Enrichment of Relevant Oxidative Degradation Products in Pharmaceuticals With Targeted Chemoselective Oxidation

Article abstract of DOI:10.1016/j.xphs.2018.10.059

The ability to produce and isolate relatively pure amounts of relevant degradation products is key to several aspects of drug product development: (a) aid in the unambiguous structural identification of such degradation products, fulfilling regulatory requirements to develop safe formulations (International Conference on Harmonization Q3B and M7); (b) pursue as appropriate safety evaluations with such material, such as chronic toxicology or Ames testing; (c) for a specified degradation product in a late-stage regulatory filing, use pure and well-characterized material as the analytical standard. Producing such materials is often a resource- and time-intensive activity, either relying on the isolation of slowly formed degradation products from stressed drug product or by re-purposing the drug substance synthetic route. This problem is exacerbated if the material of interest is an oxidative degradation product, because typical oxidative stressing (H₂O₂ and radical initiators) tends to produce a myriad of irrelevant species beyond a certain stress threshold, greatly complicating attempts for isolating the relevant degradation product. In this article, we present reagents and methods that may allow the rapid and selective enrichment of active pharmaceutical ingredient with the desired oxidative degradation product, which can then be isolated and used for purposes described above.

Full text of DOI:10.1016/j.xphs.2018.10.059

Journal of Pharmaceutical Sciences xxx (2019) 1-10
Contents lists available at ScienceDirect
Journal of Pharmaceutical Sciences
journal homepage: www.jpharmsci.org
Pharmaceutics, Drug Delivery and Pharmaceutical Technology
Enrichment of Relevant Oxidative Degradation Products in
Pharmaceuticals With Targeted Chemoselective Oxidation
Kausik K. Nanda 1 , Olivier Mozziconacci , James Small , Leonardo R. Allain ,
,
*
2
3
1
2
1
Roy Helmy , W. Peter Wuelfing
Analytical Sciences, MRL, Merck & Company, Inc., West Point, Pennsylvania 19486
Analytical Sciences, MRL, Merck & Company, Inc., Rahway, New Jersey 07065
Structure Elucidation, MRL, Merck & Company, Inc., West Point, Pennsylvania 19486
1
2
3
a r t i c l e i n f o
a b s t r a c t
Article history:
The ability to produce and isolate relatively pure amounts of relevant degradation products is key to several
aspects of drug product development: (a) aid in the unambiguous structural identification of such degra-
dation products, fulfilling regulatory requirements to develop safe formulations (International Conference
on Harmonization Q3B and M7); (b) pursue as appropriate safety evaluations with such material, such as
chronic toxicology or Ames testing; (c) for a specified degradation product in a late-stage regulatory filing,
use pure and well-characterized material as the analytical standard. Producing such materials is often a
resource- and time-intensive activity, either relying on the isolation of slowly formed degradation products
from stressed drug product or by re-purposing the drug substance synthetic route. This problem is exac-
erbated if the material of interest is an oxidative degradation product, because typical oxidative stressing
Received 22 August 2018
Revised 7 October 2018
Accepted 30 October 2018
Keywords:
drug degradation
oxidative degradation
oxidation toolkit
oxidation
2 2
(H O and radical initiators) tends to produce a myriad of irrelevant species beyond a certain stress
threshold, greatly complicating attempts for isolating the relevant degradation product. In this article, we
present reagents and methods that may allow the rapid and selective enrichment of active pharmaceutical
ingredient with the desired oxidative degradation product, which can then be isolated and used for pur-
poses described above.
®
©
2019 American Pharmacists Association . Published by Elsevier Inc. All rights reserved.
Introduction
magnetic resonance (NMR) and mass spectrometric studies. Once
structurally identified, an integral part of the identification of
One challenging aspect of the drug product (DP) development is
the degradation propensity of the active pharmaceutical ingredient
degradation products is to determine a relative response factor
(RRF), that is, the relative absorbance at a given wavelength,
compared to the API. The RRF determination enables the pharma-
ceutical scientists to calculate the weight percent formation of the
degradation products accurately. The process of determining RRF
often requires tens of milligrams of pure material. Aside from the
structure identification and RRF determination, isolated pure
degradation products, if pure enough and well characterized, could
also be used as sources of material for safety assessment qualifica-
tion studies, and even as analytical standards for the specific im-
purities supporting world market applications for new drugs, and
the eventual commercial manufacturing activities for the product.
Two most common degradation pathways observed in DP are
hydrolytic and oxidative. The focus of this article is oxidative
degradation because it presents more complex reaction profile under
simulated oxidative conditions. Oxidative degradation in DP during
storage is commonly caused by the endogenously present peroxide
impurities, which may react directly as a neutral peroxide species or
via catalytically driven radical processes. As reported in the
(
API), which may directly impact the safety of the drug. The
essential requirement to developing safe and efficacious drug
products dictates a thorough understanding of the degradation
pathways operating on the API in the DP, which often requires the
identification of the degradation product(s). This activity is guided
by ICH Q3B and ICH M7 regulatory guidance calling for identifying
degradation products that meet an established threshold, which is
based on the total daily intake of the drug of interest.
Definitive structural identification of degradation products
(
which are also referred as degradates) ideally requires isolation of
the authentic degradate in the DP in enough quantity for nuclear
This article contains supplementary material available from the authors by request
or via the Internet at https://doi.org/10.1016/j.xphs.2018.10.059.
*
Correspondence to: Kausik K. Nanda (Telephone: þ1-215-652-0371).
E-mail address: kausik_nanda@merck.com (K.K. Nanda).
https://doi.org/10.1016/j.xphs.2018.10.059
022-3549/© 2019 American Pharmacists Association . Published by Elsevier Inc. All rights reserved.
®
0

2
K.K. Nanda et al. / Journal of Pharmaceutical Sciences xxx (2019) 1-10
Figure 1. Oxidative forced stress experiment with AIBN for an MSD development candidate.
1
literature, for the purpose of understanding a molecule’s oxidative
sensitivity as well as degradate isolation, the API is enriched with
oxidative degradation products by the reaction of azonitriles, such as
elaborated, along with a few examples of enriching API with known
degradation products. The driver in choosing the reagents in the
toolkit is their effectiveness and ease of use.
2
2 2
,2’-azobisisobutyronitrile (AIBN) or H O with API. But these pro-
cesses, when performed at higher temperature or under extended
time to yield larger amount of degradates for isolation, may produce
a mixture of irrelevant degradates, exemplified in Figure 1 by a
typical forced stress experiment with AIBN for one of our develop-
ment candidates. It is also possible, and often practiced, to enrich DP
with desired oxidative degradation products under accelerated
stress (i.e., high temperature and humidity). But this takes long lead
times and gives low conversion to degradates (e.g., see Hartauer
Discussion
The main reagent for the commonly observed oxidative degra-
dation of the API in DP is oxygen. But the abundant atmospheric
molecular oxygen cannot react directly with most organic molecules
because of the difference in the electronic ground statesdtriplet
ground state for molecular oxygen and singlet ground state for the
majority of organic molecules. The selection rules derived from
quantum mechanics prohibit such reactions because these are “spin
forbidden.” For the reactions to occur, the molecular oxygen has to
2
et al. ). Stressing at higher temperatures can help the formation of
higher levels of the degradates of interest, but it is often accompa-
nied by other unintended reactions, creating less tractable prepara-
tory chromatographic isolations. Given the timelines under which
pharmaceutical scientists operate in the development area, faster
and practical ways of enriching API with relevant oxidative degra-
dates and their isolation in pure form is highly desirable.
1
2
be excited to the singlet state ( O ), or the organic molecules have to
react with other “reactive oxygen species,” such as hydroxyl radical
.
( OH) or hydrogen peroxide (H O ).
2 2
1
Oxidation of the API by the singlet oxygen ( O ) is often
encountered in photo degradation processes. Although photo
2
In this article, we present methods using oxidants used in
organic synthesis to enrich a given API with commonly observed
oxidative degradation products. This collection of methods and
reagents can be viewed as an Oxidation Toolkit for the pharma-
ceutical scientists, and we will use this terminology in the rest of
the discussion. We also present high yield conversion with com-
mon reagents used in oxidative forced stress studies (an AIBN-like
degradation is a valid degradation mechanism in DP, it is controlled
effectively, most of the time, by the use of proper protective pack-
aging. On the other hand, oxidative degradation with hydroper-
.
.
oxides (ROOH) or with hydroxy ( OH)/alkoxy ( OR)/peroxy (OOR)
radicals cannot be controlled in most instances, which leads to the
presence of degradates above ICH Q3B identification thresholds.
The oxidation toolkit presented in this article is geared toward
accessing these latter types of degradates.
2 2
reagent and H O in gas phase). Hence, this is not a collection of
1
reagents to replace those typically used to assess oxidative sensi-
tivity for APIs as described above but rather to use broader chem-
istry, not related to the DP excipients, to produce large amounts of
degradates (rather than trace amounts). Here we list reagents and
methods for oxidation reactions relevant to different functional
groups, and the pharmaceutical scientist will prioritize screening
based on the information available, which generally comes from
forced stress and accelerated stress experiments performed on the
API and on the DP. Although application of this idea could be found
Oxidative forced stress studies are commonly performed on the
API in solution phase using azonitriles such as AIBN, and H O .
2 2
Forced stress of API with AIBN emulates peroxy radicalemediated
(1-electron) oxidation (Scheme 1). On the other hand, oxidation
with H O follows a 2-electron pathway, which is generally
2
2
observed for N- or S-oxidation (Scheme 2).
Analysis of experimental data (chromatographic, spectroscopic,
mass spectrometric) obtained from accelerated stress experiments
on DP and from forced stress studies of API, along with the analysis of
structure reactivity of the API molecule often indicate the plausible
structures of degradates. Sometimes, it is possible, using compelling
chemical reactivity argument, to establish a proposed structure in
the absence of isolated pure material with comprehensive
2
-4
in isolated cases in the literature, and was introduced earlier as a
1
“
tool box” by Baertschi et al., this concept has further been
developed in this article to include a much broader array of re-
agents and methods. Principles behind this toolkit will be
Scheme 1. Formation of peroxy radicals from AIBN and subsequent oxidative degradation of API.

K.K. Nanda et al. / Journal of Pharmaceutical Sciences xxx (2019) 1-10
3
degradation products (and their isolation in selected cases) using a
set of reactions which are referred to as oxidation toolkit.
Table 1 lists some of the most common transformations
observed in the oxidative degradation of the API in DP as evidenced
1,5,6
by literature references
and by in-house examples, along with
recommended reagents for the enrichment of the API with desired
degradation products. Although process chemists will well know
these reactions to enable preparation of these degradates, it is
useful for the pharmaceutical scientist to be aware and have access
as needed. It should be noted that the reagents listed for the
enrichment of the API with desired degradation products are only a
subset of the reagents reported in the chemistry literature. The
choice of reagents, in Table 1, is based on the ease of their use and
their general effectiveness to enrich API with key degradation
products. So, it is expected that the practicing pharmaceutical sci-
entists will eventually add new tools (reagents and methods) to this
toolkit as the situation warrants.
Scheme 2. Two-electron oxidation of amines with hydrogen peroxide.
spectroscopic data. But, for the definitive structure ID, the degrada-
tion products need to be isolated in pure form. Even when it is
possible to isolate the pure degradation products from the stressed
DP or from appropriate forced stress experiments on API, the process
may be time-consuming and often the amounts of isolated degra-
dates are quite small to be used in other activities, such as RRF
determination. Ideally, pharmaceutical scientists need to have other
means (than stressing DP or relying upon limited opportunities
offered by established forced stress experimental conditions) to
enrich the API with degradation product(s), preferably in a clean
manner. Here we present examples of enrichment of API with
Entry 1 (Table 1) shows oxidation of a secondary alcohol to
ketone and entry 2 shows the oxidation of a primary alcohol to
aldehyde. For both oxidations, Dess-Martin periodinane is an
Table 1
Specific Oxidative Transformations and Recommended Reagents to Enrich API With Oxidative Degradates
Entry
1
Transformation
Reagents
(i) Dess-Martin periodinane (see preparation of 1, Scheme 3)
7
(
(
(
ii) Pyridinium chlorochromate (Corey and Suggs )
iii) Pyridinium dichromate (Corey and Schmidt )
8
(Doromy and Castro ; Baldwin et al.10)
9
iv) RuO
2
.xH2O/NaIO
4
2
3
4
(i) Dess-Martin periodinane (see preparation of 1, Scheme 3)
7
(
ii) Pyridinium chlorochromate (Corey and Suggs )
iii) Pyridinium dichromate (Corey and Schmidt )
8
(
(i) TPAP/NMO (Houri et al.12
)
4
(i) KMnO )
(Jursic13; Shaabani et al.14
2 2
O
(Coperet et al.15)
5
6
(i) MTO/H
ii) m-CPBA (Ray et al.
16
(
)
(i)
2 2
H O (see preparation of 2, Scheme 5)
2
(
(
ii) Peracetic acid (Hartauer et al. )
iii) O /RuCl
2
3
(Jain and Sain17
)
7
(i) Oxone (Trost and Curran18
ii) H /[VO F(dmpz) )
] (Hussain et al.19
)
(
2
O
2
2
2
8
9
(i) Oxone (Trost and Curran18
)
2
(i) NaNO )
/HCl (see preparation of 3, Scheme 6 and Owens20
1
1
0
1
(i) m-CPBA (March23
)
(i) FeCl
ii) AAPH (see enrichment of ibuprofen with 5 and 6, Scheme 8 and Betigeri et al.
(see enrichment of dextromethorphan with 7 and 8, Scheme 7 and Nanda et al.24)
3
25
(
)
2 2
H O )
(g) (see enrichment of rosuvastatin calcium with 4, Scheme 9 and Zhu et al.29
1
2
3
(i)
ii) Dess-Martin periodinane (see preparation of 1, Scheme 3)
(
2
(i) SeO )
(Suksamram et al.30; Khurana et al.31
1

4
K.K. Nanda et al. / Journal of Pharmaceutical Sciences xxx (2019) 1-10
Scheme 3. Oxidation of cyclandelate with Dess-Martin periodinane.
effective reagent and provides easy reaction setup and postreaction
workup. This has been demonstrated (Scheme 3) in our lab for the
enrichment of cyclandelate with its degradation product, 1, which
was isolated in 86% yield. It should be noted that the whole process
of enrichment, workup, and isolation has easily been performed in
a single day. Two other reagents, pyridinium chlorochromate and
remain soluble in dichloromethane layer, whereas the oxidant stays
in the aqueous phase. This enrichment method is fast and very easy
2 2
to execute in the laboratory. Although MTO/H O should be the
method of choice, the same transformation could also be affected
16
by H
Oxidation of tertiary amines to their corresponding N-oxides
with H is a 2-electron process, where the lone pair of electrons
2 2
O or m-chloroperoxybenzoic acid.
pyridinium dichromate, can also affect the similar trans-
2 2
O
formations.7 Combination of RuO
,8
.xH
O (catalytic) and NaIO
on the N-atom is involved (Scheme 2). So, it is important to
maintain the pH of the reaction medium at or above the pKa of the
oxidizable amine. At pH below the pKa, the amine-N will remain
protonated rendering the lone pair of electrons inaccessible toward
desired reaction. This principle has been demonstrated in the
synthesis of the N-oxide of raloxifene (2, Scheme 5) from raloxifene
hydrochloride. Raloxifene hydrochloride was neutralized in situ,
2
2
4
(
oxidant) has been used successfully for selective oxidation of
9
,10
secondary alcohol to ketone.
Although the reagents listed pro-
duce the desired degradation products most of the time, there will
be times when these might not work. Such an example is the
oxidation of lovastatin to its oxolactone, 9 (Scheme 4). All 4 re-
agents, listed under entry 1 (Table 1), failed to do the oxidation of
lovastatin in a chemoselective fashion. But, a slight variation of the
dichromate oxidant, namely, cetyltrimethylammonium dichro-
2 2
before reaction with H O , by 1 equivalent of NaOH. Recrystalliza-
tion (trituration) provided the desired degradation product 2 in 75%
yield. Application of the principles of oxidation toolkit enabled us to
enrich the API with 2 and made its subsequent isolation an efficient
process. It should be noted here that the enrichment of raloxifene
hydrochlorideecontaining tablets with only 1% of 2 was accom-
11
mate, has been reported to accomplish the desired chemo-
selective oxidation to 9.
Exhaustive oxidation of primary alcohols yields carboxylic acids
(
entries 3 and 4, Table 1). A very useful catalyst for the trans-
formation of primary aliphatic alcohol to the corresponding acid is
tetrapropylammonium perruthenate, n-Pr NRuO (TPAP), some-
ꢀ
2
plished after 8 weeks of stressing the tablets at 105 C. The syn-
2
4
4
thesis of 2 has also been reported by Hartauer et al., in multi-gram
times called Ley-Griffith reagent. The co-oxidant in this reaction is
N-methylmorpholine N-oxide (NMO). An example of the reaction of
TPAP/NMO with a primary alcohol to yield the corresponding car-
scale, from the free base of raloxifene hydrochloride using peracetic
acid. The N-oxidation of tertiary amines can also be conducted
under oxygen atmosphere with RuCl as catalyst.
3
Sulfoxides (entry 7, Table 1) and sulfones (entry 8, Table 1) are
common oxidative degradation products of sulfides, which are
formed by the reaction of hydroperoxides with API in DP. It should
be noted that the oxidation of sulfoxide group in the API will also
produce sulfone as the degradation product. An efficient way of
17
12
boxylic acid could be found in the work reported by Houri et al.
The oxidation was complete in 3 h and isolated yield of the car-
boxylic acid was 65%. Potassium permanganate, KMnO , is also an
4
efficient reagent for the exhaustive oxidation of alcohols to car-
boxylic acids, particularly benzylic alcohols to benzoic acids.
Oxidation with permanganate can be performed efficiently in
forming sulfone from sulfide or sulfoxide is to use oxone as oxidant
13
18
aqueous medium in the presence of surfactants, or if nonaqueous
reaction medium is preferred, the oxidation could be performed in
in H
2
O-MeOH reaction medium. By controlling the amount of
oxone used and reaction time, it is possible to get sulfoxide as the
major product (instead of sulfone) in some instances. A more
chemoselective way of making sulfoxides is to use a vanadium
14
the presence of an ion-exchange resin.
A common oxidative degradation is the formation of N-oxides of
pyridines (entry 5, Table 1) or of N-containing heterocycles. Such
oxidation can be carried out very effectively by the method of
Coperet et al.15 using methyltrioxo rhenium (MTO) as catalyst in the
19
catalyst [VO
An aromatic sulfide could be oxidized to sulfoxide using HNO
produced from the reaction of NaNO and HCl (entry 9, Table 1). The
operating mechanism, in this case, is via a cation radical interme-
2 2 2 2
F(dmpz) ] in the presence of H O .
,20
2
2
presence of H
2 2
O . The transformation is carried out in a biphasic
21
reaction medium, where the substrate and the reaction product
diate, not a standard 2-electron pathway, which is in play with
Scheme 4. Ineffective reagents for the chemoselective oxidation of lovastatin to lovastatin-oxolactone.

K.K. Nanda et al. / Journal of Pharmaceutical Sciences xxx (2019) 1-10
5
Scheme 5. Oxidation of the piperidine-N in raloxifene hydrochloride to the corresponding N-oxide.
peroxide- or oxone-mediated oxidation. This unique mechanism
provides an advantage to oxidize aromatic sulfides to sulfoxide
selectively, in the presence of amines or pyridines which are also
susceptible to oxidation to corresponding N-oxides. This method-
ology has successfully been applied to make the sulfoxide degra-
dation product (3) of fluphenazine (Scheme 6) in 70% isolated yield,
without oxidizing the piperidine-nitrogen to corresponding N-ox-
ide. Isolation of 3 was accomplished by preparative thin layer
chromatography (TLC), an easily executable and efficient method.
Table 1) are also affected by the action of alkoxy or peroxy radicals,
because of the low bond dissociation energy of the allylic C-H
bond. In solution, alkoxy radical can be generated by the homol-
5
ysis of 2,2’-azobisisobutyronitrile (AIBN) in acetonitrile-water in
2
6
the absence of any alcoholic solvent, whereas hydroxy radical,
which is more reactive than alkoxy radicals because of the differ-
27
ence in bond dissociation energy, can be generated in solution
2
8
using Fenton chemistry. Reaction of hydrogen peroxide in gas
2
9
phase with solid API is shown to be effective in generating
oxidative degradation products. Dess-Martin periodinane, as
described earlier, can also act as an oxidant for the oxidation of
allylic alcohol to ketone. All the above methods can be used for the
oxidation of allylic or benzylic carbons, depending on the reactivity
of individual API molecule. Here we present an example of the
oxidation of an allylic alcohol to the corresponding ketone (Scheme
2
2
Epoxides, which are mutagenic alerting structures, are readily
formed from olefinic moieties in the API, by the action of hydro-
peroxides. There are many established ways to make epoxides from
2
3
olefins. The most widely used method, also known as Prilezhaev
reaction, employs the oxidant m-chloroperoxybenzoic acid (entry
10, Table 1). This method of enriching API with its corresponding
epoxide is fast, efficient, and easy to execute.
2 2
9), using H O in gas phase. It was found that hydrogen peroxide, in
Oxidative degradation products arising from the oxidation of
benzylic position are commonly observed in DP. Usually, this kind
the gas phase, can react with solid API, rosuvastatin calcium, to
generate the oxidative degradation product 5-keto-rosuvastatin
(4). The degradate, 4, was generated cleanly (24% by peak area) in 1
week from the oxidation of the allylic alcohol in rosuvastatin. It
.
.
1,5
of oxidation is affected by alkoxy ( OR) or peroxy ( OOR) radicals,
2
4
or by transition metal ions. Hence, for the enrichment of the API
with such oxidative degradation products, any of these 2 mecha-
nisms could be exploited (entry 11, Table 1), depending on the
substrate. For example, oxidative degradation products (7 and 8) of
should be noted that the degradate 4 cannot be generated using
1
2
H O
2
in solution, under typical forced stress condtions (2-electron
oxidation). Allylic oxidation to allylic alcohol (entry 13, Table 1) can
3
0,31
dextromethorphan were obtained by using FeCl
3
(Scheme 7). In this
2
be accomplished by the use of selenium dioxide (SeO ).
instance, the oxidative mechanism goes through an aromatic cation
radical intermediate.24 On the other hand, a radical-initiated
mechanism was exploited to enrich ibuprofen with its degrada-
tion products (5 and 6, Scheme 8), and 2,2’-azobis(2-
amidinopropano) dihydrochloride (AAPH) was used to generate
the reactive peroxy radical species. AAPH is a variant of AIBN. In
this case, AAPH provided a cleaner and faster enrichment profile
than that obtained from the use of AIBN.
Qualification of degradation products is required, as per ICH
Q3B, when they reach a certain threshold. The qualification process
involves synthesis of pure degradate which is then doped into the
API to be used in the animal toxicological studies. The independent
synthesis of the degradate is a time-consuming and expensive
process in the drug development area. Application of the oxidation
toolkit may expedite this process considerably, saving time and
resources. Using the principles of the oxidation toolkit, an API can
be enriched with its desired oxidative degradation product in a
clean chemoselective fashion. This material can then be used
2
5
Oxidation of allylic carbon to its corresponding alcohol (entry
13, Table 1) and oxidation of allylic alcohol to ketone (entry 12,
Scheme 6. Oxidation of fluphenazine to fluphenazine-sulfoxide.

6
K.K. Nanda et al. / Journal of Pharmaceutical Sciences xxx (2019) 1-10
Scheme 7. Iron(III)-mediated oxidation of dextromethorphan.
directly in the animal toxicological studies for the qualification of
the degradate. We were able to use this principle to enrich rosu-
vastatin calcium with 1% of 5-keto-rosuvastatin (4), in a clean
chemoselective way. The enriched API was then successfully used
to qualify 4 in animal toxicological studies.
0%, 99.95%, and 0.05% (v:v:v) for solvent B. The following elution
conditions were set: 5% of solvent B for 0.1 min, followed by a linear
increase of B from 5% to 99% within 2 min. Column temperature
ꢀ
was maintained at 50 C. An SQD detector (Waters Corporation) was
used coupled to the UPLC system. The mass spectrometer was
equipped with an electrospray ion source. The capillary was set to
3.0 kV. The cone and extraction voltages were 20.0 eV and 3.0 eV,
Materials
ꢀ
respectively. Source and desolvation temperatures were 120 C and
ꢀ
All chemicals were used as received. Dess-Martin periodinane,
400 C, respectively.
hydrogen peroxide (H
2
O
2
) solution (35%), pyridinium dichromate,
For the LC-MS analysis of raloxifene hydrochloride, raloxifene-
pyridinium chlorochromate, sodium nitrite, ibuprofen, 1-
hydroxyibuprofen, 2-(4-isobutyrylphenyl)-propanoic acid, raloxi-
fene hydrochloride, and titanium(III) chloride were purchased from
Sigma-Aldrich (St. Louis, MO). Cyclandelate and fluphenazine hy-
drochloride were obtained from Chem-Impex International (Wood
Dale, IL) and Matrix Scientific (Columbia, SC), respectively. Lova-
statin was obtained from TCI America (Portland, OR). Rosuvastatin
calcium was obtained from Hanmi Fine Chemical Co. (Kyonggi-Do,
Korea). Trifluroacetic acid (TFA), hydrochloric acid, and HPLC-grade
solvents were obtained from Fisher Scientific (Fair Lawn, NJ).
N-oxide (2), fluphenazine, and fluphenazine-sulfoxide (3), samples
were injected (3
mm, 1.7 m) from Waters Corporation and eluted with a linear
gradient delivered at a flow rate of 300 L/min by an Acquity ultra-
mL) onto a BEH C18 RP Shield column (5 cm ꢁ 2.1
m
m
performance liquid chromatography system (Waters Corporation).
Mobile phases consisted of water/acetonitrile/TFA at a ratio of 99%,
1%, and 0.08% (v:v:v) for solvent A and a ratio of 1%, 99%, and 0.08%
(v:v:v) for solvent B. The following elution conditions were set: 2%
of solvent B for 0.5 min, followed by a linear increase in B from 2% to
ꢀ
98% within 5 min. Column temperature was maintained at 26 C. A
Synapt-G1 (Waters Corporation) hybrid mass spectrometer oper-
ated in the MSe mode and acquiring data with the time-of-flight
analyzer was used coupled to the UPLC system. The instrument
was operated for maximum resolution with all lenses optimized on
Experiment and Analysis
Reversed-phase chromatography for the analysis of the oxida-
tion experiments involving rosuvastatin was carried out on an
Agilent 1100 HPLC (Agilent Technologies, Santa Clara, CA) equipped
with a quaternary pump, heated column compartment, diode array
detector, and autosampler. The chromatographic conditions
þ
the [M þ H] ion from the Leu-enkephalin. The capillary was set to
3.0 kV. The cone and extraction voltages were 40.0 eV and 4.0 eV,
respectively, and Ar was admitted to the collision cell. Spectra were
acquired within a mass range 100-2000 amu and accumulating
data for 1.5 s per cycle. The low collision energy in the trap was set
to 12.0 V. The high collision energies were ramped from 18.0 V to
32.0 V during each scan. Source and desolvation temperatures were
included a Waters Symmetry Shield RP18 column of dimensions 4.6
ꢀ
mm ꢁ 150
m
m and 3.5
m
m particle size held at 45 C; flow rate was
maintained at 1.5 mL/min; injection volume was 60
mL; tray tem-
ꢀ
ꢀ
ꢀ
perature was 20 C; detector wavelength was 242 nm; mobile phase
A consisted of a mixture of 0.01 M potassium dihydrogen phos-
phate solution pH 2.5 and acetonitrile (95:5); mobile phase B
consisted of acetonitrile. The mobile phase gradient is listed in the
Supporting Information (Table S1).
100 C and 150 C, respectively. Time to mass calibration is made
with CsI cluster ions acquired under the same conditions.
High-resolution solution NMR data were collected on either a
Bruker Avance HD III 700 (Bruker BioSpin Corporation, Billerica,
MA) NMR spectrometer equipped with a 1.7-mm HCN TCI Micro-
Cryoprobe or a Varian VNMRS 600 MHz NMR spectrometer
equipped with a 5-mm One NMR probe. Compounds 1-3 were each
For the liquid chromatography-mass spectrometry (LC-MS)
analysis of reactions and isolated products, samples were injected
(
0.5
mm, 1.7
with a linear gradient delivered at a flow rate of 300
m
L) onto a Waters Acquity UPLC BEH C18 column (1 cm ꢁ 500
m) from Waters Corporation (Milford, MA) and eluted
L/min by an
dissolved in about 150
m
L of DMSO-d6 (CIL, 99.96þ% D) and placed
m
into a 3-mm Wilmad NMR tube and capped. The sample was used
for solution NMR experiments using standard pulse sequences
(proton, zTOCSY, ME-gHSQC, gHMBC, and ROESY experiments) at
m
Acquity ultra-performance liquid chromatography system (Waters
Corporation). Mobile phases consisted of water/acetonitrile/TFA at
a ratio of 99.95%, 0%, and 0.05% (v:v:v) for solvent A and a ratio of
ꢀ
25 C. Water suppression was used (as needed) for various experi-
ments to aid in the structure elucidation of the oxidative
Scheme 8. Benzylic oxidation of ibuprofen with AAPH.

K.K. Nanda et al. / Journal of Pharmaceutical Sciences xxx (2019) 1-10
7
Scheme 9. Oxidation of rosuvastatin calcium to 5-keto-rosuvastatin.
degradates. All data were analyzed and archived using Mnova 12.0
overnight. The reaction solution was then evaporated to dryness on
a rotary evaporator. The resulting solid was triturated in methanol-
(
Mestrelabs).
Analytical TLC was performed on EM Reagent 0.25-mm silica gel
0-F plates.
2
H O (2:1 v/v) to yield the desired degradation product 2 as an off-
white solid (28 mg, 75%). H NMR (600 MHz, DMSO-d ): d 7.62 (d,
6
1
6
Authentic commercial compounds were used to identify degra-
J ¼ 8.8 Hz, 2H, ArH) 7.32 (m, 1H, ArH), 7.31 (s, 1H, ArH), 7.12 (d, J ¼
dation products of ibuprofen by comparing retention time in HPLC
and UV-vis spectra. Oxidation product of rosuvastatin calcium was
identified by comparing HPLC retention time and UV-vis spectrum
8.6 Hz, 2H, ArH), 6.89 (d, J ¼ 8.8 Hz, 2H, ArH), 6.85, (m, 1H, ArH),
6.63 (d, J ¼ 8.4 Hz, 2H, ArH), 4.56 (t, J ¼ 4.9 Hz, 2H, OCH
2
), 3.53 (br s,
2H, NCH
2
), 3.30, 3.07 (m, 4H, CH2), 2.08, 1.54 (m, 4H, CH2), 1.58, 1.33
13
(
in-house method) from previous in-house research. Commercially
(m, 2H, CH ); C (150 MHz, DMSO-d6, ppm): d 191.9, 161.5, 158.0,
2
available compounds corresponding to the degradation products of
cyclandelate, raloxifene, and fluphenazine were not available,
prompting isolation and structural characterization by NMR and
mass spectrometry. Isolation of degradation products was accom-
plished either by recrystallization or by normal-phase preparative
chromatography using ISCO Combiflash Sg 100c instrument with
Biotage Flash 40 cartridge/column. When appropriate, isolation was
performed with preparative TLC using silica gel preparative TLC
141.0, 131.7, 131.3, 130.2, 129.5, 115.5, 114.7, 114.3, 106.9, 67.3, 61.8,
þ
64.6, 20.8, 19.7; MS m/z (M þ H) 490.2 found, 490.4 required.
Detailed analysis of NMR data is presented in the Supporting
Information.
Mass Spectrometry Analysis of Raloxifene and Its Oxidation
Product, 2
plates (1000
m
m, 20 ꢁ 20 cm) from Sorbent Technologies Inc.
LC-MS analyses of raloxifene (tel ¼ 3.0 min; Supporting
Information, Fig. S1A) and
2
(tel
¼
3.15 min; Supporting
Enrichment of API With Degradation Products, Isolation and
Structural Characterization
Information, Fig. S1B) show mass-to-charge ratios m/z ¼ 474.10
(
Supporting Information, Fig. S2A) and m/z ¼ 490.15 (Supporting
Information, Fig. S2B), respectively. Comparison of the collision-
induced dissociation (CID) spectra of raloxifene and 2 demon-
strates the presence of an N-oxide moiety in 2 (Fig. 2). The N-oxide
can easily be confirmed by the unusual thermal activation/loss of
oxygen (deoxygenation of M þ H, F1; Fig. 1b). This deoxygenation
process is associated with thermal activation and does not result
from collisional activation in the desolvation region of the API source
and provides evidence for the presence of an N-oxide in the mole-
Warning
Reagents for oxidation should be handled with care as indicated
in corresponding Material Safety Data Sheet.
Enrichment of Cyclandelate With Oxidative Degradation Product,
3,3,5-Trimethylcyclohexyl 2-oxo-2-phenylacetate (1), and Isolation
and Structural Characterization
3
2,33
To a solution of cyclandelate (300 mg, 1.1 mmol) in dichloro-
methane (6 mL) was added solid Dess-Martin periodinane (700
mg, 1.7 mmol). The resulting mixture was stirred at room tem-
perature for 2 h and filtered through 0.45-micron polytetra-
fluoroethylene syringe filter. The filtrate was concentrated on a
rotary evaporator and purified by preparative normal-phase
chromatography (40 g silica gel column, 0% to 30% ethyl acetate
in hexanes over 20 min) to yield the desired product 1 as a
cule. Such specific loss was previously reported in the literature.
In addition, the N-oxide allows for the presence of a positive charge
on the nitrogen of the piperidine, which favors gas-phase fragmen-
tation of the ether bond forming the fragment ion F2 (m/z 389.1,
Fig. 2b). In contrary, in the absence of oxidation, the protonation in
gas phase of the sulfur atom permits the ring opening of the thio-
phene moiety and the release of fragment ion F3 (m/z 362.0, Fig. 2a).
Fragment ions F4 (m/z 269.0) are the results of the fragmentation at
the carbonyl moiety in both raloxifene and 2 (Figs. 2a and 2b).
Fragmentation at the carbonyl moiety in 2 is likely to proceed from
the deoxygenated ion (M þ H, m/z 474.17). Fragment ion F5 (m/z
128.1) is unique to the fragmentation pattern of 2 (Fig. 2b). F5 is likely
to be the 1-hydroxy-vinylpiperidinium. Fragment F6 (m/z 112.1) is
likely the vinylpiperidinium ion, which results from the protonation
of the nitrogen of the piperidine moiety. F6 is common to both the
fragmentation patterns of raloxifene and 2 (Figs. 2a and 2b).
1
colorless viscous liquid (260 mg, 86%). H NMR (600 MHz, DMSO-
d
6
):
d
7.99 (dd, J ¼ 8.4, 1.3 Hz, 2H, ArH), 7.65 (m, 1H, ArH), 7.51 (dd,
J ¼ 8.3, 7.4 Hz, 2H, ArH), 7.31 (s, 1H, ArH), 5.22 (m, 1H, CH), 2.14,
1
1
(
.03 (m, 2H, CH
.00 (s, 3H CH ), 0.99 (s, 3H, CH
150 MHz, DMSO-d6):
2
), 1.84, 1.27 (m, 2H, CH
2
), 1.38, 0.84 (m, 2H, CH
2
),
C
1
3
3
3
), 0.95 (d, J ¼ 6.5 Hz, 3H, CH
3
);
d
187.0, 163.6, 134.7, 132.8, 130.0, 128.9,
þ
7
4.0, 47.4, 43.8, 40.1, 27.2, 25.4, 33.0, 25.5, 22.3. MS m/z (2M þ Na)
571.3 found, 571.3 required. Detailed analysis of NMR data is
presented in the Supporting Information.
To further confirm the presence of an N-oxide moiety, 2 is
reduced in the presence of TiCl
presence of TiCl3. Oxidized raloxifene (2) is prepared at 1 mg/mL in
50:50 milliQ H2O:ACN. An aliquot of 200 L of this solution is
mixed with 20 L of TiCl3/HCl. The solution is vortexed for 1 min.
The reaction is quenched by addition of 20 L of NH OH (5 N). The
sample is vortexed for 30 s. The latter is diluted by the addition of
400 L of 50:50 milliQ H2O:ACN solution. The sample is allowed to
3
(N-oxide moiety is reduced in the
Enrichment of Raloxifene With Oxidative Degradation Product,
Raloxifene-N-oxide (2), Isolation and Structural Characterization
To a solution of raloxifene hydrochloride (40 mg, 0.08 mmol) in
m
m
2
mL methanol was added 0.2 mL water, followed by 80
aqueous NaOH solution. To this solution was added aqueous H
solution (35%, 0.2 mL) and allowed to stir at room temperature
m
L of 1 N
m
4
2 2
O
m

8
K.K. Nanda et al. / Journal of Pharmaceutical Sciences xxx (2019) 1-10
Figure 2. CID spectra of raloxifene (a) and 2 (b). Significant fragment ions are reported as F1-F6 and their tentative structures are presented in panel (c).
stand for 5 min and then the supernatant is decanted. An aliquot of
00 L of the supernatant is transferred into a new vial for mass
spectrometry analysis.) which has been documented as a specific
reducing agent of N-oxides. The LC-MS data clearly demonstrate
the reduction of 2 into the original raloxifene molecule (Supporting
Information, Fig. S4A-S4C).
Enrichment of Fluphenazine With Oxidative Degradation Product,
Fluphenazine-sulfoxide (3), Isolation and Structural
Characterization
To a stirring solution of fluphenazine (22 mg, 0.05 mmol) in water
(1 mL) was added 2 drops of concentrated HCl, followed by 2 drops of
1
m
3
4
2
NaNO solution (100 mg/mL in water). Color of the reaction mixture
Figure 3. CID spectra of fluphenazine (a) and 3 (b). Significant fragment ions are reported as F7-F10 and their tentative structures are presented in panel (c).

K.K. Nanda et al. / Journal of Pharmaceutical Sciences xxx (2019) 1-10
9
became brown and evolution of gas occurred. After 2 min, concen-
trated aqueous NH OH solution was added dropwise until the reac-
tion solution became basic. Extract this aqueous solution with
Enrichment of Dextromethorphan With Oxidative Degradation
Products, (9 ,14 )-3-Methoxy-17-methylmorphinan-10-ol (7) and
(9 ,13 ,14 )-3-Methoxy-17-methylmorphinan-10-one (8)
The enrichment of dextromethorphan with its oxidative
degradation products, 7 and 8, was performed according to a pre-
4
a
a
a
a
a
dichloromethane (2ꢁ). Combined organic extracts were dried over
2 4
Na SO , filtered, and concentrated. Purified using preparative TLC
2
3
(
eluent: CH
2
Cl
2
-MeOH-conc. NH
4
OH 90:9:1 v/v/v) to yield the
3
viously published method. A 0.5 mg/mL FeCl solution (5 mL) in
1
desired product 3 as a white solid (16 mg, 70%). H NMR (600 MHz,
DMSO-d6) d 8.19 (d, J ¼ 7.9 Hz, 1H, ArH), 7.98 (dd, J ¼ 7.7, 1.7 Hz, 1H,
ArH), 7.94 (s, 1H, ArH), 7.81 (d, J ¼ 8.5 Hz, 1H, ArH), 7.74 (ddd, J ¼ 8.7,
0.1 N aqueous HCl was mixed with a solution of dextromethorphan
(5 mL, 0.1 mg/mL in 0.1 N aqueous HCl-MeCN 75:25 v/v), and the
resulting mixture was stirred at room temperature for 48 h to yield
9% of 7 and 16% of 8.
7
.70,1.70 Hz,1H,ArH), 7.57 (dd, J ¼ 8.0,1.6 Hz,1H, ArH), 7.35 (t, J ¼ 7.3
2
Hz,1H, ArH), 4.53, 4.50 (m, 2H, NCH ), 3.80-3.30 (broad multiplet, 8H,
NCH
NCH
2
), 3.45 (m, 1H, OCH
2
), 2.32, 2.24 (m, 2H, NCH
2
), 2.31 (m, 2H
Conclusion
13
2
), 2.33 (m, 4H, NCH ), 2.32 (m, 4H, NCH
2
2
), 1.86 (m, 2H CH
2
);
C
NMR (150 MHz, DMSO-d6)
d 139.2, 138.2, 132.9, 132.8, 133.6, 131.3,
We established here the concept of an oxidation toolkit, a
collection of oxidants and methods, for the selective enrichment of
API with a desired oxidative degradation product. Selective oxida-
tive transformations of the API have been listed, along with the
relevant reagents, with the goal to pursue such transformations in
preparative scale. To demonstrate the applicability of this concept,
several drug molecules were enriched with relevant oxidative
degradation product(s). The toolkit can also be used, as exemplified
by the qualification of rosuvastatin-5-keto acid (4), for enriching
API with desired oxidative degradation product for the use in ani-
mal toxicological studies. The oxidation toolkit presented in this
article is not meant to be a static collection of reagents and
methods, but a concept to be evolved and enriched over time with
new methods and reagents that allow chemoselective, relevant
transformations.
1
24.2,123.0,118.0,117.5,114.2,60.8, 59.0, 53.8, 44.6,23.9;MSm/z (M þ
þ
H) 454.4 found, 454.2 required. Detailed analysis of NMR data is
presented in the Supporting Information.
Mass Spectrometry Analysis of Fluphenazine and Its Oxidation
Product, 3
LC-MS analyses of fluphenazine (tel ¼ 3.0 min; Supporting
Information, Fig. S1C) and 3 (tel ¼ 2.50 min; Supporting Information
Fig. S1D) show mass-to-charge ratios m/z ¼ 438.13 (Supporting
Information, Fig. S3A) and m/z ¼ 454.11 (Supporting Information,
Fig. S3B), respectively. The LC-MS data reveal that the oxidation of
fluphenazine results in the formation of a sulfoxide moiety (Figs. 3a
and 3b). The latter is easily identified by its CID spectrum (Fig. 3b).
Indeed, the comparison of the fragment ions (F7-F10) generated
during CID of fluphenazine and 3 allows for the localization of the
oxidation site at the sulfur atom of fluphenazine. Fragment ion F7 (m/
z 406.2, Fig. 3c) corresponds to the loss of sulfoxide. During electro-
spray ionization, sulfoxide is easily protonated, which ultimately
permits the cleavage of the C-S bonds. The latter has been extensively
used in proteomic studies to develop sulfoxide-containing cleavable
Supporting Information
Chromatographic, mass spectrometric, and NMR data.
Acknowledgments
cross-linkers to ease the identification of amino acids during protein-
protein interactions.3
3
2
5-38
Fragment ions F8 (Figs. 3a and 3c, m/z
We thank Xiaoliu Geng for providing forced stress data on MSD
development candidates.
08.05) and F9 (Figs. 3b and 3c, m/z 324.06) correspond to the loss of
-(piperazin-1-yl)ethan-1-ol (131.1 Da) moieties during the CID of
fluphenazine and 3, respectively. The absence of fragment ion F8 and
the presence of fragment ion F9 during CID of 3 (Figs. 3b and 3c)
demonstrate the oxidation of the sulfur atom of fluphenazine into
sulfoxide. Fragment F10 (m/z 171.1) corresponds to the 3-(4-(2-
hydroxyethyl)piperazin-1-yl)propan-1-ylium ion (Fig. 3c). F10 is
present during CID of fluphenazine and 3, which demonstrates the
absence of oxidation of the piperazine moiety.
References
1. Baertschi SW, Alsante KM, Reed RA, eds. Pharmaceutical Stress Testing: Pre-
dicting Drug Degradation. 2nd ed. New York, NY: Informa Health Care; 2011.
2
. Hartauer KJ, Arbuthnot GN, Baertschi SW, et al. Influence of peroxide in
povidone and crospovidone on the stability of raloxifene hydrochloride in
tablets: identification and control of an oxidative degradation product. Pharm
Dev Technol. 2000;5:303-310.
3
. Jansen PJ, Smith WK, Baertschi SW, Dorman DE, Kemp CAJ, McCune KA.
Determination of the degradation chemistry of the antitumor agent peme-
trexed disodium. J Pharm Sci. 2016;105:3256-3268.
Enrichment of Rosuvastatin Calcium With Oxidative Degradation
Product, 5-Keto-rosuvastatin (4)
4. Baertschi SW, Bruner H, Bunnell CA, et al. Isolation, identification, and synthesis
of two oxidative degradation products of olanzapine (LY170053) in solid oral
formulations. J Pharm Sci. 2008;97:883-892.
2 2
Solid rosuvastatin calcium (50 mg) and 35% aqueous H O so-
lution were taken in 2 separate scintillation vials. These 2 vials were
then placed in a glass jar. After capping the glass jar, it was placed in
5. Hovorka SW, Schoneich C. Oxidative degradation of pharmaceuticals: theory,
mechanism and inhibition. J Pharm Sci. 2001;90:253-269.
6. Li M. Organic Chemistry of Drug Degradation. London, UK: Published by the
Royal Society of Chemistry; 2012.
ꢀ
a 50 C oven for 1 week. HPLC analysis showed 24% conversion (by
peak area) to the desired oxidative degradation product 4.
7
. Corey EJ, Suggs W. Pyridinium chlorochromate. An efficient reagent for
oxidation of primary and secondary alcohols to carbonyl compounds. Tetra-
hedron Lett. 1975;16:2647-2650.
Enrichment of Ibuprofen With Oxidative Degradation Products, 2-
8. Corey EJ, Schmidt G. Useful procedures for the oxidation of alcohols involving
pyridinium dichromate in aprotic media. Tetrahedron Lett. 1979;20:2647-2650.
(
(
4-(1-Hydroxy-2-methylpropyl)-phenyl)propanoic Acid (5) and 2-
4-Isobutyrylphenyl)-propanoic Acid (6)
9. Doromy J-R, Castro B. Synthesis of N-t-butoxycarbonyl-4,4-dideuterio-L-pro-
To a solution of ibuprofen (0.9 mL, 500
m
M in acetonitrile-MeOH
line. Synthesis. 1986;1986:81-82.
1
0. Baldwin JE, Bamford SJ, Fryer AM, Rudolph MPW, Wood M. Towards a versatile
synthesis of Kainoids I: introduction of the C-3 and C-4 substituents. Tetrahe-
dron. 1997;53:5233-5254.
11. Sahoo PR, Patel S, Mishra BK. Oxidation kinetics of simvastatin using cetyl-
trimethylammonium dichromate. Int J Chem Kinet. 2013;45:236-242.
2. Houri AF, Xu Z, Cogan DA, Hoveyda AH. Cascade catalysis in synthesis. An
enantioselective route to Sch 38516 (and fluvirucin B
J Am Chem Soc. 1995;117:2943-2944.
1
:1 v/v) was added solution of AAPH, 2,2’-azobis(2-
a
aminopropionamide) dihydrochloride (0.1 mL, 10 mM in water).
The resulting mixture was incubated at 37 C for 5 h and then
analyzed by HPLC. Comparison with the retention time and UV-vis
spectra of commercial standards showed that the yield of the
degradation products 5 and 6 was 26% and 15%, respectively.
ꢀ
1
1
) aglycon macrolactum.

10
K.K. Nanda et al. / Journal of Pharmaceutical Sciences xxx (2019) 1-10
1
1
1
3. Jursic B. Surfactant assisted permanganate oxidation of aromatic compounds. 27. McMillen DF, Golden DM. Hydrocarbon bond dissociation energies. Annu Rev
Can J Chem. 1989;67:1381-1383.
Phys Chem. 1982;33:493-532.
4. Shaabani A, Teimouri F, Lee DG. Ion exchange catalysis in oxidation of organic
compounds with KMnO . Synth Commun. 2003;33:1057-1065.
4
5. Coperet C, Adolfsson H, Khuong T-AV, Yudin AK, Sharpless BK. A simple and
28. Udenfriend S, Clark CT, Axelrod J, Brodie BB. Ascorbic acid in aromatic hy-
droxylation: I. A model system for aromatic hydroxylation. J Biol Chem.
1954;208:731-740.
efficient method for the preparation of pyridine N-oxides. J Org Chem. 1998;63:
29. Zhu D(A), Zhang GGZ, George KLST, Zhou D. A novel accelerated oxidative
stability screening method for pharmaceutical solids. J Pharm Sci. 2011;100:
3529-3538.
30. Suksamram A, Ponglikitmongkol M, Wongkrajang K, et al. Diarylheptanoids,
new phytoestrogens from the rhizomes of curcuma comosa: isolation, chem-
ical modification and estrogenic activity evaluation. Bioorg Med Chem Lett.
2008;16:6891-6902.
31. Khurana JM, Dawra K, Majumdar S. An efficient 1,3-allylic carbonyl trans-
position of chalcones. Monatsh Chem. 2009;140:69-72.
32. Tong W, Chowdhury SK, Chen JC, Zhong R, Alton KB, Patrick JE. Fragmen-
tation of N-oxides (deoxygenation) in atmospheric pressure ionization:
investigation of the activation process. Rapid Commun Mass Spectrom.
2001;15:2085-2090.
1740-1741.
1
1
6. Ray PC, Mittapelli V, Rohatgi A, Tyagi OD. Efficient synthesis of N-oxide de-
rivatives: substituted 2-(2-(pyridyl-N-oxide)methylsulphinyl)benzimidazoles.
Synth Commun. 2007;37:2861-2868.
7. Jain SL, Sain B. Ruthenium catalyzed oxidation of tertiary nitrogen compounds
with molecular oxygen: an easy way to N-oxides under mild conditions. Chem
Commun. 2002;10:1040-1041.
8. Trost BM, Curran DP. Chemoselective oxidation of sulfides to sulfones. Tetra-
hedron Lett. 1981;22:1287-1290.
2 2
9. Hussain S, Talukdar D, Bharadwaj SK, Chaudhuri MK. VO F(dmpz) : a new
1
1
catalyst for selective oxidation of organic sulfides to sulfoxides with H
2 2
O .
Tetrahedron Lett. 2012;53:6512-6515.
2
2
2
0. Owens ML, Juenge EC, Poklis A. Convenient oxidation of phenothiazine
salts to their sulfoxides with aqueous nitrous acid. J Pharm Sci. 1989;78:
33. Ramanathan R, Su AD, Alvarez N, et al. Liquid chromatography/mass spec-
trometry methods for distinguishing N-oxides from hydroxylated compounds.
Anal Chem. 2000;72:1352-1359.
3
34-336.
1. Bosch E, Kochi JK. Catalytic oxidation of chlorpromazine and related pheno-
thiazines. Cation radicals as the reactive intermediates in sulfoxide formation.
J Chem Soc Perkin Trans. 1995;1:1057-1064.
2. Tennant RW, Ashby J. Classification according to chemical structure, mutage-
nicity to Salmonella and level of carcinogenicity of a further 39 chemicals
tested for carcinogenicity by the U.S. National Toxicology Program. Mutat Res.
3
34. Yong RS, Beard A, Sheng H, Zhang LK, Helmy R. Applications of TiCl as a
diagnostic reagent for the detection of nitro- and N-oxide-containing com-
pounds as potentially mutagenic impurities using ultrahigh-performance
liquid chromatography coupled with high-resolution mass spectrometry. Org
Process Res Dev. 2016;20:59-64.
35. Gutierrez CB, Yu C, Novitsky EJ, Huszagh AS, Rychnovsky SD, Huang L. Devel-
oping an acidic residue reactive and sulfoxide-containing MS-cleavable
homobifunctional cross-linker for probing protein-protein interactions. Anal
Chem. 2016;88:8315-8322.
1991;257:209-227.
23. March J. Advanced Organic Chemistry: Reactions, Mechanisms and Structure. 4th
ed. New York, NY: Willey Interscience; 1992.
2
4. Nanda KK, Blincoe WD, Allain LR, Wuelfing WP, Harmon PA. Iron (III)-mediated
oxidative degradation on the benzylic carbon of drug molecules in the absence
of initiating peroxides. J Pharm Sci. 2017;106:1347-1354.
36. Kaake RM, Wang X, Burke A, et al. A new in vivo cross-linking mass spec-
trometry platform to define protein-protein interactions in living cells. Mol Cell
Proteomics. 2014;13:3533-3543.
2
5. Betigeri S, Thakur A, Raghavan K. Use of 2,2’-azobis(2-amidinopropane) dihy-
drochloride as a reagent too for evaluation of oxidative stability of drugs.
Pharm Res. 2005;22:310-317.
37. Kao A, Chiu CL, Vellucci D, et al. Development of a novel cross-linking strategy
for fast and accurate identification of cross-linked peptides of protein com-
plexes. Mol Cell Proteomics. 2011;10:1-17.
2
6. Watkins MA, Pitzenberger S, Harmon PA. Direct evidence of 2-cyano-2-
propoxy radical activity during AIBN-based oxidative stress testing in
acetonitrile-water solvent systems. J Pharm Sci. 2013;102:1554-1568.
38. Yu C, Kandur W, Kao A, Rychnovsky S, Huang L. Developing new isotope-coded
mass spectrometry-cleavable cross-linkers for elucidating protein structures.
Anal Chem. 2014;86:2099-2106.