Year

2010

Degree Name

Doctor of Philosophy

Department

School of Chemistry

Abstract

The main aim of this project was to develop a new synthetic methodology towards the total synthesis of the Stemona alkaloids oxyprotostemonine 10 and 1-hydroxyprotostemonine 11 as shown in Figure 1.
(See Thesis for figure)
Figure 1. Structures of oxyprotostemonine 10 and 1-hydroxyprotostemonine 11 and synthetic intermediates 47 and 244.

We first aimed to synthesize the tricyclic compound 244 and then the tetracyclic ether bridged compound 47. Compounds 47 and 244 are the enantiomers of compounds which would be required for the synthesis of the natural products 10 and 11. As part of our model study in Chapter 2 we report our efforts to develop a general method for the preparing cis- and trans- 4-hydroxy- or 4-alkyloxy-5-alkynylpyrrolidin-2-ones in a diastereoselective fashion. Two different synthetic approaches were examined: a) The reaction of a 4-hydroxy-5-methoxy or a 5-acetoxy-4-alkyloxy-pyrrolidin-2-one with an in situ formed zinc or copper alkynilides or a potassium alkynyltrifluoroborates; b) As an alternative method we examined the addition of lithiated terminal alkynes to a pyrrolidine-2,5-dione followed by reduction of the intermediate alcohol with NaCNBH3 or BF3.OEt2/Et3SiH. The borono-Mannich reaction of N-benzyl-5-methoxy-4-hydroxypyrrolidin-2-one with two potassium alkynyltrifluoroborates gave cis adducts with high cis selectivity but in relatively low yields due to some side reactions. The same reactions with 5-acetoxy-4-alkyloxypyrrolidin-2-ones gave good yield of adducts with high trans selectivity. We then developed an alternative method which involved the addition of lithiated terminal alkynes to a pyrrolidine-2,5-dione followed by reduction of the intermediate alcohol with NaCNBH3 or BF3.OEt2/Et3SiH. The use of NaCNBH3 as the reductant gave the desired products in moderate yields and with moderate selectivity. However when we used BF3.OEt2/Et3SiH as the reducing agent we obtained high yields and high trans selectivities. We also discovered a novel Meyer-Schuster rearrangement the enal byproduct 125.

In Chapter 3 we describe synthesis of the tricyclic compound 244 and the tetracyclic compound 47. The key intermediate, the enyne 214was prepared from the L-malic acid 96 in eight synthetic steps (Scheme 1) using some of the chemistry developed in Chapter 2. A RCEYM reaction of 214 gave the pyrrolo [1,2-a]azepine 215 in good yield (82%). Bicyclic compound 215 was coverted in highly diastereoselective fashion, by first 1,4 reduction of the enoate moiety and then a osmium-catalyzed DH reaction, to the tricyclic compound 241. Dehydration of 241 followed by reduction of the resulting a,B-unsaturated lactone 242 gave the desired tricyclic compound 244. The tricyclic compound 244 successfully synthesized in 13 steps from L-malic acid in overall yield of 1.7%. Finally, we attempted the synthesis of the tetracyclic compound 47 (Figure 1). Compound 244 was irradiated at 40 W in the presence of PhI(OAc)2/I2 in CH3CN at rt. Unfortunately only unreacted 244 was recovered. Time constraints did not allow this reaction to be examined any further.
(See thesis for figure)

Scheme 1. Synthesis of 244 from L-malic acid 96.

In Chapter 4 we report the results of our study to enhance the yield of the novel Meyer-Schuster rearrangement product which we reported in Chapter 2 as a byproduct. We obtained the Meyer-Schuster rearrangement products 125 and 276 in high yields and as 100% the E-isomers from the reactions of alcohols 274 and 275 with the Lewis acid, BF3.OEt2. This reaction could not be efficiently extended to the corresponding phenylalkyne analogue. The synthesis of two new trihydroxylated pyrrolidines 51 and 52 in a highly diastereoselective manner, has been developed using the Meyer-Schuster rearrangement as a key step. The trihydroxylated pyrrolidine 51 was obtained as its hydrochloride salt in eight synthetic steps overall in 23.3% yield, while the hydrochloride salt of 52 was obtained in seven synthetic steps and 18.3% overall yield.
(See thesis for figure)

Scheme 2. Synthesis of pyrrolidin-2-ones 51 and 52.

This methodology could, in principle, be extended to the synthesis of other natural and unnatural polyhydroxylated pyrollizidines and indolizidines. While doing a model study on the Meyer-Schuster rearrangement of the internal propargyl alcohol 122 using TiCl4 we discovered the novel rearranged product 277 in good yield. We also studied the nucleophilic addition reactions of diol 301 in the presence of BF3.OEt2 and obtained addition product 318 in moderate yield. In contrast the other potential C-nucleophiles, allyltrimethylsilane, 2-furanboronic acid, cyanotrimethylsilane and styrenylboronic acid failed to react with 301. We also examined the Ritter reaction with 301 in the presence of BF3.OEt2/CH3CN which gave 323 as a single isomer in good yield.
(See thesis for figure)

Scheme 3. Synthesis of 277, 318 and 323.

In Chapter 5 we describe the metal catalyzed cycloisomerization reactions of cis-115 and cis-116 to successfully prepare the furo[3,2-b]pyrroles 347 and 353 in good yields. AgNO3 proved to be the most effective catalyst for these cycloisomerization reactions in terms of substrate versatility, rate of reaction. We also discovered a novel CuI mediated cyclization-iodination byproduct 348 from the reaction of 115 with CuI (150 mol%) in DMF.
(See thesis for figure)

Scheme 4. Cycloisomerization of cis-pyrrolidin-2-one 53.

In Chapter 6 we report the results of our study to enhance the yield of the novel cyclization-iodination byproduct 348 from the reaction of 115 with CuI (150 mol%) in DMF at 68 – 70 oC and to extend this method to other CuX salts (X = Cl, Br, CN) to provide more usefully functionalized furo[3,2-b]pyrroles 348 and 354-356. We have successfully developed and general, direct and convenient method for the cyclization-halogenation and cyclization-cyanation reactions of cis-4-hydroxy-5-phenylethynylpyrrolidinones 115 and 372 and ortho-alkynyl phenols and ortho-alkynyl phenylacetate derivatives to give desired cyclization-halogenated and cyclization-cyanated products.
(See thesis for figure)

Scheme 5. CuX mediated cyclization-halogenation and cyclization-cyanation reactions of 115, 372 and 57.

This method showed good tolerance to electron-donating and electron withdrawing groups on the substrates. This method allows for the synthesis of 3-cyano-benzofurans in a one step process that otherwise would require two sequential steps from the same starting substrate (iodonium ion induced cyclization followed by a classical Rosenmund-von Braun reaction with CuCN). This method is also more cost effective when one compares the relatively higher costs per mole of iodine and NIS with the less expensive CuCN. Finally we attempted a cyclization-iodination of the ortho-phenylethynyl acetate 57 (R = H, R1 = H, R2) using a catalytic amount of CuI which gave 49:51 mixture of 58 (X = I) and 58 (X = H).

Share

COinS