Search tips
Search criteria 


Logo of nihpaAbout Author manuscriptsSubmit a manuscriptHHS Public Access; Author Manuscript; Accepted for publication in peer reviewed journal;
J Am Chem Soc. Author manuscript; available in PMC 2008 December 29.
Published in final edited form as:
PMCID: PMC2610460

Synthesis and Structure Verification of the Vaccine Adjuvant QS-7-Api. Synthetic Access to Homogeneous Quillaja saponaria Immunostimulants

The clinical success of conjugate anticancer and antiviral vaccines critically depends on the identification of, and access to, novel potent adjuvants with attenuated toxicity. In this context, specific fractions from extracts of the bark of Quillaja saponaria (QS) have proven to be exceedingly powerful adjuvants in immunotherapy. The QS-21 fraction,13 comprising isomeric forms of a complex triterpene saponin, is currently the most promising immuno-potentiator4 in several antitumor (melanoma, breast, SCLC, prostate)5 and infectious-disease (HIV, malaria)69 vaccine therapies. However, the tolerated dose of QS-21 in patients typically does not exceed 100 μg, above which significant local erythema and systemic flu-like symptoms arise. On the other hand, QS-7, an additional QS extract fraction, was found not only to possess significant stand-alone adjuvant activity,1,10 but also to induce remarkable synergistic immune response augmentation11 when coadministered with QS-21. Importantly, QS-7, unlike QS-21, exhibited negligible toxicity in mice.1,10,11 That QS-7 has not been advanced to the clinic likely stems from (1) a more difficult purification protocol compared to that of QS-21 and (2) uncertainty in its structural constitution, postulated to be QS-7-Api (1, Figure 1).10,11 We now report the synthesis and structural verification of QS-7-Api (1).12 Furthermore, a novel semisynthetic sequence has been developed, enabling procurement of homogeneous QS-7-Api (1) for imminent preclinical evaluation.

The synthesis of the hexasaccharide fragment within QS-7-Api (“R”, Figure 1) required initial preparation of the selectively protected monosaccharides 24, 6, and 8 (Scheme 1). While the xylo-, gluco-, and apio-derived monosaccharides 24 (Scheme 1A) were obtained in multistep sequences by previously reported procedures and modifications thereof,13,14 the novel sugars 6 and 8 were prepared from rhamnopyranose 5 and fucopyranoside 7, respectively. Silylation of the selectively protected rhamnopyranose 5 (Scheme 1B) with TIPSOTf provided the α-TIPS glycoside (96%), which subsequently underwent C4-O-debenzylation to furnish the rhamnopyranoside 6 (98%). Synthesis of the fucosyl residue within QS-7 commenced with selective C3-O-alkylation of the allyl fucopyranoside 7 (Scheme 1C) with PMBCl (56%) via its transient stannylene acetal. This allowed for sequential selective silylation of the equatorial C2-OH (97%) and acetylation of the axial C4-OH (>99%). Finally, oxidative removal of the PMB ether with DDQ provided the selectively protected fucopyranoside 8 (86%).

Scheme 1a
a Reagents and conditions: (a) TIPSOTf, 2,6-lutidine, CH2Cl2, 0→23 °C, 96%; (b) H2, Pd–C, MeOH, 23 °C, 98%; (c) n-Bu2SnO, PhMe, reflux; CsF; PMBCl, DMF, 23 °C, 56%; (d) TBSCl, imidazole, DMAP, CH2Cl2, 23 °C, ...

Convergent assembly of the branched hexasaccharide (Scheme 2) involved dehydrative glycosylation (Ph2SO·Tf2O)15 of fucopyranoside 8 with rhamnopyranose 5 (84%). The resulting α-disaccharide 9 then underwent a series of protective group exchanges, including TBS removal (95%), a novel Et2Zn/Pd(PPh3)4-mediated anomeric deallylation (68%),16 and selective anomeric silylation (75%) to afford the disaccharide 10 as a suitable glycosyl acceptor. Its glycosyl donor coupling partner was prepared by chemo- and stereoselective dehydrative glycosylation of rhamnopyranoside 6 with xylopyranose 2 to afford the β-disaccharide 11 (75%), which directly underwent glycosylation17 with the apiose-derived donor 4 to afford trisaccharide 12 (86%). The acetate esters in 12 were then exchanged for a benzylidene acetal protective group (94%), followed by selective acid hydrolysis of the rhamno-derived isopropylidene ketal to afford the corresponding vicinal diol (71%). Selective alkylation of the resulting axial rhamno-C2-OH with BnBr could then be accomplished (84%), allowing for Schmidt glycosylation18 of the C3-OH with the glucosyl imidate 3 to afford the tetrasaccharide 13 (86%). Exchange of the benzoate ester for a TES ether (91%, two steps) and conversion of the anomeric TIPS group to its α-trichloroacetimidate counterpart 14 (92%, two steps) secured a suitable donor for glycosylation of disaccharide 10. This was accomplished by treating the two components with TMSOTf to afford hexasaccharide 15 (62%), whose fucosyl-TIPS-acetal was then transformed to the α-trichloroacetimidate 16 (84%, two steps).

Scheme 2a
a Reagents and conditions: (a) Ph2SO, Tf2O, TBP, CH2Cl2, −78→23 °C, 84%; (b) TBAF, THF, 0→23 °C, 95%; (c) Et2Zn, Pd(PPh3)4, Et2O, 23 °C; 68%; (d) TIPSCl, imidazole, DMAP, DMF, 23 °C, 75%; (e) Ph ...

Late stage construction of the full QS-7-Api skeleton involved the elaborately protected triterpene–trisaccharide conjugate 17 (Scheme 3A), previously prepared in 18 steps from glucuronolactone during the course of the synthesis of QS-21.13 This C28-carboxylic acid glycosyl acceptor 17 responded well to glycosylation with trichloroacetimidate glycosyl donor 16 (BF3·OEt2) to afford fully protected QS-7-Api (71%), which underwent global deprotection under carefully managed conditions (TFA; H2, Pd–C). The resulting product (71%) was found to be identical to naturally derived QS-7-Api (1).19

Scheme 3a
a Reagents and conditions: (a) 16, BF3·OEt2, 4Å ms, CH2Cl2, −78→23 °C, 71%; (b) TFA, H2O, CH2Cl2, 0 °C; H2, Pd–C, EtOH, THF, 23 °C, 71%; (c) KOH, EtOH, H2O, 80 °C; (d) TESOTf, Py, ...

This synthesis of 1 (Scheme 3A) from de novo construction of all oligosaccharide fragments confirms the structure of QS-7-Api and provides significantly more dependable access to homogeneous samples of 1 than isolation from natural sources. This notwithstanding, the synthesis of 1 can be further augmented. Quil-A (18, Scheme 3B) is a commercially available semipurified extract from Quillaja saponaria and contains variable quantities of >50 distinct saponins,20 many of which incorporate the triterpene–trisaccharide substructure within QS-7 (and QS-21). This monodesmoside saponin 19 (Scheme 3B) can be isolated in semipure form via direct base hydrolysis of the Quil-A mixture.21 Subsequent poly(silylation) of 19 with excess TESOTf afforded the corresponding nonakis(triethylsilyl ether) (257 mg from 1.15 g of 18), whose glucuronic acid functionality could be selectively derivatized to the benzyl ester 20 (CbzCl, 68%). This triterpene–trisaccharide conjugate, obtained in only a three-step protocol from Quil-A (18), was an effective acceptor in a C28-carboxylate glycosylation (80%) with hexasaccharide 16 to provide, after global deprotection, QS-7-Api (1) (77%). The evolution of the first synthesis of 1 to this semisynthetic variant furnishes complex QS-saponin adjuvants (and likely non-natural analogues) with markedly enhanced facility, enabling heretofore untapped opportunities for novel adjuvant discovery in antitumor and antiviral vaccine development.

Supplementary Material


Supporting Information Available:

Complete ref 7; experimental details. This material is available free of charge via the Internet at


This work is dedicated to Prof. E. J. Corey on the occasion of his 80th birthday. This research was supported by the NIH (Grant GM58833), Mr. William H. Goodwin, Mrs. Alice Goodwin and The Commonwealth Foundation for Cancer Research, and The Experimental Therapeutics Center of MSKCC.


1. Kensil CR, Patel U, Lennick M, Marciani D. J Immunol. 1991;146:431–437. [PubMed]
2. Soltysik S, Wu JY, Recchia J, Wheeler DA, Newman MJ, Coughlin RT, Kensil CR. Vaccine. 1995;13:1403–1410. [PubMed]
3. Kensil CR. Crit Rev Ther Drug Carrier Syst. 1996;13:1–55. [PubMed]
4. Kim SK, Ragupathi G, Musselli C, Choi SJ, Park YS, Livingston PO. Vaccine. 2000;18:597–603. [PubMed]
5. Livingston PO, Ragupathi G. Hum Vaccines. 2006;2:137–143. [PubMed]
6. Sasaki S, Sumino K, Hamajima K, Fukushima J, Ishii N, Kawamoto S, Mohri H, Kensil CR, Okuda K. J Virol. 1998;72:4931–4939. [PMC free article] [PubMed]
7. Evans TG, et al. Vaccine. 2001;19:2080–2091. [PubMed]
8. Kashala O, Amador R, Valero MV, Moreno A, Barbosa A, Nickel B, Daubenberger CA, Guzman F, Pluschke G, Patarroyo ME. Vaccine. 2002;20:2263–2277. [PubMed]
9. Carcaboso AM, Hernández RM, Igartua M, Rosas JE, Patarroyo ME, Pedraz JL. Vaccine. 2004;22:1423–1432. [PubMed]
10. Kensil CR, Wu JY, Anderson CA, Wheeler DA, Amsden J. Dev Biol Stand. 1998;92:41–47. [PubMed]
11. Kensil CA. Saponin adjuvant compositions. U.S. Pat. 6231859. 2001.
12. The existence of an additional inseparable isomeric form, wherein the α-d-apiose is replaced by β-d-xylose, has also been postulated.
13. Kim YJ, Wang P, Navarro-Villalobos M, Rohde BD, Derryberry J, Gin DY. J Am Chem Soc. 2006;128:11906–11915. [PMC free article] [PubMed]
14. Nguyen HM, Chen Y, Duron SG, Gin DY. J Am Chem Soc. 2001;123:8766–8772. [PubMed]
15. Garcia BA, Gin DY. J Am Chem Soc. 2000;122:4269–4279.
16. Chandrasekhar S, Reddy CR, Rao RJ. Tetrahedron. 2001;57:3435–3438.
17. Roush WR, Bennett CE. J Am Chem Soc. 1999;121:3541–3542.
18. Schmidt RR, Kinzy W. Adv Carbohydr Chem Biochem. 1994;50:21–123. [PubMed]
19. Trace quantities of natural QS-7-Api (~70% purity, NMR) were obtained from exhaustive RP-HPLC purification of commercial Quil-A.
20. van Setten DC, van de Werken G, Zomer G, Kersten GFA. Rapid Commun Mass Spectrom. 1995;9:660–666. [PubMed]
21. Higuchi R, Tokimitsu Y, Fujioka T, Komori T, Kawasaki T, Oakenful DG. Phytochemistry. 1987;26:229–235.