Claudio Medana, Paola Calza, Federica Dal Bello, and Claudio Baiocchi
The occurrence of toxic drug residues is a hot topic concerning emerging environmental contaminants.
The occupational exposure to anticancer drugs could be severe for health-care personnel and general
public attending potentially polluted areas (such as oncology departments of hospitals) because of
high therapeutic concentrations. The aim of this work was to generate degradation compounds similar
to those formed in metabolic or environmental pathways by adopting a photocatalytic process and to
identify them in indoor environments alongside parent compounds. A photodegradation model was
applied to cyclophosphamide and mitomycin C. Studied substances and degradants were quantified
in environmental samples by liquid chromatography with multiple-stage mass spectrometry (LC–MS
n)
analysis using an orbital trap instrument with an electrospray interface. Various oxidative degradants
were formed using the photocatalytic simulation model of degradation of antineoplastic drugs, beside
some hydrolysis and molecule breakdown subproducts. High resolution MS
nspectra were used to
identify and confirm the proposed structures. Kinetics of formation of the main degradation products
were also studied.
LC–HRMS Determination of
Anticancer Drugs as Occupational
Contaminants Applied to
Photocatalytic Degradation of
Molecules of Different Stability
T
he frequency of recognition of toxic drug residues is an important issue concerning emerging environmental contaminants. This problem is also exacerbated by the low therapeutic index of drugs used in cancer chemotherapy. The occupational exposure to residues of anticancer drugs must be kept under strict control by safety managers because it could be severe for health care personnel and general public attending potentially polluted areas (such as oncology depart-ments and pharmacies of hospitals) because of high therapeu-tic concentrations. In addition, urinary elimination by patients represents a source of contamination. The monitoring of drug residues is necessary to guarantee occupational safety and theabsence of pollution. The presence of anticancer drug resi-dues in environmental water samples has been reported (1–3). Looking closer at the structural features of these compounds, it is possible to see that the majority of chemotherapeutic drugs hold molecular structures with a high chemical reactivity (4). As a consequence of this great aptitude to undergo chemical transformation, it could be difficult to record measurable lev-els of some of these drugs. As an example, in a previous study (5) we found evidence of the difficulty to recover traces of mi-tomycin C because of its low stability. The aim of this work is to generate degradation compounds similar to those formed in metabolic or environmental pathways by adopting a
pho-tocatalytic process and to identify them in indoor environments. An additional purpose of this study is to develop a se-lective and sensitive method as a tool to detect drug transformation and degra-dation products and to quantify them in biological and environmental samples, beside parent compounds.
I n t h i s work , t wo a nt ic a nc er drugs were studied: the N-mustard c yc l o p h o s p h a m i d e (N , N b i s (2 - chloroethyl)-1,3,2-oxazaphosphinan-2-amine 2-oxide) and the aziridine mitomycin C ([(1aS,8S,8aR,8bS)-6- amino-8a-methoxy-5-methyl-4,7-di- oxo-1,1a,2,4,7,8,8a,8b-octahydroazi-reno[2′,3′:3,4]pyrrolo[1,2-a]indol-8-yl] methyl carbamate). We chose cyclo-phosphamide and mitomycin C be-cause of the popular use of these drugs and because of their different stability. The photodecomposition of drugs was carried out by heterogeneous photo-catalysis, using titanium dioxide in water as previously reported (6,7). By this process, the photodecomposition of drugs could be due to either oxidative species or reductive conduction band electrons (8,9).
A highly sensitive method was de-veloped to measure the degradation products in wipe samples that allows detection of parent drugs at the nano-gram level (5). Occupational exposure monitoring of these drugs shows fre-quent cyclophosphamide positive mea-surements but no significant mitomycin C determination. The high instability of the mitomycin molecule is probably the main cause of these findings. One of the targets of this work is to identify new possible markers of mitomycin in environmental pollution. Identification and characterization of degradation compounds could contribute signifi-cantly to the development of more ef-fective analytical methods to monitor occupational exposure.
High performance liquid chroma-tography (HPLC) coupled to mass spectrometry (MS) via an electrospray ionization (ESI) interface is a powerful tool to identify and measure anticancer drugs in the environment (10), especially in health care settings (11). However, liq-uid chromatography—mass spectrom-etry (LC–MS) analysis performed in
full-scan mode is not sensitive enough to detect and characterize metabolites at trace levels and the excellent sensitivity provided in multiple reaction monitor-ing (MRM) mode suffers from a lack of structural information to characterize the huge number of potential degradants formed from drug compounds.
For this reason, we developed a sensi-tive liquid chromatography–high-reso-lution mass spectrometry (LC–HRMS) methodology using orbital trap tech-nology (12) with the purpose of exploit-ing the potential of untargeted analysis
for which it is fitted. From the analysis of spectra acquired at high resolving power, it was possible to hypothesize the structure of the main degradation products observed in the photocatalysis simulation model.
Experimental
Chemicals and Materials
LC–MS-grade acetonitrile was pur-chased from VWR (VWR Interna-tional PBI). Extra-pure formic acid was purchased from Fluka (Sigma-Adrich). Cyclophosphamide was from Baxter and Figure 1: Cyclophosphamide MH+ MS2 proposed fragmentation pathways.
O P N N H HN H C 5H12CI2N2O2P 233.0013 m/z C 4H8CI2N 140.0034 m/z C 4H7CIN 104.0235 m/z C 4H10CI2N 142.0182 m/z C 4H8N 70.0631 m/z C 7H16CI2N2O2P 261.0326 m/z C4H9CIN 106.0423 m/z - C 2H4 - HCI - HCI - CH 4O2P - C 3H7CINO2P - C 3H8NO2P - C3H6NO2P + O + HN + H2N + + O P N N N N H CP H H H CI CI CI CI CI CI CI CI CI CI + O +
Figure 2: Mitomycin C MH+ MS2 and MS3 proposed fragmentation pathways.
O O O O O O N N N N N O O O O NH 2 NH 2 NH 2 NH 2 NH 2 H 2N H 2N H 2N - HCN - CO - CO H 3C N O O H 2N H 3C - NH 3 - NH 3 H 3C H 3C OCH 3 H H H H H H H H + + + + + NH 2 + MC C 15H19N4O5 335.1355 m/z C 13H12N3O2 242.0930 m/z C 12H12N3O2 214.0980 m/z C 12H9N2O 197.0708 m/z C 13H9N2O2 225.0658 m/z C 12H11N2O2 215.0816 m/z - HOOCNH 2 - CH 3OH
mitomycin C was purchased from ProS-trakan. Experiments were carried out
using TiO2 Degussa P25 as the photo-catalyst. Methanol, trichloroacetic acid,
heptafluorobutanoic acid, and ammonia were purchased from Sigma-Aldrich at a
Table I: Parent drugs, degradation compounds, and LC–HRMSn
parameters Compound Molecular Ion and Formula t
R (min) MS2 MS3 CP 261.0326 (MH+) C7H16C12N2O2P 17.25 233.0013 (100) C5H12C12N2O2P 140.0034 (100) C4H8C12N 140.0034 (80) C4H8CI2N 62.9978 (68) C4H7Cl 104.0235 (100) C4H7Cl2N 106.0423 (35) C4H9ClN 70.0631 (100) C4H8Cl2N 142.0181 (25) C4H10Cl2N — CP-1 275.0121 (MH+) C7H14CI2N2O3P 18.30 221.9852 (100) C7H11ClN2O2P — 221.0012 (80) C4H11ClN2O2P 213.0194 (30) C5H11ClN2O3P CP-2 293.0224 (MH+) C7H16CI2N2O4P 20.10 221.0012 (100) C4H12CIN2O2P — CP-3 142.0185 (MHC +) 7H14CI2N2O3P 2.63 80.0188 (100)C 5H11CIN2O3P — CP-4 239.0352 (M-H)-C7H13CIN2O3P 7.66 203.0585 (100) C7H11CN2O3P 175.0268 (100) C6H12CN2O2P 180.0058 (18) C6H11CINO3 — CP-5 269.0559 (M-H)-C7H14N2O7P 1.73 209.0321 (100) C5H10N2O5P — 251.0422 (78) C7H12N2O6P 207.0056 (16) C5H8N2O5P CP-6 287.0182 (M-H)-C7H13CIN2O6P 1.93 257.0082 (100) C6H11CIN2O5P 149.9957 (100) C3H5NO4P 251.0422 (49) C7H12N2O6P 149.9957 (12) C3H5NO4P MC 335.1355 (MH+) C15H19N4O5 7.87 242.0930 (100) C13H12N3O2 214.0980 (100) C12H12N3O MC-5 351.1306 (MH+) C15H19N4O6 5.69 290.1143 (100) C14H16N3O4 244.1086 (100) C13H14N3O2 272.1037 (78) C14H14N3O3 273.0875 (42) C14H13N2O4 MC-6 351.1306 (MH+) C15H19N4O6 6.72 290.1143 (100) C14H16N3O4 247.0718 (100) C13H14N3O2 273.0875 (28) C14H13N2O4 258.0879 (24) C13H12N3O3 MC-7 349.1148 (MH+) C15H17N4O6 8.04 256.0717 (100) C13H10N3O3 228.0769 (100) C12H10N3O2
purity ≥99%. HPLC-grade water was ob-tained using a MilliQ System Academic system (Millipore).
Irradiation Technique
Irradiation was performed using a Philips TLK/05 lamp (40 W/m2) with
maximum emission at 360 nm. Irradia-tion experiments were carried out in Pyrex glass cells containing the studied drug (20 mg/L) and titanium dioxide (200 mg/L) in ultrapure water. The temperature reached during irradia-tion was 38 ± 2 °C.
Liquid Chromatography
Chromatographic separation followed by MS analysis was run on a 150 × 2.0 mm, 3-µm dp Phenomenex Luna C18 column preceded by a Luna C18 guard column, using a Dionex Ultimate 3000 HPLC instrument (Thermo Fisher Figure 3: Plots of the kinetics of (a) parent compound decrease and (b) formation of main cyclophosphamide transformation products (TPs).
1 0.8 0.6 0.4 0.2 0 30 25 20 15 10 5 0 12 10 8 6 4 2 0 0 5 10 15 20 25 30 0 30 60 90 120 150 180 210 240
Irradiation time (min) Irradiation time (min)
Mitomycin C Cyclophosphamide [M+H]+ 275 (CP–1) [M+H]+ 293 (CP–2) [M+H]+ 142 (CP–3) [M+H]– 239 (CP–4) [M–H]– 287 (CP–5) [M–H]– 269 (CP–6) C/C 0 TPs [M+H] + (area X 10 –7) TPs [M+H] – (area X 10 –6)
(a)
(b)
Figure 4: LC–ESI HRMSn chromatograms of MC-5, MC-6, and MC-7 degradation products (after 5 min of irradiation) together with hypothesized
fragmentation pathways that are able to justify the main MS3 (for isomers MC-5 and MC-6) and MS2 fragments. 100 80 60 40 20 0 Relative abundance 100 80 60 40 20 0 Relative abundance 100 80 60 40 20 0 0 5 10 20 Time (min) 8.04 6.72 5.69 15 Relative abundance H2N H2N NH2 NH2 -HOOCNH2 -HOOCNH2 -HOOCNH2 -CH3OH -CH3OH -CO -HOOCH OCH2OH OCH2OH H3C H3C O O O O O H H H H H N N H2N H3C O O N O MC-5 + H2N NH2 NH2 OCH3 H3C O O O OH H H N O + NH2 + H H NH2 + C15H19N4O6 351.1306 m/z MC-6 C 15H19N4O6 351.1306 m/z H2N NH2 NH2 OCH3 OHC O O O OH H H N O + MC-7 C 15H17N4O6 349.1148 m/z C14H16N3O4 290.1143 m/z H 2N OCH3 OHC H3C O O O H H N H 2N OHC O O N NH 2 + NH2 + C 14H16N3O4 290.1143 m/z C 13H10N3O3 256.0717 m/z H2N H2N H3C O O O H N NH2 + O N NH2 + C13H12N3O4 258.0879 m/z C 12H10N3O2 228.0769 m/z C13H14N3O4 244.1086 m/z MC-5: MS3, 351.1306 à 290.1143 à 244.1086 m/z MC-6: MS3, 351.1306 à 290.1143 à 258.0879 m/z MC-7: MS2, 349.11148 à 256.0717 m/z
Scientific). The injection volume was 20 μL and the flow rate was 200 μL/min. Mobile-phase A was 0.1% formic acid in water, and mobile-phase B was acetoni-trile. After a 2-min isocratic run at 5% B, a gradient to 100% B was concluded at 20 min. A column wash with 100% B (3 min) was followed by equilibration with 5% B starting at 25 min and main-tained until 30 min. All injection vol-umes were 20 μL, the column tempera-ture was 30 °C, and the flow rate was 0.2 mL/min. The wash solvent was 1:2 (v/v) water–acetonitrile.
High-Resolution Mass Spectrometry
An LTQ Orbitrap mass spectrometer (Thermo Scientific), equipped with an atmospheric pressure interface and an ESI ion source, was used. Analyses were carried out in both the positive and negative ion modes. The LC col-umn effluent was delivered to the ion source using nitrogen as both sheath and auxiliary gas. The source volt-age was set to 4.4 kV (-4.0 kV in the negative ion mode). The heated capil-lary temperature was maintained at 275 °C. The nitrogen f lows as sheath gas and auxiliary gas were set at 20 and 10 (arbitrary units), respectively. The acquisition method used had pre-viously been optimized in the tuning sections for the parent compound (capillary, magnetic lenses, and colli-mating octapoles voltages) to achieve maximum sensitivity. The main tun-ing parameters adopted for the ESI source in the positive ion mode were 24 V for capillary voltage and 65 V for tube lens and -15 V and -54 V in the negative ion mode, respectively. Full scan spectra were acquired in the range 50–1000 m/z. Multiple-stage mass spectrometr y (MSn
) spectra were acquired in the range between ion trap cut-off and precursor ion m/z values. Spectra were acquired with the resolution R = 30000 (full width at half maximum [FWHM]). Mass accuracy of recorded ions (vs. calculated) was ± 0.001 u (without in-ternal calibration). Calibration curves for cyclophosphamide and mitomycin C were constructed from five calibra-tion points for each compound in the range 0.10–50.0 μg/mL.
Figure 5: LC–HRMS chromatogram of the cyclophosphamide photocatalytic degradation mixture after 15 min of irradiation (positive ion mode; on the top) and 30 min of irradiation (negative ion mode; on the bottom).
100 80 60 40 20 0 22.39 18.32 20.11 2.65 7.66 1.73 1.96 0 5 10 15 20 25 30 Time (min) 0 5 10 15 20 25 30 35 Time (min) NL: 1.68E7 m/z= 261.0326 CP NL: 6.40E6 m/z= 275.0121 CP-1 NL: 8.41E5 m/z= 293.0224 CP-2 NL: 1.00E5 m/z= 269.0559 CP-5 NL: 3.07E5 m/z= 287.0289 CP-6 NL: 1.77E6 m/z= 142.0182 CP-3 NL: 1.03E5 m/z= 238.9910 CP-4 100 80 60 40 20 0 100 80 60 40 20 0 100 80 60 40 20 0 100 80 60 40 20 0 100 80 60 40 20 0 100 80 60 40 20 0 Relative abundance Relative abundance
Results and Discussion
Cyclophosphamide and Mitomycin C MS-MS Study
The MS2 and MS3 fragmentation study
of the studied compounds showed some characteristic losses that were considered in identif ying the un-known compounds formed during the photo-induced degradation of the drugs. The product ions generated from protonated molecular ions are shown in Figures 1 and 2. MSn
frag-ment identification was confirmed by the high resolution study of neutral losses. The main product ions of cy-clophosphamide were formed through an ethylene molecule loss or three different phosphoric ring breakings respectively leading to the formation of aziridine (106.0423 m/z), amine (142.0182 m/z), or imine (140.0034 m/z) nitrogen mustard structures. The successive fragmentation of these ions mainly involve simple HCl elimination (Figure 1).
The protonated ion of mitomycin C generates a single product ion (242.0930 m/z) by concerted loss of methanol and carbamic acid. The main MS3 losses
in-volve CO, HCN, and ammonia elimi-nation (Figure 2). The MS4
converg-ing losses of CO and NH3 to give the common product ion at 197.0708 m/z is noteworthy.
In the negative ion mode, cyclophos-phamide does not ionize in a noticeable way, whereas mitomycin C originates at a [M-H]– ion, about 10 times less
in-tense than the corresponding MH+ ion
formed in the positive ion mode. This deprotonated ion, by MS-MS fragmen-tation, again gives a single product ion eliminating HCNO (43.0073 u).
Laboratory Simulation
Laboratory simulation was initially performed on cyclophosphamide and mitomicyn C in ultrapure water, in the presence of titanium dioxide as photo-catalyst. The parent drugs were easily degraded and completely disappeared within 30 min of irradiation (Figure 3a). In parallel with the disappearance of parent drugs, a number of transfor-mation products were identified. Their sequential evolution is shown for cy-clophosphamide in Figure 3b and will
Figure 6: Cyclophosphamide degradation products detected after heterogeneous photocatalysis decomposition. O O P N N H CI CI CP O O O P N N H H CI CI O O P N N NH 2 CI CI CI CI CP-1 HOOC O O O P N N H O O O P N N H H H CI O H CI HO HO OH O O P N N H HO HO OH OH CP-4 CP-5 CP-6 CP-2 CP-3
Known human metabolites
Other photocatalysis products
Figure 7: MS2 spectrum of degradation product CP-1 together with the hypothesized fragmentation pathway. 100 90 80 70 60 50 40 30 20 10 0 Relative abundance 150 200 m/z 142.0183 159.9925 213.0194 233.0632 221.0012 221.9852 250 O O O P -C2H3CI -C3H2O NH3 C 7H14CI2N2O3P 275.0121 m/z C 5H11CIN2O3P 213.0194 m/z C4H12CI2N2O2P 221.0012 m/z C 7H11CIN2O2P 221.9852 m/z -H2O N N H CI CI -CI CP-1 H + O O O P N O P N N N H CI HO CI CI H + O OH O P N N H CI CI H +
be discussed in the characterization of transformation products section. Trans-formation products were characterized
by different m/z ratios of the precursor ions as well as by MS3 analysis in the
case of hydroxylated isomers, as shown
in Figure 4. This finding is in agreement with the nonselectivity of the active spe-cies involved in the photocatalytic pro-cess, that is, OH radicals (13).
Several species have been detected along with the cyclophosphamide and mitomycin C disappearance. The inter-mediates were characterized by taking into account their chromatographic be-havior and kinetics of evolution, coupled with the exact mass information, an ac-curate analysis of MS and MSn
spectra and a comparison with parent drug frag-mentation pathways. Retention times, m/z values, and elemental composition of HRMSn
observed ions are reported in Table I.
Characterization of
Cyclophosphamide Transformation Products from Irradiation Study
Six main species were detected within cyclophosphamide degradation. Ex-tracted chromatograms at 2 ppm mass tolerance level are shown in Figure 5. We evaluated both the formation of known metabolites and the generation of new unknown photocatalytic deg-radation products. The isotopic abun-dance pattern was used as a powerful additional constraint for identification of unknowns.
The cyclophosphamide degradation study showed both the formation of previously known human metabolites (14,15) (CP-1, CP-2, and CP-3 shown in Figure 6) and the generation of three new degradants whose potential structures were reported in Figure 6 (CP-4, CP-5, and CP-6). Compound CP-1 matches with the main known metabolite 4-ketocyclophosphamide whereas the quantitatively most abun-dant compound on the bases of peak areas obtained was the compound CP-3, known as NOR-nitrogen mus-tard. The maximum survival time in the presence of titanium dioxide and light was 240 min (appearance and disappearance kinetics are shown in Figure 3). For example, MS2 spectrum
(CP-1) is shown in Figure 7. In this study, the acquisition of positive and negative ions spectra was crucial be-cause CP-2 and CP-3 were detectable as positive ion only when CP-4, CP-5, and CP-6 negative ions.
Figure 8: Mitomycin C degradation products detected after heterogeneous photocatalysis decomposition (bottom box) in comparison with the known metabolites (16) (top box).
O O O O N H2N NH2 NH MC H3C OCH3 O O O O N H 2N NH2 NH2 MC-1 H3C O OH O O O N H2N NH2 OCH3 OCH3 OHC NH MC-5 MC-6 H3C O OH O O O O N H2N NH 2 NH H3C MC-7 O O O O N H 2N NH2 NH O O OH O O N H2N NH2 NH2 MC-2 (cis) MC-3 (trans) H3C O O O OH N H 2N NH2 NH MC-4 H 3C
Known human metabolites / degradants
Main photocatalysis products
Figure 9: HRMS3 spectra of isomeric compounds MC-5 and MC-6. 100 90 80 70 60 50 40 30 20 10 0 Relative abundance 100 90 80 70 60 50 40 30 20 10 0 Relative abundance
MC-5
MC-6
HCOOH loss
H
2O loss
C
2H
5N loss
CH
3OH loss
244.1086 272.1037 273.0875 229.0976 247.0718 273.0875 165.0658NH
3loss
NH
3loss
100 150 200 250 300 350 m/zCharacterization of
Mitomycin C Transformation Products from Irradiation Study
The degradation pattern of mitomycin C looks simpler than the cyclophospha-mide one. Three species were recognized and their chromatograms are shown in Figure 4.
The mitomycin C degradation study did not show any appreciable trace of known degradants. The formation of such transformation products, named mitosanes and mitosenes, is well known in the literature (16,17) (MC-1, MC-2, MC-3, and MC-4, Figure 8). On the contrary, three new compounds were obtained by heterogeneous photoca-talysis (MC-5, MC-6, and MC-7; Figure 8). The first two products are formed by hydroxylation and are isomers; the last one comes from a hydroxylation and oxidation process. To hypothesize the structures reported in Figure 4, besides the corresponding chromatograms, a high resolving power multistage MS study was performed. We hypothe-sized the structures shown in Figure 2 and Figure 4 for the couple of isomers named MC-5 and MC-6 because of the different MS3 fragmentation observed.
The elimination of methanol that char-acterizes MC-5 MS3 behavior suggested
that we leave the methoxy moiety un-modified and position the OH group on C-8 to explain the simple formamide loss. On the contrary, MC-6 was identi-fied as 8a-hydroxymethoxy-derivative to justify the MS3 base peak formed by
H2CO2 elimination. Also, in this case, we considered the more probable OH at-tack on the methoxy group than on the 5-methyl substituent because the MS2
behavior was different from the parent drug one. The two MS3 example spectra
are displayed in Figure 9.
Conclusions
In a workplace environment, just like in vivo, exogenous compounds go through many chemical and bio-transforma-tions; however, most of these transfor-mations could not be detected by cur-rent methods applied to metabolism studies or targeted on a specific drug molecule (18,19). In the present study, we demonstrated how low-abundance drug metabolites and unknown
degra-dation products could be successfully detected, analyzed, and characterized. Chemical structures of these detected metabolites were proposed, based on MS and MSn
spectra. The described methodology was applied in heteroge-neous photocatalysis experiments in which two anticancer drugs were ar-tificially decomposed. The photocata-lytic degradation model identified new transformation products from two an-ticancer drugs, cyclophosphamide and mitomycin C. In the case of cyclophos-phamide, the simulation model gives a broad scenario of the formation of de-gradants including known human me-tabolites. In contrast, the same model does not permit the formation of known degradation products of mitomycin C, formed both in vivo and in spontaneous decomposition, probably because of the very different reaction kinetics.
From the analytical point of view, LC–HRMS methodology can greatly help drug residue monitoring together with the untargeted analytes identifica-tion, and, combined with photocataly-sis, could be a suitable tool to simulate biological and chemical transformations through photochemical laboratory ex-periments. This suggests that new struc-tures for future metabolic and environ-mental studies are possible.
References
(1) E. Zuccato , S. Castiglioni, and R. Fanelli, J. Hazard. Mat. 122, 205–209 (2005).
(2) M. Gros, M. Petrovic, and D. Barceló,
Anal. Bioanal. Chem. 386, 941–952
(2006).
(3) Y. Valcárcel, S. González Alonso, J.L. Rodríguez-Gil, A. Gil, and M. Catalá,
Chemosphere 84, 1336–1348 (2011).
(4) C. Avendano and J.C. Menendez,
Me-dicinal Chemistry of Anticancer Drugs
(Elsevier, Amsterdam, The Nether-lands, 2008).
(5) C. Medana, C. Baiocchi, F. Dal Bello, V. Scarpino, and E. Davoli, unpub-lished research.
(6) C. Medana, P. Calza, V. Giancotti, F. Dal Bello, M. Pasello, M. Montana, and C. Baiocchi, Drug Test. Anal. 3, 724–734 (2011).
(7) P. Calza, C. Medana, E. Padovano, V. Giancotti, and C. Minero, Rapid
Comm. Mass Spec. 26, 1687–1704
(2012).
(8) M.R. Hoffmann, S.T. Martin, W. Choi, and D.W. Bahnemann, Chem. Rev. 95, 69–96 (1995).
(9) D.W. Bahenmann, J. Cunningham, M.A. Fox, E. Pelizzetti, P. Pichat, and N. Serpone, in Aquatic and Surface
Chemistry, G.R. Helz, R.G. Zepp, and
D.G. Crosby, Eds. (Lewis, Boca Raton, Florida, 1994), pp. 275–278.
(10) N. Negreira, M. Lòpez de Alda, and D. Barcelò, J. Chrom. A 1280, 64–74 (2013).
(11) C. Sottani, B. Porro, M. Imbriani, and C. Minoia, Toxicol. Lett. 213, 107–115 (2012).
(12) M. Scigelova and A. Makarov,
Bio-analysis 1, 741–754 (2009).
(13) A. Kunai, S. Hata, S. Ita, and K. Sasaki,
J. Am. Chem. Soc. 108, 6012–6016
(1986).
(14) C. Fenselau, J. AOAC 59, 1028–1036 (1976).
(15) L.A. Vanaerts, J.H. Copius Peere-boom, and J. Stegeman Noordhoek,
Toxicol. In Vitro 9, 677–684 (1995).
(16) J.H. Beijnen and W.J.M. Underberg,
Int. J. Pharm. 24, 219 (1985).
(17) M.M.L. Fiallo, H. Kozlowski, and A. Garnier-Suillerot, Eur. J. Pharm. Sci. 12, 487–494 (2001)
(18) M. Holcapek, L. Kolarova, and M. Nobilis, Anal. Bioanal. Chem. 391, 59–78 (2008).
(19) M.J. Nozal, J.L. Bernal, M.T. Martìn, J. Bernal, R.M. Torres, and J. Merayo, J.
Pharm. Biomed. Anal. 40, 100–104
(2006).
Claudio Medana, Federica Dal Bello, and Claudio Baiocchi
are with the Department of Molecular Biotechnology and Health Sciences at the Università degli Studi di Torino in Torino, Italy. Paola Calza is with the Department of Chemistry and the NIS Centre of Excellence at the Università degli Studi di Torino.
Direct correspondence to: claudio.medana@unito.it ◾
For more information on this topic, please visit our homepage at: www.spectroscopyonline.com
