Advances in exploring alternative Taxol sources

Abstract

The protection and sustainable utilization of natural resources are among the most pressing global problems of the 21st century. Taxol is a well-known anticancer agent with a unique active mechanism, mainly isolated from the slow-growing yew tree, which has achieved the status of “blockbuster drug” due to its popularity in the treatment of cancers. However, the source of Taxol has always been a primary concern, because its content in the plant is extremely low. In this review, we introduce the advances in exploring alternative Taxol sources aside from total synthesis, including Taxol from nursery cultivated Taxus, semi-synthesis of Taxol, Taxol from Taxus cell culture, production of Taxol by synthetic biology, Taxol from the endophytic fungi and Taxol from other non-Taxus plants. The future perspective of Taxol production is also discussed.

---

Introduction

With the approval of vincristine (1) and vinblastine (2) for the treatment of cancers in the 1960s (Fig. 1), natural products have emerged both as treatments and as leads for further drug development. Taxoids are a group of cyclic diterpenoids containing the taxadiene skeleton. These compounds have been extensively studied because of their anticancer activity, of which Taxol (generic name: paclitaxel)-(3) is the most well-known anticancer drug to date. In fact, no natural product discovered in the last decades has generated as much public interest as Taxol.

Fig. 1 Structures of compound 1–8.

The discovery and development of Taxol as an effective anticancer drug has been comprehensively summarized elsewhere.^1–3 Unlike some other tubulin-binding anticancer drugs, such as vincristine (1) and vinblastine (2), which prevent tubulin from assembling into microtubules, Taxol promotes tubulin assembly into microtubules and prevents their disassembly.^4,5 Taxol was approved for the treatment of refractory ovarian cancer by the FDA on December 29th 1992, and Bristol-Myers-Squibb (BMS, New York, NY, USA) trademarked the name Taxol® for what soon became one of the hottest cancer-fighting drugs in the world and paclitaxel was then used as a generic name. Subsequently, Taxol as well as another semisynthetic Docetaxel (Taxotere®)-(4) were approved by the FDA for other indications, such as breast cancer, squamous cancers of the bladder, prostate, head and neck, as well as for non-small cell lung cancer, small-cell lung cancer, AIDS-related Kaposi's sarcoma.^6–8 Taxol is also used for the treatment of diseases unrelated to cancer but requiring microtubule stabilization and inhibition of cell proliferation or angiogenesis.⁹ The indications include psoriasis,¹⁰ rheumatoid arthritis^11,12 and restenosis in which Taxol is used as the coronary stent coat.^13,14 Additionally, Taxol is currently being studied for the treatment of taupathies (affections in tau proteins), such as frontotemporal dementia and parkinsonism linked to chromosome 17 (FTDP-17).¹⁵ Although Taxol and its semisynthetic analogue Docetaxel have been successful in the clinic, the insolubility, selectivity, multi-drug resistance (MDR) and inability to penetrate the central nervous system or the blood–brain barrier have been the significant concerns, in which the MDR is mainly due to the overexpression of efflux transporters such as P-glycoprotein (P-gp).^16–18 To overcome the insolubility, increase the selectivity and efficacy as well as central nervous system penetrability, decrease the toxicity and alleviate the development of drug resistance, at least nine new taxoid derivatives or formulations have undergone clinical trials in the last decade, including Nab-paclitaxel (Abraxane®, approved by FDA in 2005), Cabazitaxel (Jevtana®, approved by FDA in 2010)-(5), Larotaxel (6) and the combination of Cabazitaxel with Prednisone (Rayos® or Sterapred®)-(7). For example, Nab-paclitaxel was the first FDA approved taxoid formulation based on nano-delivery systems, and a number of other novel Taxol nano-particle formulations are in clinical trials, including NK 105, Genexol-PM, Tocosol, ANG 1005, EndoTag, Paxceed and OncoGel.¹⁹ In addition to overcoming the insolubility of Taxol, some of these formulations have shown certain advantages in terms of toxicity and MDR in tumour cells.^21,22 Cabazitaxel is a semisynthetic dimethyloxy derivative of Docetaxel engineered to potentially have clinical and pharmacokinetic advantages over its precursor Docetaxel. The structural changes cause alteration of the P-gp affinity characteristic of Docetaxel, which is thought to be responsible in part for the development of resistance to Docetaxel and other taxanes. It has been reported that the biological action of Cabazitaxel is better than that of Docetaxel.²⁰ Larotaxel is a taxane analogue with a broad spectrum of activity and different toxicity profile, and with the advantages of bypassing some mechanisms of MDR and penetrating into the CNS.²¹ Larotaxel is minimally recognized by P-gp, which can explain, at least in part, its effectiveness in taxane-resistant patients. Several clinical trials confirmed the superiority of combination therapy versus monotherapy when the drugs chosen have non-overlapping resistant mechanisms and are effective as single agents.^22,23 Cabazitaxel is used in combination with Prednisone (Rayos® or Sterapred®) in the treatment of hormone-refractory metastatic prostate cancer.²⁴

Taxol is mainly produced by the genus Taxus (family Taxaceae). This genus includes the following species: T. brevifolia (Pacific yew or Western yew), T. chinensis (Chinese yew), T. floridana (Florida yew), T. globosa (Mexican yew), T. wallichiana (Himalayan yew), T. baccata (European or English yew), T. canadensis (Canadian yew) and T. cuspidata (Japanese yew), as well as also two recognized hybrids: Taxus × media = T. baccata × T. cuspidata and Taxus × hunnewelliana = T. cuspidata × T. canadensis. The concentration of Taxol in these trees is extremely low, ranging from zero to 0.069% and the highest content is found in the bark. Indeed, 10 000 kg of Taxus bark or about 3000 yew trees are necessary to produce only 1 kg of the drug and, a 100-year-old tree might yield 3 kg of bark, which provides enough paclitaxel for one 300 mg dose. A cancer patient usually requires approximately 2.5–3 g of Taxol,²⁵ so, the treatment of each patient consumes about eight 100-year-old yew trees. Accordingly, it is not realistic to supply Taxol by extraction from the wild natural resources, due to its extremely low concentration, slow growth of yew trees and the increased demand for the drug. Therefore, since the early stage of development, how to meet the clinical requirement has become a big challenge, and more attention has been paid to the alternative sources. Total synthesis of Taxol was achieved by different labs in 1990's, but the very low yields, complex multi-step reactions, costly chemical reagents, and various toxic side products do not favor it as a practical or economical way for the supply of Taxol. The present review focuses on recent advances in exploring alternative Taxol sources except the total synthesis, including Taxol from nursery cultivated Taxus, semi-synthesis of Taxol, Taxol from Taxus cell culture, production of Taxol by synthetic biology, Taxol from the endophytic fungi and Taxol from other non-Taxus plants, in which the first two approaches have become the major Taxol sources for the clinical supply.

Taxol from nursery cultivated Taxus

The destructive collection of the bark for Taxol threats the wild resources. To meet the increased demand for the drug and preserve the wild Taxus species, the Canadian Forest Service Atlantic Forestry Centre has been engaged in a program for developing ecologically sustainable harvesting protocols of yews in their natural state, by converting elite cultivars of the wild species into a commercially reared crop since 1997. Almost simultaneously, China introduced 20 000 seedlings of T. × media (a kind of shrub) from the TPL Phytogen Inc. (Vancouver, Canada) and, set up Bei-chuan and Hong-ya yew seedling bases located in Sichuan province. Eventually, the amplified seedlings from these two seedling bases were introduced to other provinces of China for plantation. In addition, the species T. chinensis was also commercially cultivated in China. Up to now, more than 150 Taxus forest farms have been established in China and some of them can provide the suitably active pharmaceutical ingredients.²⁶ Generally the twigs and/or needles are harvested for the extraction of Taxol or its precursor 10-deacetylbaccatin III (10-DAB)-(8) to keep the plant alive. This method not only avoids the destruction of the wild Taxus resource, but also permits its sustainable utilization, and it is currently considered as one of the most effective approaches to obtain Taxol and its chemical semi-synthetic precursors from Taxus in China.

Semi-synthesis of Taxol

In 1994, BMS developed a semi-synthetic process to synthesize Taxol from 10-DAB; simultaneously, the accelerated program of bark collection initially set up as a solution to the supply crisis, came to an end. The program was carried out by Hauser Chemical Research in the Pacific Northwest under contract to BMS. In fact, the semi-synthesis route to obtain Taxol from 10-DAB was first reported by Dr Denis in 1988.²⁷ The precursor 10-DAB, isolated in reasonably good yields from the needles of T. baccata in 1981 and from the bark of T. brevifolia in 1982, is a taxane diterpenoid which co-occurred with Taxol in reasonably good yield (ca. 1 g kg⁻¹, 0.1% dry weight) in the needles of T. baccata and other yew trees. This compound serves as starting material for the semi-synthesis of Taxol via a coupling reaction with a protected C-13 side chain. Since the concentration of 10-DAB is much higher than that of Taxol and besides, 10-DAB can be obtained from nursery cultivated yew trees, its conversion to Taxol is an attractive option. Currently, the semi-synthesis of Taxol from 10-DAB is one of the main approaches for the commercial supply of Taxol. BMS, a leading global supplier of Taxol, has a farm with 30 billion yews to supply the leaves necessary for the extraction of 10-DAB. In addition, in 2007, Indena developed another semi-synthetic approach to produce Taxol in Europe by a patented process based on 10-DAB, which is extracted from T. baccata trees cultivated in the company plantations. Starting from this compounds, not only Taxol but also analogs such as Docetaxel were synthesized in adequate yields.

The basic requirement to semi-synthesize Taxol from 10-DAB is to prepare the C-13 side chain and merge it to 10-DAB, which has been excellently reviewed by Borah et al.,²⁸ and it mainly includes asymmetric epoxidation routes, routes involving Sharpless asymmetric dihydroxylation, chiral auxiliary based strategies, inverse electron demand Diels–Alder reaction, enol–imine condensation, through β-lactam, chemoenzymatic synthesis and using asymmetric catalyst. Among these, the approaches through the β-lactam route and the one using Sharpless asymmetric dihydroxylation (ADH) (as shown in Fig. 2), galactose imine 9 and acid chloride 10 in the presence of Et₃N produce 11 as single diastereomer. The removal of p-methoxyphenyl (PMP) substituent then yields compound 12. The acetyl group can be selectively removed and the resulting alcohol can be protected with TES protection to form compound 13, which are considered to be relatively shorter and efficient than the others. BMS employed the β-lactam coupling method patented by Holton in Florida State University (U.S. Patents 5,229,526; U.S. Patents 5,274,124). Specifically, the protected β-lactam is coupled with the lithium alkoxide of the protected 10-DAB, and the resulting product is deprotected to yield Taxol. As Fig. 2 shows, through TES protection of the C-7 acylation and the C-10 of 10-DAB (8), 7-TES-10-DAB (14) and 7-TES-baccatin III (15) are obtained, respectively. Then, addition of n-butyllithium to a solution of compound 15 in THF at −45 °C produces alkoxide (16), which is followed by the addition of the protected β-lactam (17), and the warming of the mixture to 0 °C for 2 h to produce TES-protected Taxol (18) in nearly quantitative yield. Deprotection of the TES-protected Taxol is carried out using HF/pyridine in acetonitrile solution to produce Taxol at a yield of >98% (U.S. Patent 5,243,045). Actually, this was the method used by Wani to determine the structure of Taxol, as well as by Holton for total synthesis of Taxol. Temperature (−45 °C) and alkali (n-butyllithium) are important factors for the production of alkoxide. For instance, at −75 °C, no alkoxide is produced. Additionally, it was reported that using NaHMDS was more efficient than n-butyllithium (96% vs. 91%); while degradation of alkoxide was observed when the yield was over 90%.²⁹ The ADH reaction, also called the Sharpless bishydroxylation, is the chemical reaction of an alkene with osmium tetroxide in the presence of a chiral quinine ligand to form a vicinal diol. It is usually performed using a catalytic amount of osmium tetroxide, which is regenerated after the reaction with either potassium ferricyanide or N-methylmorpholine N-oxide. This method drastically reduces the amount of the highly toxic and high-priced osmium tetroxide required. The semi-synthetic approach starting from 10-DAB is a real breakthrough in the history of Taxol production.

Fig. 2 Semi-synthesis of Taxol from 10-DAB.

The 7-β-xylosyl-10-deacetyltaxol (XDT)-(19) compound is an analogue of Taxol that can be obtained with yield of as much as 0.5% from the dried bark of the yew tree (European Patent, EP0 ,905,130 B1). If the xylosyl moiety at C-7 is removed, as indicated in Fig. 3, by chemical or biological hydrolysis, the resultant 10-deacetyltaxol (DT)-(20) can also be used in the semi-synthesis of Taxol by acetylation at C-10 position (U.S. Patent 6,028,206).^30,31 In the semi-synthesis of Taxol from DT, the addition of TESCl (in imidazole) to DT in DMSO solution produces TES-protected DT (21), which is followed by the addition of acetic anhydride (22), and the warming of the mixture to generate TES-protected Taxol (23). Deprotection of TES-protected Taxol is carried out using HF/pyridine in acetonitrile solution to produce Taxol. Compared to 10-DAB, DT contains a C-13 side chain (Fig. 2 and 3), and it is easier to synthesize Taxol from DT than from 10-DAB. To realize the commercial Taxol production from DT, efficient conversion of XDT to DT is required. In contrast to the chemical approach, which utilizes periodate or other oxidizing agents and a substituted hydrazine in the reactions to remove the sugar, the biological approach is an enzymatic process that releases the D-xylose from XDT through the specific β-xylosidase, accordingly, it is considered to be environmentally friendly. Some bacterial isolates, such as Moraxella sp. (ATCC 55,475) (U.S. Patent 5,700,669A), Cellulosimicrobium cellulans XZ-5 (CCTCC no. M207,130) (PCT/CN2,008/000,618), and Enterobacter sp. (CGMCC 2,487),³² have been reported to have the ability to convert XDT to DT. However, these strains generated low yields of DT (ranged from 0.23–0.76 mg mL⁻¹),^32,33 which is probably due to the ubiquitously low enzyme levels in the native organisms. Recently, we have discovered and cloned the specific β-xylosidase genes from the fungus of Lentinula edodes, and the genes have been introduced into the yeast host Pichia pastoris.³¹ The recombinant yeast harboring the Lxyl-p1-2 gene could efficiently convert XDT into DT, with the yield of ∼11 g L⁻¹ DT against 15 g L⁻¹ XDT in a 1 L reaction volume, and the reaction volume was scaled up to 10 L.^34,35 Moreover, the engineered yeast has been subjected to the high-cell-density fermentation from the pilot scale to (10- to 100 L) to the demonstration/commercial scale (1000 L) and, the best process was obtained from the 1000 L demonstration/commercial scale.³⁶

Fig. 3 Semi-synthesis of Taxol from XDT.

Taxol from Taxus cell culture

Taxus cell culture has been considered as a promising means for Taxol production. This approach offers several advantages, such as not being subjected to weather and season or contamination; providing a continuous and uniform quality Taxol; a renewable and environmentally friendly resource, and a material which is being grown independently of its original, potentially remote, location. The first publication for Taxus cell culture using T. brevifolia was reported in 1989,³⁷ which was patented two years later (U.S. Patent 5,019,504). The level of Taxol was at the range of 1–3 mg L⁻¹ in the culture.³⁸ Taxol production in Taxus cell culture has been conducted with cells from many species, including T. brevifolia, T. baccata, T. cuspidate, T. chinensis, T. canadensis, T. yunnanensis, T. × media (across of T. baccata and T. cuspidata) and, more recently T. globosa. Taxol yields were enhanced by selection of high-yield cell lines, use of a two-stage culture system, optimization of carbohydrate source, using phytohormones, precursor feeding strategy, addition of adsorbants or additives, in situ product removal, immobilization, and sugars feeding strategy. Alternatively, using elicitors such as fungal culture, fungal extract plus salicylic acid, vanadyl sulfate, chitosan, methyl jasmonate, squalestatin and methyl jasmonate plus ethylene.^39–42 Also the yields were improved by reduction of lipid peroxidation of La³⁺ and implementing combined strategies, including combination of elicitation with precursor, or removing the secondary metabolites from the culture medium with elicitation. Using a two-stage process, the production of Taxol reached a maximum level of 295 mg L⁻¹ in a large-scale culture of T. × media cells.⁴³ In fact, it was reported that the cell culture systems had been successfully developed for the commercial production of Taxol in two companies, Phyton Biotech (USA) and Samyang Genex (South Korea). However, the details of the actual commercial production are not available. Nevertheless, some bottlenecks of this technology have to be resolved in the future, including the slow growth and low titres, empiricism on the (secondary) metabolism; difficulty in using stress factors and elicitors, cell aggregation and sensitivity to shearing, the high production costs and unstable product yields. Furthermore, the conditions required for each plant species are different. In particular, difficulties associated with culturing dedifferentiated plant cells (DDCs) on an industrial scale precluded the commercial viability.⁴⁴ Additionally, the nutrient manipulation and gas composition are particularly important factors for large-scale culture. In 2010, Lee et al. reported a kind of plant stem cells for the production of natural products by cell culture. They isolated and cultured innately undifferentiated cambial meristematic cells (CMCs) and identified marker genes and transcriptional programs consistent with a stem cell identity. Suspension culture of CMCs derived from T. cuspidate circumvented obstacles routinely associated with the commercial growth of DDCs. These cells may provide a cost-effective and environmentally friendly platform for sustainable production of Taxol.⁴⁵

Production of Taxol by synthetic biology or metabolic engineering

The biosynthetic pathway of Taxol is thought to contain at least nineteen steps, and great review have been contributed by experts on the subject.^40,46,47 The first committed step is the cyclization of the universal diterpenoid precursor geranylgeranyl diphosphate (GGPP) to taxa-4(5),11(12)-diene (taxadiene) catalyzed by taxadiene synthase (TS). This parental olefin is then functionalized by a series of eight cytochrome P450 (CytP450)-mediated oxygenations, three CoA-dependent acylations, and several other transformations en route to baccatin III, to which the side chain at C-13 is attached to generate Taxol as the final product (Fig. 4). Although certain catalytic enzymes have yet to be fully characterized, mainly including C1β-hydroxylase, C4, C20-epoxydase, C9α-hydroxylase, C9-oxidation, β-phenyalalanoyl CoA ligase and C2′ side chain hydroxylase, adequate information is available about the properties of the recombinant enzymes to attempt reconstructing early steps of the Taxol biosynthetic pathway in a microbial host.

Fig. 4 Biosynthetic pathway of Taxol.

Some steps of the Taxol biosynthetic pathway have been introduced into heterologous expression systems, including Escherichia coli, Saccharomyces cerevisiae and plants. For instance, Croteau's group first reported the synthesis of taxadiene in E. coli.⁴⁸ They overexpressed the genes encoding isopentenyl diphosphate isomerase (IPP isomerase), GGPP synthase (GGPPS) and TS in a single strain of E. coli, and obtained 1.3 mg taxadiene per L of culture under un-optimized conditions. Unfortunately, further development of the pathway was constrained due to the difficulties of expressing the CytP450 in microbial systems. The incorrect folding, translation and insertion of the CytP450's into the cell membrane may often lead to the loss of their functionality. Furthermore, problems with cofactor availability, together with the absence of the specific CytP450 reductases required to enzymatically recycle each CytP450 represent significant hurdles. We constructed a metabolic pathway for producing taxadiene in S. cerevisiae. The genes encoding the 3-hydroxyl-3-methylglutaryl-coenzyme A (HMG-CoA) reductase (HMGR), GGPPS, and TS were transformed into and expressed in S. cerevisiae. The synthesis of taxadiene in yeast cells was successfully characterized by gas chromatography mass spectrometry.⁴⁹ In addition, Croteau's group also reconstructed an early pathway leading from primary metabolism to taxadien-5α-acetoxy-10β-ol in S. cerevisiae. The genes encoding GGPPS, TS, CytP450 taxadiene 5α-hydroxylase (T5αH), taxadien-5α-ol-O-acetyl transferase (TAT) and CytP450 taxoid 10β-hydroxylase (T10βH) were transformed into a single yeast host. Taxadiene was readily measurable with production level of 1.3 mg L⁻¹ of culture, but a pathway restriction was encountered at the first CytP450 hydroxylation step with very low production level of ∼25 μg taxadien-5α-ol per L of culture.⁵⁰ The pathway was restricted at the first CytP450 hydroxylation. This CytP450 hydroxylation step may require co-expression with its cognate CytP450 reductase for efficient function. Until now, the highest production level of taxadiene has been achieved by Ajikumar et al. in E. coli using a multivariate-modular approach.⁵¹ They divided the taxadiene metabolic pathway into two modules, specifically a native upstream methylerythritol-phosphate (MEP) module forming isopentenyl pyrophosphate (IPP) and a heterologous downstream terpenoid-forming module. Under fed-batch cultivation, the optimized yield of taxadiene was up to 1 g L⁻¹ in 1 L scale fermenter. They also engineered the next step of a CytP450-mediated 5α-oxidation of taxadiene to taxadien-5α-ol, with a yield of 58 mg taxadien-5α-ol per L (∼116 mg L⁻¹ of two oxygenated products).⁵¹ Recently, the highest titre of the total oxygenated taxanes has been up to ∼570 mg L⁻¹ in E. coli by optimizing P450 expression, reductase partner interactions and N-terminal modifications of the enzymes.⁵²

Several plant hosts have also been exploited to produce Taxol intermediates. The overexpression of GGPPS and TS in Arabidopsis thaliana yielded 600 ng g⁻¹ taxadiene (dry weight). However, the introduction of these exogenous genes resulted in growth retardation and decreased levels of photosynthetic pigment in transgenic plants due to the accumulation of taxadiene.⁵³ The lower accumulation of endogenous plastid isoprenoid products such as carotenoids and chlorophylls in transgenic plants also suggests that the constitutive production of an active TS might alter the balance of the GGPP pool, although these phenotypes may be derived from a toxic effect of taxadiene. The TS was also overexpressed in a yellow-fruited tomato line (Yellow flesh mutant) that lacked the ability to utilize GGPP for carotenoid synthesis, leading to a yield of 160 mg taxadiene per kg of freeze dried fruit.⁵⁴ Additionally, the TS gene from T. brevifolia was constitutively expressed in the moss Physcomitrella patens with a production of taxadiene up to 0.05% of fresh weight of tissue, without significantly affecting the amounts of the endogenous diterpenoids.⁵⁵

There seems to be an obvious hurdle in engineering such a lengthy and complex pathway in its entirety in heterologous hosts, as evident from the Taxol biosynthetic pathway of Taxus sp. The transformation of genes of the entire pathway is a challenge and even more importantly, regulation of Taxol production involving epigenetic modulation and signaling crosstalk itself remains a poorly understood topic. To date, except for a few intermediates mentioned above, there is no report on obtaining the final product Taxol through synthetic biology or metabolic engineering. Because the Taxol biosynthetic pathway is highly complex and our understanding on the biosynthetic mechanisms is still incomplete, the fulfilment of this ultimate objective presents many challenges.

Taxol from the endophytic fungi

In 1993, it was first reported that an endophytic fungus, Taxomyces andreanae, could produce Taxol under in vitro axenic culture conditions, albeit the variable yield of Taxol was only as low as 24–50 ng L⁻¹.⁵⁶ This fungus was isolated from the bark of T. brevifolia. The discovery was projected as the dawn of a new era of endophyte biotechnology considering the billions of dollar's worth of global market for Taxol already in place. Agreements were immediately underway among leading pharmaceutical companies, for example, Cytoclonal Pharmaceutics, Inc. patented this process in 1994 and in 2001 signed an agreement with BMS for the development of new technology based on microbial fermentation for the production of Taxol and other taxanes. Ever since, more than 160 subsequent publications and patents have addressed the biosynthesis of Taxol and related taxanes by microorganisms, which included some reviews.^57–59 Zhang et al. described that the endophyte Cladosporium cladosporioides isolated from the bark of T. media (Taxus spp.) could produce Taxol at the level of 800 μg L⁻¹.⁶⁰ In addition, Liu et al. reported that Metarhizium anisopliae isolated from the bark of T. chinensis also produced Taxol in the amounts of up to 846 μg L⁻¹ in the culture.⁶¹ Furthermore, the endophyte Phoma betae isolated from a non-Taxus species (Ginkgo biloba) produced nearly 795 μg L⁻¹ of Taxol under suitable fermentation conditions.⁶² Thus, fermentation processes using Taxol-producing microorganisms may be an alternative promising way to produce Taxol. However, one of the biggest problems of using the endophytic fungi to produce Taxol is the extremely poor yield and unstable production. One strain of Pestalotiopsis microspor, namely CP-4, produced Taxol varying from 50 to 1487 ng L⁻¹,⁶³ suggesting that it was genetically unstable. As to the cooperation of Cytoclonal Pharmaceutics, Inc. and BMS, the inability of the fungus to deliver reproducible high-titre yields of Taxol in axenic culture, hence not feasible for industrial scale-up, led to the disappointing failure of delivering on the promises of this highly heralded discovery. Admittedly, none of the discovered endophytes have been successfully translated into industrial bioprocesses, yet. To date, more and more chemical evidence supports that some endophytic fungi can produce Taxol and its analogs, and it is believed that the biosynthetic pathway of Taxol in the endophytic fungi may have a distinctly different evolutionary pattern compared with that in the Taxus species.⁶⁴ However, the microbial biosynthetic pathway has remained elusive, and the solid genetic evidence will be the identification and functional confirmation of the related biosynthetic genes. Recently, there have been at least two contradicting reports on the independent Taxol biosynthetic capacity of endophytes.^65,66 According to a report by Heinig et al., they did not find any evidence for independent taxane biosynthesis in any of the endophytes, including the isolate described in the original publication (Taxomyces andreanae) and various recent isolates from Taxus trees, even using a combination of phytochemistry, molecular biology and genome sequencing.⁶⁵ Kusari et al. argued that it was time to rethink the production of Taxol using endophyte biotechnology.⁶⁶ Thus, the fungal production of Taxol remains an ongoing and controversial issue.

Taxol from other non-Taxus plants

Taxol and its analogs have also been discovered in non-Taxus gymnosperms and even angiosperms.^58,66,67 For instance, trace amounts of Taxol were found in Cephalotaxus manii and Pseudotaxus chienii.^68,69 In addition, baccatin III (24) was found in the stems and leaves of Cephalotaxus fortunei and Podocarpus forrestii.^70,71 With the exception of the genus Taxus, the occurrence of this diterpenoid in other reported gymnosperm taxa has not been extensively studied and seems confined to only an exceptionally limited number of species. Interestingly, Taxol and its analogs were also found in some plants whose relationship were very far from Taxus plants, such as the family Betulaceae. For example, Taxol (3), 10-deacetyltaxol (20), 10-deacetylbaccatin III (8), baccatin III (24), 7-xylosyl-10-deacetylcephalomannine (25), cephalomannine (26), 7-xylosyltaxol (27) and 7-xylose-10-deacetyltaxol (19) (Fig. 5), etc. were discovered from the hazel cell culture, or the hard shells, the green shell covers and the leaves of Tombul hazelnut (Corylus arellana L.).^72–75 Moreover, the production of taxanes in hazel cell culture increased when elicitors were used.^73–75 Then again, the molecular basis of taxane production in this angiosperm is not clear. It seems that pressures on the Taxus resources may be reduced by the extraction of Taxol and its analogs from these plants. However, the content of Taxol in these plants was even lower than in Taxus plants.

Fig. 5 Structures of compound 24–27.

Summary and future perspective

Nature has been a remarkable chemist, furnishing us with chemical structures no scientist could imagine. Taxol, one of the most powerful natural products, not only has yielded new and effective drugs, but has also provided insights into new mechanisms of action. If anyone said “the development of a new anticancer drug with a novel structure and unique mechanism of action is an important event, especially when the drug plays a clear role in improving the outcome for cancer patients”, no drug fits this description better than Taxol. For the foreseeable future, Taxol and its analogues will continue to be important cancer chemotherapeutic drugs.

Currently, chemical semi-synthesis and nursery cultivation of Taxus species prevail as the main sources for the clinical supply of Taxol. To boost the supply of Taxol, different research strategies should be combined and collaborative studies between groups need to be encouraged in the present and future endeavours, so as to drive down the price of anticancer drugs and satisfy the increasing demand of Taxol. Although endophytic fungi draw a piece of beautiful blueprint for Taxol production, no breakthrough-type progresses have been made, and the microbial biosynthetic pathway has remained elusive. Plant cell culture is one of the more promising and sustainable Taxol production approaches, but its high production costs and unstable product yields are burning problems. The total synthesis of Taxol deserved high evaluation, at least from the organic chemistry standpoint. However, there is still a long way to go for practical production. Especially, a number of continuous efforts on the Taxol biosynthetic pathway research should be addressed. A rational approach might provide a new insight into the question of how the Taxol biosynthetic pathway is regulated. With the development of synthetic biology, scientists may finally realize the objective of producing Taxol by a microbial cell factory.

Acknowledgements

This work was supported by the National Natural Science Foundation of China (Grant numbers 31270796 and 30770229), National Mega-project for Innovative Drugs (Grant number 2012ZX09301002-001-005) and PUMC Youth Fund and the Fundamental Research Funds for the Central Universities (Grant number 3332015137).

Notes and references

1. M. Wani, H. Taylor, M. Wall, P. Coggon and A. McPhail, J. Am. Chem. Soc., 1971, 18, 242–260.
2. K. Priyadarshini and U. K. Aparajitha, Med. Chem., 2012, 2, 139–141.
3. B. A. Weaver, Mol. Biol. Cell, 2014, 25, 2677–2681.
4. P. B. Schiff, J. Fant and S. B. Horwitz, Nature, 1979, 277, 665–667.
5. M. A. Ojeda-Lopez, D. J. Needleman, C. Song, A. Ginsburg, P. A. Kohl, Y. Li, H. P. Miller, L. Wilson, U. Raviv, M. C. Choi and C. R. Safinya, Nat. Mater., 2014, 13, 195–203.
6. E. K. Rowinsky, N. Onetto, R. Canetta and S. Arbuck, Semin. Oncol., 1992, 19, 646–662.
7. E. K. Rowinsky, Ann. Pharmacother., 1994, 28, S18–S22.
8. A. K. Singla, A. Garg and D. Aggarwal, Int. J. Pharm., 2002, 235, 179–192.
9. C. Herdeg, M. Oberhoff, A. Baumbach, A. Blattner, D. I. Axel, S. Schröder, H. Heinle and K. R. Karsch, J. Am. Coll. Cardiol., 2000, 35, 1969–1976.
10. A. Ehrlich, S. Booher, Y. Becerra, D. L. Borris, W. D. Figg, M. L. Turner and A. Blauvelt, J. Am. Acad. Dermatol., 2004, 50, 533–540.
11. K. Bonaventura, A. Leber, C. Sohns, M. Roser, L.-H. Boldt, W. Haverkamp and M. Dorenkamp, J. Am. Coll. Cardiol., 2012, 60, B171.
12. M. Unverdorben, C. Vallbracht, B. Cremers, H. Heuer, C. Hengstenberg, C. Maikowski, G. S. Werner, D. Antoni, F. X. Kleber, W. Bocksch, M. Leschke, H. Ackermann, M. Boxberger, U. Speck, R. Degenhardt and B. Scheller, EuroIntervention, 2015, 11, 926–934.
13. A. W. Heldman, L. Cheng, G. M. Jenkins, P. F. Heller, D.-W. Kim, M. Ware, C. Nater, R. H. Hruban, B. Rezai and B. S. Abella, Circulation, 2001, 103, 2289–2295.
14. M. Unverdorben, C. Vallbracht, B. Cremers, H. Heuer, C. Hengstenberg, C. Maikowski, G. S. Werner, D. Antoni, F. X. Kleber and W. Bocksch, Circulation, 2009, 119, 2986–2994.
15. B. Zhang, A. Maiti, S. Shively, F. Lakhani, G. McDonald-Jones, J. Bruce, E. B. Lee, S. X. Xie, S. Joyce, C. Li, P. M. Toleikis, V. M. Lee and J. Q. Trojanowski, Proc. Natl. Acad. Sci. U. S. A., 2005, 102, 227–231.
16. N. Wang, T. He, Y. Shen, L. Song, L. Li, X. Yang, X. Li, M. Pang, W. Su and X. Liu, ACS Appl. Mater. Interfaces, 2016, 8, 4368–4377.
17. F. Wang, D. Zhang, Q. Zhang, Y. Chen, D. Zheng, L. Hao, C. Duan, L. Jia, G. Liu and Y. Liu, Biomaterials, 2011, 32, 9444–9456.
18. A. H. Abouzeid, N. R. Patel and V. P. Torchilin, Int. J. Pharm., 2014, 464, 178–184.
19. P. Ma and R. J. Mumper, J. Nanomed. Nanotechnol., 2013, 4, 1000164.
20. M. D. Galsky, A. Dritselis, P. Kirkpatrick and W. K. Oh, Nat. Rev. Drug Discovery, 2010, 9, 677–678.
21. O. Metzger-Filho, C. Moulin, E. de Azambuja and A. Ahmad, Expert Opin. Invest. Drugs, 2009, 18, 1183–1189.
22. D. Waterhouse, K. Gelmon, R. Klasa, K. Chi, D. Huntsman, E. Ramsay, E. Wasan, L. Edwards, C. Tucker and J. Zastre, Curr. Cancer Drug Targets, 2006, 6, 455–489.
23. J. A. Yared and K. H. Tkaczuk, Drug Des., Dev. Ther., 2012, 6, 371–384.
24. A. Heidenreich, H.-J. Scholz, S. Rogenhofer, C. Arsov, M. Retz, S. C. Müller, P. Albers, J. Gschwend, M. Wirth and U. Steiner, Eur. Urol., 2013, 63, 977–982.
25. Y. Bedi, R. Ogra, K. Kiran, B. Kaul and R. Kapil, Yew (Taxus spp.). A new look on utilization, cultivation and conservation, 1996, pp. 443–456.
26. J. Lou, X. L. Niu, F. Yan, J. Pan and X. D. Zhu, Mycosystema, 2011, 30, 158–167.
27. J. N. Denis, A. E. Greene, D. Guenard, F. Gueritte-Voegelein, L. Mangatal and P. Potier, J. Am. Chem. Soc., 1988, 110, 5917–5919.
28. J. C. Borah, J. Boruwa and N. C. Barua, Curr. Org. Synth., 2007, 4, 175–199.
29. J. Yan, Jiangsu Chem. Ind., 2005, 33, 46–58.
30. K. V. Rao, J. Heterocycl. Chem., 1997, 34, 675–680.
31. H. L. Cheng, R. Y. Zhao, T. J. Chen, W. B. Yu, F. Wang, K. D. Cheng and P. Zhu, Mol. Cell. Proteomics, 2013, 12, 2236–2248.
32. K. Wang, T. Wang, J. Li, J. Zou, Y. Chen and J. Dai, J. Mol. Catal. B: Enzym., 2011, 68, 250–255.
33. R. L. Hanson, J. M. Howell, D. B. Brzozowski, S. A. Sullivan, R. N. Patel and L. J. Szarka, Biotechnol. Appl. Biochem., 1997, 26, 153–158.
34. W. B. Yu, X. Liang and P. Zhu, J. Ind. Microbiol. Biotechnol., 2013, 40, 133–140.
35. W. C. Liu and P. Zhu, J. Ind. Microbiol. Biotechnol., 2015, 42, 867–876.
36. W. C. Liu, T. Gong, Q. H. Wang, X. Liang, J. J. Chen and P. Zhu, Sci. Rep., 2016, 6, 18439.
37. A. Christen, J. Bland and D. Gibson, Proc. Am. Assoc. Canc. Res., 1989, 30, 566.
38. L. M. Leone and S. C. Roberts, in Natural products and cancer drug discovery, Springer, 2013, pp. 193–211.
39. R. M. Cusido, M. Onrubia, A. B. Sabater-Jara, E. Moyano, M. Bonfill, A. Goossens, M. A. Pedreño and J. Palazon, Biotechnol. Adv., 2014, 32, 1157–1167.
40. S. Howat, B. Park, I. S. Oh, Y.-W. Jin, E.-K. Lee and G. J. Loake, New Biotechnol., 2014, 31, 242–245.
41. S. A. AMINI, L. Shabani, L. Afghani, Z. Jalalpour and M. Sharifi-Tehrani, Turk. J. Biol., 2014, 38, 528–536.
42. C. Zhao, G. Song, C. Fu, Y. Dong, H. Xu, H. Zhang and L. J. Yu, Plant Cell Rep., 2016, 35, 541–559.
43. H. Tabata, in Biomanufacturing, Springer, 2004, pp. 1–23.
44. S. C. Roberts, Nat. Chem. Biol., 2007, 3, 387–395.
45. E. K. Lee, Y. W. Jin, J. H. Park, Y. M. Yoo, S. M. Hong, R. Amir, Z. Yan, E. Kwon, A. Elfick and S. Tomlinson, Nat. Biotechnol., 2010, 28, 1213–1217.
46. S. Malik, R. M. Cusidó, M. H. Mirjalili, E. Moyano, J. Palazón and M. Bonfill, Process Biochem., 2011, 46, 23–34.
47. R. Croteau, R. B. Ketchum, R. Long, R. Kaspera and M. Wildung, Phytochem. Rev., 2006, 5, 75–97.
48. Q. Huang, C. A. Roessner, R. Croteau and A. I. Scott, Bioorg. Med. Chem., 2001, 9, 2237–2242.
49. W. Wang, C. Meng, P. Zhu and K. Cheng, Chin. J. Biotechnol., 2005, 25, 103–108.
50. J. M. DeJong, Y. Liu, A. P. Bollon, R. M. Long, S. Jennewein, D. Williams and R. B. Croteau, Biotechnol. Bioeng., 2006, 93, 212–224.
51. P. K. Ajikumar, W. H. Xiao, K. E. Tyo, Y. Wang, F. Simeon, E. Leonard, O. Mucha, T. H. Phon, B. Pfeifer and G. Stephanopoulos, Science, 2010, 330, 70–74.
52. B. W. Biggs, C. G. Lim, K. Sagliani, S. Shankar, G. Stephanopoulos, M. De Mey and P. K. Ajikumar, Proc. Natl. Acad. Sci. U. S. A., 2016, 113, 3209–3214.
53. Ó. Besumbes, S. Sauret-Güeto, M. A. Phillips, S. Imperial, M. Rodríguez-Concepción and A. Boronat, Biotechnol. Bioeng., 2004, 88, 168–175.
54. K. Kovacs, L. Zhang, R. S. Linforth, B. Whittaker, C. J. Hayes and R. G. Fray, Transgenic Res., 2007, 16, 121–126.
55. A. Anterola, E. Shanle, P. F. Perroud and R. Quatrano, Transgenic Res., 2009, 18, 655–660.
56. A. Stierle, G. Strobel and D. Stierle, Science, 1993, 260, 214–216.
57. X. Zhou, H. Zhu, L. Liu, J. Lin and K. Tang, Appl. Microbiol. Biotechnol., 2010, 86, 1707–1717.
58. Z. R. Flores-Bustamante, F. N. Rivera-Orduña, A. Martínez-Cárdenas and L. B. Flores-Cotera, J. Antibiot., 2010, 63, 460–467.
59. S. Gond, R. Kharwar and J. White, Fungal Biol. Rev., 2014, 28, 77–84.
60. P. Zhang, P. P. Zhou and L. J. Yu, Curr. Microbiol., 2009, 59, 227–232.
61. K. Liu, X. Ding, B. Deng and W. Chen, J. Ind. Microbiol. Biotechnol., 2009, 36, 1171–1177.
62. R. S. Kumaran, Y.-K. Choi, S. Lee, H. J. Jeon, H. Jung and H. J. Kim, Afr. J. Biotechnol., 2014, 11, 950–960.
63. J. Y. Li, G. Strobel, R. Sidhu, W. Hess and E. J. Ford, Microbiology, 1996, 142, 2223–2226.
64. Y. Yang, H. Zhao, R. A. Barrero, B. Zhang, G. Sun, I. W. Wilson, F. Xie, K. D. Walker, J. W. Parks, R. Bruce, G. Guo, L. Chen, Y. Zhang, X. Huang, Q. Tang, H. Liu, M. I. Bellgard, D. Qiu, J. Lai and A. Hoffman, BMC Genomics, 2014, 15, 69.
65. U. Heinig, S. Scholz and S. Jennewein, Fungal Divers., 2013, 60, 161–170.
66. S. Kusari, S. Singh and C. Jayabaskaran, Trends Biotechnol., 2014, 32, 304–311.
67. R. S. Kumaran, J. Muthumary and B. K. Hur, J. Microbiol., 2009, 47, 40–49.
68. Y. F. Wang, Q. W. Shi, M. Dong, H. Kiyota, Y. C. Gu and B. Cong, Chem. Rev., 2011, 111, 7652–7709.
69. R. Stahlhut, G. Park, R. Petersen, W. Ma and P. Hylands, Biochem. Syst. Ecol., 1999, 27, 613–622.
70. S. Luo, B. Ning, D. Ruan and H. Wang, J. Plant Resour. Environ., 1993, 3, 31–33.
71. R. Zhou, D. Zhu, S. Gao and Z. Cai, J. China Pharm. Univ., 1994, 25, 259–261.
72. A. Rezaei, F. Ghanati, M. Behmanesh and M. Mokhtari-Dizaji, Ultrasound Med. Biol., 2011, 37, 1938–1947.
73. A. Hoffman and F. Shahidi, J. Funct. Foods, 2009, 1, 33–37.
74. L. Ottaggio, F. Bestoso, A. Armirotti, A. Balbi, G. Damonte, M. Mazzei, M. Sancandi and M. Miele, J. Nat. Prod., 2007, 71, 58–60.
75. F. Bestoso, L. Ottaggio, A. Armirotti, A. Balbi, G. Damonte, P. Degan, M. Mazzei, F. Cavalli, B. Ledda and M. Miele, BMC Biotechnol., 2006, 6, 45–55.

---