Zanamivir

139110-80-8

APPROVED 26-7-96……. GSK NDA 021036

A guanido-neuraminic acid that is used to inhibit neuraminidase.

Zanamivir INN /zəˈnæmɨvɪər/ is a neuraminidase inhibitor used in the treatment and prophylaxis of influenza caused by influenza A virus andinfluenza B virus. Zanamivir was the first neuraminidase inhibitor commercially developed. It is currently marketed by GlaxoSmithKline under the trade name Relenza as a powder for oral inhalation.

The drug is approved for use for the prevention and treatment of influenza in those over the age of 7 in the United States, Canada, European Union, and many other countries. It is not recommended for people with respiratory problems and ailments.

United States	6294572	APPROVED  1994-12-15	EXPIRY 2014-12-15
United States	5360817	1993-07-26	2013-07-26
Canada	2291994	2003-10-14	2011-04-24
Canada	2081356	2000-02-22	2011-04-24

Zanamivir was discovered in 1989 by scientists led by Peter Malcolm Colman and Joseph Varghese at the CSIRO, in collaboration with theVictorian College of Pharmacy, Monash University, and scientists at Glaxo, UK. Zanamivir was the first of the neuraminidase inhibitors. The discovery was initially funded by the Australian biotechnology company Biota and was part of Biota’s ongoing program to develop antiviral agents throughrational drug design. Its strategy relied on the availability of the structure of influenza neuraminidase, by X-ray crystallography. It was also known, as far back as 1974, that 2-deoxy-2,3-didehydro-N-acetylneuraminic acid (DANA), a sialic acid analogue, is an inhibitor of neuraminidase. Sialic acid (N-acetyl neuraminic acid, NANA), the substrate of neuraminidase, is itself a mild inhibitor of the enzyme, but the dehydrated derivative DANA, a transition-state analogue, is a better inhibitor.

Computational chemistry techniques were used to probe the active site of the enzyme, in an attempt to design derivatives of DANA that would bind tightly to the amino acid residues of the catalytic site, and so would be potent and specific inhibitors of the enzyme. The GRID software by Molecular Discovery was used to determine energetically favourable interactions between various functional groups and residues in the catalytic site canyon. This investigation showed that there is a negatively charged zone in the neuraminidase active site that aligns with the C₄hydroxyl group of DANA. This hydroxyl is, therefore, replaced with a positively charged amino group; the 4-amino DANA was shown to be 100 times better as an inhibitor than DANA, owing to the formation of a salt bridge with a conserved glutamic acid (119) in the active site. It was also noticed that Glu 119 is at the bottom of a conserved pocket in the active site, just big enough to accommodate a more basic functional positively charged group, such as a guanidino group, which was also larger than the amino group. Zanamivir, a transition-state analogue inhibitor of neuraminidase, was the result.

As Biota was a small company, it did not have the resources to bring zanamivir to market by itself. In 1990, zanamivir patent rights were licensed to Glaxo, now GlaxoSmithKline (GSK). In 1999, the product was approved for marketing in the US and subsequently has been registered by GSK in a total of 70 countries (GlaxoSmithKline News release, 2006). Zanamivir is delivered via Glaxo’s proprietary Diskhaler inhalation device. The license agreement entitled Biota to receive a 7% royalty on Glaxo’s sales of zanamivir.

Chemical name:	5- Acetamido- 2, 6- anhydro- 3, 4, 5- trideoxy- 4- guanidino- D- glycero- D- galacto- non- 2- enonic acid
Synonyms:	Zanamivir, GG167, 4-guanidino-Neu5Ac2en and 2,3- Didehydro- 2, 4- dideoxy- 4- guanidino- N- acetyl- D- neuraminic acid(2R,3R,4S)-4-guanidino-3-(prop-1-en-2-ylamino)-2-((1R,2R)-1,2,3-trihydroxypropyl)-3,4-dihydro-2H-pyran-6-carboxylic acid
Empirical formula:	C12H20N4O7
Structural formula:
Molecular weight:	332.31g
Beilstein number:	7083099
Normal State:	Powder
Colour:	White to ‘off white’
Melting point:	325oC
Optical rotary power:	Type [�]Conc: 0.9g/100mlSolvent: H2OOptical rotary power: 41 degWavelength: 589nmTemp: 20oC
CAS number:	139110-80-8
Solubility:	18mg/mL in water at 20oC

Zanamivir is used for the treatment of infections caused by influenza A virus and influenza B virus. There is low to moderate evidence that it decreases the risk of one’s getting influenza by 1% to 12% in those exposed. In otherwise-healthy individuals, benefits overall appear to be small.It is unclear whether it affects the risk of one’s need to be hospitalized or the risk of death. An independent analysis of its effects by the Cochrane collaboration was awaiting release of trial data as of 2012. The evidence for a benefit in preventing influenza is weak in children with concerns of publication bias in the literature. As of 2009 no influenza has shown any signs of resistance. Since then genes expressing resistance to were found in patients infected with Influenza A H7N9 and who were treated with corticosteroids.

ZANAMIVIR

Mass

1H NMR

Hydrogen	Chemical shift /ppm
(1H, d, 3-H)	5.53
(2H, 2dd, 4- and 6-H)	4.50 – 4.38
(1H, dd, 5-H)	4.21
(2H, dd+ddd, 9-Ha and 8-H)	4.00-3.88
(2H, 2dd, 9-Hb and 7-H)	3.70-3.62
(3H, s,  Ac)	2.05

13C NMR

Carbon	Shift /ppm
(C=O, Ac)	177.3
(C-1)	172.1
(guanidino)	159.9
(C-2)	152.1
(C-3)	106.8
(C-6)	78.3
(C-8)	72.6
(C-7)	71.0
(C-9)	65.9
(C-4)	54.0
(C-5)	50.6
(Me)	24.8

ref 12

ref 13for above

Although zanamivir was the first neuraminidase inhibitor to the market, it had only a few months lead over the second entrant, oseltamivir (Tamiflu), with an oral tablet formulation.

According to the CDC, Tamiflu, zanamivir’s main competitor, is not as effective at treating the influenza viruses as zanamivir, especially in H1N1 seasonal flu. In fact, tests showed 99.6% of the tested strains of seasonal H1N1 flu and 0.5% of 2009 pandemic flu were resistant to Tamiflu, while no flu samples, seasonal or pandemic, showed any resistance to zanamivir.

When first marketed in the US in 1999/2000, zanamivir captured only 25% of the influenza antiviral market, despite a huge promotional campaign. By the end of that season, Tamiflu was outselling zanamivir 3:1. During that season, zanamivir experienced worldwide safety warnings involving the risk of bronchospasm and death. Glaxo then reduced the marketing of zanamivir, and Tamiflu’s dominance increased. More than US$20 million worth of zanamivir sold by Glaxo in the first US season was returned to the company in the next two seasons because zanamivir’s sales to patients were far less than expected.

Biota commenced legal proceedings in 2004 alleging Glaxo’s reduced marketing of zanamivir to be a breach of contract. Biota claimed approximately A$700m from Glaxo. After Biota spent four years trying to progress its case, and incurring A$50m in legal costs, the company abandoned the claim in July 2008, recovering only A$20 million, including legal costs following settlement at mediation. Biota had refused an earlier tactical offer from Glaxo of A$75 million plus legal costs.

In August 2006, Germany announced it would buy 1.7 million doses of zanamivir, as part of its preparation strategy against bird flu. “Germany’s purchase shows that countries are starting to take a balanced view of influenza preparedness,” says Simon Tucker, head of research at Melbourne-based Biota, where zanamivir was originally developed.

In April 2009, many cases of swine flu (H1N1-type virus) were reported in US and Mexico. Zanamivir is one of only two drugs prescribed to treat it. A study published in June 2009 emphasized the urgent need for augmentation of oseltamivir (Tamiflu) stockpiles, with additional antiviral drugs including zanamivir, based on an evaluation of the performance of these drugs in the scenario that the 2009 H1N1 swine flu neuraminidase (NA) were to acquire the Tamiflu-resistance (His274Tyr) mutation, which is currently widespread in 99.6% of all tested seasonal H1N1 strains.n January 2011, GSK announced that it would commence phase III trials for intravenous zanamivir in a study that will span 20 countries in the Northern and Southern Hemispheres.

Recently, the reported oseltamivir-resistance H5N1 virus neuraminidase still retaining susceptibility to zanamivir indicates that the structure of zanamivir has some advantages over oseltamivir in binding to the active pocket of H5N1 neuraminidase.

As a proven anti-influenza drug target, neuraminidase continues to be attractive for the development of new inhibitors. The crystal structure of H5N1 avian influenza neuraminidase (PDB code: 2HTY) provides the three-dimensional structural information and opportunity for finding new inhibitors in this regard, because the existing inhibitors, such as oseltamivir and zanamivir, were developed based on different structures of neuraminidase, such as subtypes N9 and N2, and type B genus of influenza virus.

ZANAMIVIR

Chemistry

1. Scheigetz, J.; Zamboni, R.; Bernstein, M. A.;Roy, B. (December 1995). “A syntheses of 4-a-guanidino-2-deoxy-2,3-didehydro n-acetylneuraminic acid”. Organic Letters 27 (6): 637–644.doi:10.1021/ol901511x. Retrieved 2010-11-14.

Zanamivir synthetic process in the world

Together with oseltamivir, zanamivir is the only medicine which can prevent influenza on humans caused by H5N1 and H1N1 virus. Vietnam prepared oseltamivir (Tamiflu) medicine. But there was no zanamivir – the first influenza medicine belonging N1 kind, discovered and commercialized before oseltamivir. The scientific name of zanamivir is acid 5-acetamido-4-guanidino-6-(1,2,3-trihydroxy-propyl)-5,6-dihydro-4H-pyran-2-carboxylic. The discovery of zanamivir opens research possibilities for new medicines which have the same effect on enzyme neuraminidase inhibitor to prevent and treat influenza.

Acid sialic is an input to synthetize zanamivir. The name acid sialic (Neu5Ac2en) is used to indicate derivation at O- and N- positions of acid neuraminic, just for acid N-axetylneuraminic. Acid sialic of carbohydrate groups is on animal cells and microorganism, especially in glycoprotein and gangliosid. The commercial acid sialic is extracted from whey of the cheese and milk process as well as egg yolk, and costs about 5,000 USD per kilo.

In 1994, zanamivir was first synthesized and made public by Von Itzstein and other scientists from the Department of Pharmaceutical Chemistry under Monash University (Australia). Then, Chandler and co-workers of Glaxo company (GSK, Britain) acquired results, improved reaction steps and made them public in 1995. Accordingly, this method produced 8.3% of general output. The synthetic process is described in Figure 1.

Figure 1: Zanamivir synthetic process according to Chandler

Up to now, the research of Chandler has been the only publication about zanamivir synthetic method, the output of which is greater than milligrams, and it reproduces details about reaction conditions and physiochemical properties of the requisite substances.

Recently, a research group of Yao (China) proposed a new approach to synthetize into intermediate compound 5. Researchers started from another material – D-glucono-δ-lactone, which is cheaper than acid sialic. However, the synthetic process is longer and much complicated, including 24 steps, with lower productivity (0.2%).

Researching on synthesizing Zanamivir from Acid sialic by Institute of Chemistry

Synthetizing methyl N-acetylneuraminate (2) and O-pentaacetoxy (3) from acid sialic

Scientists from the Institute of Chemistry used acid sialic (axit N-acetylneuraminic) 98% from China as the input for the zanamivir synthetic process. They decided to use the method of Warner, using ion exchange resin Dowex-H, with the role of catalyst. Reaction was performed in the room in 10 hours. The output was metyl (2) este product of acid N-acetylneuraminic with a productivity of 99%.

Then, to synthetize O-pentaacetoxy (3), scientists applied axetyl effective chemistry method recently published, using BF₃.OEt₂catalysis at 0⁰C. Productivity in this case exceeded 95%.

Figure 2: The diagram of O-pentaacetoxy 3 derivative making

The use of catalysts which were ion exchange resin Dowex-H (for este chemical reaction) and BF₃.OEt₂ (for axetyl chemical reaction) had more advantages than the method by scientists from Glaxo.

Synthesizing intermediate compound – oxazoline (4) key from O-pentaacetoxy (3)

Figure 3: Diagram to synthesize oxazoline (4) from O-pentaacetoxy (3) according to a and b methods

Firstly, scientists conducted a survey on oxazoline (4) synthetic process according to Chandler’s process. O-pentaacetoxy (3) compound was separated from two types of OAc and formed oxazoline round thanks to the effect of strong acid Lewis, which was TMSOTf at 52⁰C in 2.5 hour. The productivity of this reaction achieved 40%. The pilot instead of TMSOTf by BF3.OEt2 catalysis in dichloromethane at room temperature at night, the productivity of the reaction to form oxazoline round from penta-acetoxy (5) was similar to the method using TMSOTf (42%). To increase productivity, scientists made a survey on one-pot method, directly from metyl este (2) to oxazoline (4), without passing O-pentaacetoxy (3), gave the highest productivity (73,3%) and was the most economic effectiveness.

Synthesizing zanamivir from oxazoline (4) intermediate compound

The next, scientists successfully conducted reactions from oxazoline (4) intermediate compound to Zanamivir (9) final product (Figure 1). Zanamivir product had IR and NMR data which were compatible with their structure.

Therefore, scientists from the Institute of Chemistry under Vietnam Academy of Science and Technology built a stable process, including seven major steps, synthesizing from acid sialic with the general productivity of 6.6% (the productivity made public in the world was 8.3%). Especially, in the first period, from acid sialic to oxazolin (4) was optimized and gave a general productivity of 74%, higher than the productivity made public by (61.7%). However, the productivity gained in the later period is still low. Now, synthesizing zanamivir influenza medicine still continues to be researched.

……………………

Beau and coworkers assembled the core dihydropyran framework of zanamivir congeners via a combination of PBM reaction and Iron(III)-promoted deprotection-cyclization sequence. A stereochemically-defined α-hydroxyaldehyde 2, diallylamine and a dimethylketal-protected boronic acid 1 is coupled to form the acyclic, stereochemically-defined amino-alcohol 3, which then undergoes an Iron(III)-promoted cyclization to form a bicyclic dihydropyran 4. Selective opening of the oxazoline portion of the dihydropyran intermediate 4 with water or timethylsilyl azide then furnish downstream products that have structures resembling the Zanamivir family members.

Reaction scheme part 1:

The commercially available N-acetyl-neuraminic acid 1 is the starting reagent for the most direct approach to the synthesis of 4-guanidino-Neu5Ac2en (Relenza). In reaction scheme 1 the steps for the conversion of N-acetyl-neuraminic acid 1 to its 4-amino analogue is shown. Step 1 is the addition of methanolic HCl (MeOH and HCl gas), which produces the methyl ester of 1, followed by acetic anhydride in pyridine with 4-(dimethylamino)pyridine catalysis, which produces the penta-acetoxy compound, 2. In step 2, 2 is converted into the oxazoline 3 at high yield using trimethylsilyl trifluoromethanesulfonate (TMSOTf) in ethyl acetate at 52oC. In step 3, the azido compound, 4, is produced by the reaction of 3 with trimethylsilyl azide in tert-butyl alcohol at 80oC. In step 4 catalytic sodium methoxide in methanol was used to remove the acetate protecting groups from 4 to give triol 5. The 4-amino analogue, 6 was made in step 5, by hydrolysis using triethylamine in water, hydrogenolysis with a Lindlar catalyst and finally the addition of Dowex 2 * 8 resin. The triethylamine salt of the 6 was made during hydrogenolysis  and the purpose of the Dowex 2 * 8 resin was to desalt this intermediate. The chemical names of the compounds are:

1: N-acetyl-neuraminic acid

2: 5- Acetamido- 3,5- dideoxy- D- glycero- �- D- galacto- 2- nonulo- pyranosonic acid methyl ester

3: Methyl (3aR, 4R, 7aR)- 2- Methyl- 4- [(1’S, 2’R)- 1′, 2′, 3′ – triacet- oxypropyl]- 3a, 7a- dihydro- 4H- pyrano [3, 4-d] oxazole- 6- carboxlate.

4: 5- Acetamido- 7, 8, 9- tri- O- acetyl- 2, 6- anhydro- 4- azido- 3, 4, 5- trideoxy- D- glycero- D- galacto- non- 2- enonic acid methyl ester.

5: 5- Acetamido- 2, 6- anhydro- 4- azido- 3, 4, 5- trideoxy- D- glycero- D- galacto- non- 2- enonic acid methyl ester.

6: 5- Acetamido- 4- amino- 2, 6- anhydro- 3, 4, 5- trideoxy- D- glycero- D- galacto- non- 2- enonic acid.

Part one of reaction scheme

Synthesis of reactant necessary for part 2 of reaction:

Aminoiminomethane-sulfonic acid (AIMSA), 7, which is necessary for the conversion of compound 6 into Relenza, 9, is synthesised in Reaction scheme 2. The oxidizing solution necessary for the reaction is prepared by the addition of peracetic acid to 30% hydrogen peroxide and then conc. sulfuric acid. This is followed by acetic anhydride and, once the reaction has completed, methanol. Thiourea is dissolved in methanol and added slowly to the oxidizing solution.to produce compound 7. Note that any crystals that form are removed and that the reaction needs to be carried out under cooled conditions. See the reference source for more experimental details.

Synthesis of AIMSA

Reaction scheme part 2:

Reaction scheme 3 shows the conversion of compound 6 into Relenza For route A, 3 mol equivalent of AIMSA, 7, and 3 mol equivalent of potassium carbonate are added in a portionwise manner to compound 6 over an eight hour period. A yield of about 48% of the crystalline product 8 should be obtained for this method. An alternative route is to treat compound 6 with 1.1 mol equivalent of cyanogen bromide in the presence of sodium acetate in methanol. Route B step 1 gives compound 9, which can be converted into the final product 8 by treating it with ammonium hydroxide and ammonium formate at 85oC. A 36% yield of the purified product can be obtained after purification with ion-exchange chromatography and crystallisation. The chemical names of the compounds in this scheme are:

8. 5- Acetamido- 2, 6- anhydro- 3, 4, 5- trideoxy- 4- guanidino- D- glycero- D- galacto- non- 2- enonic acid. (Relenza)

9. 5- Acetylamino- 2, 6- anhydro- 4- cyanoamino- 3, 4, 5- trideoxy- D- glycero- D- galacto- non- 2- enonic acid

Part 2 of reaction scheme

CLIP

On Zanamivir

Total Synthesis of Anti-Influenza Agents Zanamivir and Zanaphosphor via Asymmetric Aza-Henry Reaction

Fig 1: The diskhaler used to administer Relenza. Each blister in the Rotadisk contains 5mg of the drug

Why there is a need for a more effective influenza treatment: At present influenza is basically an uncontrolled disease and an effective method is needed for both the prevention and treatment of it. In the 20^th century there were some major pandemics such as the 1918-1919 Spanish ‘flu which killed 20 million people world wide, the 1957 Asian ‘flu, the 1968 Hong Kong ‘flu and the 1977 Russian ‘flu¹² These viruses also affect different animals, especially domesticated chickens and turkeys and in Hong Kong in 1997 a virulent bird flu virus, started infecting and killing people for the first time ever. Of the 18 people affected 6 died, although there was no evidence that the virus was able to spread between people. Given the antigenic properties of the influenza virus, in the future the virus may be passed from person to person, and because human immune systems are not prepared for avian viruses the effects on the population could be grave. It would not be possible to prepare vaccines in time and anti-viral drugs are not always adequate.

Immunology

	Fig 4: The influenza viruses as seen under the electron microscope. Neuraminidase and haemagglutin spikes are visible.
Structure of the flu virus:Influenza (Fig 4) is an RNA virus which may exist as any shape from round balls to long, spaghetti-like filaments. The genome of this virus is associated with five different viral proteins and is surrounded by a lipid membrane, which means that influenza belongs to the “enveloped” group of viruses. Eight separate pieces of ribonucleic acid (RNA) make up the influenza virus genome and each piece of RNA specifies the amino acid sequence of one and sometimes two of the virus’s proteins. The segmented nature of the RNA allows differenet flu viruses to easily “mate” with each other to form hybrid progeny viruses with bits of RNA from each parent virus.Two glycoprotein molecules, known as hemagglutinin (HA) and neuraminidase (NA) (Fig 5) are stuck onto the lipid envelope of the virus and both play a crucial role in the infection of the epithelial cells of the upper respiratory tract. HA is a rod-shaped triangular molecule.and NA exists as a mushroom shaped spike with a box-like head on top of a long stalk, containing a hydrophobic region by which it is embedded in the viral membrane..

	Fig 5: The Neuraminidase enzyme
The enzyme Neuraminidase, also known as sialidase, is a tetramer with C-4 symmetry and an approximate molecular weight of 250 000. It contains a symmetrical folding pattern of six four-stranded antiparallel �-sheets arranged like propeller blades. Nine types of neuraminidase have been identified for influenza A and only one subtype for influenza B, and only 30% of the overall amino acid sequence is conserved between all known types of neuraminidase8  – these are the amino acids which line and surround the walls of the binding pocket. If they mutate, the enzyme is inactivated, so the virus could not mutate to escape from a drug which interfered with this site. So neuraminidase offers an attractive site for therapeutic intervention in influenza infections.

How the influenza virus works:The influenza virus (like all viruses) can only replicate after invading selected living cells and growing inside them. It makes thousands of new virus particles from the cellular machinery and then goes on to infect other cells.. Hemagglutinin allows the virus to infect the epithelial cells of the upper respiratory tract by attaching it to cells through receptors on the cell containing sialic acid, it fuses the cell membrane with the membrane of the virus, allowing the RNA of the virus to get inside the cell and thus instruct the cell to make thousands of new virus particles. After this viral replication, the progeny virions must be released from the cell to repeat the cell cycle of infection.Neuraminidase removes the sialic acid receptors from the host cell and other newly made virus particles by cleavage of -glycosidic bonds. This enables the virus to escape from the cell in which it grew and spread in the body to infect other cells. The action of NA may also facilitate viral mobility through the mucus of the respiratory tract.

	Fig. 6: The life cycle of the influenza virus. Click once on this image to see a larger version
The life cycle of the influenza virusG begins with the individual virus entering the cell lining of the respiratory tract (letter a in Fig. 6), and the cell being induced to take up the virus because hemagglutinin on the virus binds to the sialic acid (b and c in Fig 6). The virus then dispatches its genetic material (made up of RNA) and its internal proteins to the nucleus of the cell (e and f). Messenger RNA is produced when some of the internal proteins duplicate the RNA (f). This messenger RNA is used by the cell as a template for making viral proteins (g and h) and genes which become new viral particles and leave the cell covered in sialic acid. This sialic acid needs to be removed so that the hemagglutinin molecules on one particle don’t attach to the sialic acid on other ones, thus causing the new viruses to clump together and stick to the cell. The sialic acid is removed from the surface of the new viral particle by neuraminidase (j) and the new viral particles are able to travel and invade other cells (k).

How Relenza works: Relenza adopts a position within the active site of the enzyme and copies the geometry of the sialoside hydrolysis transition state9. It can achieve very good binding through appropriate presentation of its four pendent substituents and contains a hydrogen bonding glycerol sidechain. The guanidino group in Relenza is believed to form salt bridges with Glu 119 in the neuraminidase active site and add a strong charge interaction with Glu 2278. Two hydroxyl groups of the 6-glycerol side chain are hydrogen bonded to Glu276 and the 4-hydroxyl is oriented towards Glu119. The NH group of the 5-N acetyl side chain interacts with a bound water molecule on the floor of the active site. The carbonyl oxygen of the same side chain is hydrogen bonded to Arg152 and the methyl group enters a hydrophobic pocket lined by Ile222 and Trp178. The glycosidic oxygen projects into bulk solvent.

	Fig 7. Relenza bound to neuraminidase
The binding involved in Fig 7 is shown more clearly in Fig 8 below. Neuraminidase can no longer remove the sialic acid receptors from the host cell and newly made virus particles because of this binding. Therefore the virsuse ‘clump’ together or to the host cell and cannot go on to effect new cells.
	Fig 8: Depiction of interaction of Relenza (GG 167) in the neuraminidase binding site6

References

1): K. J. Lui and A. P. Kendal, Am. J. Public Health, 1987, 77, 712

2): Scheiget, Zambonis, Bernstein and Roy, Org. Prep. Proced. Int., 1995, 27, 637- 644

3): Glaxo Wellcome Inc. Relenza� (zanamivir for inhalation) [package insert]. Research Triangle Park, NC: Glaxo Wellcome, Inc., 1999

4): N Seppa, Scientific American, July 10th 1999, Volume 156

5): L. Gubareva, Lancet, March 4th 2000, 355: 827-35

6): J. Medicinal Chemistry. 1999, 42, 2332-2343

7):P Smith, S Sollis, P Howes, P Cherry, I Starkey, K Cobley, H Weston, J Scicinski, A Merritt, A Whittington, P Wyatt, N Taylor, D Green, R Bethall, S Madar, R Fenton, P Morley, T Pateman, A Beresford. A. J. Med. Chem, 41, 1998, 787-797

8): C Kim, W Lew, M Williams, H Liu, L Zhang, S Swaminathan, N Bischofberger, M Chen, D Mendel, C Tai, G Laver, R Stevens, J Am Chem Soc, 1997, 119, 681-690

9): P Smith, J Robinson, D Evans, S Sollis, P Howes, N Trivedi and R Bethell, Bioorganic and Medicinal Chemistry Letters 9, 1999, 601-604

10): A. J. Hay, A. J. Wolstenholme, J. J. Skehel and M. H. Smith. EMBO J,. 1985, 4, 3021: L. J. Holsinger and R. A. Lamb, Cell, 1992, 69, 517

11): J. C. Stoof, J. Booij, B. Drukarch and E. C. Wolters, Eur. J. Pharmacol., 1992, 213, 439

12):  W. Graeme Laver, Norbert Bischofberger, and Robert G. Webster, Perspectives in Biology and Medicine 43.2 (2000) 173-192. This can be seen by visitinghttp://www.press.jhu.edu/journals/perspectives_in_biology_and_medicine/v043/43.2laver.html   nmr

13): M. Chandler, M. J. Bamford, R. Conroy, B. Lamont, B. Patel, V. K. Patel, I. P. Steeples, R. Storer, N. G. Weir, M. Wright, C. Williamson, J. Chem. Soc. Perkin Trans. 1, 1995, 1173- 1180    nmr  synth

14): A. E. Miller, J. J. Bischoff, Synthesis, 1986, 777- 779

15): G. D. Allena, S. T. Brookesa, A. Barrow, b, J. A. Dunnc and C. M. Grossec, Journal of Chromatography B: 1999, 732, 383-393