O OH

N NH
N N MeO
N NH2 O O

O MeO O N NH2
H
OH

Pharma Disconnections
NC N
N
HN Prof Edward A. Anderson N
HO

Synthesis for Biology and Medicine CDT CN

O N
H
OH October 26th-27th 2016

O
N MeO
H
N N
N N O
N F

HN
O N
N O
Aims of the Course

• Cover ‘classical’ ways to disconnect molecules, which are fundamental to organic

synthesis

• Develop an understanding of how these disconnections link to heterocycle

synthesis

• Recognize how multiple solutions are available for given target molecules, and how
to critically assess retrosynthetic options

• Interactive Problem session: Extended practice with ‘classical’ disconnection

problems through teamwork

2
Retrosynthesis: Course Outline

The principles of retrosynthesis

• Introduction to retrosynthesis: The origins of the subject, nomenclature, how we go about breaking up
molecules

• Key strategic considerations: Functional group interconversions, selectivity, protecting groups

• Synthons and Reagents: Acceptor and Donor synthons – how to recognise these disconnections and the
importance of (di)oxygenation relationships

• Focus on specific reaction types and methods to install functionality (e.g. amines, alkenes, cycloadditions,
cross-coupling)

• Worked examples throughout these topics

• Interactive Problem session: Extended practice with ‘classical’ disconnection problems

Retrosynthesis in medicinal chemistry

• Coverage of 5- and 6-membered aromatic heterocycle disconnections with worked examples

• Advanced disconnections of arenes

• Asymmetric synthesis including worked examples

• Case studies to illustrate principles

• Interactive Problem session: Extended practice with a range of pharmaceuticals

3
Retrosynthesis: Supporting Material

Stuart Warren’s ‘Disconnection

approach’ is the classic text on
this subject, and highly
recommended!

Joule and Mills’ classic

Heterocyclic Chemistry textbook
also provides a useful introduction

4
Introduction: The history of retrosynthesis
• The birth of retrosynthesis can be traced to Sir Robert Robinson’s landmark synthesis of the alkaloid tropinone
in 1917.

Recognition of available
starting materials!

Me O
N

NHMe2 O

O O

R. Robinson, J. Chem. Soc., Trans., 1917, 111, 762

5
Introduction: The history of retrosynthesis
• The birth of retrosynthesis can be traced to Sir Robert Robinson’s landmark synthesis of the alkaloid tropinone
in 1917.

Me
N O

NHMe2 O

O O

In the event, use of the calcium salt of acetone dicarboxylic acid at neutral pH facilitates the Mannich reactions:

O CO2Ca Me
N
pH 7, H2O
NHMe2 O
92%
O CO2Ca
O

R. Robinson, J. Chem. Soc., Trans., 1917, 111, 762; C. Schöpf, G. Lehmann, Leibigs Annalen, 1935, 518, 1.
6
Introduction: Getting started!
• Retrosynthesis arrows:
Forward reaction (Synthetic connection)
Retrosynthetic disconnection

• A disconnection is the reverse of a known reaction (known by YOU!)

• WHERE should we start to disconnect molecules?

Example 1: Synthons Reagents

O HO NaH,
THF or
Br DMF
a
O
d O

Br
O OH NaH,
THF or
DMF
d a Risk of E2; SN2 on 2° halide not great!

Key principle: Disconnect at a bond that is EASY TO MAKE; it may be a strong bond!

• A synthon is an imaginary (charged) species formed by the conceptual process of breaking a bond in a molecule
(retrosynthetically). Our job is to identify real-life reagents that correspond to these synthons.
• Synthons are assigned labels (a (acceptor), d (donor)) so we can talk about what type of disconnection we are using.
7
Introduction: Synthons
• A synthon is an imaginary (charged) species formed by the conceptual process of breaking a bond in a molecule (retrosynthetically). Synthons
are assigned labels (a (acceptor), d (donor)) so we can talk about what type of disconnection we are using. The subscript number refers to the
position of the charge relative to a heteroatom that forms part of the functional group.

OH OH
Acceptor Synthons:
a a1 a2

O O
Donor Synthons:

d d1 d2

Reagents (e.g.):

HO
Example 2: OH
+ / H+ (Fischer esterification)
Synthons: O

O O
or Cl / Et3N;
O O O or NaHCO3 aq. / Et2O
"Schotten-Baumann conditions"
a1
d
Bu O Bu N
or • Why use pyridine in this
reaction?
O O

8
Introductions: Functional group interconversions
• In addition to disconnections that break / make bonds, we can carry out functional group interconversions that change either the nature, or
number, of functional groups present in a molecule.

C–O
O ester O Problem:
(a / d) • We cannot directly introduce either of these
OEt OH EtOH
functionalities
H2N H2N • The acid group is meta-directing in SEAr
benzocaine
FGI

C–N
O nitration
FGI Me (a / d) Me
OH
O2N
O2N O2N
a d

Solution: Interconvert functional groups to enable ring functionalization (FGI)

Synthesis:

O
HNO3 / H2, H+ /
Me H SO Me KMnO4 CO2H cat. Pd/C CO2H EtOH
2 4 OEt

O2N O2N H2N H2N

benzocaine
• Me is o,p-directing; p-nitration
favoured by sterics • Amine does not react as it is
• Mono-nitration as product less protonated (protected!) under
reactive than sm the reaction conditions
9
Order of events
One of the crucial considerations in planning a synthesis is to consider the order of events of the planned reactions. This is particularly important in
the manipulation of aromatic rings where regioselectivity is paramount; directing effects rule!

O deactivating, meta-directing
deactivating, o,-p- directing Cl
see Appendix for
aromatic chem recap
activating, o,-p- directing

The best electron donating group (activator) is the one that ‘wins’ –
stabilisation is more important than ‘lack of destabilisation’

Q: Should we acylate or chlorinate first?

A: Sterics suggest we should acylate then chlorinate; the alkyl group favours p-substitution so chlorination first would give poor selectivity.

Q: How do we introduce the alkyl group?

A: NOT Friedel-Crafts alkylation...

AlCl3 rearrangement
Cl3Al H NOT
Cl Cl

t-Bu t-Bu
polyalkylation
t-Bu
product is more
reactive than sm

Key principle: To achieve / guarantee mono-substitution, the product must

be less reactive than the starting material...
10
Order of events
One of the crucial considerations in planning a synthesis is to consider the order of events of the planned reactions. This is particularly important in
the manipulation of aromatic rings where regioselectivity is paramount; directing effects rule!

O deactivating, meta-directing
deactivating, o,-p- directing Cl

activating, o,-p- directing

• Solution: Use Friedel-Crafts Acylation then remove carbonyl group. This again uses the important concept of FGI to reintroduce functionality into
the molecule to enable control.

C–C
O C–Cl O (a1 / d) O
Cl (a / d) F–C acylation
Cl

Cl2 / FeCl3 AlCl3

Wolff-Kishner redn.
NH2NH2 / OH– / heat

FGI or Clemmensen
Zn / HCl
C–C
(a1 / d) or H2, Pd/C
F–C acylation
O Cl O

AlCl3

11
Exercise
Plan a synthesis of the following molecule from m-cresol:

NO2

NO2
OMe OH
m-cresol
A synthetic musk (available)

Retrosynthesis:

NO2
2 x C–N C–C
nitration F-C alkylation Cl
=
NO2
OMe OMe OMe OMe

Note: this is a ‘good’ F-C disconnection

as cation rearrangement is unlikely

Synthesis:
NO2
K2CO3 / MeI
or (MeO)2SO2 AlCl3, t-BuCl HNO3

Friedel-Crafts H2SO4 NO2

OH OMe alkylation via OMe OMe

regioselectivity: ortho to
MeO, not Me (sterics)

12
Selectivity
Control / selectivity is paramount in planning a synthesis. There are three main considerations we need to think about:

1a. Chemoselectivity. WHICH group will react?

a) Differentiating between different functional groups:

O H O O
NH2 N NaBH4
Cl
? O
OMe ? OMe
HO py HO O OH

The amine is more nucleophilic than the alcohol... The ketone is more electrophilic than the ester...
(see Appendix for stereoelectronic discussion) (see Appendix for stereoelectronic discussion)

b) Differentiating between the same functional group in different environments:

TBSCl / 1 equiv. TBAF

OH OH imdazole TBSO OTBS or TBSO OH 1. PCC OH O
Classical:
R R PPTS, R 2. TBAF R H
CH2Cl2 / MeOH

OH OH OH O
Contemporary, cat. , PhI(OAc)2
chemoselective: N
R R H
O
TEMPO / BAIB

TEMPO / BAIB mech: e.g., Eur. J. Org. Chem., 2006, 4323 13

Selectivity
1b. Chemoselectivity: over-reaction.
O O
NH2 NH2
NH NH
Br2 / FeBr3 Br Br
Br2 / FeBr3

Br
Br

The aniline nitrogen atom is such a powerful activating The acetamide nitrogen atom is now a weaker
group that it is difficult to prevent over-bromination, activating group due to conjugation. Mono-
despite the lower reactivity of the product. bromination can now be achieved.

Key principle: Modifying functional groups (by FGI) can help us to control / influence selectivity

... or an alternative disconnection?

NO2 NH2
HNO3 Sn / HCl

Br Br Br

Mono-nitration occurs as product is less reactive than sm.

Reduction reveals the powerful amine activating group

Key principle: To achieve / guarantee mono-substitution, the product must

be less reactive than the starting material...

14
Selectivity
1b. Chemoselectivity: over-reaction.

O O
Br2 Br

OH– or H+ ??
The haloform reaction can be
quite a useful route to
• Under basic conditions: carboxylic acids!

O Br
O O O OH O
Br
H Br Br
OH
OH CBr3
Br Br

pKa = 25 pKa = 21

Carbonyl less basic due to

inductive EWG from Br!
• Under acidic conditions:
OH2 Br
O OH OH O
Br
H H+ H Br
–H+

Key principle: To achieve / guarantee mono-substitution, the product must

be less reactive than the starting material...

Key principle: Thinking about mechanism can also help us to control selectivity
15
Protecting groups
Protecting groups provide an obvious way to protect reactive functionality and achieve chemoselectivity. The cost is at
least one additional step in a synthesis, usually two.
• Sometimes the deprotection of protecting groups can be the most challenging step in a synthesis!
• Protecting group-free synthesis preferred – and may be achieved using a careful ordering of steps...

See Appendix for a short summary of PG methods

1. Protection for alcohols

• Silyl ethers • Benzyl ethers • Esters / carbonates • Acetals • Other ethers

O Ph
SiR3 O Ph
O Me
O R O O O O O Ph
X

2. Protection for amines 3. Protection for carbonyls (and diols!) 4. Protection for carboxyls

• Carbamates • Sulfonamides • Acetals • Esters / carbonates

O O O O
R S O O
HN O HN R O R

16
Introduction: Selectivity
2. Regioselectivity. WHERE will the molecule or functional group react?

a) Aromatic compounds: Directing effects

X
• We have already looked at this. The best cation
stabilising group wins!
Y

b) Hard vs Soft

HO H HO Me
NaBH4, CeCl3,
MeOH, –78 °C MeLi or
O MeMgBr
(Luche)

O O
[Ph3PCuH]6 Me2CuLi or
(Stryker) MeMgBr, cat. CuI
H Me

• 1,2-Addition = ‘hard’: These reactions are charge controlled and depend on electrostatic attraction of nucleophile and
electrophile. Highly charged / charge dense reagents.

• 1,4-Addition = ‘soft’: These reactions are orbital controlled and depend on the overlap and energies of the HOMO and
LUMO. Neutral / polarisable reagents.

Many other examples such as: Enolisation (see d2): Control over the site of enolisation for non-symmetrical carbonyls; Elimination (e.g.
Hofmann vs Saytzeff elimination), etc.

17
Introduction: Selectivity
3. Stereoselectivity / Stereospecificity: We may need to control stereochemistry in a given step

Stereoselective: One stereochemical outcome is favoured over two (or more) possible outcomes

Examples: generally stepwise (non-concerted) processes

E1 elimination
Felkin-Anh model (addition to α-chiral carbonyls)
Chiral auxiliary-mediated reactions
Carbonyl olefination (e.g. Wittig, etc.) O RM O attack HO Nu
RL Nu hindered RL
The facial selectivity of reactions of alkenes attack R
R favoured! RL
The facial selectivity of reactions of bicyclic systems RM RS RM RS
Nu Nu
RS
Diels-Alder reaction (endo vs exo) R
Aldol reactions (syn vs anti)

Stereospecific: A reaction where only one stereochemical outcome is possible due to the reaction mechanism.

Examples: generally concerted processes

SN2
E2 EDG EDG EDG
Ring-opening of epoxides EWG H EWG
Diels-Alder reaction (and pericyclics in general) EWG H
Electrophilic addition to alkenes
Hydrogenation of alkenes
Stereo- and regioselective reaction, but
Many rearrangements (e.g. Baeyer-Villiger)
stereospecific reaction mechanism
18
The ideal synthesis
1. Linear vs. convergent strategies: Key Principle: make disconnections in the middle of molecules!

6 4
SM 1 SM 1
13 1 4 1
Linear: SM 1 Target Convergent: Target Multicomponent: SM 2 Target
6 4
SM 2 SM 3
13 steps @ 90% = 0.913 = 25% yield
6 steps @ 90% + 1 step @ 90% 4 steps @ 90% + 1 step @ 90%
= 0.97 = 48% yield = 0.95 = 59% yield

2. Stereocontrol: Substrate stereocontrol, chiral auxiliaries, catalyst stereocontrol, chiral pool, enzymes

3. Atom economy: (B. M. Trost, Science 1991, 254, 1471) and Green chemistry – minimise waste! Compare Wittig / olefin metathesis –
Catalytic vs stoichiometric:
R'
PPh3 O R'
R O R + PPh3 vs. R R +
R' R'
Ru cat.
MW = 262 MW = 28

4. Step economy: Use of cascade strategies, multicomponent reactions, no protecting groups. Aim for e.g. 10 steps in total from cheap
starting materials (not possible for really complex molecules).

5. Redox economy: (N. Z. Burns, P. S. Baran, R. W. Hoffmann, Angew. Chem. Int. Ed. 2009, 48, 2854): Installing functional groups at the required
oxidation state avoids excessive manipulations and protecting groups strategies.

O
1. O3; PPh3 [OX]
2. Ph3P=CO2Me Grubbs II cat. Contemporary,
Classical: R R O R
redox economic:
3. LiAlH4 [RED] atoms introduced at
4. PCC [OX] ?? required oxidation level
19
Acceptor synthons

a synthons: Essentially alkyl halides, etc.

X
Synthon: R Reagents: R X = I, Br, Cl, OTs

Synthesis of these reagents is most commonly achieved from alcohols:

SOCl2
or POCl3 TsCl, py, CH2Cl2
R R R
Cl OH OTs
or CCl4 / PPh3

or CBr4 / PPh3 NaI, acetone

PBr3 (Appel reaction) (Finkelstein reaction)
or Br2, PPh3, imidazole

R R (use LiBr or LiCl to

Br I install other halides)

a1 synthons

OH OH O O

RO

O O O O O O O
Reagents =
H X Cl OR N CO2 RO X
OMe
O O
=
H X H N

20
Acceptor synthons
Synthons Reagents
a2 synthons
OH OH
OH O OH Problem:
Ph Ph Regioselectivity
O Br
Ph d a

OH O O HO
Ph Ph
a2 d

OH
NaH, THF; then
Synthesis: HO O
Ph
O
Ph

O O
R Problem:
O OH Cl R Synthesis of
Ph Ph
hydroxyketone?
O a1 O
d O
O R

O O O
O R
HO R
Ph O Ph
a2 d
Br O

O O Note: Poor nuc. so

NaHCO3 not suitable for
Synthesis: Br O R
epoxide disconn.
Ph HO R Ph
as used above
O
O
21
 Exercise
Plan a synthesis of the the Pfizer anti-fungal agent fluconazole from 1,3-difluorobenzene and 1,2,4-triazole using two a2 disconnections

C–N
N OH N
N N F H
N N
F N N
Fluconazole N
F
F

N
N
N
N N a2
N N
a2 OH O O
O FGI C–N
N N N
N N N F F
F = F
N N

F F
F F N N
N N
Synthesis: N N
H O O H
O
O Cl N N 1. Ph3P=CH2 N N
F Cl F F 2. m-CPBA F
Cl N N
Fluconazole
AlCl3 or K2CO3, DMF,
O CTAB (R4NBr) heat
F F F NaOH, tol, F
S
100 °C

O O O
Corey-Chaykovsky
S epoxide
S
epoxide synthesis O
US Patent 4,404,216 22
Donor synthons
These are typically carbanions or anionic / neutral heteroatoms. Here we have two considerations: is the reagent hard or soft
d synthons: (i.e. does it match the electrophile); and, can multiple reactions occur?

Carbon nucleophiles: Heteroatom nucleophiles:

Hard: organolithium, organomagnesium. Hard: alkoxides, amide anions.
Soft: cuprates, malonate, nitroalkane anion Soft: sulfides, thiolate anions, amines, alcohols

Amine synthesis: As amines are such good nucleophiles, polyalkylation is a big risk!

C–N alkylation
Et or Et
Ph N Ph NH2 I Ph N Ph N I
H
Et Et Et

... different solutions are required for the synthesis of highly nucleophilic amines and sulfides!

a) Synthesis of sulfides:
i) Using thiourea: S
NaOH / H2O
NH work-up
Na2S H2N NH2
S Br S NH2 SH

Not isolated

ii) Using thioacetate:

AcS Na O K2CO3, MeOH
Br S SH

Can be isolated! (and does not

smell) – protected thiolate anion
23
Donor synthons
b) Synthesis of 1° amines:

Step 1: Introduce N Step 2: Reveal Amine

NaCN LiAlH4 or H2, Pd/C

Polyalkylation can R
N
be a problem....
NH2
R

NO2 NO2 H2, Pd/C

R
or TiCl3, H2O (McMurray)
or H+/H2O (harsh)

NaN3
PPh3; H2O (Staudinger)
DMSO
R Br R N3
or H2, Pd/C

TsBocNH, Mg, MeOH

PPh3, DIAD then TFA
R NTsBoc R NH2
(Mitsunobu)
Mitsunobu: pKa must
be <10 for success! O NH2NH2,
(Gabriel) heat
R N
O
O
NaN

24
 Donor synthons
b) Synthesis of 1° amines:

NH2OH•HCl
H base (e.g. NaOH) H LiAlH4
• From oximes:
OH R NH2
R O R N

O OH– / H2O NH2

PhO P N R
• From acids (Curtius): OH PhO 3
N
R C
R O (DPPA) O H
N Ot-Bu
R
t-BuOH
O

c) Synthesis of 2° and 3° amines:

There are two important methods for the synthesis of secondary amines, both of which feature reductive processes that again avoid revealing
the amine too early in the reaction.

i) Reductive amination of aldehydes and ketones

R2 NH2 R2 NaBH3CN R2
O N HN chemoselective reduction
of imine over aldehyde
R R1 mild H+ e.g. NH4+ R R1 acid-stable mild R R1
or AcOH reducing agent

R2 R3
N R2 R3 R2 R3 R2 R3
O H N N NaBH3CN N
R mild H+ e.g. NH4+ R R R
R1 R1 R1 R1
or AcOH

Review of synthesis of 2° amines: Salvatore et al., Tetrahedron 2001, 57, 7785. 25

Donor synthons
ii) Reduction of amides

O O
O
R2 Cl LiAlH4 R3 Cl R1
R1 R1 N R2
R1 NH2 N R2 N R2
H H
O R3
Selective mono-acylation

n-BuLi or NaH LiAlH4

O
This trick is relevant for the R3 I O LiAlH4 R1
R1 N R2
alkylation of all amide derivatives N R2 R1
N R2
R3
R3

iii) Other methods:

R1 O
OH R1 N
Branched amine R1 NH
N
synthesis: the Ritter H+ / H2O
reaction

Methylamine synthesis: R1 H H R1 R1
N R2 N R2 N R2
the Eschweiler-Clarke reaction H O Me
H
O H stable iminium ion
26
Exercise
Plan a synthesis of the sedative captodiamine from thiophenol, HSC6H5

NMe2
S

S
captodiamine
C–S
d/a
Na2CO3
2-chloroethyl
dimethylamine

SH NMe2
Cl
Bu
S

thiourea; FGI
OH–

C–C C–S
FGI d / a1 d/a
Cl O
Bu Bu
Bu S S HS
S
NaBH4; PhCOCl, Na2CO3,
SOCl2 AlCl3 BuCl

27
Exercise
Plan a synthesis of the following amine from any available starting materials!

N Ph

suggests epoxide?

C–N H C–N
N Ph O HN Ph O H2N Ph

NaBH3CN / H+ NaBH3CN
/ H+
Disconnect at centre
FGI (COCl)2, DMSO;
of molecule
Et3N

C–C
a2 / d

OH Br
Mg O

28
Donor synthons 2: Enolates
An extremely useful disconnection of carbonyls via the enolate. Enolates of almost all carbonyl derivatives are used for both
d2 synthons: alkylation, aldol, and Michael addition chemistry.

Alkylation Multiple FGIs will

O be required for this

This ketone is symmetric so we do not

a1 d
need to worry about regioselectivity

O O
O

I
d2 a
Reaction conditions:
LDA, THF, –78 °C;
O O then add iodide
I
d2

a CH3
This ketone is not – so the alkylation
disconnection will never work!

Key principle: Look for symmetry to help with disconnections!

Note that alkylation doesn’t always work: the following two compounds are challenging for ‘normal’ enolate alkylation...

O
O

29
Donor synthons 2: Enolates
Specific enol equivalents: When the direct alkylation won’t give good selectivity, or in the absence of symmetry, we need to improve the
regioselectivity of alkylation. For this, we can use a ‘specific enol equivalent’ – a compound with the reactivity of an enol that affords control.

a) Silyl enol ethers

O O
vs

Kinetic lithium enolate EtI

Li
O OTMS

LDA, THF, TMSCl

–78 °C
O ‘Thermodynamic’ lithium enolate

Li
OTMS O O
MeLi EtI
Et3N,
DMF,
TMSCl

Thermodynamic silyl enol ether

• Silyl enol ethers are also useful (required!) for (Mukaiyama) aldol chemistry...

30
Donor synthons 2: Enolates
b) 1,3-dicarbonyls (malonates and beta-ketoesters). These enable regioselective enolisation and are also ‘soft’ (i.e. good!) nucleophiles

d2 d2
O FGI O O O O O

OMe OMe

O OMe
Problem: two possible a
sites of enolisation... a

Key principle: Re-introducing a directing functionality can greatly facilitate a synthesis!

Br O O O
O O NaH O O NaH;
OMe
OMe OMe Br
alkylation 1 O OMe
alkylation 2
pKa ~ 12
Allyl bromide is the
better electrophile

O O O O
H H H+ / heat
O

O O O

Key principle: Introduce the better electrophile / nucleophile in the more difficult step

31
Donor synthons 2: Enolates
b) 1,3-dicarbonyls (malonates and beta-ketoesters)

Note: A trick for alkylation at the other position: form the dienolate:

Br
O O NaH O O BuLi O O O O

OMe OMe OMe OMe

pKa ~ 12 pKa ~ 35
“Last in – First out”

c) enamines. These ‘soft’ enol equivalents can be isolated, and are useful to control site selectivity of alkylation. The resultant iminium ions
hydrolyse on work up. For aldehydes, they provide an attractive alternative to aldehyde enolates, which are prone to undergo self-aldol reaction.

O N N N O
H Et–I H+ / H2O
mild H+

1,3-allylic
strain N
This isomer is disfavoured by 1,3-allylic strain

...although enamines can be useful for alkylation, they are potentially more useful to control selectivity in aldol- or Michael-type processes...
32
Donor synthons 2: Enolates
The Aldol and Michael Reactions
This will be one of the main methods by which we assemble heterocycle precursors!
Example 1:
Reagents:
C–C
O FGI (aldol)
OH O OH O O O
Ph rehydration Ph Ph Ph H
a1 d2

This aldeyde is This ketone is

non-enolisable! symmetrical!
O O NaOH aq. O
Synthesis: This is a good aldol disconnection as we have
Ph H Ph excellent control over the reacting partners.

Example 2: Dieckmann cyclisation

(intramolecular Claisen)

O O O O O O C–N O O
C–C O
(aldol) d / a3
OEt OEt EtO OEt EtO OEt
= = OEt
N N N N
a3

O O O O O O
O EtONa
Synthesis: EtO OEt OEt OEt
MeNH2
OEt EtOH
N N N

Note: EtO– used as base to Note: Reaction driven by

‘avoid’ transesterification product deprotonation
33
Acceptor synthons 2: Conjugate Addition
In the previous slides, we used a conjugate addition reaction of an amine and an α,β-unsaturated ester to construct a
a3 synthons: piperidone. We also used a 1,5-dicarbonyl as a source of a 6-membered ring via aldol chemistry. Where did this 1,5-
dicarbonyl come from? In both cases we needed an a3 synthon – a Michael acceptor.

O O O O
Example: Problem: The d2 reagent must be soft =
HO MeO
to avoid competing 1,2-addition
a3 d2 O OMe

O C–O O
(d / a2) O OH FGI O O Solution: Use a
O O =
HO HO 1,3-dicarbonyl

Key principle: Re-introducing functionality can alter O O O O

reactivity and therefore selectivity =
HO MeO
d2 a3
MeO O
Synthesis:
LiCl,
O O O O wet DMSO, O O O
NaOMe heat NaBH4
MeO MeO MeO O
Krapcho chemoselective
O OMe O O reduction
Me Cl

O O
NaOH H+ / heat
HO

O OH 34
Exercise
Plan a synthesis of the following diketone from any available starting materials!

O C–C C–C O
aldol O Michael

O NaOEt O O NaOEt O O

O
Note: Aldol is under TD control. Na
C–C O
Enolisation at other sites leads NaOEt aldol
to strained rings (i.e. reversible) Note: product
‘protected’ as
enolate
O

OEt

Note: NaOEt ‘matches’ ester,

which is non-enolisable

35
Donor synthons 3: Umpolung
Synthons such as d1 represent the reversal of natural polarity of the functional group. O O
d1 synthons These synthons are very useful as they increase the number of methods available for R' R'
bond formation! R R
d1 a

a) Dithiane: A classical umpolung reagent, which truly converts a carbonyl into the umpolung synthon, then back to C=O.

Note: The dithiane anion is quite ‘hard’

PIFA
R' Br (PhI(O2CCF3)2)
O SH SH S n-BuLi S S S O
R' R'
R H BF3•OEt2 R S R S R or R
H Hg2+ / H2O
pKa ~ 31 O
= acyl anion Raney Ni
R'
equivalent R
R

b) Nitroalkanes: A nitroalkane, although not prepared from the original carbonyl or amine, can serve as a versatile d1 equivalent.

TiCl3 / O O
Problem: polyalkylation possible =
H2O R'
R R
NO2
Br R' R'
R NH2 NH2
=
NO2 weak base NO2 H2, Pd/C R'
R R
R R
pKa ~ 10 R
O R' R' H2, Pd/C NH2
O2N R'
R
Note: The nitroalkane anion is ‘soft’
36
 Donor synthons 3: Umpolung
Synthons such as d1 represent the reversal of natural polarity of the functional group. O O
d1 synthons These synthons are very useful as they increase the number of methods available for R' R'
bond formation! R R
d1 a

c) Cyanohydrin and related umpolung agents (Thiazolium salts, other heterocyclic carbenes)

Recall the benzoin condensation:

O

O NaCN HO CN OH– HO CN R H HO CN OH– O

R R
R H R H R R R
OH OH

This classical reaction has been superseded by soft carbene nucleophiles such as:

Cl
N-Heterocyclic carbenes
Mes N N Mes

HO These reagents can effect benzoin

Cl type reactions, Stetter reactions (d1
Thiazolium salts
Readily conjugate addition), Baylis-Hillman
S N Bn (thiamine mimics)
deprotonated at type reactions, and so on. See 1,4-
dicarbonyl synthesis, later....
O BF4
N
Triazolium salts
N N C F
6 5

Reviews: K Scheidt et al., Angew. Chem. Int. Ed. 2012, 51, 11686.
A. Grossmann and D. Enders, Angew. Chem. Int. Ed. 2012, 51, 314 37
Latent functionality
In the previous section, we saw the nitro group serving as an umpolung reagent which was later ‘revealed’ to be a ketone (or amine). Although on the
face of it this is just an FGI, the nitro group was serving as a ‘latent’ carbonyl. A more accurate definition of ‘latent’ functionality is the introduction of a
reactive functional group in unreactive form, that can be revealed at a chosen stage of a synthesis.

Example 1: Alkenes
Typical Reagents / Synthons

Br MgBr MgBr

O O O O
a2 d2 d1
O3; PPh3 R

BH3; R
R a3 d3 d2
H2O2, OH– OH OH OH OH

O O O O
a2 d2 d1
Pd(II) / Cu(II),
R
O2, H2O

The alkene serves as a latent carbonyl or alcohol, and as an acceptor or donor synthon; it is
unreactive towards conditions which the carbonyls and alcohols are reactive.

Example 2: Silanes
TBAF; then H2O2,
SiMe2Bn KHCO3, MeOH, THF OH
...note analogy to protecting groups...
R R' R R'
38
Exercise
Plan a synthesis of the following enone from any available starting materials!

Ph

O Note symmetry in the

enamine disconnection
Ph
C–C a2 d2 O

O O
C–C O
Ph + EtO
or
Ph
Br N O O
Ph O

FGI C–C

O
O HO
O
Cl Ph
Ph Ph
O d1 S N Bn
NO2 a3
O

Ph
NO2

39
Alkene / Alkyne synthesis
Alkenes are not only useful as masked functionality, or as a source of diols, epoxides, etc., but are also found in many bioactive targets, particularly
natural products. An example of this is Novartis’ everolimus. Although not directly prepared by synthesis, its parent rapamycin has been synthesised
a number of times, and chemists need to address this alkene synthesis. Here we recap some important methods for doing just that.

O
HO

MeO

Everolimus
(immunosuppressant) O O OH
(Novartis) N
O O O
O MeO
HO
O OMe

Olefination methods:

• Aldol / dehydration
• Heteroatom mediated: Wittig, Horner-Wadsworth Emmons, Julia, Peterson
• Metathesis
• Cross-coupling (for dienes etc.), Heck reaction
• Alkyne semihydrogenation
• Metal-mediated olefinations: (Takai, etc.)

Key Principle: Aside from metathesis, aldehydes are key precursors to alkenes and alkynes

40
Alkene / Alkyne synthesis
Heteroatom Method 1: Wittig

Br 1. NaH, THF
Non-stabilized ylid:
R1 PPh3 R1 R2 (Z)-selective
2. R2
O

Br 1. NaH, THF EWG Stabilized ylid:

EWG PPh3 (E)-selective
2. R2 R2
O

EWG = ester, ketone, aldehyde, nitrile, aryl

Heteroatom Method 2: Horner-Wadsworth-Emmons: A VERY NICE WAY TO MAKE MICHAEL ACCEPTORS!

O O
EtO P R2 electron-rich
EtO O
R3 phosphonate:
R1 O R1 R2 (E)-selective
base (e.g. LDA,
NaH, MHMDS) R3

O O electron-poor
EWG–O P R2 phosphonate:
EWG–O R1 (Z)-selective
R1 O
base (e.g. KHMDS) R2 O

For mechanisms:
EWG = CH2CF3 Still-Gennari
See Appendix EWG = Ph Ando
41
 Alkene / Alkyne synthesis
Heteroatom Method 3: Julia

Base (e.g. n- R1
BuLi, LDA); 1. Ac2O, py
Ph R2 Ph R2
R1 S S R1
H 2. Na/Hg
O O HO O O
R2 O

Highly (E)-selective

Heteroatom Method 4: Modified Julia

t-Bu
N N
N N R2 O R2 O N N R2 O
N Base (e.g. LDA); N N N N R2
R1 S N N R1
H R1 S N R1 S N N R1 S
O
O O t-Bu O2
O O t-Bu t-Bu O
R2 O

One step – but selectivity can be variable

A heteroaromatic group capable of acting as an

electrophile. Commonly phenyltetrazole (PT),
benzothiazole (BT), t-butyltetrazole (TBT)

Recent review: C. Aïssa, Eur. J. Org. Chem. 2009, 1831 42

 Alkene / Alkyne synthesis
Metal-Mediated reaction 1: Metathesis

Mes N N Mes Mes N N Mes

Cl Cl
Catalysts of choice: Ru Ru Drawback:
Cl Ph Cl
PCy3
i-PrO

Grubbs II Hoveyda-Grubbs II

Classes of alkene for cross-metathesis:

Type 1: Rapid homodimerization, homodimers consumable e.g. R1 Ar

R2
Type 2: Slow homodimerization, homodimers sparingly consumable e.g.
O
Type 3: No homdimerization R2
e.g.
R1
Type 4: Olefins inert to CM but do not deactivate catalyst e.g. NO2

Examples of reactions for which metathesis is particularly useful:

RCM for ring sizes ≥5. Particularly useful for medium

X X ring synthesis, and macrocycles. And normal rings.

R2 R2 High E-selectivity. For a beneficial effect of CuI on

R1 O R1 O this reaction (yield / rate) see JOC 2011, 76, 4697

For classification of alkenes in cross-metathesis, see: Grubbs et al. JACS 2003, 125, 11360
For a review of cross-metathesis, see: Blechert and Connon, Angew. Chem. Int. Ed. 2003, 42, 1900 43
Alkene / Alkyne synthesis
Metal-Mediated reaction 2: Heck Reaction Ar–I, cat. Pd(PPh3)4,
NaHCO3, Bu4NCl,
O DMF Ar O

OR OR

cat. Pd(0)
Metal-Mediated reaction 3: Cross-coupling R1 R2 R2
X M R1

See appendix for synthesis of alkenyl halides and metals

Conditions Catalysts: H2, Pd/BaSO4 or Pd/CaCO3

H H
Poisons: quinoline, ethylenediamine
Metal-Mediated reaction 4: Lindlar Hydrogenation R R'
R R' P-2-Ni: Ni(OAc)2 + NaBH4, MeOH

R O Conditions Conditions
Synthesis of Alkynes: R H R R'
H

• Corey Fuchs: CBr4 / 2 PPh3 (often w/ Zn or Et3N) n-BuLi; R’–I

(or other electrophile)
• Ohira-Bestmann: • Gilbert-Seyferth:
Pd(0), CuI, Ar–I
O O (Sonogashira)
P(OMe)2 TMS
/ n-BuLi
N2 / K2CO3, MeOH N2 See appendix for
other methods
44
 Exercise
Plan a synthesis of terbinafine (lamisil).

Terbinafine (Lamisil)
(Antifungal)
(Novartis)

cat. PdCl2(PhCN)2 expensive

e.g. X = Br: K2CO3 C–C C–C cat. CuI
e.g. X = OAc: Pd(0) allylation Sonog. piperidine

NHMe N Cl expensive

FGI PBr3
E/Z imperfect
K2CO3, KI,
DMF

NHMe Cl Cl
OH

C–C

toxic expensive
O
TL 1996, 37, 57.
EP1753770 45
 Exercise
Process route:

dialkylated amine in organic phase

desired monoalkylated in aqueous phase
silylation improves amine stability

Cl Cl TMSCl, Et3N
NH2 HN Cl N Cl
TMS
(excess)

cheap! Cl

2 eq. n-BuLi
cat. Ni(PPh3)2Cl2 cheap!
Cl Cl FeCl3 Cl Cl Li

TBAB N
Terbinafine TMS
Cl

TL 1996, 37, 57.

EP1753770 46
 Diels-Alder disconnection
Diels-Alder chemistry allows us to construct cyclohexene systems with high efficiency and stereocontrol:

EDG EDG EDG

EWG H EWG
EWG H

Selectivity:
HOMO EDG
H
• Orbital energy considerations mean ideal DA setup is electron rich diene
(HOMO) + electron poor dienophile (LUMO) EWG H
LUMO (or π4s + π2s)

OMe OMe
EDG
H
• Regioselectivity reflects overlap of the largest coefficients of
EWG these orbitals
O O H

R R

(–) EDG
H

• Endo-stereoselectivity classically explained by secondary orbital effects, EWG

but is more likely due to electrostatic or steric effects (+)
H
H
H
47
 Diels-Alder disconnection
Example 1:
O N O Ar
O N O
H Check TS: all Hs syn
H H R O H in Endo TS (correct)
N H

O H
O2N O2N

O2N O2N O2N

PPh3Br Wittig PPh3Br
O O

cross-coupling Wittig
• enal easy to make by aldol / dehydration
• stereoselectivity of Wittig is wrong! FGI

Wittig O OH

OMe

Example 2:

H
HO FGI D-A?? [2+2] O • We need a
‘ketene equivalent’
HO O H
O O
cat. OsO4,
NMO
TiCl3, H2O ‘ketene DA
Note: stereoselective exo-face
equivalent’
dihydroxylation O
e.g.
Cl CN , NO2 NO2 ‘enamine DA
H2, Pd/C equivalent’
NH2
48
Exercise
Plan a synthesis of the following molecules using Diels-Alder chemistry

HO
OH OH
CO2Me

CO2Me
Whilst we could probably
OMe OH
HO do FGIs to make this work,
there is an alternative....
Recognise potential dienophile
HO 6 HO
OH OH OH
CO2Me – C–O CO2Me ?
O
CO2Me CO2Me
OMe OMe 1 OH OH
HO HO
– H+
1,6-O
O3; NaBH4 ozonolysis
OH OH
H
CO2Me
CO2Me OH
CO2Me CO2Et
MeO CO2Me
OMe
LiAlH4 CO2Et
D-A OH

D-A

OMe CO2Me
Good electronics! Alkene ozonolysis CO2Et
(furan e-rich) O can reconnect
1,6-dioxygenation EtO2C
CO2Me
49
Saturated Heterocycles
The synthesis of saturated heterocycles can be achieved in many ways; the common theme to most of these is to use the nucleophilicity of the
heteroatom.
1. SN2 Cyclisation onto halides, epoxides, etc.

basic conditions • Works well for 3, 5, 6 and 7-membered rings, and azetidines
X Cl X • Irreversible – SN2 reactivity – stereochemistry of sm
converted into products
basic conditions • However this means stereochem of sm must be controlled

X X
O
OH

2. Cyclisation onto Michael acceptors

cat. KOt-Bu, THF, • Particularly good for pyran synthesis.

0 °C O
R1 OH • Usually operates under thermodynamic control, so equatorial
R1 O OMe product is favoured.
MeO O

3. Palladium-mediated cyclisations

• Wacker or Tsuji-Trost type reactivity; stereochemistry

cat. Pd(II)
reoxidant controlled by equatorial disposition of sidechain; or potentially
R1 X by carbonate stereochemistry
R1 X R2
R2 Reviews: Muzart, J. Mol. Cat. A: Chemical 2010, 319, 1
Wolfe, Eur. J. Org. Chem. 2007, 571
cat. Pd(0)
R1 X General reviews of oxacycle synthesis:
R1 X R2 Larossa, Romea, Urpi, Tetrahedron 2008, 64, 2683
MeO2CO R2 Wolfe, Tetrahedron 2007, 63, 261
50
 Saturated Heterocycles
4. Electrophile-promoted cyclisation onto alkenes

I2
I • Iodoetherification or iodolactonisation, and other good electrophiles,
R1 OH R1 O promote this reaction. Stereoselective for equatorial sidechain.
R2 R2

5. Reductive cyclisation processes (cyclisation onto carbonyls)

H+ / NaBH3CN R1
R1 R2 H N R2 N R1
R2 H H
R1 N R2
NH2 O H H
Nu
axial delivery of nucleophile Nu

• Nucleophile delivered axial for stereoelectronic reasons

6. RCM

n
cat. Ru=CHR n
• Particularly useful for medium-sized rings!
N N
R m
R m

7. Hetero-Diels-Alder reactions and other cycloadditions

OTES OTES R1 R1
cat. Cr(III)(salen)
N
O R1 O EWG N
R2 R2 O R1 R2 R2 O EWG

For a review, see: Pellissier, Tetrahedron 2009, 65, 2839 51

Disconnection Strategy Summary

1. Make the synthesis as short as possible!

Linear Convergent Multicomponent

6 4
SM 1 SM 1

13 1 1
4
SM 1 Target Target SM 2 Target

6 4
13 steps @ 90% = 0.913 = 25% yield SM 2 SM 3

6 steps @ 90% + 1 step @ 90% 4 steps @ 90% + 1 step @ 90%

= 0.97 = 48% yield = 0.95 = 59% yield

2. Use only known disconnections

3. Disconnect back to oxygen where possible (provides better control)

4. Disconnect C–C bonds using available (or reintroduced) FGs, and:

5. Disconnect near the middle of the molecule (e.g at branch points, or disconnect rings from chains)

6. Use symmetry

7. Use the ‘best’ disconnection / forward reaction as late as possible in the synthesis

8. Use FGIs – but as few as possible

9. Look for available sms hiding in the molecule

52
Problem Session 1
Propose syntheses of the following compounds from starting materials of your choice

O O
OH O
a) b) CO2Et c) Ph OEt d)
OEt
OH O

Ph

O N Ph O
e) f) MeO g) h)
N
OAc O

O HO Ph
t-Bu
i) from t-Bu O j) k)
NH2

O
O
O
O O
OH Ot-Bu
l) O m) n) o) BocHN OH
NH
Ph

53
Problem Session 1
Propose syntheses of the following pharmaceutical agents from starting materials of your choice

Me
O N
H
O2N N
N
NH2
CN O
Cl HO
CO2H OH O
Baclofen Entacapone Tropisetron
(Novartis) (Parkinson's) (antiemetic)
(Novartis) (Novartis)

HO
HN
N OH
N
N NH2
H
Antazoline Maprotiline Fingolimod
(Antihistamine) (antidepressant) (Immunosuppressive)
(Novartis) (Novartis) (Novartis)

OBn

OH O
N HO
MeO O O
O
O
MeO

Vernakalant (benzyl ether) Mycophenolic acid

antiarrhythmic agent (Immunosuppressant)
Merck (Novartis)
54
Retrosynthesis: Course Outline

The principles of retrosynthesis

• Introduction to retrosynthesis: The origins of the subject, nomenclature, how we go about breaking up
molecules

• Key strategic considerations: Functional group interconversions, selectivity, protecting groups

• Synthons and Reagents: Acceptor and Donor synthons – how to recognise these disconnections and the
importance of (di)oxygenation relationships

• Focus on specific reaction types and methods to install functionality (e.g. amines, alkenes, cycloadditions,
cross-coupling)

• Worked examples throughout these topics

• Interactive Problem session: Extended practice with ‘classical’ disconnection problems

Retrosynthesis in medicinal chemistry

• Coverage of 5- and 6-membered aromatic heterocycle disconnections with worked examples

• Advanced disconnections of arenes

• Asymmetric synthesis including worked examples

• Case studies to illustrate principles

• Interactive Problem session: Extended practice with a range of pharmaceuticals

55
 Aromatic heterocycle synthesis 1: Furans, Pyrroles, Thiophenes
The general theme disconnecting aromatic heterocycles is to cleave C–X bonds within the ring, with a carbonyl precursor. Here we
recognise these C–X bonds as being enols, or imines / enamines (for O, N heterocycles).

Furans: Recognise 2 x enol in target

• Although not technically correct, this manner of
R1 O R2 heterocycle analysis can be very convenient.

2 x enol

FGI C–O O
OH R1 H+
R2
R1 O R2 R1 O R2 (–H2O) R1 O R2
O
1,4-dicarbonyl
(See earlier slides on carbonyl synthesis + next section)

Extensions: As more substituents are present, modify the 1,4-dicarbonyl synthesis!

O
FGI 2 x C–O O I O O
OH OEt
R1 R1
R2 EtO R2
R1 O R2 R1 O R2 O O
O OEt

Reality!: Outcome depends on electrophile and conditions....

CO2Et CO2Et CO2Et CO2Et

EtO– / EtOH Cl O O EtO– / EtOH
HO R2 R2
R2 EtO R2 NaI R2
O O O O
Cl O O

56
 Aromatic heterocycle synthesis 1: Furans, Pyrroles, Thiophenes
The general theme disconnecting aromatic heterocycles is to cleave C–X bonds within the ring, with a carbonyl precursor. Here we
recognise these C–X bonds as being enols, or imines / enamines (for O, N heterocycles).

Pyrroles: Simply use ammonia (or primary amine) to effect cyclization!

C–N C–N
imine R2 enamine R2
+ NH3
R1 N R2
O O
H R1 NH2 R1 O

FGI (tautomerise)

C–C Br
R2 d2 / a2 R2

O O
R1 NH R1 NH2

C–Br: SOFT

O O NH3 NH2 O NH NH2

cat. H+ R2 R2
R2 R1 R1 R2 R1 R1
Br Br O O

Enamine: SOFT
H NH3

R1 N R2 R1 R2 R1 R2 R1 R2
N N N
H H H OH2 H2 OH

57
 Aromatic heterocycle synthesis 1: Furans, Pyrroles, Thiophenes
The general theme disconnecting aromatic heterocycles is to cleave C–X bonds within the ring, with a carbonyl precursor. Here we
recognise these C–X bonds as being enols, or imines / enamines (for O, N heterocycles).

Thiophenes: Recognise 2 x thioenol in target – disconnect back to 1,4-di-carbonyl again!

2 x C–S O
R1
R2
R1 S R2
O
1,4-dicarbonyl

Methods to cyclise the 1,4-dicarbonyl: H 2S

P2S5 S
MeO P S
Lawesson’s reagent
S P OMe
S

All of these heterocycles arise from 1,4-dicarbonyls, which need to be

made using unnatural synthons. Two disconnections are possible:

O O O
R2 R2 R2
R1 R1 R1
a2 d2 O O a3 O d1

umpolung umpolung
58
1,4-dicarbonyl synthesis revisited

can decarboxylate if needed

O O CO2Et O CO2Et
R2 NaH, THF
R1 R2 R2
R1 R1
a2 d2 O Br O O

O
O
R2 α-halocarbonyl = soft
R1 H+/ H2O w up R2
Br R1 Enamine / 1,3-diCO= soft
N
O

O Note: this disconnection will also

R2 1. LDA
R1 R2 work with enamine or malonate as
R1
Br O 2. O3; PPh3 nucleophile
O
Alkene = latent carbonyl

NO2 CO2Et O CO2Et

1. NaH, THF
R2 R2
R1 R1
2. H3O+
O O

Nef reaction

• Mechanistic focus: Nef reaction

H+
O O HO O HO OH HO OH
N H+ N N +/- H+ O
HO N

H OH2
59
 1,4-dicarbonyl synthesis revisited

SN2 O O + various
O
then H3O+ d1 reagents
R2 R1 Br
R1
a3 O d1 O
1. Base (e.g. Al2O3, R2
R1
OH–, etc.) Et3N, EtOH
O
O R2 O H R2
2. H3O+ or TiCl3 S
R1 NO2 N Bn R1 O
HO
Stetter reaction
Cl

• Mechanistic focus: Stetter reaction

Cl Bn Bn Bn
N Et3N, EtOH N N
H nucleophilic carbene
HO S HO S HO S

Breslow Intermediate
Bn Bn
R H N R +/– H+ N R
O O
R' S H R' S OH

R2

H O
O O R1 Bn O R1 Bn
R2 R2 N R2 N
R1
O O S O S
R' R'
Review of nitroalkane conjugate addition: Ballini et al., Chem. Rev. 2005, 105, 933. 60
For recent leading references on the Stetter reaction, see: Org. Biomol. Chem., 2011, 9, 8437; JACS 2005, 127, 14675.
 Aromatic heterocycle synthesis 1: Furans, Pyrroles, Thiophenes
Further functionalization of the heterocycle nucleus can be be achieved under a variety of conditions; these heterocycles react readily with
electrophiles as they are electron rich.

a) Direct metallation (deprotonation):

deprotonate at 2-position with t-BuLi in Et2O

O

N
H deprotonate at N with NaH (pKa = 16.5)

b) Reaction with electrophiles: generally at the 2-position

see Tam et al. Synth.
NBS / DMF Commun. 2010, 40, 2138
O Br
H
E+
O O E
Br2 / MeOH MeO OMe
O
Superior stabilisation of charge from
electrophilic attack at 2-position

c) Diels-Alder reactions:

O O O
O Br
O
O 1. Br2 O Heat
O Br O
O 2. amine O
O base O

Note exo-selective
61
 Case Study: Atorvastatin
The Pfizer synthesis of atorvastatin contains a classic example of Paal-Knorr pyrrole synthesis, combined with the highly efficient Stetter protocol
to prepare the 1,4-dicarbonyl.

HO O Sidechain prepared by chemical

CO2H CO2t-Bu synthesis or biocatalysis
HO O
d1 Stetter substrate
F
N NH2
2 x C–N O
O F
O O O C–C
Stetter H
F
NH Pivalic
acid, tol;
3 steps HN HN
O O
Cl Bn
Atorvastatin (Lipitor) Et3N / N
(cholesterol lowering)
(Pfizer) S
Recognise 1,3-dicarbonyl HO
, EtOH, 70 °C HOAc, C–C
hexane Knoevenagel

O
O
NH2 O O C–N
HN
OMe
O
toluene

http://www2.chemistry.msu.edu/courses/cem958/FS06_SS07%5CAmand.pdf 62
Aromatic heterocycle synthesis 2: Azoles
For 5-membered rings with two heteroatoms, look for ring-opening C=O disconnections which form C–Heteroatom bonds

Oxazoles:
Use dehydrating conditions R2
Example 1:
R2 N
e.g. POCl3, SOCl2,etc.
R1 R1 R3
NH O
O
Cl R3

R2 R2 C–N R2
N C–O acylation
R1 R1
NH NH2 Cl
R1 O R3
O O
O R3 O R3
enol C=O

Example 2: Using TosMIC (good for monosubstituted azoles)

C
Ts N - H+
N C O O O
N N
Ph N K2CO3
O Ph H Ph Ph
Ts
'TosMIC'

H
O O O
C N N
N
Ph Ph Ph
Ts Ts Ts

63
 Aromatic heterocycle synthesis 2: Azoles

Isoxazoles: Isoxazole formation can lead to regioselectivity issues... R2

e.g. POCl3 N
O O
R2 1,3-dipolar R2
cycloaddition Synthesis of the nitrile
N oxide is achieved by
R1 O N
R2 dehydration
R1 O
NCS
then N
HO
KHCO3
R2 C–N
C–O
R1 R2 • How can we control
R1 N H2N OH
O the regioselectivity?
O O

Regioselectivity: e.g. Hydroxylamine reacts with the most electrophilic of the two carbonyls:

MeO
O O O OH O
N

MeO NO2 MeO NO2

H2N OH
NO2
Less-populated but
more electrophilic
OH O protonated ketone R2
2 x C–N
R1 R2
R1 N H2N NH2
N O O
MeO NO2 H

Note identical disconnection for pyrazoles 64

 Aromatic heterocycle synthesis 2: Azoles
Imidazoles, Thiazoles: Similar disconnections can be
used to form these heterocycles. S
R1 N R2
Example 1:
O Et3N O OH HO S – H2O S
S R2 S R2 S R2
R1 R1 R1 R1 R2 R1 R2
N N
X NH2 NH NH H

soft-soft NH4OAc

O O
Et3N
HS R2 S R2
R1 R1
X O O

Example 2: TosMIC again...!

C
X X Ts N X
N N X = S, N
Ph Ph H K2CO3 Ph

O R3
Example 3: By exhaustive imine synthesis!
O O R3 H HN
N
R1 R2 XS NH4OAc R1
R3 R2

HN
N R3 R3 R3
R1 NH2 H2N
R2 HN NH N H2N HN
R1 N N
R3 H R1 R1
R1 R2 H2N R2 H2N H3N
R2 R2
65
 Aromatic heterocycle synthesis 2: Azoles

Further functionalization of azoles is possible with nucleophilic or electrophilic reagents:

Lithiation:
X n-BuLi X n-BuLi
X X Reaction fails for isoxazoles
Y Y Li Y Li Y

Reaction with electrophiles:

E
N E+ N E+ e.g. E = Me+, AcCl,
N N E Cu-cat. N-arylation
N N N N
H H

Largest HOMO coeff. / most

stable intermediate cation
E e.g. E = Br+,
X E+ X E+ NO2+ (forcing)
E X X
Y Y Y Y Assisted by EDGs on ring

Reaction with nucleophiles:

Cl Nu
X Nu– X Nu–
Cl Nu X X
Y Y Y Y

(prepare from azolidinone)

66
Exercise
Plan a synthesis of the non-steroidal anti-inflammatory drug tolmetin, from N-methylpyrrole.

CO2H
N
O
tolmetin

Retro:
C–C
FGI acylation
tolmetin CN CN CN
N N N
O O

Friedel-Crafts equivalent
or direct reaction with acid chloride
Synthesis:

NMe2
1. MeI
NMe2, CH2O 2. NaCN O
NMe2 CN CN
N Mannich N N POCl3 N
Villsmeier O
Reaction at C2
Reaction at C5
NaOH
via NMe3
N N

CO2H
N
O

67
Exercise

Propose a synthesis of of the anti-asthma drug broxaterol:

Br

N
O N
H
OH O
Broxaterol
OH
N
HO
H2N
Br2

C–N
Br Br Br
a2 / d FGI Br Br

N N N N
O N O O HO
H O O
OH O

Br2,
NaBH4; NaH Br cat. H+ K2CO3
Br
N
O
O Br N
O O

68
Exercise
Propose a synthesis of of the angina / hypertension drug amlopidine:

Amlopidine stepwise Cl
Cl Hantzsch
MeO2C CO2Et O H

O MeO2C CO2Et amine

N NH2 protection
H O
O O NH2 required??

Synthesis:
available
EtO2C HO CO2Et CO2Et
N3 NH4OAc
O O O
NaH O N3 H2N N3
Cl
Cl
available
MeO2C CO2Et

O
N N3
MeO2C Cl H
NaOMe
Cl MeO2C
O H2, Pd/CaCO3
O H O

Cl
Amlopidine MeO2C CO2Et

O
N NH2
H
69
 Aromatic heterocycle synthesis 3: Pyridines
The classical pyridine disconnection is to recognise a 1,5-dicarbonyl precursor:

–H2O NH2OH

R1 N R2
OH
O O Here we can use our
vast array of 1,5-di-CO
or disconnections....!!
R1 N R2 R1 R2

[ox] R1 N R2
e.g. DDQ, Ce(IV) NH4OAc
H

This is the basis of the classic Hantzsch dihydropyridine synthesis:

2x O O O O R2 O O R2 O
NH3
EtO R2 EtO R2 EtO OEt EtO OEt

R1 O R1 O R1 R1 R1 N R1
OO H
O
may proceed
via EtO

R1 NH2

70
Aromatic heterocycle synthesis 3: Pyridines
Variants: To avoid the use of overly reactive Michael acceptors or aldehydes:

NMe2
O O
NH2OH•HCl EtO OEt
heat CO2Et
OEt
O Ph O Ph N Ph NH4OAc
O O N
OEt

Mannich base – equivalent to O O

equivalent to malondialdehyde
Ph O H H

Further examples of useful pyridine chemistry...

PBr3

H2O2 Ac2O POCl3 H2NR

R R R R R
R
N N N O N Cl N N
H H
O

NaOMe
susceptible to
SEAr reactions...
Br R
NBS
R
N OMe N OMe

71
Aromatic heterocycle synthesis 4: Diazines
The synthesis of Diazines and derivatives simply involves putting the disconnections we have seen so far together! Also remember oxydiazines...

2 x C–N
R2 + ox. O
R2 H2N NH2
N R1
R1 N
Pyridazine: O Note: Oxydiazines are easy to
O convert to chlorides using POCl3.
2 x C–N
This provides a straightforward
O H2N NH2 way to perform SNAr reactions.

O
R2

N
R2
2 x C–N O O NH R1 N O
N H
R1 R2 H2N R3
Pyrimidine: R1 N R3 POCl3
Note: can also include R3 = O, N, S
R2
2 x C–N O X R2
N
R1 R2 H2N NH2 N
R1 N X
X = O, N, S R1 N Cl

Nu
2 x C–N
N R2 + ox. NH2
R2
x2
Pyrazine: R1 N R1 O N
2 x C–N R1 N Nu
R1 N R2 + ox. R1 O H2N R2

R1 N R3 R1 O H2N R3
72
Aromatic heterocycle synthesis 5: Benzannulated heterocycles
Obviously a very important class of compounds – we will again look briefly at some trends, which can equally be applied to bis-heterocyclic systems.

Benzannulated 5-membered rings

Key intermediates: Issues:
a) Indoles
R2 R2
O R1
Fischer R1 NH2 R2 Regioselectivity
N N R1 NH in [3,3]
N
H AcOH H H

OMe N
Leimgruber-Batcho Nitro reduction
N MeO N
H NO2 NO2
1. heat
2. reduce NO2 DMFDMA

R2 R1 R2
I
Larock R1 R1 R2 Pd(II) Regioselectivity in
N carbopalladation
NH2 NH2
H
Pd(0)

b) Benzofurans [3,3]
O
C–O

OH O3; SMe2 OH heat O

R R R
O
R Ac2O, NaOAc,
CHO AcOH, heat
C–C

O CO2H O
R R
73
Aromatic heterocycle synthesis 5: Benzannulated heterocycles
c) 1,3-Benzimidazoles,
C–X
Benzothiazoles, C–N X "O " R / H+ HN
X e.g.
Benzoxazoles R R reagent at acid H
oxidation level MeO OMe
N NH2 HO OMe , H2N

Ph
X C–X I
X X = S: K3Fe(CN)6 X OAc
R via
X = NH: PhI(OAc)2
N N R N R
H

LG
d) Indazoles, X N–X X Examples:
Benzisothiazoles, N OH
NH DEAD, PPh3 O
Benzisoxazoles
N
R N
R OH
R R
X N–X X
N
N
LG
R R
NMe2
R R
X C–C X N Cl Cl N
N NH2
N N
R
R N NMe2
via NH

Cl NMe2
X C–X F
X SNAr
N
N
R R
74
Aromatic heterocycle synthesis 5: Benzannulated heterocycles
2. Benzannulated 6-membered rings

a) Quinolines

R2 R2
C–C C–N O O
Combes O
R1 R2
N R1 N R1 NH2
H+ H+ or heat

R2 R2 R2
R3 C–C C–N O
Friedländer O O
R3
(acid or base) R1
N R1 NH NH2
H+ H+
R1
R3

b) Isoquinolines

FGI C–C
Bischler-Napieralski
/ POCl3
N N N HN O

R R R R

the equivalent reaction with imines is the

Pictet-Spengler tetrahydroquinoline synthesis

75
Aromatic heterocycle synthesis 6: Rings with ≥ 3 heteroatoms
To prepare triazoles, oxadiazoles, tetrazoles, etc. we simply modify our disconnections used so far with an emphasis on carbonyl recognition. The
strategies below are colour coded into cycloadditions, excision of a single atom, cyclisation (dehydration), and ‘stepwise cycloaddition’.

1,2,3-triazoles
N
N N R R N N N R
cat. Cu(I)
R

N O
N N R N N N R
R 1,2,4-triazoles
R PPh3 R'
N O O
R R R'
R R
HN NH NH2
R N N
N R
N O N N R'
N R H2N
N R N R R OR
R O R
PPh3 PPh3 N N H2N N O

H
R N R R N R
tetrazoles H2N NH
N N O O R
N R
HN N R N N N N SnBu3
N or TMSN3
R or LiCl, NaN3,
NH4Cl

R N N N R
NH
N NaN3, Tf2O
R R O

76
Aromatic heterocycle synthesis 6: Rings with ≥ 3 heteroatoms
To prepare triazoles, oxadiazoles, tetrazoles, etc. we simply modify our disconnections used so far with an emphasis on carbonyl recognition. The
strategies below are colour coded into cycloadditions, excision of a single atom, cyclisation (dehydration), and ‘stepwise cycloaddition’.

1,2,3-oxadiazoles

O
N N HO N N OH + dehydrating agent
(e.g. Ac2O)
R R R R
1,2,4-oxadiazoles and thiadiazoles

R O R O O
R R + dehydrating agent
N N (e.g. (EtO)3CH)
HN NH

O R OH
N N O
N
R NH2 Cl R
R

S
N N H2N NH2 + S2Cl2
(oxidative method)
purines R R R R

N N NH2
N O
R
N N Cl R
H N NH2

O
O
HN N Y
H2N N
R
X N N X Z
H2N N
H R

77
 Exercise
Plan a synthesis of deferasirox from starting materials of your choice

HO2C

N N

N
OH HO

Deferasirox

HO2C

HO2C

HN NH2
N N O O O
2 x C–N
N N N OH
=
H
OH HO Et3N OH HO O

Symmetry suggests
hydrazine disconnection C–N

O O
O
NH2 Cl
N OH
H O
OH HO
OH

Eur. J. Inorg. Chem. 2004, 4177 78

Exercise
Propose a synthesis of the antimalarial agent
amodiaquine OH

N
HN

Amodiaquine
Cl N (antimalarial)

OH OH C–N OH

N N N
HN C–N H2N H2N

Cl
Cl N Problem: Regio/chemoselectivity

Cl N

Et2NH, C–N
CH2O

OH OH

HN H2N
C–N
Cl O OEt
C–N
FGI C–C
Cl N O

Cl N Cl N Cl NH2 O CO2Et
H
POCl3 H+; then heat, OH–

79
 Exercise
Plan a synthesis of valatinib from starting materials of your choice

Cl

Cl
H2N

HN Cl O O
C–N FGI FGI
N Valatinib N NH OH NH2
(cancer)
N (Novartis) N N O NH2
heat POCl3

N N N N

Note elements of FGI

hydrazine and suggestive
SNAr disconnection O O
C–C
O O

NaOMe N
O N

In reality, the condensation product rearranges...

O
O O
OH
NH
O OMe OMe H2N NH2
N
N

N O N O
N

J. Med. Chem. 2000, 43, 2310 80

 Revisiting arenes – advanced disconnections
A number of specific and reliable methods are available to functionalize arenes.

1. ortho-Functionalization. This is a very large topic. Some highlights are selected below.

a) Classical methods
O O OH
i) Claisen rearrangement – a very convenient ~200 °C
method to install an allyl group at the 2-position. H

O
R very versatile functionality
O OH O for further manipulation!

ii) Fries rearrangement – also generally quite AlCl3 R

selective for functionalization at the 2-position.

iii) Other ortho-selective classical reactions such as Reimer-Teimann, Gattermann-Koch, etc.

b) Directed ortho-Metallation.
DMG DMG DMG
n- or s-BuLi Li E+ E

Mechanism: NEt2 NEt2

NEt2
O O Good Directing groups: Amides, thioamides,
O O s-BuLi, O O carbamates, oxazoles, etc.
Li H TMEDA E+ E
Li
E = RCHO, R–I, X2, TMSCl, B(OMe)3 ....

Review of directed metallation: V. Snieckus et al., Chem. Rev. 1990, 90, 879. 81
 Revisiting arenes – advanced disconnections
Regioselectivity:
TMS =
blocking
group
NEt2 NEt2 NEt2 NEt2

O O O O O O O O OH NEt2
s-BuLi, s-BuLi,
TMEDA Li TMSCl TMS TMEDA TMS Li warm to RT TMS
O

Cl Cl Cl Cl anionic Fries Cl
rearr. iterative DoM

E+ E+
inductive
effect controls NEt2 NEt2 NEt2
regioselectivity
O O O O O O Review of directed metallation:
E E H+ or TBAF TMS E V. Snieckus et al., Chem. Rev. 1990, 90, 879.
For a review of the uses of metallated
heterocycles in organic chemistry, see:
Cl Cl Cl Chem. Rev. 2004, 104, 2667.

• Heteroaromatics: DoM can be useful in heteroaromatic systems, but often the anion can be unstable.

LTMP; B(OMe)3; B(OH)2

• Lithiation ‘directed’ by
N Cl H+ N Cl halogens and alkoxy groups is
OMe OMe particularly effective

• Anions adjacent to N can be

CHO
n-BuLi; DMF unstable (lp-lp repulsion?)

MeO N OMe MeO N OMe

Reviews of heteroaromatic lithiation: Tetrahedron 2001, 57, 4059; Tetrahedron 2001, 57, 4489 82
Revisiting arenes – advanced disconnections
2. Palladium / copper catalyzed cross-coupling: An obviously important strategy. The synthesis of boronic esters and acids is key to this chemistry;
common methods include:
R B(OH)2
H+
R Bpin PdCl2dppf R Br BuLi; R B(OMe)2
R Bpin
pinB–Bpin B(OMe)3
pinacol
[Ir(OMe)cod] / dtbpy
hexane, rt R
See Hartwig, Chem. Soc. Rev., 2011, 40, 1992

pinB–Bpin

Copper-catalyzed amination / amidation is a very attractive and cheap route into aryl amines etc.:

N NH
Cu2O, ligand,
Cs2CO3, MeCN, reflux N N
Br I N N 92% See Beletskaya, Organometallics 2012, 31, 7753
N
then add NH

3. C–H activation. A very large and rapidly

expanding topic which we will not cover in detail!

Reviews:
2 x C–H activation: Chem. Rev. 2011, 111, 1215
Arene arylation by C–H activation: Angew. Chem.
Int. Ed. 2009, 48, 9792
Pd-cat C–H activation: Angew. Chem. Int. Ed.
2009, 48, 5094
Chem Soc Rev 2011 issue 4
83
Asymmetric Synthesis
Due to the breadth of this area, we will not cover in detail! General strategies to bear in mind:
1. Look for available chiral sms, such as amino acids, sugars, epoxides (epichlorohydrin)

O OH O OH O
H2N RO
OH OR Cl OMe
O
R O OH

Amino acids Carbohydrates etc. Epichlorohydrin The 'Roche' ester

(both enantiomers available) (both enantiomers available) (both enantiomers available) (both enantiomers available)

2. Look for ‘high quality’ catalytic asymmetric processes such as Sharpless AD and AE, Jacobsen epoxidation and hydrolytic kinetic
resolution, Noyori hydrogenations, CBS reduction, organocatalysis (!)

Sharpless AD Sharpless AE

DHQD ‘Down’ DHQD ligands D-tartrate ‘Down’ (–)-D-Tartate

AD-mix α

RS RS OH R1
SAD RS RM OH SAE OH
RM RM O
RL RL RL H OH
R1 R1
OH

DHQ ligands (+)-L-Tartate

AD-mix β
Large goes ‘Left’

84
Asymmetric Synthesis
2. Look for ‘high quality’ catalytic asymmetric processes such as Sharpless AD and AE, Jacobsen epoxidation and hydrolytic kinetic
resolution, Noyori hydrogenations, CBS reduction, organocatalysis (!)

Jacobsen epoxidation Jacobsen HKR

Mn(III)(S,S)Salen Cr(III)(R,R)SalenOAc OH
OH
R1
N N
Mn O H2O
R O O R R1
Cl TMSN3
R1 RL OTMS
R1 RL R R
Cr(III)(R,R)SalenN3 1
N3
aq. NaOCl R
O

Various Noyori Hydrogenations:

cat. RuCl2((R)-BINAP),
O O H2 (100 atm), MeOH OH O

R1 OMe 99% ee R1 OMe

Ph Ts
N R-BINAP gives R-stereocentre
Ru (when ester has highest priority)
Ph N
O H , i-PrOH OH
R1 R1
>95% ee cat. RuCl2((R)-BINAP)(diamine),
R2 R2 O H2 (8 atm), MeOH OH
(R,R)-Tsdpen ligand gives R- R1 99% ee R1
Ar Ar
stereocentre (when alkyne has
highest priority)
R-BINAP gives S-stereocentre
(when arene has highest priority)

85
Asymmetric Synthesis
3. Look for ‘high quality’ chiral auxiliary / reagent approaches: Asymmetric allylation, Evans auxiliary, etc.

Evans auxiliary

M
O O O O
base, M+ R1
O N O N
R1
Bn Bn

Leighton allylation / crotylation

Br

N
Si / cat. Sc(OTf)3,
N Cl CH2Cl2, –30 °C

Br

4. Look for high quality substrate stereocontrol, e.g. Felkin-Anh, Allylic strain, ring functionalisation (stereoelectronics), steric effects (less-
hindered face approach in bicyclics), stereospecific reactions (e.g. epoxide opening with amines or azides to prepare aminoalcohols)

5. Use enzymatic or chemical resolution, which can work really well at an early stage of a synthesis

86
 Case Study: Indacaterol
The synthesis of this drug requires the installation of a chiral benzylic alcohol adjacent to an amine. This could well suggest an epoxide strategy, albeit
with regiocontrol issues in the ring-opening due to benzylic stabilisation of the SN2 reaction.

FGI
For Owen: $189/kg
H2N H2N
HN C–N
O
HO HO
Br
Indacaterol C–O
(Chronic Obstructive
Pulmonary Disease)
O N (Novartis) O N O N Identify p-directing group
H H H
OH OH OH

FGI Asymmetric reduction

O
Br
C–C

N O N O N
H H
OH OBn OBn

Org. Proc. Res. Dev. 2006, 10, 135.

See also: Bioorg. Med. Chem. 2011, 19, 1136. 87
 Case Study: Indacaterol
Synthesis:
O
1. AcCl / AlCl3
1. F3C OEt 2. H2, Pd/C
Recognise symmetry
2. AcCl / AlCl3 3. NaOH / heat;
NH2 3. H2, Pd/C NH NH2•HCl
HCl
TFAc

O O O
1. AcCl / AlCl3 1. m-CBPA AcO
2. BnBr, K2CO3 2. Ac2O, 40 °C

N N O N N
H
OH OBn OBn O O OBn

Br2, BF3•OEt2
O
H Ph Ph
O
HO O O
Br N B / BH3•SMe2 Br AcO
K2CO3, H N
acetone Me
O O OBn
91% ee
O N O N O N
H H H
OBn OBn OBn – AcOH, AcO–

Ph 2-pyridone
1. Heat with amine salt O
2. H2, Pd, AcOH N Me
Ph B
H H
Indacaterol B
O
H RS
‘Small’ group positioned here to minimise 1,3-allylic strain
Interestingly RL is the CH2Br
Org. Proc. Res. Dev. 2006, 10, 135. RL
See also: Bioorg. Med. Chem. 2011, 19, 1136. 88
 Case study: Gleevec
Gleevac (Novartis) – Tyrosine kinase inhibitor used against gastrointestinal tumours.
Process synthesis developed ca.1996.; several reports on improved syntheses including OPRD paper in 2008, and flow synthesis in 2010.

• Original process route (Zimmermann)

N N
H H
N N N N FGI OH
N C–N N C–N
N N

HN Cl HO
pyridine NH2

N O O O
N
(4-hydroxymethylbenzoic
Gleevec
Gleevac (Imatinib) acid is available)
FGI

$4.6 billion sales in 2015 1

H H
N N Me2N HN N H2N

N O NH2

NO2 i-PrOH, 2 NO2 H2N N NO2

NaOH
N N

Key steps:
1. Convergent late stage aniline acylation Me2N OEt O
2. Pyrimidine ring formation used as arene coupling step
OEt

Bioorg MCL 1996, 6,1221 89

 Case study: Gleevec
• Improved ‘process’ route (Wang, 2008)

N
H H
N N N N N
N C–N Cl
N N
N
H
HN heat HN
91%
N O N O

Gleevec
Gleevac (Imatinib)
C–N 1. NH2NH2 / FeCl3 83%
2. acid chloride, Et3N, 93%

H
Cl N N C–N N NH2 Br

N N
Cl
NO2 DMEDA, CuI, NO2
O K2CO3, dioxane,
N 100 °C, 82% N

guanidine nitrate,
2 x C–N NaOH, BuOH, 86%

Me2N
Key steps / features: HN NH2
• Yields and cost improved through simplifying key steps – e.g pyrimidine O
NH2
ring formation now with guanidine, not functionalised guanidine
• Copper-catalyzed pyrimidine–arene union
N
• Avoids use of H2NCN

OPRD 2008, 12, 490 90

Case study: PKI 166

OH OMe OMe OMe Reagents: OMe

C–N
+ FGI FGI 2 x C–N

HN HN HN HN HN

N NH N Cl N O H2N O H2N O
N Ph N NH EtO

H2N H
Novartis PKI 166 NH O
(anticancer) Ph
H2N

Reagent:

MeO C–C MeO MeO

Br NH NH NH2

NH2 NH2 NH2

O CO2Et O CO2Et O CO2Et

NH3

HCl, NH2 Cl
N
EtOH
OEt
CO2Et CO2Et

91
 Case study: PKI 166
Synthesis:

Soft – reacts at bromide

HCl, NH2 Cl NH
N NH3
EtOH NaOEt
OEt NH2
CO2Et CO2Et CO2Et
MeO
NH

NH2
MeO
O CO2Et
Br
O
α-methylbenzylamine: available

OH OMe OMe O OMe

H2N
Ph H NH2
1.
POCl3 HCOOH

HN 2. BBr3 HN HN HN

N NH N Cl N O H2N O
N Ph N NH EtO

Novartis PKI 166

(anticancer)

Synthesis: http://www.chem24h.com/drug/synthesis/ykkjfgvlg.html
Traxler, P.; Bold, G.; Brill, W.K.-D.; Frei, J. (Novartis AG); Pyrrolopyrimidines and processes for the preparation thereof. EP 0836605; JP 1999508570; US 6140332; WO 9702266 . 92
 Case Study: Linezolid

O
O
O N N H
N $1.3 billion
sales in 2013
F O
O Linezolid
Cl (antibiotic)
(Pharmacia/Upjohn)

O
O FGI C–C OH
O N N H O N NH OH
N O N NH2
NH
F O F NH F
CDI, imid
O
O
Actual process synthesis: key step: FGI

O 1. NH3, PhCHO; then HCl aq.

Cl 2. Ac2O
O N NO2
3. LiOMe
O F

NH
O O
OBn C–N
O
O N N OBn O N N OLi
H LiOt-Bu
F F NAc O NH F NO2
O
F
O
O N N H
N
F O
OPRD 2003, 7, 533
J. Med. Chem. 1996, 39, 673 93
 Case Study: Telcagepant

F3C
The azepane in this Merck migraine drug features two remote
O O stereocentres. These were installed using an organocatalytic
N asymmetric conjugate addition reaction:
F
N N
F H
N N
Telcageplant EtO2CN
NH OL 2006 3307
(migraine) O OL 2006 3311
(Merck) N
O H2N
Cl
1. NaBH3CN, TFA
C–N
C–N N
2 x C–N (CDI) 2. ClO2S then H2O
O
F3C
HN C–N HN
F N O NO2
F N
N N N O
NH2 Cl F
NH Pd(OAc)2 (1 mol%)
O H2N O F
dppb (2 mol%)
aq. i-PrOH, 83 °C
FGI 99% C–N
asymmetric
conjugate addn
CF3
F3C
NH NHAc NO2 NHAc NO2
F N O C–N FGI C=C
F O
CO2H CO2H
NHAc
NH2 F F F

F F F

JOC 2010, 75, 7829

Review: Chem. Soc. Rev., 2013, 42, 774 94
 Case Study: Telcagepant
Synthesis highlights: F3C
O O
N
F
N N
F H
N N
Telcageplant
NO2 (migraine) NH
O
(Merck)

Ph
N Ph
(5 mol%)
H OTMS NO2 NO2 NHAc
O NHAc
PivOH, B(OH)3,
, DMA CO2H
aq. THF HO2C CO2H
O F F
F 73%, 95% ee
F (35 mol%)
F N F
H
91%, Z:E = 94:6
steps

CF3
F3C
1. HCl, aq. i-PrOH NH NHAc
O 1. PivCl, Et3N,
2. CDI N DMAP
Telcageplant CO2H
3. KOt-Bu, NHAc 2. NaOH / F
>99.9% ee DMSO
HN
F F
F
N N
NH
O

JOC 2010, 75, 7829

Review: Chem. Soc. Rev., 2013, 42, 774 95
Excellent reviews of recent / key methods to prepare important
pharmaceuticals, and very readable!

96
Problem Session 2

NC N O
N H
N N NMe2
N O S
HN CF3
O O
N
H
Lestrozole Leflunomide
(Breast cancer) (Arthritis) Sumatriptan
(Novartis) CN (Sanofi-Aventis) (Migraine)

O OH N
N NH
N N
N N O N
N
O
Cl
F
N N
O
H
Valsartan Clozapine Risperidone N
(Blood pressure) (Bipolar disorder) (Antipsychotic)
(Novartis) (Novartis) (J&J) N

Pazopanib OH
(Anticancer) N N
(GSK) N
N N
N N N H2N N
H CN
O
N
Vildagliptin Famciclovir
H2NO2S N N (Anti-diabetic) (Herpes virus)
H (Novartis) (Novartis) OAc OAc

97
Problem Session 2

N
NH NH2
N NC Br
N NH N
O Cl N N N O
N S H
N

Zolpidem Tizanidine Etravirine

(Insomnia) (Muscle relaxant) (HIV) CN
(Sanofi-Aventis) (Novartis) (Tibotec)

O
N N OH F
NH N
H
MeO N N N O N N N
H N
H2N N N HN O O
Fasiplon
(Anxiolytic) N
(Roussel Ucla) Abacavir O Raltegravir
(HIV) N (HIV)
(ViiV healthcare) HO (Merck)

OH O O
F
OH CN
Alvimopan
(Gastrointestinal) N N N N
N Ruxolitinib
Cl (Cancer)
Ph H2N (Novartis)
F
O NH N
Sitafloxacin
OH (Ulcer) N N
(Daiichi Sankyo) H
O
98
99
Appendix: Selectivity in Electrophilic aromatic substitution (SEAr)
Directing effects are crucial for arene functionalization:

Wheland intermediate • Electron-donating groups (R) stabilize this

R R R R intermediate so favour ortho- and para-substitution
RDS – H+ • Electron-withdrawing groups destabilize this
intermediate so favour meta-substitution where the
Wheland intermediate is less destabilized
E H E H E
E

• Activating electron-donating groups: • Electron-withdrawing groups:

π-donors: groups with lone pairs are o-, π-acceptors: groups with low-lying π*
E
p- directing E orbitals are m- directing
HO O
H e.g. OR, OC(O)R, NR2, NHAc, RS e.g. CO2R, NO2, COR, CONH2,
H SO2R, etc.

σ-donors: alkyl groups (inductive effect These intermediates are disfavoured

H E
I+ and σ-conjugation) are o-, p-
H σ-acceptors: groups with low-lying σ*
H H directing F E
orbitals are m- directing
F e.g. CF3
F H

• Deactivating electron-donating groups: Cl, Br, I, F partial positive charge on EWG also
destabilizes these intermediates.
E weak π-donor so still o-, p- directing but
Cl strong I– deactivates
H

100
Appendix: Selectivity in Nucleophilic aromatic substitution (SNAr)
We can also effect nucleophilic substitution in specific arene systems:

HNO3, O2N CF3 i-Pr2NH, O2N CF3

CF3 H2SO4 heat Trifluralin
Cl i-Pr2N (herbicide)
Cl
NO2 NO2

i-Pr2HN – H+

Note: CF3 introduction:

HF, SbF5 O SNAr requirements:

Ar CCl3
N CF3 • electronegative group (Cl, F) at
Ar CF3 O site of substitution
SF4
Ar CO2H Cl • strong EWG at o- or p- position
i-Pr2N
NO2 • F– is the best LG as attack on the
arene is the RDS, not ejection of the
Meisenheimer complex LG!

Classical aromatic SN1 is particularly useful for the synthesis of aryl iodides, phenols, nitriles

N
NH2 HONO N Nu or Cu(I)X Nu Particularly useful for Nu:

H2O, ROH
Cu(I)X (Cl, Br, CN)
N
N F KI
NaNO2,
HBF4 BF4

101
Appendix: Arenes – other disconnections
4. Miscellaneous transformations

a) ipso-substitution R O R
O SiMe3
TMS / AlCl3 O TMS
Cl R2 – TMSCl
R1 R R COR
σC–Si → p
Regiospecific substitution

b) oxidative OH O
OH O
dearomatisation
[oxidant] [Oxidant] = DDQ, [oxidant]
R1 R1 Ce(IV)(NO3)6(NH4)2 (CAN), R1 R1
PhI(OAc)2, etc.
O
OMe O
HO

c) reductive dearomatisation (Birch)

EWG EWG EWG EDG EDG EDG

Na/NH3 via Na/NH3 via

d) arynes EDG EDG

regioselective due
R2NH to anion stability
EDG EDG (inductive effect)
NR2 NR2
OTf CsF, rt/80 °C
EDG EDG OH
TMS
O
O
OMe OMe 102
OMe
Revisiting arenes – advanced disconnections
4. Miscellaneous transformations
a) reductive dearomatisation (Birch)
EWG EWG EWG EDG EDG EDG
Na/NH3 via Na/NH3 via

b) arynes EDG EDG

regioselective due
R2NH to anion stability
EDG EDG (inductive effect)
NR2 NR2
OTf CsF, rt/80 °C
EDG EDG OH
TMS
O
O
OMe OMe
OMe

c) Inverse electron-demand Diels-Alder sequences

High energy HOMO

CO2Me H
N MeO2C N
CO2Me – MeO2C
N N – N2
N
N CHCl3, N N N
N N
30 min, 45 °C, 84% N
enamine / 1,2,3-triazine Bicyclic pyridine
cycloaddition

103
Recap: Chemoselectivity based on stereoelectronics
Nucleophile selectivity: In the absence of steric effects, nucleophile selectivity is based on the relative energies of the lone pairs – the higher the
HOMO energy, the better the nucleophile...

π*C=O
O
H
NH2 N S
Cl

HO py
? O Better energy match =
HO N earlier / stronger bond
in TS
O

The amine is more nucleophilic than the alcohol...

Electrophile selectivity: Electrophile selectivity is based on sterics, electrostatics, and the relative energies of the acceptor orbitals – the lower the
LUMO energy, the better the electrophile. Another way of looking at this is to consider the degree of stabilisation of the carbonyl, which will be lost on
nucleophilic addition.

O O O O O O O
< N < < < < <
NR2 OR H O Cl

H
R Good overlap of Lower energy R Poor overlap of Cl
O N O donor orbital – O Cl
R high energy l.p. δ+ R l.p. (3p). Negative
stabilises the less stabilisation. inductive effect
amide and raises Positive inductive increases
nN π*C=O the π* energy σC–H π*C=O effect. nCl π*C=O electrophilicity
+I effect –I effect

104
 Appendix: Protecting groups
Protecting groups provide an obvious way to protect reactive functionality and achieve chemoselectivity. The cost is at
least one additional step in a synthesis, usually two.
• Sometimes the deprotection of protecting groups can be the most challenging step in a synthesis!
• Protecting group-free synthesis preferred – and may be achieved using a careful ordering of steps...

1. Protection for alcohols

• Silyl ethers Protection Deprotection

all alcohols: R3SiCl, imidazole, cat. DMAP, TBAF, HF•py, HF(aq.)

SiR3
O DMF (concentrated) H+ / MeOH (e.g. CSA, PPTS, HCl aq.)
R3SiOTf, 2,6-lutidine or Et3N, AcOH / THF / H2O (3:1:1)
CH2Cl2
1° alcohol: R3SiCl, imidazole, CH2Cl2 1° alcohol: PPTS, MeOH/CH2Cl2 (<1:1)

Order of stability (towards acid): TMS < TES < TBS (TBDMS) < TPS (TBDPS) < TIPS
• The TBDPS group is somewhat more base labile than expected – Ar–Si is more electrophilic.

• Benzyl ethers Deprotection

Protection
Bn: H2, Pd/C; Raney Ni; etc.
BnBr or PMBCl, NaH, THF or DMF Single electron transfer agents (e.g. Na/NH3, Li
O PMBTCA / cat. mild H+ (PPTS, CSA), naphthalenide)
or Sc(OTf)3
X PMB: Ce(IV)(NH4)2(NO3)6 (CAN)
NH
BnTCA / cat. TfOH DDQ; SET-agents.
Ar BBr3 or BCl3
O CCl3
'ArTCA'

Selective deprotection of silyl ethers: R. D. Crouch, Tetrahedron 2013, 69, 2383. 105
Appendix: Protecting groups
• Esters / carbonates

Protection Deprotection

O Esters: RCOCl or (RCO)2O, py, DMAP, CH2Cl2 NaOH (aq.), DIBALH. Others are case specific...
O R Particular favourites: R = Me, Ph, CCl3 Halogenated: Use Zn
Carbonates: ROCOCl, py, DMAP, Allyl: Use Pd(0) (generates pi-allyl)
Particular favourites: R = Me, Allyl, Bn, CH2CCl3 Benzyl: Use H2, Pd/C

• Acetals

Protection Deprotection
O O
THP: Dihydropyran / PPTS Mild acid (aq. or methanolic) (e.g. PPTS or
AcOH)

• Other ethers

Protection Deprotection
Me
O O
MOMCl, Hunig’s base, CH2Cl2 HCl (aq.) (fairly concentrated!)
MOMCl, NaH, THF TFA, CH2Cl2
Boron-based Lewis acids
Ph
Ph Trityl chloride, DMAP, DMF (selective for 1° OH)
O Ph Ph3C BF4 very mild acid! (e.g. a column!)

106
Appendix: Protecting groups
2. Protection for amines

• Carbamates Protection Deprotection

ROCOCl or (ROCO)2O, py, CH2Cl2 The following conditions are particularly useful as
O Particular favourites: they are orthogonal to other PGs:
R Cbz: H2, Pd/C
HN O R = Bn (Cbz)
R = t-Bu (Boc) Boc: TFA / CH2Cl2
R = Allyl (Alloc) Alloc: Pd(0), dimedone
R = Fluorenyl (Fmoc) Fmoc: Et3N
R = CH2TMS (Teoc) Teoc: TBAF

• Sulfonamides

Protection Deprotection
O O
RSO2Cl, py, CH2Cl2 Ts: Mg / MeOH, sonication.
S
HN R Particular favourites: or Na/Hg or other single electron reductants
R = p-toluene (Ts) (e.g. Na naphthalenide) or strong base
R = p-nitrophenyl (Ns) Ns: PhSH / K2CO3

3. Protection for carbonyls (and diols!)

• Acetals

Protection Deprotection
For carbonyl: MeOH, H2O, PPTS or HCl
O O HOCH2CH2OH, PPTS or TsOH
For alcohol:
ketone / aldehyde / dimethyl acetal, PPTS or
TsOH

107
 Appendix: Alkene synthesis : Wittig Mechanisms
Mechanistic focus: Wittig reaction
Br 1. NaH, THF
Non-stabilized ylid:
R1 PPh3 R1 R2 (Z)-selective
2. R2
O

Sterically disfavoured
Approach of reagents Transition state but it’s TOO LATE!!

Ph Ph R'
Ph O R' Ph O R' R
[2+2] Ph retro [2+2]
Ph P
P P R Z
O H
Ph H Ph H Ph
H R H H

Sterically favoured
Puckered EARLY TS minimises
sterics between R and R’

Ph Ph O R' Ph Ph R'
O R' H retro [2+2]
Ph [2+2] Ph P
P P H E
O H
Ph H Ph H Ph
R H R R

Sterically disfavoured Sterically disfavoured

in the transition state
• Reaction is irreversible, i.e. under kinetic control.
• Mechanism is a concerted [2+2] cycloaddition

Recent literature: Gilheany et al., JACS, 2012, 134, 9255. Aggarwal et al., JACS, 2006, 128, 2394. 108
 Appendix: Alkene synthesis : Wittig Mechanisms
Mechanistic focus: Wittig reaction

Br 1. NaH, THF EWG

Stabilized ylid:
EWG PPh3
2. R2 R2 (E)-selective
O

Note: This reaction is NOT under thermodynamic control

• The initial [2+2] is irreversible, and the geometry of the transition state is believed to be controlled by sterics and favourable dipole alignment:

cis-Transition states trans- Transition states

EtO EtO
O O
Ph Ph All structures benefit from
Planar Transition states Ph
Ph R H favourable opposed dipole
P P
H O H alignment
O
Ph H Ph
Sterically R
disfavoured

EtO EtO EtO

R O H O Ph O
Ph Ph Ph H
Puckered Transition
Ph H Ph R P
states P P
O O H O H
Ph H Ph Ph
R
EtO
Ph O O Puckered LATE TS minimises sterics
Ph between R and Ph
P R
Ph H Alternative Puckered-cis TS
H has less favourable dipoles
• Reaction is irreversible, i.e. under kinetic control.
• Late TS due to more stable ylid being used
Recent literature: Gilheany et al., JACS, 2012, 134, 9255. Aggarwal et al., JACS, 2006, 128, 2394. 109
Appendix: Alkene synthesis : HWE-type Mechanisms
Mechanistic focus: HWE reaction
M M
O O O O

Sterically favoured TS R–O PO R–O PO

R2 R2
R–O R–O
H R1 R1 H

Kinetically-favoured Aldol Aldol Aldol

This is the RDS for M M

stable (electron-poor) O O O O
phosphonates R–O PO R–O PO
R2 R2
R–O R–O
H R1 R1 H

Rotate Rotate

O O O O This is the RDS for less

R–O P stable (electron-rich)
R–O P
R–O O R1 R2 R–O O H R2 phosphonates
M M
H R1
FAST for e--poor phosphonates
due to highly electrophilic P Cyclise Cyclise Kinetically-favoured Cyclisation

O O O O
R1 O O
R–O P R–O P
R2 O R1 R2 O H R2 R2 R1

Z H H
E
H R1
110
Appendix: Methods for the synthesis of alkenyl halides and metals

Synthesis of alkenyl halides and alkenylorganometallics:

CrCl2, CHI3 JACS 1986, 108, 7408 HSnBu3, cat. Pd(0)

Takai: R O R JACS 1993, 115, 4497 or heat (radical)
dioxane I R
R SnBu3
Chem. Rev. 2000, 100, 3257
I
Ph3P=CHI
Stork-Wittig: R O R TL 1989, 30, 2173
THF H–Bpin
R R
Bpin
I2 cat. Cp2ZrHCl
R R
'M' 'M' = SnBu3, I Organometallics 1996, 14, 3127
Synthesis 2004, 1814
SiMe3, etc.

Cp2ZrHCl; I2 cat. PdCl2(dppf)

or DIBALH; I2 pinB–Bpin
R R R
I I Bpin

R
cat. Cp2ZrCl2,
Me3Al; I2 R CrCl2, CHCl2Bpin
I R O R
dioxane Bpin

Other methods for cis-reduction:

• Diimide and equivalents: KO2CN=NCO2K

Conditions H H TsNHNH2 / base
R R' NBSH = nitrobenzenesulfonyl azide
R R'
• hydrometallation / protodemetallation: BH3; HOAc

111
Case Study: Aliskiren
The aminoalcohol / diisopropyl core of this drug poses a significant chemical
MeO
challenge. Compare chiral auxiliary approaches (including those used by Novartis) NH2 O O
with a superior recent development that recognises symmetry.
MeO O N NH2
H
OH
• Approach 1: Novartis J. Med. Chem. 1997, 50, 4818 and 4832.
Aliskiren
(Hypertension)
O O (Novartis)
LiHMDS
N O
Li
Bn O O

N O O O Br
BnO BnO BnO
Br Bn N O steps

MeO MeO MeO

Evans alkylation Bn

MeO MeO MeO Schöllkopf

H N
N N
H N
N H N OMe
OMe MeO
R
Alkylation of less hindered face

MeO
N
OH O
BrMg N
BocHN BocHN OMe
H
OBn HCl BnO
RO OBn RO
steps
MeO
MeO non-selective MeO
addition (~1:1)
112
Case Study: Aliskiren
The aminoalcohol / diisopropyl core of this drug poses a significant chemical
MeO
challenge. Compare chiral auxiliary approaches (including those used by Novartis) NH2 O O
with a superior recent development that recognises symmetry.
MeO O N NH2
H
OH
• Approach 2: S. Y. Ko Helv. Chim. Acta. 2012, 95, 1937
Aliskiren
(Hypertension)
(Novartis)
Attack less-hindered face

O O O
LiHMDS;
O acetone O PCl5 O H2 / Pd/C
HO OH
O O O O
(from SAD)
O O O O

Two approaches investigated to assemble symmetrical bis-lactone O

O

Bn
O O LDA; allyl O O O O
bromide Grubbs II RO Li
N O
O N O N O N AD-mix α
O O MeO
Bn Bn Bn
desymmetrisation

O O
H2N NH2 1. H2, Pd/C O O
O
2. MsCl RO
1. O RO 3. LiBr (invert)
Aliskiren OH
2. H2, Pd/C N3 4. NaN3 (invert) MeO
MeO

113