RAPID COMMUNICATIONS IN MASS SPECTROMETRY

Rapid Commun. Mass Spectrom. 2009; 23: 899–906

Published online in Wiley InterScience (www.interscience.wiley.com) DOI: 10.1002/rcm.3935

Separation and characterization of clindamycin

phosphate and related impurities in injection by liquid
chromatography/electrospray ionization mass
spectrometry
Shao-Min Wang1, Shan-Shan Bu1, Hong-Min Liu2*, Hua-Yu Li1, Wei Liu2
and Yu-Dong Wang1
1
Department of Chemistry, Zhengzhou University, 100 Science Road, 450001 Zhengzhou, China
2
School of Pharmaceutical Sciences, Zhengzhou University, 100 Science Road, 450001 Zhengzhou, China
Received 17 November 2008; Revised 26 December 2008; Accepted 27 December 2008

A simple high-performance liquid chromatography/electrospray ionization tandem mass spectro-

metric (HPLC/ESI-MS/MS) method has been developed for the rapid identification of clindamycin
phosphate and its degradation products or related impurities in clindamycin phosphate injection.
Detection was performed by quadrupole time-of-flight mass spectrometry (Q-TOFMS) via an ESI
source in positive mode. Clindamycin phosphate and its related substances lincomycin, 7-epilinco-
mycin-2-phosphate, lincomycin-2-phosphate, clindamycin B, clindamycin B-2-phosphate, and clin-
damycin were identified simultaneously by HPLC/ESI-MS/MS results. Based on the MS/MS spectra
of their quasi-molecular ions, the fragmentation pathways of clindamycin phosphate and its related
substances were compared and proposed, which are specific and useful for the identification of the
lincosamide antibiotics and related impurities. The method was rapid, sensitive and specific and can
be used to identify clindamycin phosphate and its related impurities in clindamycin phosphate
injection without control compounds. Copyright # 2009 John Wiley & Sons, Ltd.

Clindamycin, a semi-synthetic derivative of lincomycin, is an
antibiotic that is highly effective against Gram-positive and
Gram-negative anaerobic pathogens, as well as Gram-
positive aerobes. Its phosphate ester, clindamycin-2-phos-
phate (Fig. 1), is produced by chemical modification of
clindamycin. Although the ester is not biologically active, it is
rapidly hydrolyzed to the active clindamycin in vivo.
Clindamycin phosphate injection, USP, is a sterile solution
for intramuscular or intravenous use, and is effective in the
treatment of serious infections caused by susceptible
anaerobic bacteria and by susceptible strains of streptococci,
pneumococci and staphylococci.1–3
Clindamycin phosphate injection is a heat-labile product
and unstable during a conventional high-temperature
sterilization.4–6 Common degradation products or related
impurities of clindamycin phosphate injection are free
clindamycin, lincomycin-2-phosphate, and a small amount
of clindamycin B-2-phosphate.3,7–10 The rapid identification
and characterization of minor degradation products and
related impurities is an important task.
Several gas chromatography (GC), GC/mass spectrometry
(MS), or high-performance liquid chromatography (HPLC)
*Correspondence to: H.-M. Liu, School of Pharmaceutical Sciences,
Zhengzhou University, 100 Science Road, 450001 Zhengzhou,
China.
E-mail: liuhm@zzu.edu.cn
Contract/grant sponsor: National Natural Science Foundation;
contract/grant number: 2027205.
procedures have been developed for the determination and
analysis of clindamycin phosphate and its degradation
products or related impurities in bulk drug and formu-
lations.3,7–13 However, these methods are either time-
consuming or not very sensitive. To our knowledge, HPLC
coupled with MS detection for the purpose has not been
described. Quadrupole time-of-flight mass spectrometry (Q-
TOFMS) is superior in terms of scan speed and ease of
operation, enabling fast LC/MS analysis of complex
samples. This TOF technology not only allows for accurate
mass measurement, but also provides a tool for elemental
composition suggestions of a compound of interest. Besides,
results from collision-induced dissociation (CID) give further
information about the compounds, which is valuable for
identifying minor impurities in clindamycin phosphate
injection by their fragmentation patterns.14,15
The aim of this study was to develop an analytical method
suitable for the determination of clindamycin phosphate and
its degradation products or related impurities in clindamycin
phosphate injection. In this paper, we describe an HPLC/
ESI-MS/MS method which is the first simultaneous deter-
mination of clindamycin-2-phosphate together with linco-
mycin, lincomycin-2-phosphate, 7-epilincomycin-2-phos-
phate, clindamycin B, clindamycin B-2-phosphate, and
clindamycin in clindamycin phosphate injection (Fig. 1).
Compared with previous methods, it has three principal
advantages. (1) The MS detection used allows for the
identification and characterization of minor impurities. Most

Copyright # 2009 John Wiley & Sons, Ltd.

900 S.-M. Wang et al.
Figure 1. Structures of clindamycin phosphate and its related impurities.
of the previously reported methods are not suitable for this
purpose. (2) The use of an MS-compatible mobile phase (0.1%
formic acid in water) avoids the use of an ion-pairing reagent
in the mobile phase to effect the separation such as sodium
pentane sulfonate, dioctyl sodium sulphosuccinate, as such
methods are often associated with inherent instability and
long equilibration times. (3) This method is rapid, sensitive
and specific and can be used to identify clindamycin
phosphate and its related substances without control
compounds.
EXPERIMENTAL
Chemicals and materials
The clindamycin phosphate injection was obtained from
North China Pharmaceutical Co., Ltd. (Shijiazhuang, China).
Standards of clindamycin hydrochloride, clindamycin phos-
phate and lincomycin hydrochloride were kindly provided
by Henan Topfund Pharmaceutical Co., Ltd. (Zhumadian,
China). Formic acid was from Sigma-Aldrich Co. Ltd. (Poole,
UK). Methanol (HPLC-grade) was purchased from J.T.
Baker, USA. All of the aqueous solutions were prepared
with water purified using a Milli-Q water-purification
system (Millipore, Bedford, MA, USA).
Injection sample
A heat-stressed clindamycin phosphate solution was pre-
pared by heating the clindamycin phosphate injection
(150 mg/mL) at 1008C for 30 min. Then, the prepared heat-
stressed injection was diluted in water to give a concentration
of 6.25 mg/mL for the working solution.
HPLC/ESI-MS analysis
The separation of the clindamycin phosphate and its related
impurities by reversed-phase (RP)-HPLC and the ionization
of these compounds were optimized. A Waters HPLC system
(Milford, MA, USA), consisting of a 2695 separation module
and a 2996 photodiode array (PDA) detector, was controlled
by Micromass MassLynx V4.0 SP4 software. The HPLC
column was a Waters XTerra1 MS C18 RP column (3.5 mm,
Copyright # 2009 John Wiley & Sons, Ltd.
100
2.1 mm i.d.). The mobile phase comprised solvent A
(0.1% formic acid in water) and solvent B (methanol), at a
flow rate of 0.17 mL/min, under a gradient elution
programmer from 60% solvent A, 40% solvent B linearly
diluted to 46% solvent A, 54% solvent B by 9 min, and then
the column was re-equilibrated in 60% solvent A, 40%
solvent B for 6 min. Column temperature was 408C and the
injection volume was 20 mL. The eluent was detected by a
Micromass Q-TOF MicroTM mass spectrometer with an ESI
source (Waters, Manchester, UK). Nitrogen was used as the
nebulizer and dissolvation gas. Argon was used as the
collision gas. The ESI source was operated in positive ion
mode with optimized conditions. The mass spectrometer
recorded full-scan mass spectra over the mass range 200–
1000 with a capillary voltage of 3.5 kV, a cone voltage of 35 V,
a source temperature of 808C, a dissolvation temperature of
1808C, a collision voltage of 7 V, a cone gas flow of 50 L/h and
a dissolvation gas flow of 380 L/h. Clindamycin phosphate
and its related substances were identified based on the
presence of their protonated molecule.
HPLC/ESI-MS/MS
The HPLC and ESI conditions were as described above.
During MS/MS experiments, the protonated molecule was
chosen as precursor ion and isolated in the quadruple. Next,
a collision voltage of 19 V was used to fragment the precursor
ions to produce product ion spectra.
RESULTS AND DISCUSSION
ESI-MS/MS analyses of the reference
substances
Standards of lincomycin hydrochloride (1), clindamycin
hydrochloride (6) and clindamycin phosphate (7) (Fig. 1)
were first studied. Some characteristic fragment ions, such as
the losses of neutral molecules H2O, HPO3, HCl, metha-
nethiol, 2-methylthioethenol, and the residue of 3-propyl-N-
methylpyrrolidine (Aþ), are summarized in Table 1 and
Fig. 2(c), which are specific and useful for the identification of
clindamycin phosphate and its related compounds.
Rapid Commun. Mass Spectrom. 2009; 23: 899–906
DOI: 10.1002/rcm
 Analysis of clindamycin phosphate injection by LC/MS 901

Table 1. ESI-MS/MS product ions obtained from the [MHþ] an unknown origin, an ion at m/z 335.2 corresponding to the
ions of standards 1, 6, 7 (m/z with relative abundance (%) in loss of HPO3 and CHOHCHSCH3, and an ion at m/z 126.1
parentheses) with relative abundance of 100% corresponding to the
Product ion 1 6 7
3-propyl-N-methylpyrrolidine ion after loss of the rest of the
molecule. The elemental compositions of fragment ions at
[MHþ] 407.2 (19%) 425.2 (37%) 505.2 (54%) high resolution are shown in Table 2, which closely
[MHþ–H2O] 389.2 (3%) 407.2 (1%) 487.1 (8%) corresponded to the above elucidation. The accurate mass
[MHþ–HCl] — 389.2 (1%) 469.2 (1%)
[MHþ–HSCH3] 359.2 (9%) 377.2 (9%) 457.2 (9%)
data for some of the product ions was markedly less accurate
[MHþ–CHOHCHSCH3] 317.2 (1%) 335.2 (2%) — than the precursor ion, most probably due to the low levels of
[MHþ–HPO3] — — 425.2 (2%) the product ions. Despite this, the listed calculated elemental
[MHþ–HPO3–H2O] — — 407.2 (2%) compositions were very sensible, indicating the limitations of
[MHþ–HPO3–H2O–HSCH3] — — 359.2 (1%) expressing mass accuracy in parts-per-million (ppm) at low
[MHþ–HPO3–CHOHCHSCH3] — — 335.2 (2%)
[Aþ] 126.1 (100%)126.1 (100%)126.1 (100%)
m/z ratios.
Because the fragment at m/z 487.1 corresponding to the loss
of one H2O from clindamycin phosphate can be considered
to be eliminated either from the sugar moiety or from the
phosphate group, it is necessary to establish which one is
ESI-MS of clindamycin phosphate (7) (Fig. 2(a)) showed a favored. Therefore, a control experiment using the CID
characteristic protonated isotope pattern at m/z 505.1538 and method was performed. Under the same MS/MS conditions,
507.1611 with about 3:1 ratio of relative abundance the fragmentation spectrum of the ion at m/z 99, the
corresponding to the presence of one chlorine atom in the protonated phosphoric acid, clearly indicated the presence
molecule. After scanning the fragment at m/z 505 with CID, of a prominent fragment ion at m/z 81 (corresponding to the
the product ion spectrum (Fig. 2(b)) showed an ion at loss of water) with relative abundance of 64%. However,
m/z 487.1 corresponding to the loss of H2O from an unknown compared with the MS/MS spectrum from the ion at m/z 99,
elimination location, an ion at m/z 469.2 corresponding to the MS/MS of the ion at m/z 425, clindamycin hydrochloride (6),
loss of HCl, an ion at m/z 457.2 corresponding to the loss of resulted in the dehydrated ion at m/z 407 with only 1.4%
HSCH3, an ion at m/z 425.2 corresponding to the loss of relative abundance. Thus, elimination of H2O from the
HPO3, an ion at m/z 407.2 corresponding to the loss of HPO3, phosphate group is favored over elimination from the sugar
and one H2O from an unknown origin, an ion at m/z 359.2 moiety. Therefore, the plausible fragmentation pathway of
corresponding to the loss of HPO3, HSCH3 and one H2O from clindamycin phosphate may be proposed as in Fig. 3.

Figure 2. ESI-MS/MS of clindamycin phosphate (7). (a) HRMS; (b) MS/MS of the ion at m/z 505; and
(c) observed major fragmentation of clindamycin phosphate and its related impurities.

Copyright # 2009 John Wiley & Sons, Ltd. Rapid Commun. Mass Spectrom. 2009; 23: 899–906
DOI: 10.1002/rcm
902 S.-M. Wang et al.

Table 2. Observed and calculated masses of proposed neutral loss of the ion at m/z 505.1540 of clindamycin phosphate in ESI-Q-
TOF MS/MS
Product ion m/z Proposed formula Proposed neutral loss Theoretical m/z Error (ppm)

505.1540 C18H35N2O8PSCl — 505.1540 0

487.1427 C18H33N2O7PSCl H2O 487.1435 1.6
469.1721 C18H34N2O8PS HCl 469.1774 11.3
457.1493 C17H31N2O8PCl HSCH3 457.1507 3.1
425.1869 C18H34N2O5SCl HPO3 425.1877 1.9
407.1769 C18H32N2O4SCl HPO3þH2O 407.1771 0.6
359.1701 C17H28N2O4Cl HPO3þH2OþHSCH3 359.1738 10.3
335.1774 C15H28N2O4Cl CHOHCHSCH3þHPO3 335.1738 10.7
126.1278 C8H16N — 126.1283 3.9

Moreover, compared with the precursor ion at m/z 407 of
lincomycin hydrochloride (1), the relative abundance of the
dehydrated ion at m/z 389.2 was 3%, which was higher than
that of the dehydrated ion of clindamycin hydrochloride (6)
(1%). It is implied that the ion at m/z 389.2 was formed mostly
from the elimination of H2O at position 7, not at the sugar
moiety of lincomycin hydrochloride.
Analyses of clindamycin phosphate injection
by HPLC/ESI-MS/MS
Figure 4(a) showed the HPLC/ESI-MS scan of the heat-
stressed clindamycin phosphate injection. The peaks num-
bered represent those that were suspected of arising from
clindamycin phosphate and its related substances on the
basis of retention times, accurate mass data and the
fragmentation patterns arising from their MS/MS compared
with the response of authentic standards to these exper-
iments. For some of the degradation products no commercial
standards were available, so only the MS analysis could be
used to determine the identification of these substances.
Additionally, positive ion mode extracted ion current (EIC)
analyses were performed in order to confirm these assign-
ments. The substances for which commercial standards were
available for comparison will be discussed first.

Figure 3. Proposed fragmentation pathway of clindamycin phosphate.

Copyright # 2009 John Wiley & Sons, Ltd. Rapid Commun. Mass Spectrom. 2009; 23: 899–906
DOI: 10.1002/rcm
 Analysis of clindamycin phosphate injection by LC/MS 903

Figure 4. HPLC/MS analysis of heat-stressed clindamycin phosphate injection:

(a) positive ion mode ESI-MS TIC monitoring and (b) the corresponding EIC chro-
matograms with SIM at m/z 407, 487, 411, 491, 425 and 505. Peak identification: 1.
lincomycin, 2. 7-epilincomycin-2-phosphate, 3. lincomycin-2-phosphate, 4. clindamy-
cin B, 5. clindamycin B-2-phosphate, 6. clindamycin, 7. clindamycin-2-phosphate.

As shown in Fig. 4(b), the EIC chromatogram at m/z 407,
425, and 505 shows peak 1 at 2.28 min, peak 6 at 4.82 min, and
peak 7 at 7.41 min, which were matched with the retention
time (tR) of standard lincomycin, clindamycin, and clinda-
mycin phosphate, respectively. Accurate mass analyses of
m/z 407, 425, and 505 gave empirical formulae
of C18H35N2O6S, C18H34N2O5SCl, and C18H35N2O8PSCl,
which are identical to those obtained from the protonated
molecules of lincomycin, clindamycin, and clindamycin
phosphate, respectively (Table 3). Furthermore, MS/MS
spectra of these peaks (Table 4) are in close agreement with
the characteristic fragments of standards. Thus, peaks 1, 6,
Copyright # 2009 John Wiley & Sons, Ltd.
and 7 are identified as lincomycin, clindamycin, and
clindamycin phosphate, respectively.
The identification of the substances for which no
comparable standards were available relied on the MS data
alone to elucidate the suspected structures of the peaks
obtained from the total ion current (TIC) and EIC
chromatograms (Fig. 4). Table 3 shows all accurate masses
of protonated molecules were within 5 ppm. Product ions
measured are shown in Table 4 and compared with the data
found in previous studies. The available accurate mass data
for these substances and its product ions confirmed our
structures.
Rapid Commun. Mass Spectrom. 2009; 23: 899–906
DOI: 10.1002/rcm
904 S.-M. Wang et al.

Table 3. Q-TOF accurate mass analysis data of identification of clindamycin phosphate and its related impurities
Peak Identification Empirical formula of ion seen [identity] Theoretical m/z Experimental m/z Error (ppm)

1 lincomycin C18H35N2O6S 407.2216 407.2234 4.5

2 7-epilincomycin-2-phosphate C18H36N2O9PS 487.1879 487.1901 4.5
3 lincomycin-2-phosphate C18H36N2O9PS 487.1879 487.1864 3.1
4 clindamycin B C17H32N2O5SCl 411.1720 411.1732 2.8
5 clindamycin B-2-phosphate C17H33N2O8PSCl 491.1384 491.1369 3.0
6 clindamycin C18H34N2O5SCl 425.1877 425.1893 3.8
7 clindamycin-2-phosphate C18H35N2O8PSCl 505.1540 505.1535 1.0

Table 4. Fragmentation pathways of identification of clindamycin phosphate and its related impurities (m/z with relative
abundance (%) in parentheses) (tR: retention time)
Peak tR (min) Identification MS [identity] m/z MS/MS [identity] m/z

1 2.28 lincomycin 407 [MHþ] 389 [MHþ-H2O] (3%), 359 [MHþ-HSCH3] (7%), 126 [Aþ] (100%).
2 2.33 7-epilincomycin -2-phosphate 487 [MHþ] 469 [MHþ-H2O] (9%), 439 [MHþ -HSCH3] (9%), 389 [MHþ-HPO3-H2O]
(7%), 126 [Aþ] (100%).
3 2.71 lincomycin -2-phosphate 487 [MHþ] 469 [MHþ-H2O] (8%), 439 [MHþ-HSCH3] (6%), 407 [MHþ-HPO3] (4%),
389 [MHþ-HPO3-H2O] (2%), 126 [Aþ] (100%).
4 3.23 clindamycin B 411 [MHþ] 393 [MHþ-H2O] (11%), 363 [MHþ-HSCH3] (21%), 112 [Bþ] (100%).
5 4.32 clindamycin B-2-phosphate 491 [MHþ] 473 [MHþ-H2O] (11%), 443 [MHþ-HSCH3] (25%), 411 [MHþ-HPO3] (7%),
393 [MHþ -HPO3-H2O] (7%), 321 [MHþ-HPO3-CHOHCHSCH3] (7%), 112
[Bþ] (100%).
6 4.82 clindamycin 425 [MHþ] 407 [MHþ-H2O] (1%), 389 [MHþ-HCl] (2%), 377 [MHþ-HSCH3] (11%),
341 [MHþ-HCl-HSCH3] (3%), 126 [Aþ] (100%).
7 7.41 clindamycin -2-phosphate 505 [MHþ] 487 [MHþ-H2O] (8%), 457 [MHþ-HSCH3] (9%), 425 [MHþ-HPO3] (3%),
407 [MHþ-HPO3-H2O] (3%), 359 [MHþ-HPO3-H2O-HSCH3] (2%), 335
[MHþ-HPO3-CHOHCHSCH3] (2%), 126 [Aþ] (100%).

The EIC chromatogram at m/z 487 (Fig. 4(b)) showed two
peaks: peak 2 at 2.33 min and peak 3 at 2.71 min. The full-scan
mass spectrum of the two peaks both showed the ion at m/z
487 (Figs. 5(a) and 5(b)). The two ions at m/z 487 had accurate
masses of 487.1901 and 487.1864, and an empirical formula
of C18H36N2O9SP (calculated mass 487.1879). MS/MS of the
two ions at m/z 487 both produced the diagnostic fragments
corresponding to the neutral losses of H2O, HSCH3,
[HPO3þH2O], and the residue of 3-propyl-N-methylpyrro-
lidine, which are in close agreement with the fragmentation
pathway of clindamycin phosphate shown in Fig. 3. There-
fore, the ions at m/z 487 were deduced as degradation
products of clindamycin phosphate. Oesterling and Rowe
studied the hydrolysis of clindamycin-2-phosphate and
suggested scission of 7(S)–Cl to form the 7(R)-OH analog
predominating at pH greater than 6 at 90.48C.7 Munson and
Kubiak also described lincomycin-2-phosphate as one of the
principal degradation products in clindamycin-2-phosphate
bulk drug and formulations.3 It has been reported that
lincomycin is the major degradation product of clindamycin
at pH greater than 5, and the formation of a trace of 7-
epilincomycin after refluxing an aqueous solution of
clindamycin at pH 7.8.16 It was therefore deduced that the
substances which matched the m/z 487 ion were protonated
lincomycin-2-phosphate and its isomer 7-epilincomycin-2-
phosphate. Comparing the peak areas of peaks 2 and 3 in
Fig. 4(b), it is obvious peak 2 represents the minor
degradation product, 7-epilincomycin-2-phosphate, and peak
3 represents the major degradation product, lincomycin-
Copyright # 2009 John Wiley & Sons, Ltd.
2-phosphate. The ratio of the amount of 7-epilincomycin-2-
phosphate to lincomycin-2-phosphate was 3:83, calculated
by measuring the peak areas for each obtained from the EIC
chromatogram.
Clindamycin B-2-phosphate was the other one related
impurity in clindamycin-2-phosphate bulk drug or formu-
lations.3,9 The EIC chromatogram at m/z 491, protonated
clindamycin B phosphate, showed the single peak 5 at
4.32 min. High-resolution mass spectrometry (HRMS) of the
peak exhibited the characteristic protonated isotope pattern
at m/z 491.1369 and 493.1286 with about 3:1 ratio of relative
abundance, indicating the presence of a chlorine atom in the
molecule. The ion at m/z 491 had an empirical formula
of C17H33N2O8PSCl, suggesting this substance might be
clindamycin B phosphate. The fragmentation spectrum
could be interpreted by comparison with the fragmentation
spectra of clindamycin-2-phosphate and its related sub-
stances, as studied previously. MS/MS of the ion at m/z 491
yielded ions at m/z 473, 443, 411, 393, 321, and 112, with
relative abundances of 11, 25, 7, 7, 7, and 100%, representing
the losses of H2O, HSCH3, HPO3, [HPO3þH2O], [HPO3þ
CHOHCHSCH3], and the residue of 3-ethyl-N-methylpyr-
rolidine, respectively.
Since clindamycin was one major hydrolysis product after
heating clindamycin phosphate injection at 1008C for 30 min,
treating clindamycin B-2-phosphate under the same operat-
ing conditions may give the hydrolysis product clindamycin
B. It was found that the EIC chromatogram at m/z 411,
protonated clindamycin B, showed a single peak 4 at
Rapid Commun. Mass Spectrom. 2009; 23: 899–906
DOI: 10.1002/rcm

Figure 5. ESI-HRMS and ESI-MS/MS of peaks 2, 3, 4, and 5 in positive mode: (a–d) HRMS of peak
2, 3, 4; and 5, respectively and (e–h) MS/MS fragmentation spectra of [MHþ] ion of peaks 2, 3, 4, and
5, respectively.
3.23 min. The ion at m/z 411 had accurate mass of 411.1732
and an empirical formula of C17H32N2O5SCl. MS/MS of the
ion at m/z 411 yielded diagnostic ions at m/z 393, 363, and 112,
which are in close agreement with the characteristic
fragments of clindamycin B found in clindamycin bulk drug
interpreted by Zhou et al.17 Therefore, the substance was
confirmed as clincomycin B, which has never been reported
previously in the analysis of clindamycin phosphate bulk
drug or formulations.
CONCLUSIONS
A HPLC/ESI-MS/MS method has been developed for the
determination of clindamycin phosphate and its related
impurities. A heat-stressed clindamycin phosphate injection
was examined and six related lincosamide compounds have
been identified. The experimental results demonstrate
HPLC/ESI-MS/MS is a powerful analytical tool for the
rapid identification of the lincosamide compounds.
Copyright # 2009 John Wiley & Sons, Ltd.
Analysis of clindamycin phosphate injection by LC/MS 905
The positive ESI-MS/MS of clindamycin phosphate and its
derivative compounds showed some diagnostic fragments,
such as the neutral losses of H2O, HCl, methanethiol,
2-methylthioethenol, HPO3, [HPO3þH2O], and the residues
of 3-propyl-N-methylpyrrolidine and 3-ethyl-N-methylpyr-
rolidine, which are specific and useful for the identification of
the lincosamide antibiotics and related impurities.
Acknowledgement
This research was supported by the National Natural Science
Foundation (No. 2027205).
REFERENCES
1. Cherlet M, Croubels S, De Backer P. J. Mass Spectrom. 2002;
37: 848.
2. Liu CM, Chen YK, Yang TH, Hsieh SY, Hung MH, Lin ET.
J. Chromatogr. B 1997; 696: 298.
3. Munson JW, Kubiak EJ. J. Pharm. Biomed. Anal. 1985; 3: 523.
4. Nahata MC, Morosco RS, Hipple TF. Am. J. Hosp. Pharm.
1993; 50: 2559.
Rapid Commun. Mass Spectrom. 2009; 23: 899–906
DOI: 10.1002/rcm
906 S.-M. Wang et al.
5. Li LC, Parasrampuria J, Bommireddi A, Pec E, Dudleston A,
Mayoral J. Drug Dev. Ind. Pharm. 1998; 24: 89.
6. Lou DB, Duan GL, Gao MJ, Sun GY. Chin. J. New Drugs Clin.
Rem. 2001; 35: 278.
7. Oesterling TO, Rowe EL. J. Pharm. Sci. 1970; 59: 175.
8. Brown LW. J. Pharm. Sci. 1978; 67: 1254.
9. Xia ZH, Mao CJ, Han XF, Zhou ZM. Chin. J. Pharm. Anal.
2002; 22: 292.
10. Ye YR, Bektic E, Buchta R, Houlden R, Hunt B. J. Sep. Sci.
2004; 27: 71.
11. Guo P, Ding YM, Wu C, Wu XZ, Du Y, Zhu HY, Zhan GJ.
J. Chin. Pharm. Univ. 2000; 31: 210.
Copyright # 2009 John Wiley & Sons, Ltd.
12. Zhang DS, Li J, Xu MZ, Wang C, Yao L, Xiang BR, Hu CQ.
Chin. J. Antibiot. 2008; 33: 347.
13. Hornedo-Nunez A, Getek TA, Korfmacher WA, Simenthal F.
J. Chromatogr. 1990; 503: 217.
14. Li HY, Wang SM, Liu HM, Li J, Han D, Bu SS, Zhang MZ.
Rapid Commun. Mass Spectrom. 2008; 22: 1.
15. Pelzing M, Macht M. The Applications Book: Determination of
Elemental Composition of Phytohormones using LC-ESI-TOF MS
for Accurate Mass Measurements, Bruker Daltonik GmbH:
Germany, September 2004; 24–229.
16. Oesterling TO. J. Pharm. Sci. 1970; 59: 63.
17. Zhou H, Zheng Z, Wu H, Tai Y, Cao X, Pan Y. J. Pharm.
Biomed. Anal. 2006; 41: 1116.
Rapid Commun. Mass Spectrom. 2009; 23: 899–906
DOI: 10.1002/rcm