Food Chemis y 444 (2024) 138643
A ailable online 4 Feb ua y 2024
0308-8146/© 2024 The Au ho (s). Published by Else ie L d. This is an open access a icle unde he CC BY-NC-ND license (h p://c ea i ecommons.o g/licenses/by-
nc-nd/4.0/).
Mul i a ge and suspec -sc eening o an imic obials in ege ables samples:
Up ake expe imen s and iden i ica ion o ans o ma ion p oduc s
I. Ve ga a-Luis
a
,
b
,
*
, M. Jin
a
, J.C. Baez-Mill´
an
a
, B. Gonz´
alez-Gaya
a
,
b
, I. Iju co
c
, M. Lacues a
c
,
M. Oli a es
a
,
b
, A. P ie o
a
,
b
a
Depa men o Analy ical Chemis y, Facul y o Science and Technology, Uni e si y o he Basque Coun y (UPV/EHU), Leioa, Basque Coun y, Spain
b
Resea ch Cen e o Expe imen al Ma ine Biology and Bio echnology (PIE), Uni e si y o he Basque Coun y (UPV/EHU), Plen zia, Basque Coun y, Spain
c
Depa men o Plan Biology and Ecology, Facul y o Pha macy, Uni e si y o he Basque Coun y (UPV/EHU), Vi o ia-Gas eiz, Basque Coun y, Spain
ARTICLE INFO
Keywo ds:
An imic obials
Vege ables
UHPLC-MS/MS
q-O bi ap
T ans o ma ion p oduc s
ABSTRACT
This wo k p o ided an accu a e analy ical me hod o pe o m a mul i a ge analysis o a a ie y o an imic obials
(AMs) including sul onamides, e acyclines, mac olides, luo oquinolones and quinolones, one imidazole and
one ni oimidazole, one iazole, one diaminopy idine and one de i a i e o Penicillium s oloni e um in ege a-
bles. The analysis is pe o med using liquid-ch oma og aphy coupled o a low- esolu ion iple quad upole mass
spec ome e (UHPLC-MS/MS) o de ec he a ge analy eso coupled o a high- esolu ion q-O bi ap (HRMS) o
moni o he o med ans o ma ion p oduc s (TPs). Bo h ins umen s we e compa ed in e ms o limi s o
quan i ica ion and ma ix e ec a he de ec ion. The me hod was applied o de e mine he p esence o AMs in
o ganic and non-o ganic ege ables, whe e sul adiazine and mycophenolic acid we e de ec ed. On he o he
hand, he ans e ence o ou AMs ( ime hop im, sul ame hazine, en o loxacin, and chlo e acycline) om
soils o le uces was e alua ed h ough con olled up ake expe imen s. The choice o AMs was based on he
classi ica ion in o di e en amilies, and on he ac ha hose AM amilies a e he mos equen ly de ec ed in
he en i onmen . In his case, each o he AMs wi h which he soils we e con amina ed we e ound in he exposed
le uces. Mo eo e , in bo h s udies, speci ic TPs o he AMs we e iden i ied, posing he necessi y o assessing hei
e ec s in ela ion o ood and human sa e y.
1. In oduc ion
The g owing global popula ion and he p og ess o socie y ha e
esul ed in he widesp ead use o chemicals, which se e o enhance ou
e e yday li es and p omo e human wel a e. Howe e , he con inued use
o chemicals has led o hei accumula ion in he en i onmen , esul ing
in d ugs such as an imic obials (AMs) being conside ed en i onmen al
pollu an s (Ga cía e al., 2020).
AMs ha e been ound in nume ous en i onmen al samples, including
wa e (Valc´
a cel e al., 2011), sludge (Li e al., 2013), animal manu e
(Zalewska e al., 2021), soil (Hang e al., 2021), and e en ege ables
(Kang e al., 2013) as a esul o hei en y in o he ood chain. How-
e e , AMs a e equen ly epo ed nea he limi s o quan i ica ion in
scien i ic li e a u e due o hei high biodeg adabili y unde ligh o
empe a u e condi ions (Cyco´
n e al., 2019). Acco dingly, mo e and
mo e scien i ic s udies a e ocusing he sea ch and de e mina ion o
ans o ma ion p oduc s (TPs) (Sunye -Caldú & Diaz-C uz, 2021; Tadi´
c
e al., 2019). The main obse a ions o hose s udies sugges ha he
concen a ion o TPs in ege able samples migh ac ually exceed he
le els o hei p ecu so compounds. Fo ins ance, Tadi´
c e al. epo ed a
i e- old highe concen a ion o ime hop im 304 (TMP304), a
deg ada ion p oduc o TMP, in a le uce sample compa ed o i s p e-
cu so (Tadi´
c e al., 2019), while di e en TPs o sul onamides (SAs)
we e ound by Sunye e al. (Sunye -Caldú & Diaz-C uz, 2021) in le -
uces i iga ed wi h eclaimed wa e .
The p esence o AMs in e en low concen a ions can be de imen al
o bo h human and ecosys em heal h. This includes he po en ial sp ead
o AM esis ance, alle gic eac ions, oxici y, and mo e (Jadeja & Wo -
ich, 2022). The e o e, i is essen ial o es ablish a eliable analy ical
me hodology o e ec i ely moni o bo h AMs and TPs in complex
en i onmen al samples, such as ege ables.
The high concen a ion o pigmen s and cellulose in ege able sam-
ples poses a signi ican analy ical challenge o he analysis o AMs, as
hese majo in e e ences can signi ican ly impac he sensi i i y o he
* Co esponding au ho a : Depa men o Analy ical Chemis y, Facul y o Science and Technology, 48490 Leioa, Basque Coun y, Spain.
E-mail add ess: [email p o ec ed] (I. Ve ga a-Luis).
Con en s lis s a ailable a ScienceDi ec
Food Chemis y
jou nal homepage: www.else ie .com/loca e/ oodchem
h ps://doi.o g/10.1016/j. oodchem.2024.138643
Recei ed 24 Oc obe 2023; Recei ed in e ised o m 17 Janua y 2024; Accep ed 29 Janua y 2024
Food Chemis y 444 (2024) 138643
2
me hod (Anumol e al., 2017). Liquid ch oma og aphy andem mass
spec ome y (LC-MS/MS) enables he comp ehensi e analysis o mul-
iclass AMs wi hin complex ma ices a e an e ec i e sample p e-
ea men o emo e any in e e ence while p ese ing he a ge com-
pounds. (Dasenaki & Thomaidis, 2015). Nowadays, ul a high-
pe o mance LC (UHPLC) has o e aken high-pe o mance LC (HPLC)
in e ms o esolu ion and speed o analysis (Rod iguez-Alle e al.,
2013) and i is o en coupled o low esolu ion iple quad upole (QqQ)
mass analyse (Mijangos e al., 2019) when mul i esidue d ug analysis a
egula o y con ol concen a ion le els is equi ed. In o de o ex end
he analysis o unknown TPs (Jongedijk e al., 2023), a g owing numbe
o s udies a e employing high- esolu ion MS (HRMS) mass analyse s
such as q-O bi ap (Cas ellani e al., 2023) and/o Time-o -Fligh (TOF)
(Va enina e al., 2022). Rega ding he ionisa ion, elec osp ay ionisa ion
(ESI) (Sun e al., 2016) is o en used o low- and high- esolu ion
analysis.
While ecen esea ch has made signi ican s ides in he ad ance-
men o me hodologies enabling he simul aneous analysis o AMs and
TPs in a ious ma ices, he analysis is s ill equen ly cons ained o
speci ic AM amilies (Ba on e al., 2008; Meng e al., 2017). In such
cases, when mul iple AM g oups a e s udied oge he , he accu acy e-
qui emen s o he egula ions a e no ully sa is ied (da Sil a e al.,
2020).
Wi hin his con ex , he p esen wo k aimed o ex end he me hod
p e iously de eloped in he esea ch g oup o SAs and e acyclines
(TCs) de e mina ion (Ve ga a-Luis e al., 2023) o co e he simul a-
neous analysis o i e SAs, ou TCs, ou mac olides (MCs), nine luo-
oquinolones and quinolones (FQs), one imidazole (IM) and one
ni oimidazole (NIM), one iazole (TZ), one diaminopy idine (DAP) and
one de i a i e o Penicillium s oloni e um (DP) in ege ables (le uce,
onion, oma o, and ca o ) equen ly consumed by humans. In o de o
achie e his goal, we ully s udied he e ec i eness o he clean-up
p o ocol and we compa ed he sensi i i y and selec i i y o he
UHPLC-MS/MS and UHPLC-HRMS ins umen s. Fu he mo e, conce n-
ing o HRMS, wo acquisi ion modes (Disco e y and Con i ma ion) we e
expe imen ed wi h in o de o s ike a balance be ween he numbe o
compounds de ec ed and he sensi i i y and sui abili y equi ed o
suspec sc eening analysis. The analy ical me hods we e used o iden i y
AMs in se e al ege able ma ices (le uce, onion, oma o, and ca o )
and o in es iga e he accumula ion and deg ada ion o AMs in le uce
samples cul i a ed in pollu ed soils.
2. Expe imen al p ocedu e
2.1. Reagen s and ma e ials
The physicochemical p ope ies o he a ge AMs and su oga e
s anda ds and hei dis ibu o s a e ga he ed in Table S1. Indi idual
solu ions o all o hem we e mon hly p epa ed a 1000–3000 mg⋅kg
−1
in UHPLC-g ade me hanol (MeOH, 99.9 %, Scha lau, Sen mena , Ca a-
lonia, Spain), UHPLC-g ade ace oni ile (ACN, 99.9 %, A an o Pe o -
mance Ma e ials, Gliwice, Silesia, Poland) o dime hyl sul oxide (DMSO,
Pan eac AppliChem, Da ms ad , Ge many) (see Table S1). The indi-
idual s ock solu ions o FQs we e p epa ed adding d ops o NaOH (2 M)
(99 %, Me ck, Da ms ad , Hesse, Ge many) as epo ed in a p e ious
wo k (Ve ga a-Luis e al., 2023). Fu he combined dilu ions we e
weekly p epa ed in ACN a 100 mg⋅kg
−1
and 5 mg⋅kg
−1
o sample
spiking. Mos concen a ed solu ions (1000–3000 mg⋅kg
−1
and 100
mg⋅kg
−1
) we e s o ed a −20 ◦C, while 5 mg⋅kg
−1
ones we e kep a 4 ◦C,
using silanised ambe ials (Bu henne e al., 1999).
The sample ex ac ion p ocedu e employed he ollowing sal s: NaCl
(100 %) acqui ed om PanReac AppliChem (Cas ella del Vall´
es, Ca a-
lonia, Spain), anhyd ous ci ic acid H
3
Ci (99.5 %) and anhyd ous
Na
2
HPO
4
(98 %) om Scha lau and anhyd ous Na
2
SO
4
(99 %) om
Me ck. UHPLC-g ade ACN was se as he ex ac ion sol en .
The dispe si e clean-up s ep (dSPE) me hod o ege able samples
in ol ed he use o P ima y Seconda y Amine (PSA), Bondesil-C
18
(40
µm, Agilen Technologies, San a Cla a, CA, EEUU) and G aphi ised
Ca bon Black (GCB) (37–125 µm, Supe clean ENVI-Ca b, Me ck) so -
ben s; whe eas Oasis HLB ca idges (200 mg, 6 cc, 30
μ
m) pu chased
om Wa e s (Mil o d, Massachuse s, USA) we e employed o he
clean-up o soil samples. A ci a e bu e consis ing o an aqueous so-
lu ion o anhyd ous NaH
2
Ci (99 %) and Na
2
HCi ⋅1.5H
2
O (99 %)
(Honeywell Fluka, Cha lo e, No h Ca olina, USA) was also used in soil
analysis. Oxalic acid (100 %, Me ck) was used in he inal ex ac
econs i u ion.
Du ing he sample ea men p ocedu e, a Mul i Reax shake by
Heidolph (Schwabach, Ba a ia, Ge many) and a 5840R cen i uge by
Eppendo (San Sebas i´
an de Los Reyes, Mad id, Spain) we e used.
2.2. Sample ea men p ocedu e: Ex ac ion and clean-up
This s udy ex ends he p e iously de eloped me hod by he esea ch
g oup (Ve ga a-Luis e al., 2023) o simul aneously de e mine wen y-
se en AMs in ege ables (le uce, onion, oma o, and ca o ). B ie ly,
esh, c ushed and homogenised ege able (le uce, onion, oma o o
ca o ) samples (10 g) we e spiked wi h 200 µL o a 5 mg⋅kg
−1
s ock
solu ion, o exed (2000 cycles⋅min
−1
, 10 min) and kep in he da kness
o 30 min a oom empe a u e. ACN (10 mL), a ce amic homogenise
and he sal s (4 g anhyd ous Na
2
SO
4
, 1 g NaCl, 0.5 g anhyd ous H
3
Ci
and 0.049 g anhyd ous Na
2
HPO
4
) we e added. The mix u e was hen
shaken manually and degasi ied by opening he cen i uge ube un il no
gas was eleased. All samples we e o exed (2000 cycles⋅min
−1
, 8 min)
and cen i uged (4000 cycles⋅min
−1
, 5 min) a 10–15 ◦C.
The p esence o co-elu ing elemen s in he ma ix can lead o in-
e e ences du ing analysis, making i essen ial o ca y ou a clean-up
p ocedu e. Fo ha pu pose, dSPE app oach was employed and basing
on he li e a u e and p e ious expe ience o he esea ch g oup (He
e al., 2018; Ve ga a-Luis e al., 2023), wo so ben combina ions we e
e alua ed: PSA (10 mg) and C
18
(25 mg), wi h o wi hou GCB (2.5 mg)
addi ion, oge he wi h 150 mg anhyd ous Na
2
SO
4
in all he cases.
Unde op imal condi ions, an aliquo o 1 mL o he ex ac an was
ans e ed o a 50 mL cen i uge ube con aining 10 mg PSA, 25 mg C
18
and 150 mg Na
2
SO
4
. The mix u e was hen o exed (2000
cycles⋅min
−1
, 1 min) and cen i uged (4000 cycles⋅min
−1
, 5 min) a
10–15 ◦C. Aliquo s o 500 µL we e econs i u ed in 1 mL o 1:1 ( / )
ACN:oxalic acid (aq., 0.01 mol⋅L
-1
, pH 2) and il e ed h ough 0.22 µm
polyp opylene il e s (Cla i y-PP, Phenomenex, To ance, Cali o nia,
USA) be o e UHPLC-MS/MS and UHPLC-HRMS analysis.
2.3. Analysis
2.3.1. UHPLC-MS/MS
An Agilen 1290 In ini y II UHPLC coupled o an Agilen 6430 T iple
Quad andem mass-spec ome e (QqQ) (Agilen Technologies) was
used o mul i a ge analysis. P e ious esea ch conduc ed by ou g oup
has al eady documen ed he mobile phase, ch oma og aphic column,
and a ious de ec o pa ame e s used (Ve ga a-Luis e al., 2023; Ve -
ga a-Luise al., 2023) which a e indica ed in he sec ion 1. o he sup-
plemen a y ma e ial.
2.3.2. UHPLC-HRMS
A The mo Scien i ic Dionex Ul iMa e 3000 UHPLC coupled o a
The mo Scien i ic Q Exac i e Focus quad upole-O bi ap mass spec-
ome e (UHPLC-q-O bi ap) equipped wi h a hea ed ESI sou ce (HESI,
The mo-Fishe Scien i ic, CA, USA) was used o pe o m a suspec
analysis o mo e han 22,278 suspec s. To achie e his, we e alua ed he
ch oma og aphic sepa a ion esolu ion and sensi i i y using wo col-
umns (i.e., C
18
and C
18
pola columns) and h ee o ganic mobile phases
(i.e., ACN wi h 0.1 % o HCOOH, MeOH wi h 0.1 % o HCOOH and
MeOH wi h 0.1 % o oxalic acid). The sensi i i y o he de ec ion was
es ed by he op imisa ion o HESI condi ions: (i) capilla y empe a u e
I. Ve ga a-Luis e al.
Food Chemis y 444 (2024) 138643
3
(200 ◦C, 320 ◦C and 400 ◦C) and, (ii) sp ay ol age (2.5 kV, 3.2 kV and 4
kV). Mo eo e , ull scan – da a dependen MS2 (Full MS-ddMS2) dis-
co e y and con i ma ion acquisi ion modes we e e alua ed o u he
AM iden i ica ion.
Acco ding o he esul s in sec ion 3.1, he ACE Ul aCo e XB-C
18
ch oma og aphic column (2.1 mm x 150 mm, 1.7 µm) wi h a p e- il e
(2.1 mm ID, 0.2 µm) om Phenomenex was se as op imal o he
analysis and Milli-Q wa e (0.1 % HCOOH) (A) and MeOH (0.1 %
HCOOH) (B) we e used as mobile phases. The HESI sou ce pa ame e s
we e se o 4 kV o he sp ay ol age and 400 ◦C o he capilla y
empe a u e. Me hod de ails and pa ame e s a e included in sec ion 2.
o he supplemen a y ma e ial.
2.4. Me hod alida ion
In his wo k, UHPLC-MS/MS and UHPLC-HRMS analysis echniques
we e e alua ed in e ms o linea i y, p ecision (ins umen al and p o-
cedu al epea abili y), ins umen al (LOQ
INS
) and p ocedu al (LOQ
PROC
)
limi s o quan i ica ion and ma ix e ec a he de ec ion.
LOQ
INS
we e calcula ed using a hi een-poin ex e nal calib a ion
cu e (0.25–200
μ
g⋅kg
−1
), as he lowes ex e nal calib a ion poin wi h
RSD % and a sys ema ic e o in ela ion o he heo e ical alue below
30 %. Fo ha aim, he calib a ion cu e poin s be ween 0.25 and 25
μ
g⋅kg
−1
concen a ion le els we e measu ed in iplica e. This da a was
also employed o es ima e ins umen al epea abili y (in he same day)
and in e media e epea abili y (in di e en days). Linea i y anges we e
de ined conside ing he de e mina ion coe icien s (
2
) o he calib a ion
cu es buil be ween LOQ
INS
and uppe limi . The la e was es ablished
as he highes concen a ion possible ha p o ided
2
alue closes o
one, a oiding quad a ic i ing o he calib a ion cu es.
LOQ
PROC
alues, de ined as he lowes concen a ion ul illing he
c i e ia o an RSD % and a sys ema ic e o lowe han 30 %, we e ob-
ained by spiking ege able ex ac s ( inal ex ac s, a e been submi ed
o he whole p ocedu e) a di e en concen a ion le els (1, 2.5, 5, 7.5,
10, 15, 25 and 50
μ
g⋅kg
−1
) and analysing hem in iplica e.
Ma ix e ec a he de ec ion was de e mined using Eq. (1) as
desc ibed in p e ious wo ks (Ve ga a-Luis e al., 2023).
Ma ix e ec (%) = (A easample
A eaRe e ence
−1)x 100 (1)
T ueness was de e mined using iso opically labelled compounds and
ma ix-ma ched calib a ion app oach (using a six-poin calib a ion
cu e in he ange o 1–75
μ
g⋅kg
−1
p epa ed in each o he ou ege able
ma ices). The epea abili y o he p ocedu e was also e alua ed in e ms
o RSD %, using h ee eplica es o each eal sample p ocessed in he
same day.
Finally, o suspec analysis, ins umen al (LOI
INS
) and p ocedu al
(LOI
PROC
) limi s o iden i ica ion we e calcula ed using he same con-
cen a ion le els selec ed o he calcula ion o he LOQs. LOIs we e
es ima ed as he lowes concen a ion o which he expe imen al and
heo e ical MS2 spec a ma ch was equal o g ea e han 70 % in a leas
wo o he h ee eplica es o each concen a ion le el and he e en ion
ime di e ence was less han ±0.1 min.
2.5. Da a ea men
2.5.1. UHPLC-MS/MS
The aw iles ob ained om he UHPLC-MS/MS we e subjec ed o
da a p ocessing using he Agilen MassHun e Wo ks a ion so wa e
(Quan i a i e Analysis o QQQ, e sion 10.0) by Agilen . Analy e
iden i ica ion and quan i ica ion c i e ia we e es ablished acco ding o
guidelines de ailed in he Council Di ec i e 96/23/EC (Commission
Decision, 2002). The p esence o he compound was e ec i ely e i ied
by compa ing i s e en ion ime wi h a e e ence s anda d, alongside he
de ec ion o he wo mos dis inc i e ansi ions o each a ge
compound.
2.5.2. UHPLC-HRMS: Ta ge analysis and suspec sc eening
E en hough he UHPLC-HRMS used a non- a ge me hod o
acquisi ion, he da a was p ocessed using wo app oaches: a ge anal-
ysis and suspec sc eening. Fo he UHPLC-HRMS a ge analysis,
T aceFinde 5.1 (The mo-Fische Scien i ic) so wa e was used o
iden i y and quan i y he s udied AMs aking in o accoun he e en ion
ime, p ecu so and p oduc ions o he selec ed and p e iously known
AMs (Table S2). A 5 ppm a iabili y was conside ed accep able o
monoiso opic mass and agmen s, and a 70 % i o heo e ical iso opic
pa e n.
Fo he UHPLC-HRMS suspec sc eening, he Compound Disco e e
3.2 (The mo-Fishe Scien i ic) so wa e was used. Fo candida es’
iden i ica ion, i s ly ACN blanks we e used as e e ence o noise
elimina ion. As o he peak selec ion c i e ia, he “peak in ensi y”
il e ing c i e ion was se a 10,000 o conside only candida es wi h a
minimum peak a ea equal o o g ea e han 10,000. In addi ion, he
Lo en zian shape o he ch oma og aphic peaks and a mass e o o less
han 5 ppm we e es ablished as manda o y c i e ia o u he anno a-
ion. Di e en mass lis s, such as he COMPTOX lis , which includes AMs
and TPs, and hose gene a ed wi h he BioT ans o me 3.0 so wa e
we e used, o b oaden he sea ch o me abolic TPs o 22,278 com-
pounds in o al. Only he ea u es included in he mass lis s we e
conside ed and a spec al ma ch be ween he expe imen al and he
heo e ical MS1 o mo e han he 70 % was equi ed. Fu he mo e, peaks
wi h ch oma og aphic a eas h ee imes la ge han he blanks and wi h
a ela i e s anda d de ia ion (RSD %) lowe han 30 % wi hin he h ee
injec ion eplica es we e aken in o accoun . Mo eo e , due o he spe-
ci ic iso opic p o iles o molecules con aining O, N, Cl, B , S and/o F, he
analysis was limi ed o compounds con aining hese a oms. mzCloud
da abase (h ps://www.mzcloud.o g/) was used o MS2 compa ison
and a h eshold alue o 70 % o highe was conside ed o posi i e
iden i ica ion. When mass spec as we e no a ailable in he da abase
men ioned abo e, Mass F on ie Spec al In e p e a ion So wa e
(The mo-Fishe Scien i ic) was used o In-silico agmen a ion. Re en-
ion imes o he candida es mus ma ch hose p edic ed by he Re en-
ion Time Index (RTI) pla o m (h ps:// i.chem.uoa.g /) and he
candida es we e ejec ed o accep ed depending on whe he o no he e
was a s a is ical di e ence wi h he es ima ed alue wi hin he unce -
ain y o he buil model (only box 1 and box 2 candida es we e
conside ed). Fo compounds’ anno a ion he con idence le el es ab-
lished by Schymanski e al. (Schymanski e al., 2014) was used and, in
his wo k, only hose candida es anno a ed a a con idence le el 1
(candida e con i med by MS1, MS2 and e en ion ime), 2 (candida e
con i med by MS
2
lib a y ma ching (Le el 2a) o diagnos ic MS
2
in-silico
agmen a ion (Le el 2b) when no s anda d o expe imen al MS
2
da a-
base is a ailable) and 3 (candida e con i med by MS1 and in-silico MS2,
being all he candida es s uc u al isome s) we e epo ed.
Rega dless o he analysis mode, when i was equi ed, s a is ical
analysis was ca ied ou pe o ming an ANOVA es in Excel (2021
e sion).
2.6. Me hod applica ion: De ec ion o AM con amina ion and le uce
up ake expe imen
The alida ed me hod was applied o wo di e en case s udies:
i. The analysis o ou een le uces, h ee onions, h ee oma oes and
en ca o s om o ganic (se en een) and non-o ganic ( hi een)
ag icul u e, pu chased in local ma ke s in he Basque Coun y.
ii. The ans e o ou AMs ( ime hop im, sul ame hazine, en o-
loxacin, and chlo e acycline) om soils o le uces and he e al-
ua ion o hei deg ada ion in bo h ma ices. Fo his pu pose,
uni e sal soil subs a e and seeds o he comme cial le uce cul i a
’Ba a ia’ (Lac uca sa i a a . longi olia, Vilmo in seeds), one o he
I. Ve ga a-Luis e al.
Food Chemis y 444 (2024) 138643
4
mos widely used in some a eas o Eu ope (Ryde , s. .), we e used.
Six po s, each con aining 180 g o uni e sal subs a e, we e p epa ed
o each o he s udied AMs. Th ee seeds we e sown pe po , p e i-
ously s e ilised in o de o ensu e ha a leas one plan co e-
sponded o each po . All po s we e kep in a g ow h chambe
(Ibe cex®), unde con olled condi ions o 23 ◦C/18 ◦C (day/nigh )
empe a u e, 70/80 % (day/nigh ) ela i e humidi y, 450 µmol m
−2
s
−1
o pho osyn he ically ac i e adia ion (he eina e , PAR), and a
14-hou pho ope iod. Th oughou hei de elopmen , he plan s
we e i iga ed wi h dis illed wa e h ee imes a week. Two weeks
a e sowing, he numbe o plan s pe po was homogenised o one,
elimina ing he emaining plan s. Twen y- i e days a e sowing, he
s udied AMs we e applied o he soil subs a e, indi idually, a a
concen a ion o 1 mg⋅kg
−1
( h ee eplica es pe AM and h ee blanks
we e p epa ed). Soil and le uce samples we e aken in he i s and
hi d week a e spiking and we e analysed h ough UHPLC-MS/MS
and UHPLC-HRMS. Fo soil analysis he me hod desc ibed in a p e-
ious wo k was applied (Ve ga a-Luis e al., 2023).
3. Resul s and discussion
3.1. UHPLC-HRMS: Se ing up he ch oma og aphic condi ions
3.1.1. Ch oma og aphic column and mobile phase
The ch oma og aphic column (C
18
and C
18
pola columns) and mo-
bile phase (ACN wi h 0.1 % o HCOOH, MeOH wi h 0.1 % o HCOOH
and MeOH wi h 0.1 % o oxalic acid) we e op imised o ob ain he
maximum ch oma og aphic peak a ea and op imal ch oma og aphic
sepa a ion o he selec ed compounds. In a p e ious wo k (Ve ga a-Luis
e al., 2023), he p esence o oxalic acid has been epo ed o a oid
ch oma og aphic peak ailing o TCs, he e o e, i s use as a mobile phase
componen has been e alua ed in his wo k. The ch oma og aphic peak
a eas o he a ge AMs (excep miconazole wi h was la e included in
he me hod) injec ed in each selec ed column and mobile phase a e
shown in Figu e S1.
The lowes ch oma og aphic peak a eas we e ob ained wi h he
column C
18
pola and ACN wi h 0.1 % HCOOH as mobile phase.
Opposi e, he highes peak a eas we e obse ed o eigh AMs (see
Figu e S1) when he o ganic mobile phase pola i y was inc eased wi h
MeOH. This las mobile phase in C
18
column e ie ed be e esul s o
ele en compounds in compa ison o he pola column whils when
oxalic acid was included only ou AMs ( lumequine, oxy e acycline,
hiabendazole and sul ame hazine) showed a signi ican signal
imp o emen . Rega ding he e en ion imes, he sho e imes we e
eco ded when he C
18
pola column was used o bo h high and low
molecula weigh compounds. The e o e, conside ing he mo e sui able
esul s in e ms o ch oma og aphic sepa a ion, elu ion ime and peak
shape o he a ge AMs, he C
18
column was se as op imal, using MeOH
wi h 0.1 % HCOOH as mobile phase.
3.1.2. Sou ce pa ame e s
To op imise he sou ce pa ame e s, h ee di e en capilla y em-
pe a u es (200 ◦C, 320 ◦C, and 400 ◦C) we e s udied while main aining a
sp ay ol age o 3.2 kV o he a ge AMs, excep o miconazole. Once
he op imum capilla y empe a u e was selec ed, h ee di e en ol -
ages (2.5 kV, 3.2 kV and 4 kV) we e e alua ed.
As i can be obse ed in Figu e S2, he less in ensi e signals o he
a ge analy es we e ob ained when he capilla y empe a u e was se a
200 ◦C. S a is ically compa able esul s we e eco ded wi h 320 ◦C and
400 ◦C, howe e , a 320 ◦C SAs and FQs showed a lowe ionisa ion.
The e o e, he capilla y empe a u e was se a 400 ◦C.
Rega ding he sp ay ol age (Figu e S3), al hough no s a is ical
di e ences we e obse ed a he h ee es ed alues, he highes signal
in ensi ies o se en een o he wen y-six AMs we e ob ained a a
ol age o 4 kV. Thus, 4 kV was he selec ed as he op imal ol age.
3.2. E alua ion o he clean-up s ep
Fo he clean-up s ep, he addi ion o GCB (2.5 mg) o he dSPE
so ben s PSA (10 mg), C
18
(25 mg) and Na
2
SO
4
(150 mg) was e alua ed
in e ms o ma ix e ec a he de ec ion and eco e ies o he a ge
analy es. The analysis was pe o med using UHPLC-MS/MS. Acco ding
o he esul s (see Figu e S4), no s a is ical di e ences o eco e y
alues we e no iced when GCB was used (43–118 % wi h s 44–121 %
wi hou GCB). Howe e , GCB p esence led o a highe posi i e ma ix
e ec o some compounds, being especially no iceable o TCs. The e-
o e, he use o GCB was disca ded in u he assays and we p oceeded o
de e mine he adequacy o ins umen al se -up.
3.3. UHPLC-MS/MS s UHPLC-HRMS: E alua ion o he sui abili y o
mul i a ge analysis
In o de o assess hei sui abili y o mul i a ge analysis, he line-
a i y, LOQ
INS
, and de ec ion ma ix e ec o UHPLC-MS/MS and
UHPLC-HRMS we e compa ed. Bo h me hodologies we e e alua ed
using he op imum ex ac ion and clean-up p o ocol ou lined in sec ion
2.3.
3.3.1. Linea i y, epea abili y and LOQ
INS
Bo h analysis echniques, UHPLC-MS/MS and UHPLC-HRMS,
showed adequa e linea i y o e 0.25–200
μ
g⋅kg
−1
concen a ion ange
wi h de e mina ion coe icien s (
2
) highe han 0.97 (see Table S3).
Rega ding ins umen al and in e media e epea abili y, o e all,
mo e epea able alues we e ge using UHPLC-MS/MS. Conc e ely, RSD
<20 % we e ob ained using UHPLC-MS/MS o all compounds a low
concen a ion le els, excep o e y h omycin (RSD alues o 30–47 % in
he concen a ion ange o 0.25–1.00
μ
g⋅kg
−1
). In he case o UHPLC-
HRMS analysis, hose low RSD alues (i.e., <20 %) we e only ob-
ained o some FQs and TCs a 1.00
μ
g⋅kg
−1
and abo e concen a ions.
Rega ding LOQ
INS
alues, using UHPLC-MS/MS p o ided LOQ
INS
in
he ange o 0.2–1.4 µg⋅kg
−1
, simila o he ones epo ed by Tadi´
c e al.
(Tadi´
c e al., 2019) by UHPLC-MS/MS analysis (0.4–1.7 µg⋅kg
−1
) and
lowe o hose ge by UHPLC-HRMS in his wo k, especially in he case o
e acyclines and mac olides, i espec i e o he use o disco e y
(LOQ
INS
0.2–20 µg⋅kg
−1
) o con i ma ion (LOQ
INS
be ween 0.2 and 59
µg⋅kg
−1
) acquisi ion modes. Fo example, e y h omycin could no be
quan i ied due o he lack o MS2 when he analysis was pe o med using
UHPLC-HRMS a he disco e y acquisi ion mode; and he e o e, no
LOQ
INS
could be gi en o his compound unde hose condi ions. Thus,
when a ge analysis o AMs a low concen a ion le els is equi ed,
UHPLC-MS/MS using he DMRM acquisi ion mode p o ides be e e-
sul s in compa ison o UHPLC-HRMS.
3.3.2. Ma ix e ec a he de ec ion
Rega dless o he ins umen al echnique used, ma ix componen s
can dis up he ionisa ion o a ge AMs a he ionisa ion sou ce, leading
o dec eased sensi i i y and ep oducibili y, mani es ed as signal sup-
p ession o enhancemen (Van De S eene & Lambe , 2008; Zhou e al.,
2017). The de ec ion ma ix e ec alues ob ained o all he e alua ed
ma ices a e depic ed in Fig. 1 (UHPLC-MS/MS), Fig. 2a and 2b
(UHPLC-HRMS in con i ma ion (a) and disco e y (b) modes).
In he case o UHPLC-MS/MS, signal enhancemen was mainly
obse ed o onion, ca o and oma o ma ices, wi h his e ec being
pa icula ly no iceable o some TCs and SAs, which showed a ME % >
30 % (i.e., sul ame hazine 42–96 %, oxy e acycline 61–110 % and
doxycycline 89–147 % among o he s). In he case o le uce, a nega i e
ma ix e ec p edomina es, wi h o loxacin (-41 %) and oxi h omycin
(-46 %) s anding ou . These esul s a e in ag eemen wi h hose epo ed
in he li e a u e o FQs, SAs and ime hop im analysis in le uce, as
hey obse ed a ma ix e ec o less han 15 % wi h he excep ion o
o loxacin, and also en o loxacin in his wo k (Tadi´
c e al., 2019). The
esul s ob ained by He e al. in cabbage a e simila o he ones epo ed
I. Ve ga a-Luis e al.
Food Chemis y 444 (2024) 138643
5
in his wo k o FQs, cla i h omycin and oxi h omycin in le uce (He
e al., 2018).
A simila phenomenon was obse ed in UHPLC-HRMS con i ma ion
mode, whe e a posi i e ma ix e ec was obse ed o compounds
de ec ed in all ma ices, excep le uce, bu being less p onounced han
using UHPLC-MS/MS, as ME % <30 % we e ob ained wi h he excep-
ion o chlo e acycline (136 %) and enoxacin (39 %) in onion. How-
e e , in his acquisi ion mode a signi ican nega i e ma ix e ec o
e y h omycin ((−61)-(−64) %) was obse ed in all he ege able
ma ices. As o he UHPLC-HRMS disco e y mode, an o e all posi i e
ma ix e ec was obse ed o a ge s de ec ed in all ege able ma ices,
bu less p onounced han using UHPLC-MS/MS (ME % <30 % excep o
le uce (3–61 %)). As excep ion, a signal supp ession was obse ed o
e y h omycin ((-4)-(−49) %) and mycophenolic acid ((-8)-(−52) %).
Al hough UHPLC-HRMS p o ides a highe selec i i y in he analysis,
showing a lowe ma ix e ec in he de ec ion compa ed o UHPLC-MS/
MS, he signi ican ly lowe LOQ
INS
ob ained using he la e echnique
demons a ed i s highe sensi i i y o quan i y AMs a lowe concen-
a ion le els. Mo eo e , he ma ix e ec associa ed wi h UHPLC-MS/
MS can be co ec ed by di e en s a egies, such as he use o
Fig. 1. Ma ix e ec a he de ec ion by UHPLC-MS/MS analysis o he a ge AMs in each o he s udied ma ices (n =3, unce ain y exp essed as RSD %).
Fig. 2. Ma ix e ec a he de ec ion by UHPLC-HRMS analysis o he a ge AMs in each o he s udied ma ices: in con i ma ion (a) and disco e y (b) modes (n =3,
unce ain y exp essed as RSD %).
I. Ve ga a-Luis e al.
Food Chemis y 444 (2024) 138643
6
deu e a ed analogues as su oga es, which has been applied in his wo k
(see sec ion 3.3.1). These esul s a e in line wi h hose epo ed in he
li e a u e. Fo ins ance, he compa ison o UHPLC-QqQ and UHPLC-
qTOF o de e mine e e ina y AMs in animal issues ca ied ou by
Anumol e al. showed ha QqQ p o ides low limi s o quan i ica ion as
long as he de ec ion ma ix e ec was no oo p onounced whe eas q-
TOF was a p ime s a egy o b oaden he analy ical co e age by he
moni o ing un a ge ed AMs (Anumol e al., 2017). In ano he s udy,
Giusepponi e al. compa ed MS/MS and HRMS pe o mance cha ac e -
is ics o he analysis o six y AMs in bo ine muscle and milk. A highe
selec i i y was a ibu ed o HRMS in compa ison o MS/MS, howe e ,
he use o HRMS a ec ed he sensi i i y o he me hod o milk analysis
due o he massi e p esence o in e e ing subs ances (Giusepponi e al.,
2019). The e o e, conside ing all o he abo e, i was decided o use
UHPLC-MS/MS o mul i a ge analysis and a ge compounds quan i-
ica ion and UHPLC-HRMS o la e ex end he me hod o he iden i i-
ca ion o unknown compounds o TPs.
3.4. Mul i a ge d-SPE-UHPLC-MS/MS analysis
3.4.1. T ueness and p ecision
Vege able ma ices (i.e., le uce, onion, ca o and oma o) we e
spiked a 5 µg⋅kg
−1
, 25 µg⋅kg
−1
and 50 µg⋅kg
−1
wi h all he AMs p io o
he sample ea men explained in sec ion 1.2. The absolu e eco e ies
o he mul i a ge me hod in each o he s udied ege ables a e ga he ed
in Figu e S5. T ueness was e i ied by he de e mina ion o analy es’
appa en eco e ies using bo h, su oga e co ec ion and ma ix-
ma ched calib a ion app oaches, whe eas epea abili y was calcula ed
in e ms o RSD (%) (see he esul s in Tables S4-S7).
Absolu e eco e ies anged om 45 o 187 %, 41–212 %, 52–187 %
and 30–129 % in compounds analysed in le uce, onion, oma o, and
ca o , espec i ely. T ueness de e mined by su oga e co ec ion in he
ange o 35–178 % o a ge s measu ed in le uce, 34–188 % in onion,
40–141 % in ca o and 39–169 % in oma o. Compa ed o he alues
epo ed by Tadi´
c e al. a concen a ion le els o 10 µg⋅kg
−1
and 100
µg⋅kg
−1
, in his wo k accu a e esul s we e ob ained o sul adiazine and
sul a hiazole a low concen a ion in le uce, as well as o en o loxacin
a low and high le els in le uce and oma o. Howe e , hey e ie ed
be e esul s o ime hop im in le uce and sul adiazine in oma o
(Tadi´
c e al., 2019). Simila esul s we e also epo ed by He e al. in
cabbage a 5 µg⋅kg
−1
and 50 µg⋅kg
−1
o SAs, TCs, FQs and MCs (He
e al., 2018).
Using ma ix-ma ched calib a ion app oach mo e u h ul esul s
we e ob ained acco ding o he Commission Implemen ing Regula ion
(EU) 2021/808 o 22 Ma ch 2021 (i.e., 70–120 % o 1–10
μ
g⋅kg
−1
concen a ions, 80–120 % o concen a ions >10
μ
g⋅kg
−1
and p eci-
sion, exp essed as RSD %, ≤30 %) (Commission Implemen ing Regu-
la ion (EU) 2021/808 o 22 Ma ch 2021 on he Pe o mance o
Analy ical Me hods o Residues o Pha macologically Ac i e Subs ances
Used in Food-P oducing Animals and on he In e p e a ion o Resul s as
Well as on he Me hods o Be Used o Sampling and Repealing Decisions
2002/657/EC and 98/179/EC (Tex wi h EEA Rele ance), 2021). These
eco e ies anged om 67 o 134 % o a ge s de e mined in le uce
(wi h he excep ion o sul a hiazole a 5
μ
g⋅kg
−1
), 83–121 % in onion
(wi h he excep ion o e y h omycin a 5
μ
g⋅kg
−1
), 84–132 % in ca o
(wi h he excep ion o e y h omycin a 5
μ
g⋅kg
−1
), and 75–136 % in
oma o.
Chuang e al. (Chuang e al., 2015) epo ed appa en eco e y
alues calcula ed wi h ma ix-ma ched calib a ion o sul adiazine,
sul ame hoxazole, oxy e acycline and ime hop im in le uce a 200
μ
g⋅kg
−1
concen a ion le el (74 %, 74 %, 72 % and 82 %, espec i ely)
which a e in conco dance wi h he ones de e mined in his wo k a he
highes alida ion le el, 50
μ
g⋅kg
−1
(86 %, 92 %, 91 % and 101 % o
sul adiazine, sul ame hoxazole, oxy e acycline and ime hop im,
espec i ely).
Rega dless o he s a egy used o he calcula ion o appa en
Table 1
Compounds anno a ed a he ege ables samples a le els 1–5 using suspec sc eening by UHPLC-HRMS.
Fea u e Accu a e
mass
Exac mass e o
(ppm)
Fo mula Name RT (min) Con idence
le el
O ganic Vege ables Non-O ganic Vege ables
Le uce Onion Ca o Toma o Le uce Onion Ca o Toma o
1 280.16719 −0.96 C
16
H
24
O
4
B e eldin A 12.724 2b ✓
2 475.30643 −3.03 C
21
H
41
N
5
O
7
Ne ilmicin 7.741 2a ✓ ✓
3 145.05278 0.10 C
9
H
7
NO Indole-4-ca boxaldehyde 7.071 2a ✓ ✓ ✓ ✓ ✓ ✓
4 150.10445 −0.11 C
10
H
14
O Ca one 13.177 2a ✓ ✓ ✓ ✓ ✓ ✓
5 306.11003 −0.99 C
16
H
18
O
6
TP o mycophenolic acid 13.793 2b ✓ ✓
I. Ve ga a-Luis e al.
Food Chemis y 444 (2024) 138643
7
eco e ies, RSD alues lowe han 30 % we e ob ained o all o he
analy es and ma ices, indica ing good p ecision and in compliance wi h
he egula ion.
3.4.2. LOQ
PROC
LOQ
PROC
ob ained in his wo k a e ga he ed in Table S8. All in all,
his wo k o e ed be e LOQ
PROC
wi h a ange be ween 0.1 and 2.8
µg⋅kg
−1
compa ed o hose ob ained by He e al, (2.0–5.0 µg⋅kg
−1
) (He
e al., 2018) and Yu e al. (1.1–5.8 µg⋅kg
−1
) (Yu e al., 2018).
3.5. Suspec sc eening using d-SPE-UHPLC-qO bi ap
Disco e y acquisi ion mode was p e e ably selec ed o he iden i i-
ca ion o unknown compounds, as i is di icul o ga he he necessa y
in o ma ion om unknown compounds o pe o m an analysis in
con i ma ion acquisi ion mode (i.e. molecula o mula, exac mass and
e en ion ime). LOI
INS
and LOI
PROC
we e calcula ed in disco e y mode
o he wen y-se en AMs and a e included in Table S9.
LOI
INS
anged om 0.2
μ
g⋅kg
−1
o 8.9
μ
g⋅kg
−1
wi h he excep ion o
doxycycline, chlo e acycline and azi h omycin o which alues o 18
μ
g⋅kg
−1
, 36
μ
g⋅kg
−1
and 38
μ
g⋅kg
−1
, espec i ely, we e ob ained. In he
p esence o ma ix, he lowes LOI
PROC
we e obse ed o oma o (i.e.,
1.6–13.0
μ
g⋅kg
−1
excep o chlo e acycline and azi h omycin), o
which he lowes LOQ
PROC
we e also calcula ed o some TCs, MCs and
FQs, especially compa ed o le uce, whe e he highes LOI
INS
we e
es ima ed. This may be due o he chlo ophyll con en o le uce (Cos-
ache, M. A. e al., 2012), which could in e e e wi h he sample ea -
men p ocess leading o a loss o sensi i i y. Compa ing he AM amilies,
he lowes LOIs, bo h LOI
INS
and LOI
PROC
, we e es ima ed o SAs ol-
lowed by FQs, while MCs and TCs showed he highes alues. Nei he
LOI
INS
no LOI
PROC
could be calcula ed o e y h omycin due o he lack
o agmen a ion spec a. E y h omycin was unsuccess ully ionised in
he UHPLC-HRMS in acco dance wi h he poo esul s ob ained also o
he s anda ds and he e o e, when acqui ed in ddMS2-disco e y mode,
could no be selec ed o u he agmen a ion in he O bi ap.
Compa ing wi h he li e a u e, Gonz´
alez-Gaya e al. ob ained LOI
INS
anging om 2.4 o 263
μ
g⋅kg
−1
(Gonz´
alez-Gaya e al., 2021) o
di e en an ibio ics in wa e using he qO bi ap ins umen , being
highe han hose ob ained in his wo k. Howe e , no wo k has been
ound epo ing LOI
PROC
in ege ables o he han hose alida ed he e,
al hough as seen in his wo k o he di e en ypes o ege ables ana-
lysed, LOI
PROC
can a y g ea ly be ween ma ices. These esul s u he
inc ease he need o mo e wo k applying suspec sc eening analysis o a
a ie y o en i onmen al samples.
4. APPLICATION TO REAL SAMPLES
The alida ed me hods o de e mine AMs and TPs we e used in wo
case s udies in o de o e alua e hei applicabili y in en i onmen al
conce ns: (i) occu ence and deg ada ion o AMs in ege able samples,
and (ii) up ake and deg ada ion o AMs in ege able samples cul i a ed
in con amina ed soils.
4.1. E alua ion o AMs and hei TPs p esence in ege ables om he
Basque Coun y
We conduc ed a ho ough analysis o hi y samples o le uce, onion,
ca o , and oma o ob ained om local ma ke s in he Basque Coun y.
The aim was o quan i y AMs using UHPLC-MS/MS and o sc een o any
po en ial suspec using UHPLC-HRMS. The a ge analysis in ca o s
om non-o ganic and o ganic ag icul u e by UHPLC-MS/MS allowed
he de ec ion o wo AMs, sul adiazine and mycophenolic acid, below
he LOQ
PROC
.
Rega ding suspec sc eening, by applying he il e s p e iously
men ioned in sec ion 2.5.2, he numbe o candida es was educed om
he ini ial 13,798 o 77. Following his, manual peak picking was ca ied
ou , esul ing in 35 ea u es ha me he cons ain s o peak shape.
A e ca e ul conside a ion o mul iple quali y c i e ia, including a MS2
ma ch o o e 70 % based on he mzCloud da abase, in-silico agmen-
a ion by Mass F on ie , and/o en ies in ChemSpide , as well as a
ma ching e en ion ime p edic ed by he RTI model, we anno a ed 5 key
ea u es. Table 1 includes he ea u es anno a ed a le els 1–5 acco ding
o he Schymanski scale. Ne ilmicin (semisyn he ic aminoglycoside
an ibio ic) was iden i ied a le el 2a and B e eldin A (an i i al d ug) a
le el 2b. Ne ilmicin was ound in h ee ca o s, wo om o ganic ma -
ke s and one om non-o ganic ma ke s, while B e eldin A was ound in
esh oma oes bough in non-o ganic ma ke s. The p esence o bac e ia
esis an o ne ilmicin in ege ables ha e been epo ed (Schwaige
e al., 2011), which migh indica e he p e ious p esence o his com-
pound in hose ege able ma ices o in hei nea en i onmen , such as
sedimen s o compos used as e ilise s (Quaik e al., 2020). Indole-4-
ca boxaldehyde (used o pha maceu ical syn hesis), iden i ied a le el
2a, was ound in le uces and ca o s om o ganic and non-o ganic
ma ke s as well as in wo oma oes and one onion om non-o ganic
ma ke s. Simila ly, he p esence o he an i ungal agen ca one, iden-
i ied a le el 2a, was de ec ed in le uces and ca o s om o ganic and
non-o ganic ma ke s as well as in wo oma oes and one onion om non-
o ganic ma ke s. Ca one has been epo ed o be used agains oma o
oo -kno caused by Meloidogyne incogni a, which could explained he
p esence o his compound in such ma ix (Elsha kawy e al., 2022).
Fig. 3. Chemical s uc u es o mycophenolic acid and a deme hyla ed TP o mycophenolic acid.
I. Ve ga a-Luis e al.
Food Chemis y 444 (2024) 138643
8
Finally, a TP de i ed om he deme hyla ion o mycophenolic acid
(see Fig. 3), (4E)-6-(4,6-dihyd oxy-7-me hyl-3-oxo-1,3-dihyd o-2-
benzo u an-5-yl)-4-me hylhex-4-enoic acid, was en a i ely iden i ied a
le el 2b and ound in ca o s om o ganic and non-o ganic ag icul u e,
consis en wi h he p esence o i s p edecesso mycophenolic acid in he
same samples. The e is limi ed in o ma ion on he occu ence o
mycophenolic acid in ege ables. Howe e , i has been epo ed o be
syn he ised by Byssochlamys ni ea, which is a ungi species esponsible
o he spoilage o ege ables and ui s. Mo eo e , as i can su i e hea
ea men s used o ood p ocessing and can g ow du ing s o age a
oom empe a u e (Puel e al., 2005), i is no unusual o de ec i s
p esence e en in he ege ables. See he ch oma og ams o some o he
anno a ed compounds in Figu es S6-S8.
4.2. S udy o he ans e o ams o le uces g own in con amina ed soils
The quan i ied concen a ions o each AM in bo h ma ices, soil and
he co esponding le uce, can be seen in Table 2.
The esul s indica e a signi ican decline in he p esence o AMs in he
soil du ing he ini ial week a e spiking. No ably, sul ame hazine
expe ienced an imp essi e deg ada ion a e o 95 %, closely ollowed by
ime hop im which showed a subs an ial deg ada ion o 88 % wi hin
he same ime ame. O e ime, he deg ada ion con inued, esul ing in
signi ican ly lowe de ec ion o AMs in he soils du ing he hi d week.
None heless, he e was an obse ed ans e o AMs om he soil o he
le uce. Simila o he soil, a highe ini ial concen a ion was de ec ed
compa ed o he concen a ion obse ed in he hi d week. This sugges s
ha deg ada ion o AMs also occu s in he plan .
The beha iou o each AM in soil and i s mobili y owa ds he plan
can be be e unde s ood conside ing he hal -li e and adso p ion coe -
icien (K
oc
) o AMs in soil ( alues ob ained om he En i onmen al
P o ec ion Agency (EPA) (US EPA, 2023) a e ga he ed in Table 2). Fo
example, conside ing he deg ada ion ime o chlo e acycline, i makes
sense ha i is he compound de ec ed in he highes concen a ion in
soil and, conside ing i s K
oc
, i is he one mos ans e ed o le uce. As
o sul ame hazine, i s K
oc
and deg ada ion ime also ag ee wi h he
esul s, as i is he compound de ec ed a he lowes concen a ion in he
soil and i s ans e o le uce is minimal.
The samples we e also analysed using suspec sc eening app oach
and a TP o en o loxacin, 1-cyclop opyl-7-{[(1Z)-2-(e hylamino)
e henyl]amino}-6- luo o-4-oxoquinoline-3-ca boxylic acid, was de ec-
ed in he exposed le uce samples. The chemical s uc u e o he TP, he
ob ained MS2 and p oposed s uc u es o he agmen ions a e ga h-
e ed in Fig. 4. Sul ame hazine was also iden i ied a 2a le el in he
spiked soils and he le uces g own in he e.
5. Conclusions
This wo k success ully alida ed an analy ical me hod capable o
accu a ely and simul aneously de e mining a wide ange o AMs a ace
concen a ion le els in se e al ege able samples by mul i a ge anal-
ysis. In addi ion, i has also demons a ed i s po en ial o iden i y un-
known AMs and hei TPs in ege able ma ices. The lowe LOQ
INS
ob ained and he possibili y o con olling he ma ix e ec a he
Table 2
Exposi ion expe imen concen a ions in soils and he co esponding le uces, oge he wi h he soil adso p ion coe icien (K
oc
) and hal -li e imes o each o he
s udied an imic obials.
a
Soil adso p ion coe .
(Koc)
L⋅kg
−1
a
Hal -li e
(days)
Soil (µg⋅kg
−1
) Le uce (µg⋅kg
−1
)
1s week 3 d week 1s week 3 d week
Sul ame hazine 143 3.4 50 ±3 8 ±1 0.3 ±0.1 0.2 ±0.1
En o loxacin 481 3.4 183 ±7 82 ±3 9.1 ±0.1 1.3 ±0.1
T ime hop im 115 4.2 124 ±9 43 ±3 2.9 ±0.4 0.4 ±0.1
Chlo e acycline 31 148 593 ±12 303 ±6 15 ±3 2.0 ±0.5
a
(US EPA, 2023).
Fig. 4. Chemical s uc u es o en o loxacin and he iden i ied TP and he s uc u es p oposed o he ion agmen s in he MS2 o he TP.
I. Ve ga a-Luis e al.
Food Chemis y 444 (2024) 138643
9
de ec ion wi h he use o su oga es, made us o p e e ably conside
UHPLC-MS/MS echnique o pe o m he mul i a ge analysis. Howe e ,
his wo k highligh s he impo ance o wo king wi h bo h low- and high-
esolu ion analysis echniques o pe o m a comp ehensi e AM
con amina ion moni o ing, as he use o UHPLC-HRMS has allowed o
de ec he o ma ion and p esence o speci ic TPs o AMs in ege ables,
which is no easible wi h he a ge analysis.
The me hods applied in se e al case s udies ha e yielded ema kable
esul s. On he one hand, he obse a ions ound in up ake expe imen s
showed ha AMs ans e ence occu om con amina ed soils o le uce.
Al hough his las conclusion, he occu ence o AMs s udied in ege-
able samples ga he ed in se e al comme cial poin s in he Basque
Coun y was negligible and only sul adiazine and mycophenolic acid
we e de ec ed in ca o s om non-o ganic and o ganic ag icul u e below
LOQ
PROC
. The e o e, ege ables do no seem o be an impo an sou ce
o AM con amina ion in he Basque Coun y. Howe e , he disco e y o
AMs’ by-p oduc s highligh s he need o u he esea ch o iden i y he
po en ial isks o hese compounds and o es ablish he necessa y eg-
ula ions o con ol hem. Al hough he e a e cu en ly some h esholds
o he p esence o AMs in en i onmen al samples, he e a e no egu-
la ions ega ding hei TPs. Thus, e o s should be ocused on he
in es iga ion and iden i ica ion o TPs due o he a ying en i onmen al
condi ions ha can lead o a ious eac ions and he o ma ion o
di e en p oduc s.
CRediT au ho ship con ibu ion s a emen
I. Ve ga a-Luis: In es iga ion, Da a cu a ion, Valida ion, Fo mal
analysis, Me hodology, W i ing – o iginal d a , Visualiza ion. M. Jin:
In es iga ion, Da a cu a ion. J.C. Baez-Mill´
an: In es iga ion, Valida-
ion. B. Gonz´
alez-Gaya: In es iga ion, Supe ision. I. Iju co: Sample
acquisi ion. M. Lacues a: Sample acquisi ion, Funding acquisi ion. M.
Oli a es: Me hodology, Concep ualiza ion, Fo mal analysis, Funding
acquisi ion, W i ing – e iew & edi ing. A. P ie o: Supe ision, Me h-
odology, Concep ualiza ion, Fo mal analysis, W i ing – e iew & edi -
ing, Funding acquisi ion, P ojec adminis a ion.
Decla a ion o compe ing in e es
The au ho s decla e ha hey ha e no known compe ing inancial
in e es s o pe sonal ela ionships ha could ha e appea ed o in luence
he wo k epo ed in his pape .
Da a a ailabili y
Da a will be made a ailable on eques .
Acknowledgmen s
Au ho s acknowledge inancial suppo om he Elka ek p ojec
en i led “Eme gencia y diseminaci´
on de esis encia a los an ibi´
o icos:
ínculos en e salud humana, ganade ía, alimen aci´
on y medioambien e
(Elka ek 20/88)”, he p ojec s “E aluaci´
on del iesgo de apa ici´
on y
diseminaci´
on de esis encias a an ibi´
o icos en p oduc os ege ales es-
cos y suelos de cul i o de la comunidad au ´
onoma del País Vasco (PA21/
05 and PA22/03)” inside he “Resea ch p ojec s a ge ed o ag icul u e
2020 p og am” o he Basque Go e nmen (Basque Coun y, Spain).
Au ho s also hanks he inancial suppo o he Uni e si y o he Basque
Coun y (UPV/EHU) om he collabo a i e p ojec COLAB 20/14
“Assessmen and p elimina y diagnosis o dissemina ion o an ibio ic
esis ance genes h ough he ood p oduc ion chain in he Basque
Coun y”; and he Basque Go e nmen h ough he inancial suppo as
consolida ed g oup o he Basque Resea ch Sys em (IT1446-22 and
IT1682-22). I. Ve ga a-Luis is g a e ul o he Uni e si y o he Basque
Coun y (UPV/EHU) o he p e-doc o al ellowship.
Appendix A. Supplemen a y da a
Supplemen a y da a o his a icle can be ound online a h ps://doi.
o g/10.1016/j. oodchem.2024.138643.
Re e ences
Anumol, T., Leho ay, S. J., S e ens, J., & Zweigenbaum, J. (2017). Compa ison o
e e ina y d ug esidue esul s in animal issues by ul ahigh-pe o mance liquid
ch oma og aphy coupled o iple quad upole o quad upole– ime-o - ligh andem
mass spec ome y a e di e en sample p epa a ion me hods, including use o a
comme cial lipid emo al p oduc . Analy ical and Bioanaly ical Chemis y, 409(10),
2639–2653. h ps://doi.o g/10.1007/s00216-017-0208-y
Ba on, L., Tobin, J., & Paull, B. (2008). Mul i- esidue de e mina ion o pha maceu icals
in sludge and sludge en iched soils using p essu ized liquid ex ac ion, solid phase
ex ac ion and liquid ch oma og aphy wi h andem mass spec ome y. Jou nal o
En i onmen al Moni o ing, 10(3), 353. h ps://doi.o g/10.1039/b717453e
Bu henne, J., Ludwig, M., & Spi elle , M. (1999). Pola pho odeg ada ion p oduc s o
quinolones de e mined by HPLC/MS/MS. Chemosphe e, 38(6), 1279–1286. h ps://
doi.o g/10.1016/S0045-6535(98)00525-6
Cas ellani, F., Ricci, M., Colag ande, M. N., Sco ichini, G., & Salu i, G. (2023).
De elopmen and Valida ion o a Con i ma o y Me hod o he De e mina ion o 12
Coccidios a Residues in Eggs and Muscle by Means o Liquid Ch oma og aphy
Coupled o Hyb id High Resolu ion Mass Spec ome y. Sepa a ions, 10(3), 202.
Scopus. h ps://doi.o g/10.3390/sepa a ions10030202.
Chuang, Y.-H., Zhang, Y., Zhang, W., Boyd, S. A., & Li, H. (2015). Compa ison o
accele a ed sol en ex ac ion and quick, easy, cheap, e ec i e, ugged and sa e
me hod o ex ac ion and de e mina ion o pha maceu icals in ege ables. Jou nal
o Ch oma og aphy A, 1404, 1–9. h ps://doi.o g/10.1016/j.ch oma.2015.05.022
2002/657/EC: Commission Decision o 12 Augus 2002 implemen ing Council Di ec i e
96/23/EC conce ning he pe o mance o analy ical me hods and he in e p e a ion
o esul s (Tex wi h EEA ele ance) (no i ied unde documen numbe C(2002)
3044), 221 OJ L (2002). h p://da a.eu opa.eu/eli/dec/2002/657/oj/eng.
Commission Implemen ing Regula ion (EU) 2021/808 o 22 Ma ch 2021 on he
pe o mance o analy ical me hods o esidues o pha macologically ac i e
subs ances used in ood-p oducing animals and on he in e p e a ion o esul s as
well as on he me hods o be used o sampling and epealing Decisions 2002/657/
EC and 98/179/EC (Tex wi h EEA ele ance), 180 OJ L (2021). h p://da a.eu opa.
eu/eli/ eg_impl/2021/808/oj/eng.
Cos ache, M. A., Campeanu, G., & Nea a, G. (2012). S udies conce ning he ex ac ion o
chlo ophyll and o al ca o enoids om ege ables. 8. 17(5), 7702-7708.
Cyco´
n, M., M ozik, A., & Pio owska-Sege , Z. (2019). An ibio ics in he Soil
En i onmen —Deg ada ion and Thei Impac on Mic obial Ac i i y and Di e si y.
F on ie s in Mic obiology, 10, 338. h ps://doi.o g/10.3389/ micb.2019.00338
da Sil a, J. J., da Sil a, B. F., S adio o, N. R., Pe o ic, M., Gago-Fe e o, P., & G os, M.
(2020). P essu ized Liquid Ex ac ion (PLE) and QuEChERS e alua ion o he
analysis o an ibio ics in ag icul u al soils. Me hodsX, 7, A icle 101171. h ps://doi.
o g/10.1016/j.mex.2020.101171
Dasenaki, M. E., & Thomaidis, N. S. (2015). Mul i- esidue de e mina ion o 115
e e ina y d ugs and pha maceu ical esidues in milk powde , bu e , ish issue and
eggs using liquid ch oma og aphy– andem mass spec ome y. Analy ica Chimica
Ac a, 880, 103–121. h ps://doi.o g/10.1016/j.aca.2015.04.013
Elsha kawy, M. M., Al-Aska , A. A., Behi y, S. I., Abdelkhalek, A., Saleem, M. H.,
Kam an, M., & De balah, A. (2022). Resis ance induc ion and nema icidal ac i i y o
ce ain mono e penes agains oma o oo -kno caused by Meloidogyne incogni a.
F on ie s in Plan Science, 13, A icle 982414. h ps://doi.o g/10.3389/
pls.2022.982414
Ga cía, J., Ga cía-Gal´
an, M. J., Day, J. W., Boopa hy, R., Whi e, J. R., Wallace, S., &
Hun e , R. G. (2020). A e iew o eme ging o ganic con aminan s (EOCs), an ibio ic
esis an bac e ia (ARB), and an ibio ic esis ance genes (ARGs) in he en i onmen :
Inc easing emo al wi h we lands and educing en i onmen al impac s. Bio esou ce
Technology, 307, A icle 123228. h ps://doi.o g/10.1016/j.bio ech.2020.123228
Giusepponi, D., Paole i, F., Ba ola, C., Mo e i, S., Salu i, G., Ianni, F., Sa della, R., &
Gala ini, R. (2019). T ans e o a mul iclass me hod o o e 60 an ibio ics in ood
om high esolu ion o low esolu ion mass spec ome y. Molecules, 24(16), 2935.
Scopus. h ps://doi.o g/10.3390/molecules24162935.
Gonz´
alez-Gaya, B., Lopez-He guedas, N., Bilbao, D., Mijangos, L., Ike , A. M.,
E xeba ia, N., I azola, M., P ie o, A., Oli a es, M., & Zuloaga, O. (2021). Suspec
and non- a ge sc eening: The las on ie in en i onmen al analysis. Analy ical
Me hods, 13(16), 1876–1904. h ps://doi.o g/10.1039/D1AY00111F
Hang, L., Zhao, Y., Liu, C., Yu, Y., He, Y., Xu, J., & Lu, Z. (2021). De e mine Mul iple
Classes o Ve e ina y An ibio ics in Soil: Compa ing Dispe si e and Solid-Phase
Ex ac ion o Sample Cleanup. Ch oma og aphia, 84(9), 833–844. h ps://doi.o g/
10.1007/s10337-021-04064-5
He, Z., Wang, Y., Xu, Y., & Liu, X. (2018). De e mina ion o An ibio ics in Vege ables
Using QuEChERS-Based Me hod and Liquid Ch oma og aphy-Quad upole Linea Ion
T ap Mass Spec ome y. Food Analy ical Me hods, 11(10), 2857–2864. h ps://doi.
o g/10.1007/s12161-018-1252-8
Jadeja, N. B., & Wo ich, A. (2022). F om gu o mud: Dissemina ion o an imic obial
esis ance be ween animal and ag icul u al niches. En i onmen al Mic obiology, 24
(8), 3290–3306. h ps://doi.o g/10.1111/1462-2920.15927
Jongedijk, E., Fi eik, M., A izabalaga-La a˜
naga, A., Polze , J., Blokland, M., & S e k, S.
(2023). Use o high- esolu ion mass spec ome y o e e ina y d ug mul i- esidue
I. Ve ga a-Luis e al.
