolate in THF provides the corresponding pivaloyl esters Compounds 142a and 142b. The pivaloyl ester Compound 142a was prepared as discussed above after a five minute reaction time at 25°C in 99 percent yield.

Similarly, reaction of either of Compounds 141a and 141b with 1.05 equivalents of NaH and of a-nitrobenzyl bromide and 0.1 equivalents of ⁿBu₄NI in THF provides the corresponding photolabile ONBn group and Compounds 143a and 143b. Compound 143a was prepared in 90 percent yield after a one hour reaction time.

Any of Compounds 142a, 142b, 143a or 143b can be used to form a fused-ring enediyne compound of the invention using the steps outlined in the prior reaction schemes, such as Compounds 59a and 59b.

The use of Compound 143b to prepare fused-ring enediyne compounds having a spacer portion in the benzo ring para to the carbamate nitrogen atom is illustrated in Scheme XVIII. Thus, Compound 143b is reacted as discussed elsewhere herein to form fused-ring enediyne Compound 150. Those prior reactions are indicated schematically by the two arrows between Compounds 143b and 150. Compound 150 is a para-hydroxy analog of the compound formed in step b of Scheme VIII.

Compound 150 is then reacted in step a with

2-bromoethanol and NaH in the THF to form Compound 151. Compound 151 is oxidized in step b as with mCPBA to form Compound 153, whose ethoxyethanol hydroxyl group can be used to form a glysidic link to a saccharide as

discussed previously.

Compound 150 can also be reacted with ethyl bromoacetate and cesium carbonate in acetonitrile as in step c to form the corresponding ethyl carboxymethyl derivative. Hydrolysis of that ester with lithium hydroxide in THF-water and neutralization provides the free acid Compound 152 in step d. Oxidation of the free acid as in step b provides Compound 154. The carboxylic acid group of Compound 154 can be used to form a chimera via an ester link to a before-discussed saccharide, or an ester or amide link to a before-discussed monoclonal antibody.

The results of an exemplary DNA-cleaving study using Compound 40 in a method of the invention are illustrated in Figure 1. Other compounds such as

Compound 41 Formula X wherein R¹ is phenoxycarbonyl, R²=R³=R⁶=R⁷=R⁸=H, A is a saturated bond, R⁴ is hydroxyl and R⁵ is methoxy at the "e" position of the benzo ring] also cleaved DNA. Compound 41 was utilized at a 2 mM concentration at pH 5.0. Compounds 40, 42 and 54 were further found to cause significant DNA cleavage when incubated at 5mM with supercoiled øX174 DNA at pH 8.0 (Figure 2).

Noteworthy in these studies is the observation of double strand cuts as well as single strand cuts, as is the case with dynemicin A (Compound 1). The methoxy

derivatives 47, 55, 58 and 62 exhibited diminished activity against DNA.

On the other hand, the compounds exhibit anti-microbial activity and all of those assayed exhibit some activity in vitro against tumor cells. A graph

illustrating the inhibition of MIA PaCa-2 human

pancreatic carcinoma cell growth using Compound 2 (DY-1) is shown in duplicate in Figure 3. Graphs illustrating the inhibition of MB-49 murine bladder carcinoma cell growth for Compounds 2 and 20 (DY-2) are shown in

Figures 4 and 5, in quadruplicate and duplicate

respectively. Values of IC₅₀ obtained from the two wider range studies shown in Figure 4 were 43 nM and 91 nM. In a comparative study. Compound 2 was also shown to be more active against the cancerous MB-49 cells than against non-transformed CV-1 African green monkey kidney cells (ATCC CCL 70) or WI-38 human lung cells (ATCC CCL 75).

Each of the dynemicin A analog Compounds 24a-g exhibited anti-tumor activity against MIA PaCa-2 cells, with the esters (Compounds 24b-g) being more potent than the free acid (Compound 24a). Compounds 24a-g were also active against MB-49 cells and inactive against CV-1 and WI-38 cells. Compounds 2b-d shown in below each

exhibited weaker anti-tumor activity against MB-49 cells than did Compound 2.

The described chemistry supports the viability of two paths as triggering mechanisms for the dynemicin A-type cascade by showing that a lone pair of electrons on a heteroatom (N or O) strategically positioned on the aromatic ring in relation to the epoxide moiety serves to initiate the cycloaromatization reaction. Such reactive species can be generated within the cell by enzymatic reactions, or as shown above, be released from suitable precursors under mild conditions in the

laboratory. In addition, the relative stability and observation of Compounds 42, 54, 58 and 62 is

interesting in that it allows for the nario of bioreduction prior to intercalation as well as for the possibility of DNA interacting nucleophilically against quinone methide species. Thus the proposition that dynemicin A may be interacting with DNA by a dual mechanism (nucleophilic and radical) appears attractive, not only because of the observed preference for cleavage of adenine and guanine, but also in view of the

chemistry of Compound 42.

Further biological evaluation data are

provided hereinafter in Tables 1-3. Best Mode for Carrying out the Invention

Methods

DNA Cleavage Studies

The ethereal solution of Compound 40 produced by the LiAlH₄ reduction of Compound 21 (39 mg, 0.10 mmol) was evaporated in vacuo to dryness and the residue dissolved in THF (4 mL) to give a 25 mM solution of Compound 40, assuming complete conversion of Compound 21. Analysis of Compound 40-induced damage to

supercoiled, covalently closed, circular (form I) øX174 DNA was performed by incubation at varying

concentrations of Compound 40 (100 μM - 5000 μM) in phosphate buffered aqueous solution at 37°C for 12-24 hours followed by agarose gel electrophoresis to

separate the various DNA products.

Thus, a vial containing a 50 micromolar per base pair solution of 0X174 Form I double stranded DNA in 2.0 microliters of pH 7.4 phosphate (50 mM) buffers were added 6.0 microliters of the same buffer solution and 2.0 microliters of a 5.0 millimolar ethanol solution of Compound 40.

The vials were then placed in a 37°C oven for 12-24 hours. A 2.0 microliter portion of glycerol loading buffer solution containing bromothymol blue indicator was added to each vial. A 10 microliter aliquot was then drawn from each. Gel electrophoresis analysis of the aliquots was performed using a 1.0 percent agarose gel with ethidium bromide run at 115 volts for 1 hour. DNA cleavage was indicated by the formation of nicked relaxed circular DNA (form II) or linearized DNA (form III), which was detected by visual inspection of the gel under 310 nanometer ultraviolet light using ethidium bromide. Procedure for 6-Well Cvtotoxicitv Assay

MIA PaCa-2 cells, MB-49, CV-1 or WI-38 cells were loaded into each well of a 6-well plate at a density of 100,000 cells/well in 3 ml culture medium. They were incubated for 4 hours (37°C, 7 percent CO₂).

Then 6 microliters of solution containing a compound to be assayed were added into 3 ml of medium (RPMI-1640, with 5 percent fetal bovine serum and 1 percent

glutamine) in a 500X dilution so that in one well ethanol was added to take a 0.2 percent ethanol control. The plates were then cubated for 4 days (37°C, 7 percent CO₂). The megium was then drained, crystal violet dye (Hucker formula) was added to cover the well bottoms and then they were rinsed with tap water until rinses were clear. The stained cells were solubilized for quantitation with Sarkosyl solution (N-Lauryl sarcosine, 1 percent in water) at 3 ml/well. The absorbance of the solution was then read at 590-650 nm. Large Scale Screening Against Cancerous Cell Lines

In addition to the screening already discussed, several of the before-described compounds and chimeras were screened against several or all of a panel of ten cancerous cell lines as target cells. This screening utilized a sulforhodamine B cytotoxidity assay as discussed below.

SULFORHODAMINE B CYTOTOXICITY ASSAY

1. Preparation of target cells in 96-well plates

a. Drain media from T₇₅ flask of target cell line(s) and carefully wash cell monolayer two times with sterile PBS (approximately 5 mL per wash)

b. Add 5 mL trypsin/EDTA solution and wash monolayer for approximately 15 seconds c Drain all but approximately 1 mL of trypsin/EDTA from flask, cap flask tightly, and incubate at 37°C for

approximately two to five minutes until cells come loose.

d. Add 10-15 mL tissue culture (T.C.) medium (RPMI 1640 plus 10 percent fetal calf serum and 2 mM L-glutathione) to flask and pipet gently up and down to wash cells.

e. Remove a 1/2 mL aliquot of the cell

suspension and transfer to a glass 12 X 75 mm culture tube for counting.

f. Count cells on a hemacytometer using

trypan blue, and determine percent viability.

g. Adjust volume of cell suspension with

T.C. media to give a density of 1 X 10⁵ cells/mL.

h. Add 100 μL of T.C. medium to wells A1 and

B1 of a 96-well plate for blanks, i. Add 100 μL of cell suspension to the

remaining wells of the 96-well plates, j. Incubate plates for 24 hours at 37°C,

5-10 percent CO₂ in a humidified

incubator.

2. Preparation of sample drugs and toxic control

a. Stock drug solutions were prepared by

dissolving drug in the appropriate solvent (determined during chemical characterization studies) and sterile filtering the drug-solvent solution through a sterile 0.2 μ filter unit. An aliquot was taken from each filtered drug solution and the O.D. was measured to determine the drug concentration.

b. Dilute the stock drug solution prepared above with T.C. medium to the desired initial concentration (10^-2-10-₄M). A minimum volume of 220 μL of diluted drug is required per 96-well plate used in the assay,

c Prepare toxic control by diluting stock doxorubicin solution to 10^-7 to 10^-9M in

T.C. medium. A minimum volume of 300 μL is required per 96-well plate.

3. Addition of Sample Drugs, Compounds, Chimeras and Controls to 96-well Plates

a. Remove and discard 100 μL of T.C. medium from the wells in Column #2 of the 96- well plate using a multi-channel pipettor and sterile tips.

b. Add 100 μL of the initial compound

dilution to adjacent duplicate wells in Columns #2. (Four materials can be tested in duplicate per 96-well plate.) c. Remove 10 μL of diluted compound from the wells in Column #2 and transfer to the corresponding wells in Column #3. Mix by pipetting up and down gently approximately five times.

d. Transfer 10 μL to the appropriate wells in Column #4 and continue to make 1:10 dilutions of compound across the plate through Column #12.

e. Remove and discard 100 μL of medium from wells F1, G1, and H1. Add 100 μL of toxic control (Doxorubicin diluted in T.C. medium) to each of these wells, f. Incubate (37°C, 5-10 percent CO₂ in

humidified incubator) plates for a total of 72 hours. Check plates at 24 hour intervals microscopically for signs of cytotoxicity.

4. Cell Fixation

a. Adherent cell lines:

1. Fix cells by gently layering 25 μL of cold (4°C) 50 percent

trichloroacetic acid (TCA) on top of the growth medium in each well to produce a final TCA concentration of

10 percent.

2. Incubate plates at 4°C for one hour. b. Suspension cell lines:

1. Allow cells to settle out of

solution.

2. Fix cells by gently layering 25 μL of cold (4°C) 80 percent TCA on top of the growth medium in each well.

3. Allow cultures to sit undisturbed for five minutes.

4. Place cultures in 4°C refrigerator for one hour.

c. Wash all plates five times with tap

water.

d. Air dry plates.

5. Staining Cells

a. Add 100 μL of 0.4 percent (wt./vol.)

Sulforhodamine B (SRB) dissolved in 1 percent acetic acid to each well of 96- well plates using multichannel pipettor. b. Incubate plates at room temperature for 30 minutes.

c. After the 30 minute incubation, shake

plates to remove SRB solution.

d. Wash plates two times with tap water and 1 × with 1 percent acetic acid, shaking out the solution after each wash. Blot plates on clean dry absorbent towels after last wash.

e. Air dry plates until no standing moisture is visible.

f. Add 100 μL of 10mM unbuffered Tris base (ph 10.5) to each well of 96-well plates and incubate for five minutes on an orbital shaker.

g. Read plates on a microtiter plate reader at 540 nM.

IC₅₀ values; i.e., the concentration of

Compound required to kill one-half of the treated cells, were then calculated.

The cell lines assayed are listed below along with their respective sources:

Cell Line Source and Type

NHDF Normal human dermal fibroblasts, as

controls - Clonetics Corporation

Capan-1 American Type Culture Collection (ATCC)

Molt-4 (All are human cancer cell lines as

OVCAR-3 described by the ATCC)

OVCAR-4

Sk-Mel-28

U-87

U-251 UCLA M-14 Dr. R. Reisfeld of The Scripps Research UCLA M-21 Institute, and originally obtained from UCLA P-3 Dr. D. Morton, University of California,

Los Angeles. M-14 and M-21 are human melanoma cell lines, whereas P-3 is a human non-small cell lung carcinoma cell line.

Control studies were also carried out using the following well known anticancer drugs with the following IC₅₀ values for NHDF and cancer cells.

Range of Average IC₅₀ Values (Molarity)

Drug NHDF Cancer Cells

Dexorubicin - - 1.6X10^-10 - 9.8X10^-8

Dynemicin A 10^-8 1.6X10^-8 - 9.8X10^-10*

Calicheamicin 2.5X10^-9 5X10^-5 - 10^{- 12**}

Morpholinodoxorubicin - - 1.6X10^-7 - 9.8X10^-9

Taxol 10^-8 10^-7 - 10^-9

Methotrexate 5X10^-5 >10^-4 - 10^-8

Cis-Platin 5X10^-5 10^-4 - 10^-6

Melphelan 10^-4 10^-4 - 10^-6

* UCLA-P3 cells were susceptible at 10 ^-12M. All other cells were susoceptible at 1.56X10^-10 M or higher

concentrations.

** Molt-4 cells were susceptible at 10^-12M All other cells were susceptible at 3.9X10^-9M or higher

concentrations.

Tables 1 and 2 herein below provide average

IC ₅₀ data from five to all ten of the above cancer cells lines for fused ring enediyne compounds disclosed herein. Those tables provide a generic structure and a description of the R group of that generic structure for each compound. Compound numbers as provided

hereinbefore are also provided. Table 3 contains data for two fused-ring enediyne compounds (Compounds 47 and 120) as well as data for six chimers.

As will be seen from the data of the tables, the compounds and chimeras had activities similar to those of the well known anticancer drugs.

Compounds having an R⁴ hydrogen tended to be more active than those having an R⁴ hydroxyl or other oxygen-containing group. Compounds whose carbamate portion contained a phenylsulfonylethoxy or

naphthylsulfonylethoxy group were about 10 to 100 times more active than similar compounds having a phenoxy group as part of the carbamate. Compounds having an electron releasing group relative to hydrogen para to the carbamate nitrogen atom tended to be equal to less active than those with hydrogen, whereas compounds with an intracellular-formed electron releasing group (e.g. Compounds 59a and 59b) relative to hydrogen meta to that nitrogen atom tended to be more active than compounds having hydrogen at that position.

Turning to the chimeras, the data Table 3 indicate that the chimeras are effective. Those data also indicate that the presence of the FMOC group inhibits activity, but that presence of the oxime does not. Those data also indicate that chimeras having the stereochemistry of the calicheamicin oligosaccharide are more active than those having the epimeric

stereochemistry at C-4 of the A ring.

The data of the tables also show compounds and chimeras described herein to be particularly active against Molt-4 leukemia cells. Thus, for those cells. Compounds 59b, 41b, 4lc and 41d exhibited IC₅₀ values 10,000 times more potent than the potency observed against the other cell lines. The activity of chimeric Compound 127a against Molt-4 cells was about 100-times that of the average of the other cell lines examined. Those IC₅₀ values against Molt-4 cells were also 10,000- 100,000 times smaller than the IC₅₀ values for those compounds against NHDF cells.

Compound Preparation and Data

Example 1: 7,8,9,10-Tetrahydrophenanthridine

N-oxide (Compound 4a.

A solution of 4 (1.42 g, 7.76 mmol) in

dichloromethane (50 mL) was treated at 25°C with mCPBA (1.58 g of an 85 percent sample, 7.76 mmol) and stirred for 1 hour. The solution was poured into saturated sodium bicarbonate solution (50 mL) and extracted. The aqueous layer was extracted with further dichloromethane (2 × 50 mL), the combined organic layers were dried (Na₂SO₄), evaporated in vacuo, and the residue was purified by flash chromatography on silica eluting with 25 percent MeOH in EtOAc to give the pure N-oxide

Compound 4a (1.24 g, 80 percent) as an off-white

crystalline solid: R_f=0.34 (25 percent MeOH in EtOAc); mp 131.7°C (from EtOAc); IR (CDCl₃) ν_max 2950, 1580, 1390, 1300, 1210, 1140 cm^-1; ¹H NMR (500MHz, CDCl₃) : δ 8.72 (d, J=8.3 HZ, 1 H, H-4), 8.31 (S, 1 H, H-6), 7.91 (d, J=8.3 Hz, 1 H, H-l) , 7.68 (t, J=8.3 Hz, 1 H, H-2 or H-3), 7.61 (t, J=8.3 HZ, 1 H, H-2 or H-3), 3.02 (t, J=6.3 Hz, 2 H, H-10) , 2.79 (t, J=6.3 Hz, 2 H, H-7), 1.98-1.84 (m, 4 H) ; MS (FAB+) m/e (relative intensity) 200 (M+H, 100), 184 (12); HRMS calcd for C₁₃H₁₄NO (M+H) 200.1075, found

200.1055. Example 2: Compound 2

A solution of enediyne 3 (205 mg, 0.50 mmol) in dry THF (10 mL) was cooled to -78°C and treated with lithium diisopropylamide (0.37 mL of a 1.5 M solution in cyclohexane, 0.56 mmol). After stirring 1 hour at

-78°C, the reaction was quenched with saturated ammonium chloride solution (2 mL) , allowed to warm to room temperature, poured into saturated sodium bicarbonate solution (30 mL), and extracted with dichloromethane (3 × 50 mL). The combined organic extracts were dried (Na₂SO₄) , evaporated in vacuo, and purified by flash chromatography on silica eluting with 50 percent ether in petroleum ether to give recovered 3 (48 mg, 23 percent), followed by the ten-membered enediyne Compound 2 (120 mg, 59 percent) as a white crystalline solid: R_f=0.42 (50 percent ether in petroleum ether); mp=228-230°C dec. (from ether) ; IR (CDCl₃) v_maX

3420,2360,2330,1720 cm^-1; ¹H NMR (500 MHz, CDCl₃) : S 8.60 (dd, J=8.1, 1.3 Hz, 1 H, aromatic), 7.47-7.10 (series of multiplets, 8 H, aromatic), 5.83 (d, J=10.1 Hz, 1 H, olefinic), 5.67 (dd, J=10.1, 1.6 Hz, 1 H, olefinic), 5.53 (d, J=1.6 Hz, 1 H, C H-N) , 2.35-1.71 (series of multiplets, 6 H, C H₂) ; ¹³C NMR (125 MHz, CDCl₃) : δ

151.0, 135.8, 131.3, 129.3, 128.0, 127.8, 126.3, 125.8, 125.3, 124.0, 122.2, 121.6, 100.4, 94.3, 93.9, 88.8, 74.1, 73.2, 64.4, 50.5, 35.4, 23.2, 19.1; MS: m/e

(relative intensity) 409 (26, M+) , 368 (18), 236 (11),

162 (13), 131 (100); HRMS: calcd. for C₂₆H₁₉NO₄: 409.1314, found: 409.1314; Anal, calcd. for C₂₆H₁₉NO₄.H₂O: C, 73.06; H, 4.95; N, 3.28. Found: C, 73.44; H, 5.04; N, 3.26.

Compounds 2-d were prepared from Compound 2 by reaction of Compound 2 with a mixture of the appropriate alcohol and its sodium salt as shown in Figure 15.

Example 3: N-Phenyloxycarbonyl-6-(3(Z)-hexene-1,5- diynyl)-6a:10a-epoxy-10-oxo- 5,6,6a,7,8,9,10,10a- octahydrophenanthridine (Compound 3.

Silver nitrate (5.28 g, 31.2 mmol) was added to solution of the silyl acetylene Compound 13 (4.50 g, 9.36 mmol) in 100 mL of a H 20-EtOH-THF mixture (1:1:1) at 25°C, and the mixture was stirred until TLC analysis (30 percent ether in petroleum ether) indicated the consumption of 13 (approximately 1 hour). Potassium cyanide (4.32 g, 57.6 mmol) was then added and the mixture was stirred for 10 minutes. The mixture was poured into saturated sodium bicarbonate solution (100 mL) and extracted with dichloromethane (3 × 100 mL) . The combined organic layers were dried (Na₂SO₄), evaporated in vacuo, and purified by flash chromatography on silica eluting with 30 percent ether in petroleum ether to give enediyne Compound 3 (2.30 g, 60 percent) as a colorless gum: R_f=0.38 (30 percent ether in petroleum ether); IR (CDCl₃) v_max. 3304, 2940, 2240, 1720, 1492, 1378, 1321, 1206 cm^-1; ¹H NMR (500 MHz, CDCl₃) : δ 8.87 (dd, J=7.8, 1.4 Hz, 1 H, H-4), 7.53-7.09 (m, 8 H, aromatic), 5.93 (d, J=1.2 Nz, 1 H, H-6), 5.78 and 5.79 (AB quartet,

J=10.1 Hz, 2 H, olefinic), 3.16 (d, J=1.2 Hz, 1 H, C≡C- H) , 2.79-2.66 (m, 2 H, H-9), 2.38-2.29 (m, 2 H, H7), 2.04-1.89 (m, 2 H, H-8); ¹³C NMR (125 MHz, CDCl₃) : δ 150.9, 135.9, 130.0, 129.3, 128.8, 127.6, 125.9, 125.8, 123.0, 121.4, 120.4, 120.2, 90.6, 85.1, 82.8, 80.1, 75.1, 57.4, 48.1, 38.9, 23.9, 18.3; MS m/e (relative intensity) 409 (2, M+), 262(15), 212(18), 162(59),

58(100); HRMS: calcd. for C₂₆H₁₉NO₄: 409.1314, found:

409.1308.

Example 4: 10-Acetoxy-7,8,9,10- tetrahydrophenanthridine (Compound 5)

A solution of 7,8,9,10-tetrahydrophenanthridine N-oxide (Compound 4a) (1.23 g, 6.18 mmol) in acetic anhydride (50 mL) was heated to 100°C for 20 hours, evaporated to dryness, dissolved in dichloromethane (50 mL) and washed with saturated sodium bicarbonate solution (50 mL). The aqueous layer was extracted with dichloromethane (2 × 50 mL), the combined organic layers were dried (Na₂SO_A), evaporated in vacuo. and the residue was purified by flash chromatography on silica eluting with Et₂O to give pure Compound 5 (1.15 g, 77 percent) as a white crystalline solid: R f=0.33 (ether) ; mp=128-129°C (from ether); IR (CDCl₃) v_max 2970, 1728, 1241 cm^-1; ¹H NMR (500 MHz, CDCl₃) : δ 8.70 (s, 1 H, H-6), 8.08 (d, J=9.5 HZ, 1 H, H-4), 7.76 (d, J=9.5 HZ, 1 H, H-1), 7.63 (t, J=7.2 Hz, 1 H, H-2 or H-3), 7.52 (t, J=7.2 HZ, 1 H, H-2 or H-3) , 6.57 (bs, 1 H, CH-OAc), 3.02 (bd, J=17.5 Hz, 1 H, H-7) , 2.88-2.80 (m, 1 H, H-7), 2.27 (bd, J=13.8 Hz, 1 H, H-9) , 2.05 (s, 3 H, OAc) , 2.01-1.88 (m, 3 H) ; ¹³C NMR (125 MHz, CDCl₃) : δ 170.2, 152.3,

147.5, 137.8, 131.8, 130.1, 127.9, 127.0, 126.8, 122.2, 64.5, 29.1, 27.8, 21.7, 18.4; MS (FAB+) m/e (relative intensity) 242 (M+H, 100), 182 (23); HRMS calcd. for C₁₅H₁₆NO₂ (M+H) 242.1181, found 242.1181; Anal, calcd. for C₁₅H₁₅NO₂; C, 74.67; H, 6.27; N, 5.80. Found: C, 74.59; H, 6.31; N, 5.82.

Example 5: 10-Hydroxy-7,8,9,10- tetrahydrophenanthridine (Compound 6.

A solution of Compound 5 (1.15 g, 4.77 mmol) in methanol (50 mL) was treated with potassium carbonate (200 mg, catalytic) and stirred for 1 hour. The

solution was poured into saturated sodium bicarbonate solution (100 mL) and extracted with dichloromethane (1 × 100 mL, 2 × 50 mL). The combined organic layers were dried (MgSO₄) , evaporated in vacuo, and the residue was purified by flash chromatography on silica eluting with ethyl acetate to give alcohol Compound 6 (0.95 g, 100 percent) as a white crystalline εr.lid: R_f=0.20

(ether); mp=176-177°C (from ether); IR (CDCl₃) v_max 3600, 2950, 1510 cm^-1; ¹H NMR (500 MHz, CDCl₃) : δ 8.51 (s, 1 H, H-6), 8.20 (d, J=9.1 Hz, 1 H, H-4) , 8.00 (d, J=9.1 Hz, 1 H, H-1), 7.61 (t, J=6.8 Hz, 1 H, H-2 or H-3), 7.55 (t, J=6.8 Hz, 1 H, H-2 or H-3), 5.39 (bs, 1 H, CH-OH) , 2.89 (bd, J=16.1 Hz, 1 H, H-7), 2.80-2.72 (m, 1 H, H-7) , 2.80-2.60 (bs, 1 H, OH), 2.24 (bd, J=12.5 Hz, 1 H, H-8 or H-9), 2.07-1.88 (m, 3 H) ; MS (FAB+) m/e (relative intensity) 200 (M+H, 100), 154 (41), 136 (37), 109 (24); HRMS calcd. for C₁₃H₁₄NO (M+H) 200.1075, found 200.1085.

Example 6: 10-tert-Butyldimethylsilyloxy-7,8,9, 10- tetrahydrophenanthridine (Compound 7)

A solution of Compound 6 (3.70 g, 18.6 mmol) in dry dichloromethane (100 mL) was treated with 2,6-lutidine (3.4 mL, 27.9 mmol) and tert-butyl-dimethylsilyl triflate (5.35 mL, 22.3 mmol). After stirring for 1 hour at 25°C, methanol (2 mL) was added, stirring was continued for a further 5 minutes, and then the solution was poured into saturated sodium

bicarbonate solution (100 mL) and extracted. The aqueous layer was extracted with further dichloromethane (2 × 50 mL), the combined organic layers were dried (Na₂SO₄), evaporated in vacuo and the residue was

purified by flash chromatography on silica eluting with 50 percent ether in petroleum ether to give pure silyl ether Compound 7 (5.38g, 92 percent) as a white solid. 7: colorless oil; R_f=0.50 (70 percent ether in petroleum ether); IR (CDCl₃) v_max 2970, 2930, 2860 cm^-1; ¹H NMR (500 MHz, CDCl₃) : δ 8.68 (s, 1 H, H-6) , 8.08 (d, J=4.7 Hz, 1 H, H-l or H-4), 8.05 (d, J=4.7 Hz, 1 H, H-l or H-4) , 7.62 (t, J=4.7 Hz, 1 H, H-2 or H-3) , 7.53 (t, J=4.7 Hz, 1 H, H-2 or H-3), 5.45 (t, J=2.8 Hz, 1 H, H-10) , 3.00 (dd, J=5.5, 16.6 Hz, 1 H, CH-Ar), 2.81 (m, 1 H, CH-Ar) , 2.23-2.10 (m, 2 H, CH₂) , 1.88-1.78 (m, 2 H, CH₂) , 0.84 (s, 9 H, ^tBu) , 0.22 (s, 6 H, SiMe₂); ¹³C NMR (125 MHz, CDCl₃) : δ 152.8, 147.0, 141.2, 129.8, 129.3, 127.9, 126.9, 126.1, 123.6,-63.2, 31.8, 27.0, 25.8, 18.2, 16.4, -3.6, -4.5; MS (FAB+) m/e (relative intensity) 314 (M+H, 100), 256 (7), 182 (11); HRMS calcd. for C₁₉H₂₈NOSi (M+H) 314.1940, found 314.1951.

Example 7: N-Phenyloxycarbonyl-10-tert- butyldimethylsilyloxy-6-ethyl- 5,6,7,8,9,10-hexahydrophenanthridine

(Compound 8.

A solution of quinoline Compound 7 (5.20 g, 16.6 mmol) in dry THF (166 mL) was cooled to -78°C and treated with ethynylmagnesium bromide (36.5 mL of a 0.5 M solution in THF, 18.3 mmol) followed by phenyl chloroformate (2.3 mL, 18.3 mmol). The solution was allowed to warm up slowly to 25°C over 1 hour, quenched with saturated ammonium chloride solution (10 mL), poured into saturated sodium bicarbonate solution (150 mL) and extracted. The aqueous layer was extracted with dichloromethane (2 × 100 mL), the combined organic layers were dried (Na₂SO₄) , evaporated in vacuo, and purified by flash chromatography on silica eluting with 10 percent ether in petroleum ether to give pure

carbamate Compound 8 (6.96 g, 92 percent) as a colorless oil (about 3:1 mixture of isomers as judged by NMR). R_f=0.85 (30 percent ether in petroleum ether); IR

(CDCl₃) v_max 3300, 2952, 2858, 2250, 1715, 1473, 1204 cm ^{-1 1}H NMR (500 MHZ, CDCl₃) : δ 7.68 (d, J=7.5 Hz, 1 H,

H-4), 7.40-7.12 (m, 8 H) , 5.68, 5.61 (2 × s, 1 H, H-6) , 5.00, 4.69 (2 × bs, 1 H, H-10) , 2.50-1.50 (m, 7 H) , 0.80, 0.92 (2 X S, 9 H, ^tBu) , 0.28, 0.19, 0.10, 0.09 (singlets, 6 H, SiMe₂) ; ¹³C NMR (125 MHz, CDCl₃) : δ

151.1, 136.3, 132.9, 129.8, 129.3, 127.2, 126.0, 125.4, 125.1, 124.2, 124.1, 123.9, 122.0, 80.2, 72 3, 65.0 and

64.2, 48.7 and 48.2, 32.3 and 31.4, 28.0, 26.1, 18.4 and

16.3, -4.1 and -4.8; MS m/e (relative intensity) 459 (M\ 10), 402 (100), 366 (10), 308 (24), 206 (26), 151 (27), 75 (29); HRMS calcd. for C₂₈H₃₃O₃NSi (M⁺) : 459.2230, found: 459.2233.

Example 8: N-Phenyloxycarbonyl-10-tert- butyldimethylsilyloxy-6a:1 Oa-epoxy- 6-ethyl-5,6,6a,7,8, 9,10,10a- octahydrophenanthridine (Compound 9.

A solution of Compound 8 (8.60 g, 18.8 mmol) in dichloromethane (120 mL) was treated with mCPBA (8.08 g of a 60 percent sample, 37.6 mmol) and stirred at 25°C for 3 hours. The solution was poured into saturated sodium bicarbonate solution (100mL) , extracted, and the aqueous layer extracted with further dichloromethane (2 × 100 mL). The combined organic layers were dried (Na₂SO₄) , evaporated in vacuo. and the residue was purified by flash chromatography on silica eluting with 10 percent ether in petroleum ether to give epoxide Compound 9 (8.20 g, 92 percent) as a white foam (mixture of two isomers, about 3:1 ratio); R_f=0.73 (30 percent ether in petroleum ether); IR v_max (CDCl₃) 3307, 2953, 2250, 1721, 1494, 1384, 1322, 1250, 1207 cm^-1; ¹H NMR (500 MHz, CDCl₃) : δ 7.88 (d, J=7.1 Hz, 1 H, H-4) , 7.50-7.10 (m, 8 H) , 5.58 (bs, 1 H, H-6) , 4.82 (dd, J=10.0, 5.7 Hz, 1 H, H-10) , 2.34 (dd, J=14.8, 5.6 Hz, 1 H) , 2.09 (bs, 1 H) , 1.95-1.85 (m, 2 H) , 1.78-1.62 (m, 2 H) , 1.40-1.30 (m, 1 H) ; ¹³C NMR (125 MHz, CDCl₃) : δ 153.9 and 151.1, 135.5, 129.2, 129.2, 129.1, 128.3, 128.1, 127.9, 127.1, 125.5, 121.6, 78.5, 73.8, 72.8, 69.9, 60.4, 48.0, 31.0 and 29.6, 26.0 and 25.8, 24.0, and 26.5, 18.2 and 20.3, -0.28, -0.28, -0.37; MS: m/e (relative intensity) 475(M⁺,2), 419 (100), 325 (28), 268 (10), 222 (14), 151 (18), 73 (42); HRMS: calcd. for C₂₈H₃₃O₄NSi (M⁺) :

475.2179, found: 475.2175. Example 9: N-Phenyloxycarbonyl-6a:10a-epoxy-6-ethyl- 10-hydroxy-5,6,6a,7,8,9,10, 10a- octahydrophenanthridine (Compound 10. A solution of epoxide Compound 9 (8.20 g, 17.4 mmol) in THF (100mL) was treated with TBAF (20.9 mL of a 1 M solution in THF, 20.9 mmol) and heated to 42°C for three hours. The solution was evaporated in vacuo and purified by flash chromatography on silica eluting with 50 percent Et₂O/petroleum ether to give pure alcohol

Compound 10 (6.00 g, 100 percent) as a white crystalline solid (about 3:1 mixture of isomers as shown by NMR). R_f=0.31 (50 percent ether in petroleum ether); mp

78-79°C (from Et₂O) ; IR (CDCl₃) v_max 3580, 3306, 2951, 2250, 1720, 1595, 1494, 1382, 1322, 1206 cm^-1; ¹H NMR (500 MHz, CDCl₃) : δ 7.91 and 7.88 (d, J=8.0 Hz, 1 H, H-4), 7.50-7.08 (m, 8 H) , 5.62 and 5.59 (d, J=1.0 Hz, 1 H, H-6), 4.89 and 4.70 (m, 1 H, H-10) , 2.47-1.35 (m, 8 H) ; ¹³C NMR (125 MNz, CDCl₃) : δ 150.9, 135.5, 129.3, 128.7, 128.6, 128.4, 127.7, 127.3, 126.1, 125.8, 121.5, 78.7 and 78.2, 74.8 and 70.8, 73.2, 66.6, 65.9 and 64.4, 60.9 and 58.2, 47.8, 30.3 and 27.0, 24.1 and 19.0, 15.2 and 13.8; MS m/e (relative intensity) 361 (M⁺,65), 224 (100), 196 (24), 180 (29), 167 (30), 94 (40), 77 (45); HRMS: calcd. for C₂₂H₁₉NO₄ (M⁺) ; 361.1314, found:

361.1317.

Example 10: N-Phenyloxycarbonyl-6a:10a-epoxy-6-ethyl-

10-oxo-5,6,6a,7,8,9,10,10a- octahydrophenanthridine (Compound 11)

Alcohol Compound 10 (6.009, 17.4 mmol) was dissolved in dry dichlorometh ne (180 mL) and treated with powdered, activated 4A molecule... sieves (l g) and pyridinium chlorochromate (6.25 g, 29.0 mmol). The suspension was stirred for 1 hour at 25°C, filtered through celite, concentrated in vacuo, and the residue was purified by flash chromatography on silica eluting with 30 percent Et₂0/petroleum ether to give ketone Compound 11 (4.49 9,75 percent) as a white foam: R_f-=0.51 (50 percent ether in petroleum ether); IR (CDCl₃) ν_max 3306, 2259, 1721, 1491, 1321, 1206 cm^-1; % NMR (500 MHz, CDCl₃) : δ 8.50 (d, J=7.8 Hz, 1 H, H-4) , 7.53-7.10 (m, 8 H, aromatic), 5.73 (d, J=2.4 Hz, 1 H, H-6), 2.76 (dt, J=15.2, 4.9 Hz, 1 H, H-9), 2.60 (ddd, J=15.2, 10.4, 6.1 Hz, 1 H, H-9), 2.37-2.28 (m, 2 H, H-7), 2.21 (bs, 1 H, C≡C-H), 2.04-1.90 (m, 2 H, H-8) ; ¹³C NMR: (125 MHz, CDCl₃) : δ 201.0, 153.9, 151.0, 135.8, 129.9, 129.3, 129.0, 127.6, 126.1, 125.9, 123.0, 121.5, 77.7, 74.9, 74.2, 57.4, 47.3, 38.9,23.8, 18.3; MS m/e (relative intensity) 359(M⁺,100), 266 (52), 222 (65), 194 (54), 180 (51), 146 (45), 69 (80); HRMS calcd. for C₂₂H₁₇NO₄ (M⁺) : 359.1158, found: 359.1154; Anal, calcd. for

C₂₂H₁₇NO₄: C, 73.53; H, 4.77; N, 3.90. Found: C, 73.27; H, 4.79; N, 3.91. Example 11: N-Phenyloxycarbonyl-6-[6-trimethylsilyl-

3 (Z) -hexene-1,5-diynyl]-6a:10a- epoxy-10-oxo-5,6,6a,7,8,9,10,10a- octahydrophenanthridine (Compound 13. Palladium¹¹ acetate (192 mg, 0.86 mmol) and triphenylphosphine (832 mg, 3.17 mmol) in dry, degassed benzene (10 mL) were heated under argon at 60°C for 1 hour. The resulting dark red solution was cooled to 25°C, and the (Z)-chloroenyne Compound 12 (2.88 g, 18.2 mmol) in dry, degassed benzene (20 mL) was added, followed by n-butylamine (1.92 mL, 19.4 mmol). The solution was stirred for 15 minutes at 25°C, cooled to zero degree C, and the acetylene 11 (4.49g, 12.5 mmol) in dry, degassed benzene (50 mL) was added, followed by copper (I) iodide (512 mg, 2.69 mmol). The solution was stirred for 2 hours at 25°C, poured into saturated sodium bicarbonate solution (100 mL) and extracted. The aqueous layer was extracted with dichloromethane (2 × 50 mL) , the combined organic layers were dried (Na₂SO₄) , evaporated in vacuo, and the residue was purified by flash chromatography on silica eluting with 20 percent ether in petroleum ether to give the coupled product Compound 13 (4.50 g, 74 percent) as a colorless gum: R_f=0.51 (30 percent ether in petroleum ether); IR

(CDCl₃) V_max 2962, 1720, 1492, 1378, 1322, 1252, 1206, 846 cm^-1; ¹H NMR (500 MHz, CDCl₃) : δ 8.36 (d, J=8.5 Hz, 1 H, H-4), 7.52-7.09 (m, 8 H, aromatic), 5.99 (d, J=1.6 Hz, 1 H, H-6), 5.82 (d, J=11.2 Hz, 1 H, olefinic), 5.66 (dd, J=11.2, 1.6 Hz, 1 H, olefinic), 2.76 (dt, J=15. , 4.7 Hz, 1 H, H-9), 2.71 (ddd, J=15.3, 10.8, 6.1 Hz, 1 H, H-9), 2.39-2.30 (m, 2 H, H-7) , 2.07-1.89 (m, 2 H, H-8), 0.25 (s, 9 H, SiMe₃) ; ¹³C NMR (125 MHz, CDCl₃) : δ 201.1, 150.9, 135.8, 129.9, 129.2, 128.9, 128.4, 127.7, 126.0, 125.8, 122.9, 121.4, 120.8, 118.9, 103.6, 101.5, 90.4, 83.0, 74.9, 57.5, 48.3, 38.9, 23.9, 18.2, 0.00; MS m/e (relative intensity) 481 (M⁺,11), 360 (100), 146 (10); HRMS: calcd. for C₂₉H₂₇O₄NSi (M⁺) : 481.1709, found:

481.1705.

Example 12: Compound 2a

A solution of enediyne Compound 2 (100.1 mg,

0.224 mmol) in pyridine (2 mL) was treated with acetic anhydride (0.50 mL, 5.31 mmol) and DMAP (10 mg,

catalytic) at 25°C. After 2 hours, the reaction mixture was poured into saturated sodium bicarbonate solution (25 mL), extracted with dichloromethane (3 × 25 mL), the combined organic layers were dried (Na₂SO₄), evaporated in vacuo, and the residue was purified by flash

chromatography (silica, 30 percent ether in petroleum ether) to give acetate Compound 2a (110.7 mg, 100 percent). 2a: white crystalline solid, mp 212-214°C dec (from ether); R_f=0.55 (50 percent ether in

petroleum ether); IR (CDCl₃) v_max 3075, 2950, 2215, 1742, 1720, 1500, 1216, 769 cm^-1; ¹H NMR (500 MHz, CDCl₃) : δ 7.92 (d, J=8.1 Hz, 1 H, aromatic), 7.50-7.09 (m, 8 H, aromatic), 5.83 (d, J=10.1 Hz, 1 H, olefinic) , 5.65 (d, J=10.1 Hz, 1 H, olefinic), 5.53 (s, 1 H, N-CH(C)-C), 2.51-1.70 (m, 6 H, CH₂) , 2.18 (s, 3 H, OAc) ; ¹³C NMR (125 MHz, CDCl₃) : δ 169.1, 150.8, 130.0, 129.3, 128.4, 128.1, 128.0, 127.2, 126.8, 125.7, 125.2, 124.3, 122.9, 121.5, 97.9, 96.5, 93.7, 88.9, 78.0, 73.5, 62.6, 50.3, 29.8,

22.7, 21.8, 18.8; MS (FAB+) m/e (relative intensity) 452 (M+H, 52), 410 (37), 392 (100), 316 (32), 272 (43), 242 (30), 154 (77), 136 (70); HRMS calcd. for C₂₈H₂₈NO₅ (M+H) 452.1498, found 452.1469.

A solution of enediyne Compound 2a (93.0 mg,

0.206 mmol) and 1,4-cyclohexadiene (1.0 mL) in benzene (3.0 mL) was treated with p-toluene-sulfonic acid (39 mg, 0.23 mmol) and stirred at 60°C for 2 hours. The solvent was removed in vacuo and the residue purified by flash chromatography (silica, 50 percent ether in petroleum ether) to give diol acetate Compound 15a (80.2 mg, 83 percent). 15a: while crystalline solid, mp 198-200°C (from ether). Example 13: Compound 15

Colorless solid: R_f=0.22 (50 percent ether in petroleum ether); IR (CDCl₃) v_max 3360, 3072, 2950, 1738, 1715, 1500, 1192 cm^-1; ¹H NMR (500 MHz, CDCl₃) : δ 7.57 (d, J=7.8 Hz, 1 H, aromatic), 7.40-7.01 (series of multiplets, 12 H, aromatic), 6.83 (bs, 1 H, O H), 5.59 (s, 1 H, N-C H(C)-C), 3.17 (m, 1 H, CH₂), 2.28 (m, 1 H, CH₂) , 2.26 (s, 3 H, OAc) , 1.80-1.40 (series of

multiplets, 3 H, CH₂), 0.72 (m, 1 H, CH₂) ; ¹³C NMR (125 MHz, CDCl₃) : δ 174.9, 150.8, 137.7, 134.8, 133.5, 129.7, 129.3, 129.2, 129.0, 128.8, 128.5, 128.2, 127.8, 127.7, 125.5, 124.8, 123.4, 121.8, 93.8, 75.1, 70.6, 61.4, 32.5, 31.4, 22.6, 19.8; MS: m/e (relative intensity) 471 (M+, 19), 245 (100), 162 (100), 94 (42); HRMS: calcd. for C₂₈H₂₅O₆N (M⁺) : 471.1682, found: 471.1683; Anal.

calcd. for C₂₈H₂₅O₆N: C, 71.33; H, 5.34; N, 2.97. Found: C, 71.36; H, 5.54; N, 2.84.

Example 14: Compound 15b

Dry HCl gas was bubbled through a solution of acetate Compound 2a (32 mg, 0.071 mmol) and 1,4-cyclohexadiene (40 mg, 0.32 mmol) in dichloromethane (4 mL) at 25°C for 30 seconds. The solvent was removed in vacuo and the residue purified by flash chromatography (silica, 50 percent ether in petroleum ether) to give the chloride, Compound 15b (25 mg, 80 percent).

Colorless gum; R_f=0.21 (50 percent ether in petroleum ether); IR (CDCl₃) v_max 3500, 2945, 1710, 1492, 1400, 1225, 789 cm^-1; ¹H NMR (500 MHz, CDCl₃) : δ 7.72 (d, J=8.1 Hz, 1 H, aromatic), 7.45-6.96 (m, 12 H, aromatic), 5.85 (s, 1 H, benzylic), 2.56 (bs, 1 H, OH), 2.37 (bs, 1 H, OH), 2.34-1.42 (m, 6 H, CH₂) ; ¹³C NMR (125 MHz,

CDCl₃) : δ 151.2, 134.7, 132.6, 130.4, 129.4, 129.3,

128.6, 128.5, 128.2, 128.2, 128.1, 127.5, 125.7, 124.5, 124.2, 124.0, 121.7, 81.2, 80.4, 70.2, 62.7, 35.4, 33.7, 18. ε; MS (FAB+) : m/e (relative intensity) 580 (M+Cs, 100), 419 (42), 286 (100), 154 (37); HRMS: Calcd. for C₂₆H₂₂O₄NClCs (M+Cs) : 580.0291, found: 580.0286.

Example 15: Compound 17

Compound 17 has been prepared by several methods as indicated below.

Method (i) : A solution of the cobalt complex

Compound 19 (42 mg, 0.060 mmol) in CH₂Cl₂ (1 mL) was treated with Et₃N⁺-o- (32.7 mg, 0.29 mmol) and stirred at 25°C for 4 hours. The solution was poured into saturated sodium bicarbonate solution (25 mL) and extracted with CH₂Cl₂ (3 × 25 mL). The combined organic layers were dried (MgSO₄) , evaporated in vacuo and purified by flash chromatography to give the aromatized product Compound 17 (17.7 mg, 83 percent).

Method (ii) : A solution of enediyne Compound 2 (57.8 mg, 0.141 mmol) and 1,4-cyclohexadiene (0.5 mL) in benzene (1.5 mL) was treated with p-toluenesulfonic acid (29.6 mg, 0.155 mmol) and stirred at 25°C for 24 hours. The solvent was removed in vacuo and the residue

purified by flash chromatograph (silica, 50 percent ether in petroleum ether) to give keton Compound 17 (53.7 mg, 92 percent).

Method (iii): Trimethylsilyl triflate (15 μL, 0.08 mmol) was added to a solution of enediyne Compound 2 (32 mg, 0.078 mmol) and triethylsilane (40 mg, 0.32 mmol) in dichloromethane (2 mL) at -78°C. After five minutes, the mixture was quenched at -78°C with

saturated ammonium chloride solution (1 mL), diluted with either (10 mL), washed with water (2 × 3 mL), brine (3 mL) and dried (MgSO₄). The organic solvent was removed in vacuo and the residue was purified by flash chromatograph (silica, 50 percent ether in petroleum ether) to give ketone Compound 17 (22 mg, 68 percent) .

Method (iv) : Dry HCl gas was bubbled through a solution of enediyne Compound 2 (32 mg, 0.078 mmol) and 1,4-cyclohexadiene (40 mg, 0.32 mmol) in dichloromethane (4 mL) at 25°C for 30 seconds. The solvent was removed in vacuo and the residue purified by flash

chromatography (silica, 50 percent ether in petroleum ether) to give Compound 17 (25 mg, 78 percent).

White crystals; R_f=0.63 (70 percent ether in petroleum ether); mp=191-193°C (from methylene

chloride/ether); IR (CDCl₃) v_max 3480, 3080, 2935, 1712, 1490, 1264, 1192 cm^-1; ¹H NMR (500 MHz, CDCl₃) : δ 8.33 (dd, J=7.9, 1.3 Hz, 1 H, aromatic), 8.09 (d, J=7.5 Hz, 1 H, aromatic), 7.54-7.02 (m, 11 H, aromatic), 5.65 (s, 1 H, benzylic), 2.75 (bs, 1 H, OH), 2.69-1.80 (m, 6 H, CH₂) ; ¹³C NMR (125 MHz, CDCl₃) : δ 207.5, 153.0, 150.9, 148.2, 137.1, 134.2, 129.8, 129.5, 128.5, 128.2, 127.8, 127.1, 126.1, 126.0, 124.3, 122.8, 121.8, 121.3, 82.5, 65.0, 64.1,40.0, 30.2, 23.5; MS: m/e (relative

intensity) 411 (M⁺,100), 318 (58), 274 (49), 246 (12), 217 (55), 94 (29); HRMS: Calcd. for C₂₆H₂₁O₄N (M+) :

411.1471, found: 411.1468; Anal. Calcd. for C₂₆H₂₁O₄N: C, 75.90; H, 5.14; N, 3.40. Found: C, 75.66; H, 5.45; N, 3.14.

Example 16: Compound 18

A solution of the enediyne Compound 2 (124 mg,

0.30 mmol) in CH₂Cl₂ (4 mL) was treated with Co₂(CO)₈ (260 mg, 0.76 mmol) and stirred at 25°C for 5 minutes. The solution was concentrated in vacuo and the residue was purified by flash chromatography to give the cobalt complex Compound 18 (291 mg, 98 percent).

Green crystalline solid; mp>300°C (from ether); R_f=0.80 (50 percent ether in petroleum ether); IR

(CDCl₃) v_max 3500, 2950, 2872, 2095, 2070, 2025, 1725, 1492, 1207 cm^-1; ¹H NMR (500 MHz, CDCl₃) : δ 8.87 (bs, 1 H, aromatic), 7.61-7.02 (m, 8 H, aromatic), 6.47 [bs, 1 H, N-CH(C)-C], 6.38 (bd, J=10.7 Hz, 1 H, olefinic), 6.19 (bd, J=10.7 Hz, 1 H, olefinic), 3.50 (bs, 1 H, OH), 2.70-1.71 (m, 6 H,CH₂); ¹³C NMR (125 MHz, CDCl₃) : δ

199.1, 198.6, 197.8, 151.3, 134.9, 132.9, 130.9, 129.4, 128.8, 127.2, 125.8, 125-3, 125.1, 124.8, 123.4, 121.6, 98.5, 88.9, 81.5, 80.1, 78.0, 73.9, 63.1, 59.0, 44.2, 24.9, 17.1; MS (FAB+) m/e (relative intensity) 1114 (M+Cs, 11), 1086 (M+Cs-CO, 18), 1058 (M+CS-2CO, 6), 1030 (M+CS-3CO, 19) , 1002 (M+CS-4CO, 11) , 943 (M+Cs-4CO-Co, 10), 918 (11), 890 (24), 862 (34), 813 (100); HRMS calcd. for C₃₈H₁₉O₁₆NCo₄Cs (M+Cs) 1113.7086, found

1113.7001.

Example 17: Compound 19

A solution of the cobalt complex Compound 18

(291 mg, 0.30 mmol) in CH₂Cl₂ (4 mL) was treated at zero degree C with trifluoroacetic acid (68.6 μL, 0.89 mmol). After 5 minutes, the mixture was poured into saturated sodium bicarbonates solution (25 mL) and extracted with CH₂Cl₂ (3 × 25 mL). The combined organic layers were dried (MgSO₄), evaporated in vacuo. and purified by flash chromatography (silica, 50 percent ether in petroleum ether) to give the ketone Compound 19 (167.4 mg, 81 percent).

Brown crystalline solid; mp >300°C (from ether);

R_f=0.25 (50 percent ether in petroleum ether); IR

(CDCl₃) ν_max 3408, 2945, 2100, 2065, 2032, 1875, 1735, 1680, 1512, 1217 cm^-1; ¹H NMR (500 MHz, CDCl₃) : δ 7.81 (d, J=8.2 Hz, 1 H, aromatic), 7.42-7.11 (m, 8 H,

aromatic), 7.00 (d, J=10.2 Hz, 1 H, olefinic), 6.39 [s, 1 H, N-C H(C)-C], 5.52 (d, J=10.2 Hz, 1 H, olefinic), 3.35-1.82 (m, 7 H, CH₂, OH); ¹³C NMR (125 MHz, CDCl₃) : δ 202.9, 198.9, 198.1, 154.3, 150.9, 144.0, 133.2, 132.8, 129.6, 128.3, 128.1, 126.7, 126.0, 125.8, 123.3, 121.8, 108.7, 93.2, 92.5, 82.1, 81.0, 68.5, 56.2, 38.0, 30.2,

21.6; MS (FAB+) m/e (relative intensity) 828 (M+Cs, 17), 800 (18), 688 (74), 639 (20), 555 (32), 527 (100); HRMS calcd. for C₃₂B₁₉NO₁₀Co₂Cs (M+Cs) 827.8727, found

827.8730; Anal, calcd. for C₃₂H₁₉NO₁₀Co_{2 :} C, 55.27; H, 2.75; N, 2.01; Co, 16.97. Found: C, 54.98; H, 2.79; N, 1.86; Co, 15.22.

Example 18: Compound 20

Thiocarbonyldiimidazole (180 mg, 0.99 mmol) was added to a solution of the alcohol Compound 2 (137 mg. 0.335 mmol) and 4-dimethylaminopyridine (DMAP) (25 mg, 0.18 mmol) in dichloromethane (2 mL) at 25°C. After 48 hours, the solution was concentrated in vacuo and the residue purified by flash chromatography (silica, 80 percent ether in petroleum ether) to give

thionoimidazolide 20 (160 mg, 95 percent). 20: white crystalline solid, mp 178-179°C dec. (from

ether/dichloromethane) ; R_f=0.62 (70 percent ether in petroleum ether); IR (CDCl₃) v_max 3042, 2912, 2195, 1710, 1500, 1495, 1212, 1105 cm^-1; ¹H NMR (500 MHz, CDCl₃) : δ 8.49 (s, 1 H, N-CH=N), 7.71-7.05 (m, 11 H, aromatic), 5.93 (d, J=10.3 Hz, 1 H, olefinic), 5.73 (dd, J=10.3, 1.6 Hz, 1 H, olefinic), 5.60 (d, J=1.6 Hz, 1 H, N-CH-C≡C) , 3.08 (d, J=11.1 Hz, 1 H, CH₂), 2.46-1.70 (m, 5 H, CH₂) ; ¹³C NMR (125 MHz, CDCl₃) : δ 179.1, 153.4, 151.0,

137.0, 135.9, 130.9, 129.4, 129.3, 128.2, 127.0, 126.4, 125.8, 125.4, 123.9, 123.2, 121.3, 117.7, 100.6, 94.3, 93.9, 88.9, 85.4, 74.5, 65.9, 63.2, 50.3, 28.0, 22.7, 18.4; MS (FAB+) m/e (relative intensity) 653 (M+Cs, 21), 419 (19), 379 (15), 286 (100), 154 (30); HRMS Calcd. for C₃₀H₂₁N₃O₄SCs (M+Cs) 653.0385, found 653.0360; Anal, calcd. for C₃₀H₂₁N₃O₄S: C, 69.35; H, 4.07; N, 8.09; S, 6.17. Found: C, 69.01; H, 4.17; N, 7.91; S, 6.19. Example 19: Compound 21

A solution of the imidazolide Compound 20 (112 mg, 0.24 mmol), azobisisobutyronitrile (AIBN; 3 mg) and tri-n-butylstannane (n-Bu₃SnH) (94 μL, 0.36 mmol) was heated at 75°C for 2 hours, the solvent was removed in vacuo. and the residue was purified by flash

chromatography to give the deoxygenated product Compound 21 (71 mg, 75 percent).

White crystals; R_f=0.62 (30 percent ether in petroleum ether); mp 248-250°C dec. (from ether);

IR(CDCl₃) ν_max 2945, 2872, 2232, 2205, 1712, 1465, 1325, 1185 cm^-1; % NMR (500 MHz, CDCl₃) : δ 7.67 (d, J=7.5 Hz, 1 H, aromatic), 7.6-7.14 (m, 8 H, aromatic), 5.84 (dd, J=10.5, 1.6 Hz, 1 H, olefinic), 5.72 (dd, J=10.5 , 1.6 Hz , 1 H, olefinic) , 5.57 (d, J=1.6 Hz , 1 H, N-CH-C≡), 3.85 (d, J=1.6 Hz, 1 H, C≡C-CH-C), 2.49 (m, 1 H, CH₂) , 2.30 (m, 1 H, CH₂) , 2.12-1.60 (m, 4 H, C H₂) ; ¹³C NMR (125 MHZ, CDCl₃) : δ 151.0, 135.5, 129.4, 129.4, 128.2, 127.3, 125.8, 125.8, 125.4, 125.0, 122.0, 122.0, 121.5, 101.8, 94.9, 91.4, 88.8, 70.5, 61.1, 50.0, 29.8, 22.9, 22.5, 15.5; MS: m/e (relative intensity) 393 (20,M⁺), 294(9), 262(15), 212 (11), 149 (42); HRMS: Calcd. for C₂₆H₁₉O₃N (M⁺) : 393.1365, found: 393.1332.

Example 20: Compound 23

A solution of the enediyne Compound 21 (30 mg,

0.076 mmol) and 1,4-cyclohexadiene (0.5 mL) in benzene (2 mL) was treated with TsOH.H₂O (18 mg, 0.09 mmol) and stirred at 25°C for 24 hours. The solvent was removed in vacuo and the residue was purified by flash

chromatography to give the diol Compound 23 (26 mg, 85 percent).

Colorless gum; R_f=0.35 (50 percent ether in petroleum ether); IR (CDCl₃) v_max 3310, 3082, 2925, 1705, 1592, 1395, 1200 cm^-1; ¹H NMR (500 MHz, CDCl₃) : δ 7.58 (bd, J=4.4 Hz, 1 H, aromatic), 7.47 (bd, J=7.8 Hz, 1 H, aromatic), 7.40-7.09 (m, 10 H, aromatic), 6.81 (d, J=8.1 Hz, 1 H, aromatic), 5.78 (s, 1 H, N-benzylic), 4.00 (bs, 2 H, OH), 3.24 (s, 1 H, benzylic), 2.42-0.72 (m, 6 H, CH₂) ; ¹³C NMR (125 MHz, CDCl₃) : δ 151.0, 138.2, 134.7, 129.4, 129.4, 129.3, 128.8, 128.4, 128.3, 127.1, 126.9, 125.7, 125.0, 124.4, 121,8, 121.8, 121.5, 83.0, 66.2, 65.1, 51.2, 33.5, 27.1, 18.7; MS(FAB+): m/e (relative intensity) 546 (15, M+Cs), 379 (31), 312 (30), 286

(100); HRMS: Calcd. for C₂₆H₂₃O₄NCs (M+Cs): 546.0681, found: 546.0691. Example 21: Compound 23a

HCl gas was bubbled through a solution of the enediyne Compound 21 (30 mg, 0.076 mmol) and 1,4-cyclohexadiene (0.5 mL) in CH₂Cl₂ at 25°C for 30 seconds. The solvent was removed in vacuo and the residue was purified by flash chromatography to give the chloride Compound 23a (27 mg, 84 percent).

Pale yellow solid; mp=114-116°C; R_f=0.62 (50 percent ether in petroleum ether); IR (CDCl₃) v_max 3500, 3065, 2932, 1712, 1495, 1382, 1200 cm^-1; ¹H NMR (500 MHz, CDCl₃) : δ 7.71-6.73 (m, 13 H, aromatic), 5.87 (s, 1 H, N-benzylic), 3.62 (s, 1 H, benzylic), 2.52 (bs, 1 H, OH), 2.50-1.60 (m, 6 H, CH₂) ; MS: m/e (relative

intensity) 564 (5, M+Cs), 419 (100), 379 (58); HRMS: Calcd. for C₂₆H₂₂O₃NClCs (M+Cs): 564.0343, found:

564.0351.

Example 22: Compound 24c

Ethyl bromoacetate (27 mg, 0.16 mmol) was added to a mixture of the alcohol Compound 2 (34 mg, 0.08 mmol) and Cs₂CO₃ (67 mg, 0.2 mmol) in anhydrous DMF (2 mL) at 60°C. The mixture was heated for 3 hours, diluted with ether (10 mL), washed with NH₄CO (2 × 3 mL), water (2 × 2 mL) , and brine (2 mL). The organic layer was dried (MgSO₄) and purified by flash

chromatography to give the ester Compound 24c (37.5 mg, 91 percent.

Colorless oil; R_f=0.55 (50 percent ether in petroleum ether); ¹H NMR (500 MHz, CDCl₃) : δ 8.63 (d, J=10.3 Hz, 1 H aromatic), 7.45-7.10 (m, 8 H, aromatic), 5.82 (d, J=10.5 Hz, 1 H, olefinic), 5.68 (dd, J=10.5, 1.5 Hz), 1 H, olefinic), 5.51 [d, J=1.5 Hz, 1 H,

C≡C(C)CHN], 4.34 [d, J=15.6 Hz, 1 H, C(O)CH₂O], 4.28

[d=15.6 HZ, 1 H, C(O)CH₂O], 4.24 (m, 2 H, OCH₂CH₃) , 2.35-1.70 (m, 6 H, CH₂) , 1.30 (5, J*8.2 Hz, 3 H, OCH₂CH₃) ; ¹³C NMR (125 MHz, CDCl₃) : δ 169.5, 150.9, 135.6, 131.0, 129.3, 129.3, 128.1, 127.7, 125.7, 125.6, 123.9, 122.4, 121.6, 121.6, 98.2, 96.3, 94.5, 89.0, 80.1, 73.2, 68.1, 63.2, 61.5, 50.6, 30.0, 29.8, 19.2, 14.5: HRMS calcd. C₃₀H₂₅O₆NCs (M+Cs), 628.0736; found: 628.0750.

Compounds 24a, b, d, e, f and g are similarly prepared.

Example 23: Alternative Preparation of Compound 5

(a) Toluenesulfonic acid (TsOH.H₂O; 8 mg,

0.05 mmol) was added in one portion to a solution of the alcohol Compound 2 (18 mg, 0.045 mmol), 1,4-cyclohexadiene (0.2 mL) and benzene (0.5 mL) at 25°C, and the solution was stirred for 24 hours. The organic solvent was removed in vacuo and the residue was

purified by flash chromatography to give the ketone Compound 5 (15 mg, 90 percent).

(b) Trimethylsilyl trifluoromethylsulfonate (TMSOTf; 15 μL, 0.08 mmol) was added to a solution of the alcohol Compound 2 (32 mg, 0.078 mmol) and

triethylsilane (Et₃SiH) (40 mg, 0.32 mmol) in CH₂Cl₂ (2 mL) at -78°C. After 1 minute, the mixture was quenched at -78°C with saturated NH₄Cl solution (1 mL), diluted with ether (10 mL) , washed with water (2 × 3 mL), brine (3 mL) and then dried (MgSO₄). The organic solvent was removed in vacuo. and the residue was purified by flash chromatography to give the ketone Compound 5 (22 mg, 75 percent).

(c) HCl gas was bubbled through a solution of the alcohol Compound 2 (32 mg, 0.078 mmol) and 1,4-cyclohexadiene (40 mg, 0.32 mmol) in dichloromethane (4 mL) at 25°C for 30 seconds. The solvent was removed in vacuo and the residue was purified by flash

chromatography to give the ketone Compound 5 (25 mg, 80 percent). Example 24: Compound 2b

A solution of the phenyl carbamate Compound 2 (42 mg, 0.103 mmol) in dry methanol (4 mL) was treated with sodium methoxide (17 mg, 0.31 mmol) and heated at 60°C for 2 hours. The reaction mixture was diluted with dichloromethane (25 mL), washed with sodium bicarbonate solution (25 mL), dried (Na₂SO₄), evaporated in vacuo and the residue purified by flash chromatography

(silica, 40 percent ether in petroleum ether) to give methyl carbamate 2b (28.5 mg, 80 percent). 2b: white crystalline solid, mp 126-127°C (from ether/petroleum ether); R_f=0.43 (50 percent ether in petroleum ether); IR (CDCl₃) V_max 3600, 3450, 2957, 2257, 2250, 1706 cm^-1; ¹H NMR (500 MHz, CDCl₃) : δ 8.65 (d, J=8.0 Hz, 1 H, aromatic), 7.25-7.10 (m, 3 H, aromatic), 5.81 (d, J=10.1 Hz, 1 H, olefinic), 5.69 (d, J=10.1 Hz, 1 H, olefinic) , 5.45 (s, 1 H, CH-N), 3.82 (s, 3 H, OMe), 2.79 (s, 1 H, OH), 2.27-1.72 (m, 6 H, CH₂) ; ¹³C NMR (125 MHz, CDCl₃) : δ 135.9, 131.3, 127.8, 127.5, 126.1, 124.9, 123.9, 122.1, 100.7, 94.1, 88.3, 74.2, 73.1, 65.8, 64.2, 53.7, 50.1, 35.2, 23.2, 19.2, 15.2; MS m/e (relative intensity) 347 (M⁺, 100) 291 (35), 204 (50); HRMS calcd. for C₂₁H₁₇NO₄ (M⁺) 347.1158, found 347.1159. Anal, calcd. for C₂₁H₁₇NO₄: C, 72.61; H, 4.93; N, 4.03. Found: C, 72.63; H, 5.24; N, 3.79.

Example 25: Compound 40

Carbamate Compound 21 (39 mg, 0.10 mmol) in THF (3 mL) was treated at zero degrees C with LiAlH₄ (0.25 mL of a 1.0 M solution in ether, 0.25 mmol). After stirring for 30 minutes, the reaction Was quanched with saturated sodium bicarbonate solution (1 mL) , diluted with ether (20 mL), washed with 1.0 M aqueous LiOH solution (2 × 5 mL) in order to remove phenol, dried (Na₂SO₄) , filtered and stored under argon at -78°C until required. MS (FAB+) m/e (relative intensity) 290 (97), 278 (75), 274 (M+H, 98) , 235 (100); HRMS calcd. for C₁₉H₁₆NO (M+H) 274.1232, found 274.1247.

Physical data for further selected compounds:

Compound 42

R_f=0.63 (50 percent diethyl ether in benzene); ¹H NMR (300 MHz, d₈-THF/D₂O, 10:1): 5 8.53 (d, J=8.8 Hz, 1 H, aromatic), 7.45-7.10 (m, 5 H, aromatic), 6.88 (dd, J=2.5 Hz, 1 H, aromatic), 6.63 (dd, J=8.8, 2.5 Hz, 1 H, aromatic), 5.97 (d, J=10.0 Hz, 1 H, olefinic), 5.78 (dd, J=10.0, 1.6Hz, 1 H, olefinic), 5.46 (bs, 1 H, CHN), 2.35-1.55 (m, 6 H, CH₂) . Compound 43

R_f=0.78 (50 percent diethyl ether in benzene); ¹H NMR (300 MHZ, C₆D₆) : δ 8.97 (d, J=9.0 Hz, 1 H,

aromatic), 7.72 (d, J=8.3 Hz, 1 H, aromatic), 7.52 (d, J=7.7 Hz, 1 H, aromatic), 7.13-7.01 (m, 5 H, aromatic), 6.97-6.87 (m, 2 H, aromatic), 6.81 (bd, J=8.9 Hz, 1 H, aromatic), 6.66 (t, J=7.9 Hz, 1 H, aromatic), 5.90 (bs, 1 H, C HN) , 5.31 (d, J=10.1 Hz, 1 H, olefinic), 5.17 (dd, J=10.1, 1.7 Hz, 1 H, olefinic), 5.13 and 5.04 (AB, J=16.0 Hz, 2 H, ArCH₂O), 2.29 (bs, 1 H, OH), 2.15-1.85 (m, 4 H, CH₂) , 1.70-1.60 (m, 1 H, CH₂) , 1.37-1.29 (m, 1 H, CH₂) ; IR (C₆H₆) V_max 3554, 2954, 2927, 1728, 1615, 1579, 1529, 1506, 1494, 1378, 1343, 1302, 1280, 1252, 1239, 1202 cm^-1; HRMS Calcd for C₃₃H₂₅N₂O₇ (M+H): 561.1162, found 561.1162.

Compound 44c

R_f=0.33 (5 percent methanol in methylene chloride); ¹H NMR (300 MHz, CDCl₃):δ 7.43 (dd, J=5.3, 3.5 Hz, 1 H, aromatic), 7.25-7.10 (m, 4 H, aromatic), 6.56 (dd, J=8.4, 2.2 Hz, 1 H, aromatic), 6.53 (d, J=2.2 Hz , 1 H, aromatic, 5.58 (s , 1 H, CHN) , 5.46 (t, J=5. 6 Hz , 1 H, CH₂NHCON) , 3. 19-2 .95 (m, 2 H, CH₂NHCON) , 2.80 (dt, J=11.3 , 6.3 HZ , 1 H, CH₂CH₂NH) , 2.53-2.26 (m, 5 H, CH₂CH₂NH, CH₂, OH) , 1.84 (s , 1 H, OH) , 1. 68 (dd, J=12.4 , 4 .7 HZ , 1 H, CH₂) , 1. 61-1.36 (m, 5 H, CH₂CH₂NHCON,

CH₂CH₂NH, CH₂) , 0.94 (t, J=7.3 Hz , 3 H, CH₃CH₂CH₂NHCON) , 0.89 (dd, J=4.8 , 2.4 Hz , 1 H, CH₂) , 0.83 (t, J=7.3 Hz , 3 H, CH₃CH₂CH₂NH) , 0.78-0.59 (m, 1 H, CH₂) ; IR (CHCl₃) V_max 3591, 3439 , 3270 , 2964 , 2935 , 1645 , 1613 , 1500 , 1459 , 1416 , 1252 , 1188 cm^*1 ; HRMS Calcd. for C₂₆H₃₄N₃O₄

(M+H) : 452.2549 , found: 452.2549.

Compound 45

R_f=0.22 (silica, 70 percent diethyl ether in petroleum ether) ¹H NMR (500 MHz, CDCl₃): δ 7.92-7.1 (m, 9H, aromatic), 5.73 (d, J=10.1 Hz, 1 H, olefinic), 5.63 (d, J=10.1 Hz, 1 H, olefinic), 5.37 (bs, 1 H, NC H) , 4.65-4.22 (m, 2 H, SO₂CH₂CH₂) , 3.73 (S, 1 H, CHCH₂) , 3.48 (m, 2H, SO₂CH₂CH₂), 2.43-1.52 (m, 6 H, CH₂) ; ¹³C NMR (125 MHz, CDCl₃) : δ 134.01, 129.3, 128.5, 128.1, 127.9,

127.1, 125.2, 124.8, 121.9, 101.7, 93.7, 91.2, 88.6, 70.1, 60.9, 59.3, 55.0, 49.4, 29.3, 23.1, 22.3, 15.6; IR (CHCl₃) V_max 2975, 2950, 1715, 1360, 1300, 1150 cm^-1; HRMS Calcd. for C₂₈H₂₃SNCs (M+Cs^⊕) : 618.0351, found: 618.0352.

Compound 48

R_f=0.61 (silica, 70 percent diethyl ether in petroleum ether) ¹H NMR (500 MHz, CDCl₃):5 8.32 (d, J=7.34 Hz, 1 H, aromatic), 7.11 (t, J=7.34 Hz, 1H, aromatic), 6.82 (t, J=7.34 Hz, 1 H, aromatic), 6.55 (d, J-7.34 Hz, 1 H, aromatic), 5.83 (d, J=8.8 Hz, 1 H, olefinic), 5.73 (dd, J=8.8, 1.74 Hz, olefinic), 4.32 (d, J=1.74 Hz, 1 H, NC H) , 4.00 (bs, 1 H, NH), 3.50 (s, 3 H, OC H₃) , 2.36-1.68 (m, 6 H, CH₂) ; ¹³C NMR (125 MH_Z, CDCl₃) : 142.2, 131.0, 128.3, 123.1, 122.5, 122.0, 119.3, 115.9, 100.2, 96.9, 94.6, 87.3, 79.6, 72.9, 62.9, 29.0, 24.6, 19.0; IR (CHCl₃) V_max 3400, 2950, 2850, 1100, 1080 cm^"1; HRMS Calcd. for C₂₀H₁₇O₂N (M^⊕) : 303.1337, found 303.1348. Compound 50

R_f=0.40 (silica, 70 percent diethyl ether in petroleum ether) ¹H NMR (500 MHz, CDCl₃) : δ 7.63 (d, J=7.5 Hz, 1 H, aromatic), 7.35 (m, 2 H, aromatic), 7.18 (d, J=7.0 Hz, 1 H, aromatic), 7.14 (t, J=7.0 Hz, 1 H aromatic), 7.09 (t, J=7.0 Hz, 1 H, aromatic), 7.03 (m, 3 H, aromatic), 6.87 (d, J=7.0 Hz, 1 H, aromatic), 6.80 (t, J=7.0 Hz, 1 H, aromatic), 6.61 (t, J=7.0 Hz, 1 H, aromatic), 6.31 (d, J=7.5 Hz, aromatic), 4.12 (2s, 2H, NH and N-CH), 3.73 (S, 1 H, OH), 3.62 (t, J=2.82 Hz, 1 H, CH-CH₂), 2.80 (ddd, J=12.8, 12.8, 5.64 Hz, 1 H, CH₂) , 2.41 (ddt, J=12.8, 12.8, 4.5 Hz, 1 H, CH₂) , 1.75 (dd, J=13.16, 4.88 HZ, 1 H, CH₂), 1.54 (m, 1 H, CH₂), 1.39 (bd, J=13.16 Hz, 1 H, CH₂) , 0.9 (m, 1 H, CH₂) ; ¹³C NMR (125 MHz, CDCl₃) : 141.6, 139.2, 137.1, 135.1, 133.7, 130.6, 128.1, 127.8, 127.4, 127.0, 126.8, 126.4, 119.9, 115.8, 70.4, 62.1, 55.6, 33.3, 29.8, 28.0, 18.8: IR (CHCl₃) V_max 3450, 3390, 3070, 2930, 2870, 1490, 1470 cm^-1; HRMS Calcd. for C^H^OSN (M+) :385.1500, found

385.1500.

Compound 55

R_f=0.3 (50 percent diethyl ether in petroleum ether); 'H NMR (500 MHz, CDCl₃) : δ 8.21 (d, J=8.8 Hz, 1 H, aromatic), 7.32 (t, J=7.6 Hz, 2 H, aromatic), 7.18 (t, J=7.6 Hz, 1 H, aromatic), 7.10 (bd, J=6.9 Hz, 2 H, aromatic), 6.89 (bs, 1 H, aromatic), 6.65 (dd, J=8.8, 2.7 Hz, 1 H, aromatic), 5.82 (d, J=10 Hz, 1 H,

olefinic), 5.67 (dd, J=10, 1.7 Hz, 1 H, olefinic), 5.48 (S, 1 H, OH), 5.28 (bs, 1 H, CHN) , 3.47 (s, 3H, OCH₃) , 2.28 (dd, J=15.1, 8.2 Hz, 1 H, CH₂) , 2.15 (ro, 2 H, CH₂) , 1.92 (m, 2 H, CH₂) , 1.75 (m, 1 H, CH₂) ; ¹³C NMR (125 MHz, CDCl₃): δ 155.1, 150.9, 136.8, 133.2, 131.0, 130.1,

129.3, 125.8, 124.2, 124.1, 122.2, 121.6, 113.0, 99.5, 94.9, 93.9, 88.4, 79.3, 72.1, 63.2, 52.1, 50.5, 28.5, 23.2, 18.9: IR(CHCl₃) ν_max 3400, 3004, 2979, 2936, 2876,

1719, 1384 cm^-1; HRMS Calcd. for C₂₇H₂₁NO₅Cs (M+Cs):

572.0474, found 572.0429.

Compound 58

R_f=0.31 (diethyl ether); ¹H NMR (500 MHz, CDCl₃) : δ 7.37 (d, J=10.3 Hz, 1 H, olefinic), 6.36 (dd, J=10.4, 2.0 Hz, 1 H, olefinic), 5.85 (d, J=9.8 Hz, 1 H, olefinic), 5.81 (dd, J=1.7 Hz, 1 H, olefinic), 5.10 (d, J=2.0 Hz, 1 H, olefinic), 4.13 (d, J=4.0 Hz, 1 H, NH), 4.07 (d, J=3.0 Hz, H, olefinic), 3.72 (dd, J=4.2, 1.7 Hz, 1 H, CHN), 3.41 (s, 3H, OCH₃) , 3.20 (m, 1 H, CH₂) , 2.36 (m, 1 H, CH₂) , 2.10 (m, 1 H, CH₂) , 1.88 (m, 1 H, CH₂) , 1.74 (m, 1 H, CH₂) ; ¹³C NMR (125 MHz, C₆D₆) : δ

184.4, 157.2, 137.7, 134.8, 123.3, 122.6, 113.3, 98.8, 98.4, 90.7, 87.2, 78.1, 74.8, 74.0, 58.4, 57.8, 51.5,

27.5, 27.4, 14.3; IR(CHCl₃) ν_max 3527, 3385, 2956, 2928, 2855, 1656, 1597 cm^-1; UV (CHCl₃, C=2.2 X 10-4): λ(log ∊) 330 (3.09), 285 (3.37), 256 (3.7), 244 (3.5); HRMS

Calcd. for C₂₀H₁₇O₄NCs (M+Cs): 468.0212, found 468.0254.

Compound 76

R_f = 0.44 (33 percent ethyl ether in petroleum ether); ¹H NMR (300 MHz, CDCl₃) δ 7.68 (dd, J - 7.8, 1.1 Hz, 1 H, aromatic), 7.52 (br s, l H, aromatic), 7.37 (t, J = 7 Hz, 2 H, aromatic), 7.30 (dd, J - 7.4, 1.1 Hz, 1 H, & matic), 7.26-7.13 (m, 8 H, aromatic), 5.61 (s, 1 H, CHN), 3.87 (t, J = 2.8 Hz, 1 H, CH₂CH-) , 2.49 (dd, J - 15.0, 7.8 Hz, 1 H, CH₂) , 2.32-2.19 (m, 1 H, CH₂) , 2.08-1.98 (m, 2 H, CH₂) , 1.90-1.82 (m, 1 H, CH₂) , 1.68-1.55 (m, 1 H, CH₂) ; ¹³C NMR (125 MHz, CDCl₃) δ 150.9, 135.6, 130.0, 129.2, 129.2, 128.9, 128.8, 128.6, 128.3, 128.3, 128.1, 128.1, 127.7, 127.2, 126.8, 125.6, 125.2, 121.5, 121.5, 96.9, 90.7, 90.6, 88.8, 70.1, 60.7, 49.7, 28.9, 22.7, 22.6, 15.5; UV (EtOH) λ_max (log ∊) 282

(3.55), 260 (sh, 3.74), 237-210 (br, 4.36-4.32) nm; HRMS

Calcd. for C₅₀H₂₁NO₃Cs: 576.0576 (M+Cs^⊕), found 576.0611 (M+Cs^⊕).

Compound 87

R_f = 0.52 (33 percent ethyl ether in petroleum ether); ¹H NMR (300 MHz, CDCl₃) δ 7.75 (s, 1 H,

aromatic), 7.73 (s, 1 H, aromatic), 7.72-7.66 (m, 3 H, aromatic), 7.53 (br s, 1 H, aromatic), 7.50-7.42 (m, 2 H, aromatic), 7.38 (t, J = 7.7 Hz, 2 H, aromatic), 7.30-7.13 (m, 5 H, aromatic), 5.64 (s, 1 H, CHN), 3.92 (br s, 1 H, CH₂CH) , 2.53 (dd, J = 15.4, 7.5 Hz, 1 H, CH₂) , 2.34-2.22 (m, 1 H, CH₂), 2.12-2.00 (m, 2 H, CH₂) , 1.91-1.84 (m, 1 H, CH₂) , 1.70-1.60 (m, 1 H, CH₂) ; ¹³C NMR (125 MHz, CDCl₃) δ 150.9, 135.8, 132.3, 131.7, 130.3, 129.3, 129.3, 128.7, 128.4, 128.3, 128.1, 127.7, 127.6, 127.6, 127.3, 127.2, 125.6, 125.2, 123.8, 122.7, 121.5, 96.4, 90.7, 90.5, 88.8, 70.2, 60.8, 49.8, 28.9, 22.7, 22.6, 15.5; HRMS Calcd. for C₃₄H₂₃NO₃Cs: 626.0732 (M+Cs^⊕) , found 626.0732 (M+Cs^⊕) .

Compound 89

R_f = 0.53 (50 percent ethyl ether in benzene); ¹H NMR (300 MHz, CDCl₃) δ 8.40 (s, 1 H, aromatic), 8.23 (s, 1 H, aromatic), 8.22 (s, 1 H, aromatic), 7.99-7.88 (m, 2 H, aromatic), 7.75 (dd, J = 7.7, 1.7 Hz, 1 H, aromatic), 7.54 (s, 1 H, aromatic), 7.50-7.40 (m, 5 H, aromatic), 7.30 (d, J = 7.4 Hz, 3 H, aromatic), 7.14-7.00 (m, 2 H, aromatic), 6.11 (s, 1 H, CHN), 3.62 (br s, 1 H, CH₂CH) , 2.46 (dddd, J = 10.8, 10.8, 3.3, 3.3 Hz, 1 H, CH₂) , 2.30 (ddd, J = 13.6, 13.6, 6.0 Hz, 1 H, CH₂) , 1.94 (dd, J = 13.3, 4.6 Hz, 1 H, CH₂) , 1.53 (br t, J = 13.5 Hz, 2 H, CH₂) , 0.98 (ddddd, J = 13.9, 13.9, 13.9, 4.3, 4.3 Hz, 1 H, CH₂). Compound 115a

Pale yellow oil; R_f= 0.39 (silica, 10 percent methanol in dichloromethane), [a]_D ²⁵ = -87.3° (c = 0.48, CHCl₃), ¹H NMR, (500MHZ, C₆D₆) : δ - 8.97 (dd, 1 H, J = 4.2, 0.7 Hz, Dyn-Ar), 7.52 (m, 1 H, Dyn-Ar), 7.41-6.98 (m, 12 H, 11 Ar, propargylic H) , 6.98 (dd, 1 H, J = 7.1, 7.1 Hz, Dyn-Ar), 5.89 (bs, 1 H, OH), 5.83 (bs, 1 H, O-N-H) , 5.78 (s, 1 H, E-1), 5.28 (d, 1 H, J = 10.0 Hz, vinylic H) , 5.10 (dd, 1 H, J = 10.0, 1.7 Hz, vinylic H), 4.58-4.51 (m, 2 H, CH₂-Ph), 4.50 (d, 1 H, J = 7.4, A-1), 4.48-4.40 (m, 1 H, E-5ax), 4.25-4.17 (m, 1 H, E-5eq), 4.13-4.02 (m, 3 H, A-3, OCH₂CH₂O) , 4.01-3.94 (m, 1 H, OCH₂CH₂O) , 3.93-3.90 (m, 1 H, E-3), 3.77 (dd, 1 H, J = 9.5, 7.3 HZ, A-2), 3.69-3.52 (m, 1 H, A-5), 3.26 (s, 3 H, OCH₃), 2.78-2.66 (m, 2 H, E-4, N-CH₂), 2.65-2.57 (m, 1 H, N-CH₂), 2.47 (dd, 1 H, J = 12.0, 2.2 Hz, E-2eq),

2.44 (dd, 1 H, J = 9.5, 9.5 Hz, A-4), 2.31 (dd, 1 H, J = 14.5, 6.5 Hz, Dyn-CH₂) , 2.04 (d, 1 H, J = 6.7 Hz, Dyn-CH₂) , 1.95-1.83 (m, 4 H, CH₂) , 1.43 (dd, 1 H, J = 14.5, 9.3 Hz, E-2ax), 1.33 (d, 3 H, J = 6.1 Hz, A-6), 1.04 (t, 3 H, J = 6.5 Hz, N-CH₂-CH₃) ; IR (CHCl₃) ν_max = 2965, 2931, 1733, 1380, 1323, 1146, 1098, 1071 cm^*1; HRMS calcd. for C₄₉H₅₅N₃O₁₁ (M+Cs^⊕) 994.2891; found 994.2904.

Compound 115b

Pale yellow oil; R_f = 0.38 (silica, 10 percent methanol in dichloromethane), [a]_D ²⁵ = +125.7° (c= 0.68, CHCl₃) , ¹H NMR (500MHZ, C₆D₆) : δ - 8.93 (dd, 1 H, J = 4.2, 0.7 Hz, Dyn-Ar), 7.56 (dd, J = 7.4, 1.4 Hz, Dyn-Ar), 7.32-7.02 (m, 12 H, 11 Ar, propagylic H) , 6.90 (dd, 1 H, J - 7.1, 7.1 Hz, Dyn-Ar), 5.90 (bs, 1 H, O-N-H), 5.88 (bs, 1 H, OH), 5.82 (s, 1 H, E-1), 5.29 (d, 1 H, J = 10.2 Hz, vinylic H), 5.11 (dd, 1 H, J = 10.2, 1.7 Hz, vinylic H), 4.56-4.51 (m, 2 H, CH₂Ph), 4.49 (dd, 1 H, J = 11.1, 9.0 Hz, E-5ax), 4.48 (d, 1 H, J = 7.4 Hz, A-1), 4.22 -4.17 (m, 2 H, OCH₂CH₂O) , 4.14 (dd, 1 H, J = 11.1,

4.7 Hz, E-5eq), 4.09 (dd, 1 H, J = 9.5, 9.5 Hz, A-3), 4.06-4.01 (m, 1 H, OCH₂CH₂O) , 3.93-3.88 (m, 1 H,

OCH₂CH₂O), 3.87-3.81 (m, 1 H, E-3), 3.78 (dd, 1 H, J = 9.5, 7.1 HZ, A-2), 3.58 (dq, 1 H, J = 9.5, 6.1 Hz, A-5), 3.27 (s, 3 H, OCH₃) , 2.84 (ddd, 1 H, J = 9.0, 9.0, 4.7

HZ, E-4), 2.77 (m, 2 H, N-CH₂), 2.54 (dd, 1 H, J = 12.2, 2.5 Hz, E-2eq), 2.42 (dd, 1 H, J = 9.5, 9.5 Hz, A-4), 2.30 (dd, 1 H, J = 14.6, 10.5 HZ, Dyn-CH₂), 2.06 (dd, 1 H, J = 14.6, 7.1 Hz, Dyn-CH₂), 1.97-1.83 (m, 4 H, Dyn-CH₂), 1.51 (dd, 1 H, J = 12.2, 9.2 Hz, E-2ax), 1.33 (d, 3 H, J = 6.1 HZ, A-6), 1.10 (t, 3 H, J = 6.5 Hz, N-CH₂-CH₃) ; IR (CHCl₃) : ν_max = 2962, 2957, 2929, 1733, 1386, 1323, 1146, 1097, 1070

cm^-1; HRMS calcd. for C₄₉H₅₅N₃O₁₁ (M+Cs^⊕) 994.2891: found 994.2904.

Although the present invention has now been described in terms of certain preferred embodiments, and exemplified with respect thereto, one skilled in the art will readily appreciate that various modifications, changes, omissions and substitutions may be made without departing from the spirit thereof.

Claims (34)

CLAIMS:

1. A fused ring compound corresponding to the structural formula

wherein A is a double or single bond;

R¹ is selected from the group consisting of H, C₁-C₆ alkyl, phenoxycarbonyl, benzyloxycarbonyl, C₁-C₆ alkoxycarbonyl, substituted C₁-C₆ alkoxycarbonyl, and 9-fluorenylmethyloxycarbonyl;

R² is selected from the group consisting of H, carboxyl, hydroxylmethyl and carbonyloxy-C₁-C₆ alkyl;

R³ is selected from the group consisting of H and C₁-C₆ alkoxy;

R⁴ is selected from the group consisting of H, hydroxyl, C₁-C₆ alkoxy, oxyacetic acid, oxyacetic C₁-C₆ hydrocarbyl or benzyl ester, oxyacetic amide, oxyethanol, oxyimidazilthiocarbonyl and C₁-C₆ acyloxy;

R⁶ and R⁷ are each H or together form with the intervening vinylene group form a one, two or three fused aromatic six-membered ring system;

W together with the bonded vinylene group forms a substituted aromatic hydrocarbyl ring system containing 1, 2 or 3 six-membered rings such that said fused ring compound contains 3, 4 or 5 fused rings, all but two of which are aromatic, and in which W is joined [a, b] to the nitrogen-containing ring of the structure shown; and

R⁸ is hydrogen or methyl, with the proviso that R⁸ is hydrogen when W together with the intervening vinylene group is 9,10-dioxoanthra.

2. The fused ring compound according to claim 1 wherein R⁶ and R⁷ are H, or together with the

intervening vinylene group form a benzo or naphtho ring system.

3. The fused ring compound according to claim 1 wherein said substituted aromatic hydrocarbyl ring system W is selected from the group consisting of a substituted or unsubstituted benzo ring, a substituted or unsubstituted naphtho ring and a substituted 9,10-dioxoanthra ring.

4. The fused ring compound according to claim 3 wherein the formed aromatic hydrocarbyl ring system is an otherwise unsubstituted benzo ring.

5. The fused ring compound according to claim

3 wherein the formed aromatic hydrocarbyl ring system is a benzo ring substituted at one or two of the remaining positions by a radical selected from the group

consisting of hydroxyl, C₁-C₆ alkoxy, o-nitrobenzyloxy, benzyloxy, C₁-C₆-acyloxy, oxyethanol, oxyacetic acid and halo.

6. The fused ring compound according to claim 3 wherein the formed aromatic hydrocarbyl ring system is a naphtho ring having a 4-position radical selected from the group consisting of hydroxyl, C₁-C₆ alkoxy, C₁-C₆ acyloxy, benzyloxy and halo, and radicals at the 5- and 8-positions selected from the group consisting of hydroxyl, C₁-C₆ alkoxy, benzyloxy, C₁-C₆ acyloxy, oxo and halo.

7. The fused ring compound according to claim

3 wherein the formed aromatic hydrocarbyl ring system is a 9,10-dioxoanthra ring substituted at one or more of the 4-, 5- and 8-positions by a radical selected from the group consisting of hydroxyl, C₁-C₆ alkoxy, C₁-C₆ acyloxy, and halo.

8. The fused ring compound according to claim 1 wherein A is a single bond.

9. A fused ring compound corresponding in structure to the formula

wherein A is a double or single bond; R¹ is selected from the group consisting of H, C₁-C₆ alkyl, phenoxycarbonyl, benzyloxycarbonyl, C₁-C₆ alkoxycarbonyl, substituted ethoxycarbonyl and 9-fluorenylmethyloxycarbonyl;

R² is selected from the group consisting of H, carboxyl, hydroxylmethyl and carbonyloxy-C₁-C₆ alkyl;

R³ is selected from the group consisting of H and C₁-C₆ alkoxy;

R⁴ is selected from the group consisting of H, hydroxyl, oxyacetic acid, oxyacetic C₁-C₆

hydrocarbyl or benzyl ester, oxyacetic amide,

oxyethanol, oxyimidazilthiocarbonyl and C₁-C₆ acyloxy;

R^s is selected from the group consisting of H, hydroxyl, C₁-C₆ alkoxy, o-nitrobenzyloxy,

benzyloxy, and C₁-C₆ acyloxy;

R⁶ and R⁷ are each H or together form with the intervening vinylene group form a one, two or three fused aromatic six-membered ring system; and

R⁸ is methyl or hydrogen.

10. The fused ring compound according to claim

9 wherein R², R³, R⁵, R⁶ and R⁷ are H.

11. The fused ring compound according to claim

10 wherein R¹ is phenoxycarbonyl,

phenylsulfonylethoxycarbonyl or

naphthylsulfonylethoxycarbonyl.

12. The fused ring compound according to claim

11 wherein R⁴ is selected from the group consisting of H, hydroxyl, C₁-C₆ alkoxy, oxyacetic acid,

imidazylthiocarbonyloxy, oxyacetic amide and oxyacetic C₁-C₆ hydrocarbyl or benzyl esters.

13. A fused ring compound corresponding in structure to the formula

wherein R¹ is selected from the group consisting of H, phenylsulfonylethoxycarbonyl,

naphthylsulfonylethoxycarbonyl, phenoxycarbonyl and benzyloxycarbonyl; and

R⁴ is selected from the group consisting of H, hydroxyl, oxyacetic acid, oxyacetic C₁-C₆ hydrocarbyl or benzyl ester, oxyacetic amide, oxyethanol,

oxyimidazilthiocarbonyl and C₁-C₆ acyloxy.

14. The fused ring compound according to claim

13 wherein R¹ is phenoxycarbonyl,

phenylsulfonylethoxycarbonyl or

naphthlsulfonylethoxycarbonyl.

15. The fused ring compound according to claim

14 wherein R⁴ is selected from the group consisting of H, hydroxyl, and oxyethanol.

16. A fused-ring compound corresponding to the formula

wherein R¹ is selected from the group consisting of H, phenoxycarbonyl, benzyloxycarbonyl, phenylsulfonylethoxycarbonyl and

naphthylsulfonylethoxycabony;

R⁴ is selected from the group consisting of H, hydroxyl, oxyacetic acid, oxyacetic amide, oxyacetic

C₁-C₆ hydrocarbyl or benzyl ester and oxyethanol; and

R⁵ is situated meta or para to the nitogen atom bonded to R¹ and is selected from the group

consisting of hydroxyl, C₁-C₆ alkoxy, benzyloxy, C₁-C₆ acyloxy, oxyethanol, oxyacetic acid, an oxyacetic C₁-C₆ hydrocarbyl or benzyloxy ester and o-nitrobenzyloxy.

17. The fused ring compound according to claim 16 wherein R¹ is phenylsulfonylethoxycarbonyl.

18. The fused ring compound according to claim

17 wherein R⁴ is H.

19. The fused ring compound according to claim 18 wherein R⁵ is hydroxyl or C₁-C₆ acyloxy.

20. A pharmaceutical composition that comprises a DNA-cleaving or cytotoxic amount of a fused ring compound having the structural formula shown below dissolved or dispersed in a physiologically tolerable diluent

wherein A is a double or single bond;

R¹ is selected from the group consisting of H, C₁-C₆ alkyl, phenoxycarbonyl, benzyloxycarbonyl, C₁-C₆ alkoxycarbonyl, substituted C₁-C₆ alkoxycarbonyl, and 9-fluorenylmethyloxycarbonyl;

R² is selected from the group consisting of H, carboxyl, hydroxyImethyl and carbonyloxy-C₁-C₆ alkyl;

R³ is selected from the group consisting of H and C₁-C₆ alkoxy;

R⁴ is selected from the group consisting of H, hydroxyl, C₁-C₆ alkoxy, oxyacetic acid, oxyacetic C₁-C₆ hydrocarbyl or benzyl ester, oxyacetic amide, oxyethanol, oxyimidazilthiocarbonvl and C₁-C₆ acyloxy;

R⁶ and R⁷ are each H r together form with the intervening vinylene group form a one, two or three fused aromatic six-membered ring system;

W together with the bonded vinylene group forms a substituted aromatic hydrocarbyl ring system containing 1, 2 or 3 six-membered rings such that said fused ring compound contains 3, 4 or 5 fused rings, all but two of which are aromatic, and in which W is joined [a, b] to the nitrogen-containing ring of the structure shown; and

R⁸ is hydrogen or methyl, with the proviso that R⁸ is hydrogen when W together with the intervening vinylene group is 9,10-dioxoanthra.

21. The composition according to claim 20 wherein R⁶ and R⁷ are H, or together with the

intervening group form a benzo or naphtho ring system, and R², R³ and R⁸ are H.

22. The composition according to claim 21 wherein said substituted aromatic hydrocarbyl ring system W is selected from the group consisting of a substituted or unsubstituted benzo ring, a substituted or unsubstituted naphtho ring and a substituted

9,10-dioxoanthra ring.

23. The composition according to claim 21 wherein the formed aromatic hydrocarbyl ring system is a benzo ring substituted at one or two of the remaining positions by a radical selected from the group

consisting of hydroxyl, C₁-C₆ alkoxy, benzyloxy, C₁-C₆-acyloxy, oxyethanol, oxyacetic acid and halo.

24. The composition according to claim 21 wherein A is a single bond.

25. The composition according to claim 21 wherein R¹ is pehnoxylcarbonyl,

phenylsulfonylethoxycarbonyl or

naphthylsulfonylethoxycarbonyl.

26. The composition according to claim 25 wherein W is subsituted or unsubstituted benzo.

27. The composition according to claim 26 wherein R⁶ and R⁷ are both H.

28. The composition according to claim 27 wherein said benzo group, W, is substituted meta or para to the nitrogen atom bonded to R¹ with a moiety selected from the group consisting of hydroxyl, C₁-C₆ alkoxy, benzyloxy, C₁-C₆ acyloxy, oxyethanol, oxyacetic acid and an oxyacetic C₁-C₆ hydrocarbyl ester.

29. A chimeric compound comprised of an aglycone portion bonded to (i) an oligosaccharide portion or (ii) a monoclonal antibody or antibody binding site portion thereof that immunoreacts with target tumor cells,

wherein said aglycone poriton is a fused ring compound corresponding to the structural formula

wherein A is a double or single bond; R¹ is selected from the group consisting of H, C₁-C₆ alkyl, phenoxycarbonyl, benzoxycarbonyl, C₁- C₆ alkoxy carbonyl, substituted C₁-C₆ alkoxycarbonyl, and 9-fluorenylmethyloxycarbonyl;

R² is selected from the group consisting of H, carboxyl, hydroxylmethyl and carbonyloxy-C₁-C₆ alkyl;

R³ is selected from the group consisting of H and C₁-C₆ alkoxy;

R⁴ is selected from the group consisting of H, hydroxyl, C₁-C₆ alkoxy, oxyacetic acid, oxyacetic C₁-C₆ hydrocarbyl or benzyl ester, oxyacetic amide, oxyethanol, oxyimidazilthiocarbonyl and C₁-C₆ acyloxy;

R⁶ and R⁷ are each H or together form with the intervening vinylene group form a one, two or three fused aromatic six-membered ring system;

W together with the bonded vinylene group forms a substituted aromatic hydrocarbyl ring system containing 1, 2 or 3 six-membered rings such that said fused ring compound contains 3, 4 or 5 fused rings, all but two of which are aromatic, and in which W is joined [a, b] to the nitrogen-containing ring of the structure shown; and

R⁸ is hydrogen or methyl, with the proviso that R⁸ is hydrogen when W together with the intervening vinylene group is 9,10-dioxoanthra;

said oligosaccharide portion comprising a sugar moiety selected from the group consisting of ribosyl, deoxyribosyl, fucosyl, glucosyl, galactosyl,

N-acetylglucosaminyl, N-acetylgalactasaminyl, a

saccharide whose structure is shown below, wherein a wavy line adjacent a bond indicates the position of linkage said monoclonal antibody or combining site portion thereof being bonded to said fused ring compound aglycone portion through an R⁴ oxyacetic acid amide or ester bond or an oxyacetic acid amide or ester bond from W, and said oligosaccharide portion being glycosidically bonded to the aglycone portion through the hydroxyl of an R⁴ oxyethanol group or the hydroxyl of an oxyethanol- substituted W.

30. The chimeric compound according to claim

29 wherein A is a single bond, R², R³, R⁶, R⁷ and R⁸ are hydrogen, and W is benzo.

31. The chimeric compound according to claim 30 wherein said aglycone portion is bonded to said oligosaccharide portion.

32. The chimeric compound according to claim 31 wherein the glycosidic bond between the aglycone and oligosaccharide portions is formed form an R⁴ oxyethanol group.

33. The chimeric compound according to claim

31 wherein R¹ is phenoxycarbonyl,

phenylsulfonylethoxycarbonyl or

naphthylsulfonylethoxycarbonyl.

34. The chimeric compound according to claim

32 wherein said oligosaccharide portion is selected from the group of oligosaccharides shown.