Evaluation of the Edman degradation product of vancomycin bonded to core-shell particles as a new HPLC chiral stationary phase

Article abstract of DOI:10.1002/chir.22985

A modified macrocyclic glycopeptide-based chiral stationary phase (CSP), prepared via Edman degradation of vancomycin, was evaluated as a chiral selector for the first time. Its applicability was compared with other macrocyclic glycopeptide-based CSPs: TeicoShell and VancoShell. In addition, another modified macrocyclic glycopeptide-based CSP, NicoShell, was further examined. Initial evaluation was focused on the complementary behavior with these glycopeptides. A screening procedure was used based on previous work for the enantiomeric separation of 50 chiral compounds including amino acids, pesticides, stimulants, and a variety of pharmaceuticals. Fast and efficient chiral separations resulted by using superficially porous (core-shell) particle supports. Overall, the vancomycin Edman degradation product (EDP) resembled TeicoShell with high enantioselectivity for acidic compounds in the polar ionic mode. The simultaneous enantiomeric separation of 5 racemic profens using liquid chromatography-mass spectrometry with EDP was performed in approximately 3?minutes. Other highlights include simultaneous liquid chromatography separations of rac-amphetamine and rac-methamphetamine with VancoShell, rac-pseudoephedrine and rac-ephedrine with NicoShell, and rac-dichlorprop and rac-haloxyfop with TeicoShell.

Full text of DOI:10.1002/chir.22985

Received: 3 May 2018
Revised: 15 May 2018
Accepted: 17 May 2018
DOI: 10.1002/chir.22985
R E G U L A R A R T I C L E
Evaluation of the Edman degradation product of
vancomycin bonded to core‐shell particles as a new HPLC
chiral stationary phase
Garrett Hellinghausen1
| Diego A. Lopez2 | Jauh T. Lee2 | Yadi Wang1 |
Choyce A. Weatherly1 | Abiud E. Portillo1 | Alain Berthod3
| Daniel W. Armstrong1,2
1 Department of Chemistry and
Abstract
Biochemistry, The University of Texas at
A modified macrocyclic glycopeptide‐based chiral stationary phase (CSP), pre-
pared via Edman degradation of vancomycin, was evaluated as a chiral selector
for the first time. Its applicability was compared with other macrocyclic glyco-
peptide‐based CSPs: TeicoShell and VancoShell. In addition, another modified
macrocyclic glycopeptide‐based CSP, NicoShell, was further examined. Initial
evaluation was focused on the complementary behavior with these glycopep-
tides. A screening procedure was used based on previous work for the enantio-
meric separation of 50 chiral compounds including amino acids, pesticides,
stimulants, and a variety of pharmaceuticals. Fast and efficient chiral separa-
tions resulted by using superficially porous (core‐shell) particle supports. Over-
all, the vancomycin Edman degradation product (EDP) resembled TeicoShell
with high enantioselectivity for acidic compounds in the polar ionic mode.
The simultaneous enantiomeric separation of 5 racemic profens using liquid
chromatography‐mass spectrometry with EDP was performed in approximately
3 minutes. Other highlights include simultaneous liquid chromatography sep-
arations of rac‐amphetamine and rac‐methamphetamine with VancoShell,
rac‐pseudoephedrine and rac‐ephedrine with NicoShell, and rac‐dichlorprop
and rac‐haloxyfop with TeicoShell.
Arlington, Arlington, Texas, USA
2 AZYP, LLC, Arlington, Texas, USA
3 Institute of Analytical Sciences CNRS,
University of Lyon 1, Villeurbanne,
France
Correspondence
Daniel W. Armstrong, The University of
Texas at Arlington, Arlington, TX 76019,
USA.
Email: sec4dwa@uta.edu
Funding information
French National Center for Scientific
Research, Grant/Award Number: ISA‐
CNRS‐UMR5280; Robert A. Welch Foun-
dation, Grant/Award Number: Y0026
KEYWORDS
complementary behavior, Edman degradation, electrospray ionization liquid chromatography‐mass
spectrometry, enantiomer separations, superficially porous particles, teicoplanin, vancomycin
1 | INTRODUCTION
CSPs are multimodal, meaning they are stable and effi-
cient in normal phase (NP), reversed phase (RP), polar
Macrocyclic glycopeptide antibiotics were first introduced
as chiral selectors for liquid chromatography by Arm-
strong in the early 1990s.1 These natural products are pro-
duced by bacterial fermentation. Purified and bonded to
silica particles, they make useful chiral stationary phases
(CSPs) with a broad spectrum of interactions and there-
fore applicability.2 The macrocyclic glycopeptide‐based
organic mode (POM), and polar ionic mode (PIM).3,4
The most distinctive feature of macrocyclic glycopeptides
as chiral selectors is their ionic character. All macrocyclic
glycopeptides are ionizable, bearing primary or secondary
amines rendering them positively charged at neutral and
acidic pH values.5-7 They also have a carboxylic acid bear-
ing a negative charge at neutral and high pH values so
Chirality. 2018;1–12.
wileyonlinelibrary.com/journal/chir
© 2018 Wiley Periodicals, Inc.
1

2
HELLINGHAUSEN ET AL.
that the net charge is adjustable according to the mobile
phase pH. This is the foundation of PIM, which utilizes
100% methanol containing trace amounts of acid and
base or a nonvolatile salt to tune the charges on the chiral
selector to effect ionizable enantiomers' retention and
separation.6 Many ionizable compounds can be separated
in PIM, but sometimes it is beneficial to adjust the hydro-
gen bonding interactions by switching to POM, which
contains a mixture of acetonitrile and methanol with acid
and base.3 Others favor RP, in which methanol is gener-
ally mixed with an ammonium salt with pH adjustment
to enhance ionic interactions. In RP, a low pH is gener-
ally preferred for amines, while higher pH is favored for
acids. All these modes are compatible with mass spec-
trometry, and usually NP is not necessary for enantio-
meric separations with macrocyclic glycopeptides, while
other CSPs depend on it. This is especially important to
biological analysis, which depends on mass spectrometry
sensitivity for thermally liable and complex samples.
Another feature of the macrocyclic glycopeptide class
of chiral selectors is their complementary behavior.5,8 If a
separation of an enantiomeric pair is observed on a mac-
rocyclic selector, say teicoplanin, chances are that a base-
line separation of this pair will be observed on a different
selector, say vancomycin. The large number of possible
interactions and structural similarities between the differ-
ent macrocyclic glycopeptides explain the observed com-
plementary behavior, which provides an ease of method
development. A plethora of native macrocyclic glycopep-
tides have been explored for their use as chiral selectors,
not only in liquid chromatography (LC) but also capillary
electrophoresis and super critical fluid chromatogra-
phy.1,4,5,7-19 Of these, vancomycin, teicoplanin, and
ristocetin A were commercialized as the CHIROBIOTICs
as well as teicoplanin's aglycone.20 Since the recent devel-
opment of superficially porous particles or core‐shell par-
ticles, which offer high throughput and more effective
separations, several studies have been explored using
core‐shell
macrocyclic
glycopeptide‐based
CSPs
(TeicoShell [TS] and VancoShell [VS]).21-25 This has been
particularly useful to ultrafast chiral separations needed
in second dimension (2D) LC.24-29 However, many glyco-
peptides are costly and have limited availability, which
has led to a need to understand the applicability and lim-
itations of more available glycopeptides so further explo-
ration can be made concerning useful modifications.
Comprehensive studies have indicated that vancomy-
cin is most useful for the separation of basic amines,
while teicoplanin is most useful for the separation of
acids, specifically amino acids.17,20,23 When exploring
the structural interactions driving these separations, it is
difficult to assess their separation mechanisms due to
the diverse and complex interactions of each macrocyclic
glycopeptide.6 However, it is thought that the carboxylic
acid located in the vancomycin structure might play an
important role for the interaction with amines, while
the primary amine in teicoplanin might be important to
chiral recognition for acids.6 Some studies have been
done with modified macrocyclic glycopeptides, such as
the crystalline degradation of vancomycin, which incor-
porates a second carboxylic acid moiety in the struc-
ture.30-32 Recently, a modified macrocyclic glycopeptide‐
based CSP, NicoShell (NS), was used for the novel LC
enantiomeric separation of nicotine from tobacco e‐liq-
uids and several nicotine‐related compounds, including
carcinogenic tobacco‐specific nitrosamines.21,22 NicoShell
was further utilized for the separation of several chiral
amines.23 An effective methodology was proposed for
core‐shell CSPs and was used in this study to evaluate a
new selector, the vancomycin Edman degradation prod-
uct (EDP).23 The EDP differs from native vancomycin
by the loss the N‐terminus leucine residue, leaving a pri-
mary amine33 (Figure 1). A set of 50 biologically active
chiral compounds including stimulants, nonsteroidal
FIGURE 1 Structures of vancomycin
and the vancomycin Edman degradation
product. The five‐aromatic ring
association in the peptidic aglycone
“basket” is labeled 1‐5. The red arrow
indicates the structural modification
leading to the vancomycin Edman
degradation product. See Section 2 for
information concerning preparation of
macrocyclic glycopeptide‐based chiral
stationary phases

HELLINGHAUSEN ET AL.
3
anti‐inflammatory drugs, pesticides, and a variety of
acidic and basic pharmaceuticals were subjected to LC
enantiomeric separation. Edman degradation product
results were then compared with three other macrocyclic
glycopeptide‐based core‐shell CSPs: TS, VS, and NS.
set as follows: nebulizer gas flow, 3 L/min; dryer gas flow,
15 L/min; desolvation line temperature, 250°C; heat
block temperature, 400°C. Multiple ultraviolet (UV)
wavelengths, 220, 230, and 254 nm, were utilized for
detection and identification of enantiomers. All separa-
tions were carried out at room temperature, unless other-
wise noted, using an isocratic method. Mobile phases
were degassed by ultrasonication under vacuum for
5 minutes. Each analyte was screened in PIM, POM,
RP, and NP. The screening mobile phase conditions refer-
ring to Table 1 were as follows: PIM: MeOH‐NH4Formate
(100:0.1, v/w), POM: ACN‐MeOH‐AA‐TEA (60:40:0.3:0.2,
v/v/v/v), RP: MeOH‐NH4Formate (pH 3.6; 16 mM)
(30:70, v/v), and NP: Hex‐EtOH‐TFA‐TEA (70:30:0.3:0.2,
v/v/v/v).
The dead time, t0, was determined by the peak of
the refractive index change due to the unretained sam-
ple solvent. Retention factors (k) were calculated using
k = (tR−t0)/(t0), where tR is the retention time of the
first peak and t0, the dead time of the column. Selectiv-
ity (α) was calculated using α = k2/k1, where k1 and k2
are retention factors of the first and second peaks,
respectively. Resolution (Rs) was calculated using the
peak width at half peak height, Rs = 2(tR2−tR1)/
(w1 + w2). Two EDP columns were produced and had
a relative standard deviation (%RSD) within 5.0% for
all Rs factors obtained.
2 | MATERIALS AND METHODS
Macrocyclic
glycopeptide‐based
core‐shell
CSPs
(100 × 4.6 mm [i.d.]): vancomycin (VS), teicoplanin
(TS), NS, and the vancomycin EDP were obtained from
AZYP, LLC. (Arlington, Texas). The EDP selector was
synthesized by reacting vancomycin with phenyl isothio-
cyanate in pyridine/water (50/50, v/v) followed by treat-
ment with trifluoroacetic acid to selectively remove the
N‐terminal residue highlighted in red33 in Figure 1. Thus,
the hexapeptide derivative with a free primary amine (red
arrow, Figure 1) was produced. The hexapeptide selector
was then bonded to 2.7‐μm core‐shell particles, like the
other CSPs, discussed in past literature, with a surface
area of 120 m2/g, a pore diameter of 12 nm, and a shell
thickness of 0.5 μm.1,25,29
Analytes were purchased as racemic standards or
individual enantiomer standards (then mixed to form
racemates) from Cerilliant Corporation (Round Rock,
Texas), Sigma‐Aldrich (St Louis, Missouri), and LKT Lab-
oratories Inc (Minneapolis, Minnesota). Racemic stan-
dards were prepared with methanol at 1 mg/mL for
analysis. In the set of 50 selected analytes, 48 were ioniz-
able compounds, mostly bases, since only 10 acidic com-
pounds did not contain a nitrogen atom. Twenty‐six
analytes were amines or have an amine group in their
structure. The remaining 14 nitrogen containing com-
pounds were mostly amides (nine analytes) and a pyr-
rolizidine, pyran, benzoxazole, and two pyridine‐
containing compounds.
Solvents and additives including HPLC grade acetoni-
trile (ACN), methanol (MeOH), ethanol (EtOH), hexane
(Hex), acetic acid (AA), trifluoroacetic acid (TFA),
trimethylamine (TEA), formic acid (FA), ammonium for-
mate (NH4HCO2), and ammonium trifluoroacetate
(NH4TFA) were obtained from Sigma‐Aldrich (St Louis,
Missouri). Water was purified by a Milli‐Q water purifica-
tion system (Millipore, Billerica, Massachusetts).
An Agilent 1260 (Agilent Technologies, Palo Alto,
California) HPLC was used. It consisted of a 1200 diode
array detector, autosampler, and quaternary pump. The
mass spectrometer used in this study was a Shimadzu tri-
ple quadrupole liquid chromatography‐mass spectrome-
try (LC‐MS) instrument, LCMS‐8040 (Shimadzu, Tokyo,
Japan). All MS was operated in positive ion mode with
an electron spray ionization source. The parameters were
3 | RESULTS AND DISCUSSION
3.1 | Screening results
Preliminary screening with all four CSPs using 50 chiral
compounds in each compatible chromatographic mode
(PIM, POM, RP, and NP) was performed, making 200
analyses per CSP. When a partial separation of the enan-
tiomers was obtained, this separation could be signifi-
cantly improved by modulating the mobile phase as
shown in previous studies, but it should be noted that this
was not the aim of this study.20,23 The best screening
result (in terms of Rs) by each CSP are tabulated accord-
ing to each compound in Table 1. In 46 cases (184 analy-
ses), the compounds could not be separated on a CSP by
all four mobile phases assayed. These are reported in
Table 1 with α = 1.00 and Rs = 0.0 in the Table 1. No
k1 was listed since four different values were obtained
in the four modes tested, all four producing a single peak
for the enantiomeric pair.
Overall, the screening procedure resulted in 40 race-
mic compounds (80%) baseline separated (Rs ≥ 1.5)
(Table 1). Several had Rs ≥ 1.5 with more than one CSP;
one compound (methylphenidate) separated on all four
CSPs, five compounds on three of the CSPs, 17

4
HELLINGHAUSEN ET AL.
TABLE 1 Chiral separation comparisons using core‐shell macrocyclic glycopeptide‐based CSPs
Namea
Structurea
CSPb
MPc
k1
αd
Rs
d
d
(a) Chemical amines
α‐Methylbenzylamine
VS
PIM
POM
PIM
…
0.9
5.8
0.4
…
1.07
1.17
1.05
1.00
0.8
2.4
0.3
0.0
NS
EDP
TS
α,4‐Dimethylbenzylamine
VS
POM
POM
POM
…
3.0
2.8
1.0
…
1.12
1.09
1.08
1.00
1.4
1.0
0.6
0.0
NS
EDP
TS
α‐Methyl‐4‐nitrobenzylamine
4‐Methoxymethylbenzylamine
N,N‐α‐Trimethylbenzylamine
VS
PIM
PIM
PIM
NP
1.5
4.3
0.6
7.0
1.07
1.06
1.04
1.02
0.9
1.2
0.3
0.4
NS
EDP
TS
VS
PIM
PIM
PIM
NP
1.0
2.2
0.4
4.6
1.45
1.08
1.40
1.03
5.1
1.6
2.4
0.5
NS
EDP
TS
VS
RP
PIM
…
0.3
2.7
…
1.23
1.11
1.00
1.03
1.3
1.6
0.0
0.6
NS
EDP
TS
NP
5.5
(b) Stimulants
Amphetamine
VS
PIM
…
NP
…
1.0
…
2.0
…
1.17
1.00
1.12
1.00
1.7
0.0
1.6
0.0
NS
EDP
TS
Methamphetamine
VS
PIM
PIM
NP
…
1.3
4.0
1.8
…
1.11
1.02
1.14
1.00
1.6
0.4
1.8
0.0
NS
EDP
TS
β‐Ketoamphetamine (cathionine)
(1 RS; 2 SR)‐ephedrine
(1 RS; 2 RS)‐pseudoephedrine
Norephedrine
VS
PIM
PIM
PIM
NP
0.9
2.3
0.4
5.4
1.18
1.80
1.12
1.11
1.6
8.3
0.6
1.1
NS
EDP
TS
VS
POM
POM
NP
2.5
6.2
1.9
4.8
1.02
1.13
1.01
1.03
0.4
2.0
0.2
0.6
NS
EDP
TS
POM
VS
POM
POM
NP
2.5
4.7
2.5
5.4
1.08
1.38
1.12
1.09
1.4
5.0
1.6
1.4
NS
EDP
TS
POM
VS
NP
PIM
NP
3.2
2.4
2.3
2
1.03
1.07
1.04
1.02
0.4
1.3
0.6
0.4
NS
EDP
TS
PIM
Epinephrine
VS
…
POM
…
POM
…
1.7
…
8.7
1.00
1.06
1.00
1.04
0.0
1.0
0.0
0.5
NS
EDP
TS
(Continues)

HELLINGHAUSEN ET AL.
5
TABLE 1 (Continued)
Namea
Structurea
CSPb
MPc
k1
αd
Rs
d
d
Citalopram
VS
PIM
PIM
NP
…
1.8
3.4
4.2
…
1.13
1.05
1.13
1.00
1.4
0.9
2.0
0.0
NS
EDP
TS
Fluoxetine
VS
PIM
POM
RP
1.2
3.3
2.0
…
1.26
1.05
1.32
1.00
2.5
1.1
1.8
0.0
NS
EDP
TS
…
Methylphenidate
Mianserin
VS
PIM
POM
PIM
PIM
0.9
2.6
0.4
2.5
1.48
1.10
1.36
1.12
3.3
1.6
1.7
1.7
NS
EDP
TS
VS
PIM
PIM
PIM
PIM
0.6
0.9
0.4
1.7
2.07
1.21
1.38
1.09
3.6
1.8
1.6
1.0
NS
EDP
TS
Lorazepam
Temazepam
VS
RP
…
PIM
PIM
11.1
…
0.3
0.4
1.03
1.00
1.10
3.60
0.5
0.0
0.6
6.3
NS
EDP
TS
VS
RP
RP
PIM
NP
7.4
6.7
0.3
2.8
1.12
1.04
1.13
1.12
1.0
0.5
0.6
1.0
NS
EDP
TS
(c) Pharmaceuticals
Carbinoxamine
VS
PIM
PIM
NP
…
1.3
2.3
5.0
…
1.08
1.06
1.14
1.00
0.8
1.0
2.1
0.0
NS
EDP
TS
Propranolol
VS
POM
POM
POM
POM
2.2
5.3
1.2
3.1
1.13
1.59
1.07
1.15
1.7
5.0
0.6
2.3
NS
EDP
TS
Phensuximide
VS
RP
RP
NP
RP
1.3
1.2
0.5
1.3
1.11
1.05
1.10
1.16
1.4
0.6
0.9
1.9
NS
EDP
TS
Proglumide
VS
RP
RP
PIM
RP
4.1
3.9
0.4
2.9
2.10
2.10
1.16
1.16
3.5
3.9
0.7
1.9
NS
EDP
TS
(Continues)

6
HELLINGHAUSEN ET AL.
TABLE 1 (Continued)
Namea
Structurea
CSPb
MPc
k1
αd
Rs
d
d
Hexobarbital
VS
RP
RP
RP
RP
2.0
1.6
2.5
1.3
1.18
1.11
1.14
1.14
1.8
1.4
1.2
1.4
NS
EDP
TS
Warfarin
VS
RP
RP
NP
RP
8.5
9.5
0.8
3.0
1.10
1.04
1.05
1.32
1.5
1.4
0.5
3.5
NS
EDP
TS
(d) Amino acids and derivatives
Phenylalanine
VS
NP
…
PIM
5.1
…
0.5
1.08
1.00
1.20
1.40
0.6
0.0
0.8
2.3
NS
EDP
TS
RP
0.7
FMOC phenylalanine
VS
RP
…
PIM
PIM
0.9
…
1.0
1.07
1.00
1.27
2.33
0.5
0.0
1.7
2.5
NS
EDP
TS
0.2
4‐Nitrophenylalanine
Kynurenine
VS
RP
RP
RP
RP
0.6
1.0
0.8
1.3
1.11
1.07
1.13
1.19
1.0
0.6
0.7
1.4
NS
EDP
TS
VS
RP
…
RP
RP
0.8
…
1.3
1.0
1.98
1.00
1.40
3.67
3.2
0.0
1.9
8.5
NS
EDP
TS
DOPA
VS
…
…
RP
RP
…
…
0.4
0.5
1.00
1.00
1.50
1.75
0.0
0.0
1.7
2.4
NS
EDP
TS
2‐Amino‐2‐phenylbutyric acid
VS
RP
0.3
1.6
0.5
0.7
1.14
1.12
1.57
2.12
0.7
0.6
2.7
3.6
NS
EDP
TS
POM
PIM
RP
(e) Nonsteroidal anti‐inflammatory drugs
Benoxaprofen
VS
…
…
PIM
…
…
…
0.9
…
1.00
1.00
1.35
1.00
0.0
0.0
2.0
0.0
NS
EDP
TS
Etodolac
VS
…
…
PIM
RP
…
…
0.7
0.8
1.00
1.00
1.17
1.08
0.0
0.0
1.1
0.9
NS
EDP
TS
Flurbiprofen
VS
…
…
PIM
RP
…
…
0.9
3.3
1.00
1.00
1.37
1.12
0.0
0.0
2.1
1.4
NS
EDP
TS
(Continues)

HELLINGHAUSEN ET AL.
7
TABLE 1 (Continued)
Namea
Structurea
CSPb
MPc
k1
αd
Rs
d
d
Ibuprofen
VS
…
…
PIM
RP
…
…
0.6
1.00
1.00
1.26
1.15
0.0
0.0
1.4
1.4
NS
EDP
TS
3.2
Ketoprofen
VS
…
…
PIM
RP
…
…
0.9
1.7
1.00
1.00
1.31
1.06
0.0
0.0
1.8
0.7
NS
EDP
TS
Ketorolac
VS
…
…
PIM
PIM
…
…
1.1
0.4
1.00
1.00
1.10
2.40
0.0
0.0
0.9
3.5
NS
EDP
TS
Loxoprofen
VS
…
…
PIM
RP
…
…
0.5
2.6
1.00
1.00
1.17
1.19
0.0
0.0
1.5
1.8
NS
EDP
TS
(f) Pesticides
2‐(4‐chlorophenoxy) propionic acid
VS
…
…
1.00
1.08
1.36
3.64
0.0
0.5
2.1
3.5
NS
EDP
TS
NP
PIM
PIM
0.3
0.9
0.1
2‐phenylpropionic acid
Bromacil
VS
…
…
PIM
RP
…
…
0.8
0.8
1.00
1.00
1.18
1.11
0.0
0.0
1.2
1.1
NS
EDP
TS
VS
RP
RP
RP
RP
2.9
2.7
1.7
2.2
1.14
1.04
1.05
1.18
1.6
0.6
0.6
2.1
NS
EDP
TS
Dichlorprop
VS
…
…
PIM
PIM
…
…
0.9
0.1
1.00
1.00
1.37
1.70
0.0
0.0
2.1
3.5
NS
EDP
TS
Haloxyfop
VS
RP
…
PIM
PIM
9.6
…
0.5
0.1
1.05
1.00
1.26
1.70
0.7
0.0
1.4
3.6
NS
EDP
TS
Mecoprop
VS
…
…
PIM
PIM
…
…
0.7
0.1
…
…
1.36
1.70
0.0
0.0
2.1
2.9
NS
EDP
TS
Mecoprop methyl ester
VS
RP
RP
…
6.6
6.8
…
1.10
1.17
…
1.4
2.0
0.0
0.0
NS
EDP
TS
…
…
…
(g) Nicotine and metabolites
Nicotine
VS
NP
PIM
NP
15.1
0.8
4.0
1.06
1.81
1.05
1.04
0.6
3.5
0.6
0.4
NS
EDP
TS
PIM
1.6
(Continues)

8
HELLINGHAUSEN ET AL.
TABLE 1 (Continued)
Namea
Structurea
CSPb
MPc
k1
αd
Rs
d
d
Nornicotine
VS
PIM
PIM
…
2.6
10.7
…
1.08
1.19
1.00
1.00
1.1
3.1
0.0
0.0
NS
EDP
TS
…
…
Cotinine
VS
PIM
…
NP
0.3
…
2.4
1.11
1.00
1.02
1.12
0.6
0.0
0.3
1.1
NS
EDP
TS
PIM
0.8
5‐(3‐pyridyl)tetrahydrofuran‐2‐one
Rac‐(*,*)‐4‐trans‐cotinine carboxylic acid
VS
NP
NP
NP
RP
12.2
7.1
3.0
1.13
1.14
1.01
1.10
1.4
1.5
0.2
1.6
NS
EDP
TS
3.4
VS
PIM
NP
PIM
RP
1.0
5.9
1.3
1.0
1.12
1.33
1.39
1.2
0.9
1.8
2.1
1.4
NS
EDP
TS
(h) Oxazolidinone
4‐Phenyl‐2‐oxazolidinone
VS
NP
…
NP
RP
2.3
…
1.1
3.2
1.26
1.00
1.12
1.79
2.5
0.0
1.4
4.2
NS
EDP
TS
Abbreviations: CSPs, chiral stationary phases; EDP, Edman degradation product; MP, mobile phase; NP, normal phase; NS, NicoShell; PIM, polar ionic mode;
POM, polar organic mode; RP, reversed phase; TS, TeicoShell; VS, VancoShell.
aSee Section 2 for sample information.
bCore‐shell chiral stationary phases were all 100 × 4.6 mm (i.d.): VS, NS, Vancomycin EDP, and TS. See Section 2 for more information.
cSee Section 2 for MP conditions: PIM, POM, RP, and NP.
dSee Section 2 for chromatographic calculations of retention factor of the first peak (k1), selectivity (α), and resolution (Rs).
compounds with two CSPs, and 17 compounds with
only one CSP (Table 1). Of the remaining 10 com-
pounds, all had a partial separation (Rs > 0.0) with at
least one CSP (Table 1). The data from Table 1 is illus-
trated in Figure 2A. Figure 2A depicts the number of
respective CSP: EDP, TS, VS, and NS, 54%, 30%, 38%,
and 35% (Table 1 and Figure 2B). Reversed phase was
the next most efficient mode, utilized for 31% of the
best separations, dominantly for VS and TS, 38% and
48%, respectively. Reversed phase was less useful for
NS and EDP, 26% and 13%, respectively (Table 1 and
Figure 2B). Normal phase and POM were less utilized
as the best mobile phases, only used for 17% and 12%
of all the best separations obtained by the four CSPs
(Table 1 and Figure 2B).
separations in terms of Rs
> 0.0 (bar 1, red),
0.0 > Rs > 1.5 (bar 2, blue), and Rs ≥ 1.5 (bar 3, green)
for each CSP. Each macrocyclic glycopeptide‐based
core‐shell CSP was able to separate (Rs > 0.0) at least
60% of the 50 chiral compounds (Figure 2A). Edman
degradation product had the highest efficacy of the four
CSPs, separating 46 of 50 (92%) of the set (Rs > 0.0)
with only four chiral compounds with Rs = 0.0; the
three basic amines: trimethylbenzylamine, epinephrine,
and nornicotine as well as the nonionizable methyl
ester of mecoprop (Figure 2A and Table 1). In
Figure 2B, the number of separations with Rs > 0.0
(bar 1, red) were distinguished by which chromato-
graphic mode was utilized. Polar ionic mode was the
most successful chromatographic mode, utilized overall
to perform 40% of the best separations and for each
3.2 | Complementary behavior and best
applications
As expected, VS and NS were highly effective for
separating basic amines, while TS was more effective
for separating acidic compounds, which highlights their
complementary behavior. Edman degradation product
was most like the teicoplanin chiral selector as it sepa-
rated most chiral acids, like the amino acids, herbicides,
and nonsteroidal anti‐inflammatory drugs (Table 1). To

HELLINGHAUSEN ET AL.
9
FIGURE 2 Percentage of racemic compounds separated by each macrocyclic glycopeptide‐based chiral stationary phase (CSP): VancoShell
(VS), Vancomycin Edman degradation product (EDP), TeicoShell (TS), and NicoShell (NS). A, The highest resolution (Rs) for all 50 compounds
obtained during screening (from Table 1) by each CSP is indicated by each bar. Bar 1 (red) represents the percentage of racemic compounds
with Rs > 0.0, while bar 2 (blue) indicates the percentage of racemic compounds with 0.0 < Rs < 1.5, and bar 3 (green) shows the percentage of
baseline separations obtained (Rs ≥ 1.5). B, Bar 1 (Rs > 0.0 during screening from Table 1) from Figure 2A for each CSP is distinguished into the
chromatographic modes utilized. From the bottom to the top of each bar, it is divided into polar ionic mode (diagonal lines), polar organic
mode (crisscross), reversed phase (horizontal lines), and normal phase (grid). See Section 2 for chromatographic parameters and information
illustrate this, Figure 3 compares the Rs obtained from a
selection of 10 chiral compounds between each CSP.
Edman degradation product was able to differentiate
all 10 compounds, exhibiting the broadest spectrum of
the four CSPs. However, the most effective selector was
not EDP for each of the 10 compounds. TeicoShell was
more selective than EDP and had the highest Rs for 19
of the 50 compounds (Table 1). As shown in Figure 3,
the Rs was 3 to 10 times higher for TS compared with
the other CSPs for the neutral lorazepam and the two
acids, kyurenine and ketorolac. Clearly, the acidic enan-
tiomers are most easily recognized by TS. Similarly, when
examining the results of VS, it was clearly the most appli-
cable CSP for basic enantiomers. However, certain com-
pounds were more selective to NS and EDP. For
example, EDP was the only CSP to separate the nonste-
roidal anti‐inflammatory benoxaprofen (Table 1). Also,
NS was by far the best CSP for cathinone, pseudoephed-
rine, propranolol, and nicotine with Rs factors 3 to 10
times higher than the other CSPs (Figure 3). Overall,
the screening procedure showed that 19, 13, 10, and 8
compounds were best separated by TS, NS, EDP, and
VS, respectively (Table 1).
Specific applications of these CSPs based on their
complementary behavior are shown in Figure 4. The
simultaneous separations of rac‐amphetamine and rac‐
methamphetamine in 5 minutes, rac‐pseudoephedrine
and rac‐ephedrine in 10 minutes, and rac‐dichlorprop
and rac‐haloxyfop in 5 minutes are shown using VS, NS,
and TS, respectively (Figure 4A,B,C). Edman degradation
product was the most effective CSP for the separation of
FIGURE
3
Comparison of resolution (Rs) obtained from
VancoShell (VS, purple and vertical lines), the vancomycin Edman
degradation product (EDP, green and crisscross), TeicoShell (TS,
light blue and horizontal lines), and NicoShell (NS, orange and
diagonal lines) to emphasize the broad‐spectrum recognition of VS
and EDP compared with the high chiral selectivity obtained from
TS and NS for 10 selected chiral compounds (full data in Table 1).
See Section 2 for chromatographic parameters and information

10
HELLINGHAUSEN ET AL.
FIGURE 4 Highlighted applications of core‐shell macrocyclic glycopeptide‐based chiral stationary phases. A, Separation of rac‐
Amphetamine and rac‐Methamphetamine using VancoShell (100 × 4.6 mm [i.d.]) with MeOH‐AA‐TEA (100:0.2:0.1, v/v/v) at 0.5 mL/min,
25°C, UV 254 nm. (1): (+)‐Amphetamine, (2): (−)‐Amphetamine, (3): (+)‐Methamphetamine, (4): (−)‐Methamphetamine. B, Separation of
rac‐Pseudoephedrine and rac‐Ephedrine using NicoShell (100 × 4.6 mm [i.d.]) with MeOH‐AA‐NH4OH (100:0.2:0.05, v/v/v) at 0.5 mL/min,
25°C, UV 220 nm. (1): (+)‐Pseudoephedrine, (2): (−)‐Pseudoephedrine, (3): (+)‐Ephedrine, (4): (−)‐Ephedrine. C, Separation of rac‐
Dichlorprop and rac‐Haloxyfop using TeicoShell (100 × 4.6 mm [i.d.]) with MeOH‐NH4Formate (pH 3.6; 16 mM) (30:70, v/v) at 0.8 mL/min,
45°C, UV 230 nm. D, Simultaneous LC‐MS separation of rac‐Ibuprofen (red), rac‐Flurbiprofen (green), rac‐Ketoprofen (blue), rac‐
Benoxaprofen (brown), and rac‐Indoprofen (pink) using the vancomycin Edman degradation product (100 × 4.6 mm [i.d.]) with MeOH‐
NH4Formate (100:0.1, v/w) at 1.0 mL/min, 25°C, UV 230 nm. The total ion chromatogram (TIC) is shown in black, and each profen is shown
according to their m/z in the extracted ion chromatograms (EICs). See Section 2 for more information
racemic profens (nonsteroidal anti‐inflammatory drugs)
with fast analysis times using PIM, which is shown in
Figure 4D. The simultaneous LC‐MS separation of five
profens was performed within approximately 3 minutes,
which is shown in the total ion chromatogram, then each
profen's m/z extracted in the subsequent extracted ion
chromatograms (Figure 4D). This instrument was not
optimized for its extra column band broadening,
explaining the lower efficiency observed compared with
the screening results, such as for rac‐ibuprofen.
behavior with their native analogs, indicating the value
and need for investigation of new macrocyclic glycopep-
tide chiral selectors.
ACKNOWLEDGMENTS
We thank AZYP, LLC, for their technical support for
HPLC chiral column technology. This work was sup-
ported by the Robert A. Welch Foundation (Y0026) and
the French National Center for Scientific Research (ISA‐
CNRS‐UMR5280).
4 | CONCLUSIONS
ORCID
The screening procedure used to evaluate the new van-
comycin EDP as a chiral selector demonstrated its abil-
ity to discriminate the enantiomers of 46 chiral
compounds of a set of 50. The other recent modified
macrocyclic glycopeptide‐based core‐shell CSP, NS, was
shown to separate some nonionizable and acidic com-
pounds but was most useful for amines, like beta
blockers and stimulants. These modified macrocyclic
glycopeptides provide examples of complementary
Garrett Hellinghausen http://orcid.org/0000-0002-5427-327X
Alain Berthod http://orcid.org/0000-0002-7452-9527
Daniel W. Armstrong http://orcid.org/0000-0003-0501-6231
REFERENCES
1. Armstrong DW, Tang Y, Chen S, Zhou Y, Bagwill C, Chen JR.
Macrocyclic antibiotics as a new class of chiral selectors for liq-
uid chromatography. Anal Chem. 1994;66(9):1473‐1484.

HELLINGHAUSEN ET AL.
11
2. Boehm RE, Martire DE, Armstrong DW. Theoretical
considerations concerning the separation of enantiomeric
solutes by liquid chromatography. Anal Chem. 1988;60(6):
522‐528.
16. Zhang X, Bao Y, Huang K, Barnett‐Rundlett K, Armstrong DW.
Evaluation of dalbavancin as chiral selector for HPLC and com-
parison with teicoplanin‐based chiral stationary phases.
Chirality. 2010;22(5):495‐513.
3. Beesley TE. Description and evaluation of chiral interactive sites
on bonded chiral stationary phases for liquid chromatography.
In: Berthod A, ed. Chiral recognition in separation methods, Hei-
delberg. Germany: Springer; 2010:53‐76.
17. Armstrong DW, Liu Y, Ekborg‐Ott KH. A covalently bonded
teicoplanin chiral stationary phase for HPLC enantiose-
parations. Chirality. 1995;7(6):474‐497.
18. Dolzan MD, Shu Y, Smuts JP, et al. Enantiomeric separation of
citalopram analogues by HPLC using macrocyclic glycopeptide
and cyclodextrin based chiral stationary phases. J Liq Chromatog
Rel Technol. 2016;39(3):154‐160.
4. Xiao TL, Rozhtov RV, Larock RC, Armstrong DW. Separation of
the enantiomers of substituted dihydrofurocoumarins by HPLC
using macrocyclic glycopeptides CSPs. Anal Bioanal Chem.
2003;377(4):639‐654.
19. Liu Y, Berthod A, Mitchell CR, Xiao TL, Zhang B, Armstrong
DW. Super/subcritical fluid chromatography chiral separations
with macrocyclic glycopeptide stationary phases. J Chromatogr
A. 2002;978(1‐2):185‐204.
5. Gasper MP, Berthod A, Nair UB, Armstrong DW. Comparison
and modeling study of vancomycin, ristocetin A, and
teicoplanin for CE enantioseparations. Anal Chem.
1996;68(15):2501‐2514.
20. Beesley TE, Lee JT. Method development strategy and
applications update for CHIROBIOTIC chiral stationary
6. Berthod A, Qiu HX, Staroverov SM, Kuznestov MA, Armstrong
DW. Chiral recognition with macrocyclic glycopeptides:
mechanisms and applications. In: Berthod A, ed. Chiral
phases.
J
Liq Chromatog Rel Technol. 2009;32(11‐12):
1733‐1767.
recognition in separation methods, Heidelberg.
Springer; 2010:203‐222.
Germany:
21. Hellinghausen G, Lee JT, Weatherly CA, Lopez DA, Armstrong
DW. Evaluation of nicotine in tobacco‐free‐nicotine commercial
products. Drug Test Anal. 2017;9(6):944‐948.
7. Ekborg‐Ott KH, Kullman JP, Wang X, Gahm K, He L, Arm-
strong DW. Evaluation of the macrocyclic antibiotic
22. Hellinghausen G, Roy D, Wang Y, et al. A comprehensive meth-
odology for the chiral separation of 40 tobacco alkaloids and
their carcinogenic E/Z‐(R,S)‐tobacco specific nitrosamine
metabolites. Talanta. 2018;181:132‐141.
avoparcin as
a new chiral selector for HPLC. Chirality.
1998;10(7):627‐660.
8. Chen S, Liu Y, Armstrong DW, Borell JI, Martinez‐Terpel B,
Matallama JL. Enantioresolution of substituted 2‐methoxy‐6‐
oxo‐tetrahydropyridine‐3‐carbonitriles on macrocyclic antibiotic
and cyclodextrin stationary phases. J Liq Chromatog Rel Technol.
1995;18(8):1495‐1507.
23. Hellinghausen G, Roy D, Lee JT, et al. Effective methodolo-
gies for enantiomeric separations of 150 pharmacology and
toxicology related 1°, 2°, and 3° amines with core‐shell
chiral stationary phases.
J Pharm Biomed Anal. 2018;
9. Peter A, Vѐkes E, Armstrong DW. Effects of temperature on
retention of chiral compounds on a ristocetin A chiral stationary
phase. J Chrom a. 2002;958(1‐2):89‐107.
155:70‐81.
24. Barhate CL, Breitbach ZS, Costa Pinto E, Regalado EL, Welch
CJ, Armstrong DW. Ultrafast separation of fluorinated and
desfluorinated pharmaceuticals using chiral selectors bonded
to superficially porous particles. J Chromatogr A. 2015;1426:
241‐247.
10. Karlsson C, Karlsson L, Armstrong DW, Owens PK. Evaluation
of
a
vancomycin chiral stationary phase in capillary
electrochromatography using polar organic and reversed‐phase
modes. Anal Chem. 2000;72(18):4394‐4401.
25. Barhate CL, Wahab MF, Breitbach ZS, Bell DS, Armstrong DW.
High efficiency, narrow particle size distribution, sub‐2 μm
based macrocyclic glycopeptide chiral stationary phases in
HPLC and SFC. Anal Chim Acta. 2015;898:128‐137.
11. Rundlett KL, Gasper MP, Zhou EY, Armstrong DW. Capillary
electrophoretic enantiomeric separations using the glycopeptide
antibiotic, teicoplanin. Chirality. 1996;8(1):88‐107.
12. Maier V, Ranc V, Švidrnoch M, et al. Study on the use of
boromycin as a chiral selector in capillary electrophoresis.
J Chromatogr A. 2012;1237:128‐132.
26. Patel DC, Breitbach ZS, Wahab MF, Barhate CL, Armstrong
DW. Gone in seconds: praxis, performance, and peculiarities of
ultrafast chiral liquid chromatography with superficially porous
particles. Anal Chem. 2015;87(18):9137‐9148.
13. Berthod A, Yu T, Kullman JP, Armstrong DW. Evaluation of the
macrocyclic glycopeptide A‐40,926 as a high‐performance liquid
chromatographic chiral selector and comparison with
27. Patel DC, Wahab MF, Armstrong DW, Breitbach ZS. Advances
in high‐throughput and high‐efficiency chiral liquid chromato-
graphic separations. J Chromatogr A. 2016;1467:2‐18.
teicoplanin chiral stationary phase.
J
Chromatogr A.
2000;897(1‐2):113‐129.
28. Barhate CL, Joyce LA, Makarov AA, et al. Ultrafast chiral sepa-
rations for high throughput enantiopurity analysis. Chem
Commun. 2016;53:509‐512.
14. Berthod A, Nair UB, Bagwill C, Armstrong DW. Derivatized
vancomycin stationary phases for LC chiral separations.
Talanta. 1996;43(10):1767‐1782.
29. Spudeit DA, Dolzan MD, Breitbach ZS, Barber WE, Micke GA,
Armstrong DW. Superficially porous particles vs. fully porous
particles for bonded high performance liquid chromatographic
chiral stationary phases: isopropyl cyclofructan 6. J Chromatogr
A. 2014;1363:89‐95.
15. Xiao TL, Tesarova E, Anderson JL, Egger M, Armstrong DW.
Evaluation and comparison of a methylated teicoplanin agly-
cone to teicoplanin aglycone and natural teicoplanin chiral
stationary phases. J Sep Sci. 2006;29(3):429‐445.

12
HELLINGHAUSEN ET AL.
30. Ghassempour A, Aboul‐Enein HY. Vancomycin degradation
products as potential chiral selectors in enantiomeric separation
of racemic compounds. J Chromatogr A. 2008;1191(1‐2):182‐187.
33. Booth PM, Stone DJM, Williams DH. The Edman degradation of
vancomycin: preparation of vancomycin hexapeptide. J Chem
Soc Chem Commun. 1987;709:1694‐1695.
31. Ghassempour A, Abdollahpour A, Tabar‐Heydar K, Nabid MR,
Mansouri S, Aboul‐Enein HY. Crystalline degradation products
of vancomycin as a new chiral stationary for liquid chromatog-
raphy. Chromatographia. 2005;61(3‐4):151‐155.
How to cite this article: Hellinghausen G, Lopez
DA, Lee JT, et al. Evaluation of the Edman
degradation product of vancomycin bonded to core‐
shell particles as a new HPLC chiral stationary
phase. Chirality. 2018;1–12. https://doi.org/
10.1002/chir.22985
32. Mojtahedi MM, Chalavi S, Ghassempour A, et al. Chiral separa-
tion of three agrochemical toxins enantiomers by high‐
performance liquid chromatography on a vancomycin crystal-
line degradation products‐chiral stationary phase. Biomed
Chromatogr. 2007;21(3):234‐240.