Talan a 273 (2024) 125816
A ailable online 1 Ma ch 2024
0039-9140/© 2024 The Au ho s. Published by Else ie B.V. 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/).
G een in an o mula analysis: Op imizing headspace solid-phase
mic oex ac ion o ca bonyl compounds associa ed wi h lipid pe oxida ion
using GC-MS and pen a luo ophenylhyd azine de i a iza ion
Jo ge A. Cus odio-Mendoza
a
,
b
,
**
, Ana Lopez Blanco
b
, Ana M. A es-Fuen es
c
, An onia M. Ca o
Díaz
b
,
d
,
e
,
*
a
Depa men o Technique and Food De elopmen , Ins i u e o Human Nu i ion Sciences, Wa saw Uni e si y o Li e Sciences (WULS-SGGW), Nowou synowska 159 c,
02-776, Wa szawa, Poland
b
Depa men o Analy ical Chemis y, Nu i ion and Food Science. Uni e si y o San iago de Compos ela, 15782, San iago de Compos ela, Spain
c
Cen e o Applied Chemis y and Bio echnology (CQAB), Uni e si y o Alcal´
a, 28805, Alcal´
a de Hena es, Spain
d
Heal h Resea ch Ins i u e o San iago de Compos ela (IDIS). Uni e si y o San iago de Compos ela, 15782, San iago de Compos ela, Spain
e
Ins i u o de Ma e iais (iMATUS). Uni e si y o San iago de Compos ela, 15782, San iago de Compos ela, Spain
ARTICLE INFO
Handling edi o : J.-M. Kau mann
Keywo ds:
GC-MS
Glyoxal
HS-SPME
In an o mula
Malondialdehyde
Me hylglyoxal
ABSTRACT
The e inemen and op imiza ion o a me hod combining headspace solid-phase mic oex ac ion (HS-SPME) wi h
gas ch oma og aphy-mass spec ome y (GC-MS) was success ully pe o med o he i s ime o de e mine se en
ca bonyl and dica bonyl compounds, including glyoxal, me hylglyoxal, dime hylglyoxal, and malondialdehyde
in in an o mulae, ela ed o lipid pe oxida ion. HS-SPME was u ilized o simul aneous ex ac ion and de i -
a iza ion wi h pen a luo ophenylhyd azine (PFPH). C i ical pa ame e s such as empe a u e, pH, ex ac i e
phase, and sal ing-ou we e me iculously in es iga ed and ine- uned by an asymme ical 2
2
3
2
//9 sc eening
design o ensu e he me hod’s e icacy and eliabili y. Op imal condi ions included a PFPH concen a ion o 5 g/
L, pH 5.0, head-space ex ac ion a 60 ◦C wi hin 10 min, u ilizing a DVB/CAR/PDMS coa ing, and a 20% w/w
sal ing-ou . The analy ical alida ion o his me hod, complian wi h FDA guidelines, demons a ed excep ional
linea i y, sensi i i y, speci ici y, p ecision (RSD ≤13.8%), and accu acy (84.8% ≤ eco e y ≤111.5%). The
me ic app oach AGREEp ep con i ms i s eco- iendliness, ma king a signi ican s ep owa ds an en i onmen ally
conscious app oach in in an o mula analysis. An occu ence s udy conduc ed on 25 in an o mula samples
e ealed widesp ead ca bonyl and dica bonyl compounds in bo h powde ed and liquid a ian s. ANOVA esul s
exhibi ed a ia ions in compound concen a ions among di e en sample g oups. Clus e ing analyses delinea ed
dis inc g oups based on ca bonyl con en , indica ing he po en ial o hese compounds as ma ke s o lipid
pe oxida ion and ood quali y assessmen . This me hod se es as a aluable ool o e alua ing in an o mula
quali y, s abili y owa ds oxida ion, and sa e y.
Abb e ia ions:
α
-Dica bonyl Compounds (
α
-DDC), Accep able Daily In ake (ADI); Ace aldehyde (ACE), Analysis o Va iance (ANOVA); Benzaldehyde (PhCHO),
Ca boxen (CAR); Dime hylglyoxal (DMGO), Di inylbenzene (DVB); Eu opean Food Sa e y Au ho i y (EFSA), Flame Ioniza ion De ec o (FID); Food Addi i es,
Fla ou ings; P ocessing Aids, and Ma e ials in Con ac wi h Food (AFC); Food and D ug Adminis a ion (FDA), Fo maldehyde (FCHO); Gas Ch oma og aphy (GC),
Gene ally Recognized as Sa e (GRAS); Glyoxal (GO), Head-Space (HS); High-Pe o mance Liquid Ch oma og aphy (HPLC), In an Fo mula (IF); In e na ional Agency
o Resea ch on Cance (IARC), Join FAO/WHO Expe Commi ee on Food Addi i es (JECFA); Limi o De ec ion (LOD), Liquid Follow-up IF (LFIF); Liquid S a e
IF (LSIF), Malondialdehyde (MDA); Mass Spec ome y (MS), Me hylglyoxal (MGO); Polyac yla e (PA), Polydime hylsiloxane (PDMS); Polyunsa u a ed Fa y Acids
(PUFA), Powde ed Follow-up IF (PFIF); Powde ed S a e IF (PSIF), Quali y Con ol Samples (QCs); Reac i e Oxygen Species (ROS), Rela i e S anda d De ia ion (%
RSD); Solid Phase Mic oex ac ion (SPME), Lowe Limi o Quan i ica ion (LLOQ); Th eshold o Toxicological Conce n (TTC), Ul a-high empe a u e p ocessing
(UHT); Uppe Limi o Quan i ica ion (ULOQ), Wo ld Heal h O ganiza ion (WHO).
* Co esponding au ho . Depa men o Analy ical Chemis y, Facul y o Chemis y, Uni e si y o San iago de Compos ela, 15782, San iago de Compos ela, Spain.
** Co esponding au ho . Depa men o Technique and Food De elopmen , Ins i u e o Human Nu i ion Sciences, Wa saw Uni e si y o Li e Sciences (WULS-
SGGW), Nowou synowska 159 c, 02-776, Wa szawa, Poland.
E-mail add esses: [email p o ec ed] (J.A. Cus odio-Mendoza), [email p o ec ed] (A.M. Ca o Díaz).
Con en s lis s a ailable a ScienceDi ec
Talan a
jou nal homepage: www.else ie .com/loca e/ alan a
h ps://doi.o g/10.1016/j. alan a.2024.125816
Recei ed 18 Decembe 2023; Recei ed in e ised o m 16 Feb ua y 2024; Accep ed 18 Feb ua y 2024
Talan a 273 (2024) 125816
2
1. In oduc ion
In an o mula (IF) plays a pi o al ole in ensu ing he nu i ion and
well-being o in an s, pa icula ly when b eas eeding is no easible,
acco ding o guidelines se o h by he Wo ld Heal h O ganiza ion
(WHO) [1]. IF se es as a i al al e na i e o supplemen a y sou ce o
nu i ion du ing he c i ical i s 12 mon hs o a child’s li e [1,2].
Ensu ing IF sa e y is c ucial, gi en hei exposu e o igo ous
manu ac u ing p ocesses, including Ul a-high empe a u e p ocessing
(UHT) a up o 135 ◦C o 2–5 s o elimina e biological haza ds [2].
Howe e , hese p ocesses can gene a e haza dous o ganic compounds
h ough lipid pe oxida ion and he Mailla d eac ion. Lipid pe oxida ion
is a highly agg essi e eac ion in ood, a ec ed by ex e nal ac o s like
ligh , UV adia ion, hea , and s o age condi ions. I esul s in he
b eakdown o polyunsa u a ed a y acids (PUFA) by eac i e oxygen
species (ROS) h ough a adical pa hway [3]. Simul aneously, he
Mailla d eac ion, a complex p ocess be ween amino g oups and
educing suga s like lac ose, occu s a ying wi h pH bu sha e a common
Amado i ea angemen [4]. The p ima y conce n associa ed wi h hese
eac ions is no only he educ ion in he nu i ional alue o ood
p oduc s bu also he o ma ion o
α
-dica bonyl compounds (
α
-DDC),
including glyoxal (GO), me hylglyoxal (MGO), and dime hylglyoxal
(DMGO), as well as he occu ence o a ious unsa u a ed and sa u a ed
ca bonyl compounds, including malondialdehyde (MDA), o maldehyde
(FCHO), ace aldehyde (ACE), and benzaldehyde (PhCHO), aising con-
ce ns abou hei po en ial heal h isks since hey ha e been documen ed
o be linked o a ious in lamma o y condi ions and a e po en ially
implica ed in he de elopmen o ch onic diseases such as diabe es,
kidney disease, and Alzheime ’s disease, among o he s [4,5].
Regula o y o ganiza ions such as he In e na ional Agency o
Resea ch on Cance (IARC) ca ego ize compounds based on hei
ca cinogenic po en ial o humans. Fo maldehyde is in Ca ego y 1,
indica ing s ong e idence o i s ca cinogenici y. ACE is in Ca ego y 2B,
sugges ing possible ca cinogenici y o humans. MDA, GO, and MGO a e
Unclassi iable in e ms o human ca cinogenici y (Ca ego y 3) due o
insu icien oxicological da a [6,7]. The Eu opean Food Sa e y Au-
ho i y (EFSA) has es ablished a Th eshold o Toxicological Conce n
(TTC) o MDA consump ion a 30
μ
g/kg o body weigh (bw) pe day
[8]. Simila ly, he WHO has se a ole able in ake o 0.2 mg/kg o bw pe
day o li elong o al exposu e o GO [9]. The WHO has also de ined a TDI
a 0.15 mg/kg o bw, con i med by he Scien i ic Panel on Food Addi-
i es, Fla ou ings, P ocessing Aids, and Ma e ials in Con ac wi h Food
(AFC) o he EFSA [10]. ACO, DMGO, and PhCHO a e Gene ally
Recognized as Sa e (GRAS) by he Food and D ug Adminis a ion (FDA),
while ACE and PhCHO a e accep ed as la o ing subs ances in he Eu-
opean Union [11–13]. Addi ionally, he Join FAO/WHO Expe Com-
mi ee on Food Addi i es (JECFA) has se an Accep able Daily In ake
(ADI) o 5 and 900 mg/kg o bw pe day o PhChO and DMGO,
espec i ely [10,14]. Addi ionally, li e ime accep able in ake o 185
μ
g/day has been es ablished o ACE [15].
Gi en he ood sa e y conce ns associa ed wi h hese eac ions,
adop ing a p ecise analy ical app oach is c ucial in ood analysis.
Mic oex ac ions o e accu a e ca bonyl compound analysis wi h ben-
e i s like sample p ese a ion, ace compound sensi i i y, eco- iendly
sol en use, as e ex ac ion, au oma ion, and selec i i y [16]. Solid
Phase Mic oex ac ion (SPME) is a eliable me hod o ola ile and
semi- ola ile compound analysis in complex oods, simpli ying sample
p epa a ion, educing sol en use, and p o iding high sensi i i y and
quan i a i e analysis [16,17]. I s e iciency, sho e ex ac ion imes,
and lowe con amina ion isk make i ideal o p ecise ca bonyl com-
pound analysis in complex ood [16–18].
SPME also enables in- ibe and in-solu ion de i a iza ion eac ions,
commonly used o ca bonyl compound analysis [16–18]. De i a iza-
ion o ca bonyl compounds commonly use hyd azine and
hyd azine-de i a es due o hei eac i i y, s able de i a i es, inc eased
sensi i i y, and compa ibili y wi h a ious analy ical echniques,
including high-pe o mance liquid ch oma og aphy (HPLC) and gas
ch oma og aphy (GC) wi h di e en de ec ion sys ems [16,19]. SPME
has al eady demons a ed i s e ec i eness in analyzing o he ola ile
compounds in in an ood p oduc s. Ga cía-Lla as e al., employed SPME
o GC-mass spec ome y (MS) analysis o pen anal, hexanal, hep anal
and pen ane in liquid milk-ce eal-based in an oods [20]. Subsequen ly,
SPME was used o analyzing hexanal and pen ane in liquid in an oods
and powde ed o mulas by GC- lame ioniza ion de ec o (FID), ocusing
on moni o ing lipid pe oxida ion. Ga cía-Lla as e al. u ilized a ca box-
en/polydime hylsiloxane SPME ibe o 45-min head-space (HS)
ex ac ion in bo h s udies [21].
He e, we p esen a no el app oach o simul aneously de e mining
se en ca bonyl compounds in IF. This me hodology in ol es ex ac ing
and de i a izing o o maldehyde, malondialdehyde, ace aldehyde,
glyoxal, me hyl glyoxal, dime hyl glyoxal and benzaldehyde using HS-
SPME and GC-MS. This wo k p esen s a di e en SPME app oach o e
he exis ing ones o analyzing ola ile compounds in in an o mula and
baby ood ia GC-MS and GC-FID [20,21]. The s udy explo es a ious
SPME ibe coa ings and di e en hyd azines as de i a i e eagen s.
No ably, i e alua es in-sample de i a iza ion simul aneously wi h an-
aly e ex ac ion o enhance sensi i i y, de ia ing om he common
p e-ex ac ion de i a iza ion p ocedu e. The HS-SPME-GC-MS me hod
was alida ed ollowing he guidelines se by he FDA [22,23]. I ach-
ie ed excellen esul s in e ms o linea i y, speci ici y, accu acy, and
p ecision. The g eenness o he me hod is con i med by he no el
AGREEp ep ool. An occu ence s udy was conduc ed on 25 in an o -
mula samples, shedding ligh on hei ca bonyl compound con en . This
allowed hei seg ega ion in o h ee di e en classes h ough he use o
Hie a chical Clus e Analysis and P incipal Componen Analysis. The
p esen me hodology can be used o s udy he deg ee o lipid pe oxi-
da ion o in an o mulae and he impac o he mal ea men by
assessing po en ial a ia ions in hei ca bonyl compound con en .
2. Ma e ials and me hods
2.1. Chemicals and ma e ials
A ange o high-pu i y (≥97%) chemicals and eagen s we e pu -
chased o his s udy, including hype g ade ace ic acid (AAc, Cas No.
64-19-7), ammonium ace a e (AA, Cas No. 631-61-8), ammonium
o ma e (AF, Cas No. 540-69-2), ammonium sul a e (AS, CAS No: 7783-
20-2), dini ophenylhyd azine (DNPH, CAS No. 119-26-6), o mic acid
(FAc, Cas No. 64-18-6), me hanol (MeOH, CAS No: 6-56-1), pen a-
luo ophenylhyd azine (PFPH, CAS No. 828-73-9), and eagen g ade,
and phenylhyd azine (PH, Cas No. 100-63-0) om Me ck (Da ms ad ,
Ge many). Addi ionally, hyd ochlo ic acid (HCl, CAS no. 7647-07-0)
wi h a pu i y o 37% (w/w) was pu chased om VWR Chemicals
(Pennsyl ania, USA), while Milli-Q wa e was gene a ed using a Milli-
po e pu i ica ion sys em (Millipo e, Bille ica, MA, USA). To c ea e
s anda d solu ions o analysis, a ious analy ical s anda ds including
ace aldehyde (ACE, CAS no. 75-07-0); benzaldehyde (PhCHO, CAS no.
100-52-7); dime hylglyoxal (DMGO, CAS no. 431-03-8); o maldehyde
(FCHO, CAS no. 50-00-0); glyoxal (GO, CAS no. 107-22-2); malondial-
dehyde (MDA, CAS no. 100683-54-3), and me hylglyoxal (MGO, CAS
no. 78-98-8) we e pu chased om Me ck. These chemicals we e dis-
sol ed in me hanol (MeOH) a speci ic concen a ions and s o ed a
−20 ◦C. Deu e a ed ace aldehyde (ACEd4, CAS no. 1632-89-9), and
ace one (ACOd6, CAS no. 666-52-4); we e also pu chased om Me ck
(Da ms ad , Ge many), used as in e nal s anda ds (IS) a a concen a ion
o 0.5 ng/mL, and s o ed a −20 ◦C.
SPME ibe s wi h di e en coa ing ma e ials we e acqui ed om
Me ck, including ibe s wi h coa ings o 100
μ
m polydime hyl siloxane
(PDMS), 85
μ
m polyac yla e (PA), 75
μ
m Ca boxen/PDMS (CAR/
PDMS), 50/30
μ
m di inylbenzene/CAR/PDMS (DVB/CAR/PDMS), and
65
μ
m PDMS/DVB. To ensu e hei op imal pe o mance, all ibe s un-
de wen a condi ioning p ocess as pe he manu ac u e ’s ins uc ions
J.A. Cus odio-Mendoza e al.
Talan a 273 (2024) 125816
3
be o e hei ini ial use. Manual sampling was ca ied ou using a manual
holde , also p ocu ed om Me ck. This s udy employed di e en labo-
a o y equipmen , including a Cen omix II-BL Cen i uge om J. P.
Selec a (Ba celona, Spain), a pH me e model Basic 20 om C ison In-
s umen s (Ba celona, Spain) a 2510EMTH ul asonic ba h om B anson
Ul asonics (Danbu y, USA), and a Reax op o ex mixe om In-
s umen s GmbH & Co (Schwalbach, Ge many) we e used o de elop all
expe imen s.
2.2. Samples
A o al o 25 in an o mula (IF) samples we e acqui ed o his s udy,
comp ising powde ed s a e IF (PSIF, n =7), powde ed ollow-up IF
(PFIF, n =7), liquid s a e IF (LSIF, n =4), and liquid ollow-up IF
(LFIF, n =7). These samples we e pu chased om pa apha macies and
local supe ma ke s in San iago de Compos ela, Spain. All samples we e
s o ed in hei o iginal packaging un il he s udy. Liquid samples we e
s o ed a 4 ◦C and opened jus be o e analysis, while powde samples
we e sa egua ded om ligh and econs i u ed a a concen a ion o 15%
w/ using boiled d inking wa e o o m he sample solu ion, adhe ing
o he manu ac u e ’s guidelines. Addi ionally, quali y con ol samples
(QCs) we e eshly p epa ed by adding s anda d solu ions o an ini ial
in an o mula sample a 250, 500, and 1000 ng/mL. These QCs we e
hen s o ed a −20 ◦C o no mo e han one week, p ese ing hei
in eg i y and consis ency o subsequen analy ical assessmen s.
2.3. Head-space solid-phase mic oex ac ion
Simul aneous ex ac ion o ACE, DMGO, FCHO, GO, MDA, MGO, and
PhCHO om in an o mulae was accomplished using HS-SPME. In a
concise desc ip ion o he p ocedu e, 2.5 mL o he sample solu ion (o
liquid in an o mula), which included he in e nal s anda ds, was
placed in o a conical-bo om cen i uge ube. Subsequen ly, 20% w/ o
ammonium sul a e was added, and he mix u e was o exed o 1 min
and hen cen i uged a 3500 pm o 5 min. Following his, a 0.5 mL
aliquo o he aqueous phase was ans e ed o a 6 mL head-space ial.
In his ial, 0.1 mL o PFPH (5 g/L in 215 mmol/L HCl) solu ion was
added, and he olume was adjus ed o 2.15 mL wi h MilliQ wa e
ob aining a pH 5.0. A magne ic ba was placed inside he ial, which was
hen sealed wi h an aluminum cap and sep um. The sys em was o exed
a oom empe a u e o 1 min. The SPME holde ’s needle pie ced he
sep um, and he ial was placed in a wa e ba h se a 60 ◦C. The 50/30
μ
m CAR/PDMS/DVB ibe was exposed o he headspace o 10 min. The
analy es we e deso bed a he GC-MS inle , which was main ained a
270 ◦C, o 5 min. A schema ic ep esen a ion o he HS-SPME p ocess is
depic ed in Fig. 1.
2.4. GC-MS analysis
GC-MS analysis was ca ied ou employing an Agilen Technologies
7890B–5977B-MSD sys em based in Cali o nia, USA. Fo he inle , a
s aigh ul a-ine line (Model 5190-4048) wi h a 0.755 mm inne
diame e was used in spli less mode, wi h he inle empe a u e main-
ained a 270 ◦C and a He low a e o 1 mL/min. The analy e sepa a ion
was achie ed using an Agilen Technologies 2A J&W HP-5MS column,
measu ing 30 m in leng h, 0.25 mm in inne diame e , and wi h a ilm
hickness o 0.25
μ
m. The o en empe a u e p og am ini ially s a ed a
80 ◦C and was held o 1 min, hen amped up a a a e o 3 ◦C pe minu e
o each 150 ◦C, and inally inc eased o 280 ◦C wi h a a e o 30 ◦C pe
minu e, whe e i was held o 3 min.
Fo analy e de ec ion, a single quad upole mass analyze was u i-
lized, ope a ing in he posi i e ion mode (EI+) wi h an elec on impac
sou ce se a 70 eV. The empe a u es o he ans e line, sou ce, and
quad upole we e se a 280 ◦C, 230 ◦C, and 150 ◦C, espec i ely. Da a
acquisi ion and analysis we e ca ied ou using MassHun e Wo ks a ion
So wa e, e sion B.07.00, p o ided by Agilen Technologies in Cali-
o nia, USA. The o al analysis ime o his p ocedu e was 15.33 min.
2.5. S a is ical analysis
An asymme ic 2
2
3
2
//9 sc eening design was employed o in es i-
ga e he in luen ial ac o s o he me hodology in a o al o 9 expe i-
men s [24]. This design choice signi ican ly educed he numbe o
equi ed uns compa ed o a ull ac o ial design, which would ha e
necessi a ed o 36 expe imen s. The ac o s unde sc u iny encompassed
ex ac ion/de i a iza ion empe a u e (40 and 60 ◦C), SPME coa ing
ype (PDMS/DVB and DVB/CAR/PDMS), pH le els (3.0, 5.0, and 7.0),
and he concen a ion o he sal ing-ou agen (0.17, and 20% w/ ).
This design cha ac e izes how subs i u ing one le el o a pa icula
ac o o ano he a ec s he SPME me hod, wi h esponse as he ob-
ained peak a ea in he GC-MS analysis. The e ec s we e ep esen ed
g aphically in he o m o o al e ec s and del a weigh plo s. The a i-
ables indica e he p esence o absence o speci ic ac o s. Fo ins ance,
he coe icien b
1/21
de ines he impac o eplacing an ex ac ion em-
pe a u e o 40 ◦C wi h 60 ◦C, and simila ly, b
2/21
desc ibes he e ec o
eplacing PDMS/DVB coa ing wi h DVB/CAR/PDMS, and so o h. To
Fig. 1. Scheme o he head-space solid-phase mic oex ac ion o ca bonyl and
α
-dica bonyl compounds o simul aneous GC-MS de e mina ion in in an o mulae. IS,
in e nal s anda d; PFPH, pen a luo ophenylhyd azine.
J.A. Cus odio-Mendoza e al.
Talan a 273 (2024) 125816
4
gene a e he expe imen al design, assess he expe imen al da a, and
isualize i s e ec s, Nem odW® s a is ical so wa e ( e sion 2011) was
u ilized.
The expe imen al da a ela ed o he p esence o ca bonyl com-
pounds in in an o mula unde wen a comp ehensi e s a is ical analysis
using IBM SPSS S a is ics so wa e. This analysis aimed o cha ac e ize
he da a and unco e insigh s in o he di e ences and g oupings wi hin
he da ase . Analysis o Va iance (ANOVA) was u ilized o de e mine
whe he signi ican a ia ions exis ed among di e en ypes o b ands o
in an o mula. A Box and Whiske Plo was employed o summa ize he
sp ead o coun -based da ase s by displaying qua iles, he median, and
any ou lie s. The box ep esen s he in e qua ile ange, while whiske s
show da a wi hin 1.5 imes he in e qua ile ange, wi h ou lie s being
ma ked. K-means clus e ing was used o iden i y na u al g oupings in
he occu ence da a, di iding i in o clus e s based on simila i ies.
Addi ionally, hie a chical clus e analysis (HCA) c ea ed a hie a chical
s uc u e o clus e s o e eal ela ionships be ween ca bonyl compound
p o iles in di e en in an o mulae.
2.6. Analy ical pe o mance o he me hod
A PSIF sample se ed as he blank specimen o conduc ing analy -
ical alida ion o he me hod, ollowing he guidelines ou lined by he
Food and D ug Adminis a ion (FDA) o he Valida ion o Chemical
Me hods in Food, Feed, Cosme ics, and Ve e ina y P oduc s [22], as well
as he Bioanaly ical Me hod Valida ion Guidance o Indus y [23].
To es ablish he me hod’s speci ici y, a quan i ie ion and wo
quali ie ions we e chosen o all analy es and in e nal s anda ds, each
elu ing a di e en e en ion imes. The absence o any in e e ence in
hese elu ion egions con i med he me hod’s speci ici y. Indi idual
s anda d addi ion was pe o med alongside in e nal s anda d calib a-
ion cu e cons uc ion, anging om he lowe limi o quan i ica ion
(LLOQ) o he uppe limi o quan i ica ion (ULOQ) o each analy e.
These cu es we e u ilized o de e mine he limi o de ec ion (LOD) and
LLOQ acco ding o eq. (2).
YLOD
/LLOQ
=Yblank +
3.3
/
10
σ
blank (eq 2)
Addi ionally, LLOQ and ULOQ ep esen he lowes and highes
concen a ions o each analy e ha can be quan i a i ely de e mined
wi h accep able p ecision and accu acy. The e o e, calib a ion cu es
we e es ablished a six concen a ion le els (n =3) and used o e alua e
he linea i y o he me hod. Fu he mo e, he sensi i i y a LLOQ was
assessed as he calib a ion slope di ided by he ela i e s anda d de i-
a ion (%RSD) o each analy e in iplica e. QCs ( e e o Samples sec-
ion) we e employed o e alua e he accu acy o he me hod, in ol ing
eco e y (n =3), and p ecision h ough in aday and in e day assays,
measu ed in e ms o %RSD (n =5).
Ma ix e ec was assessed h ough a compa ison o slope alues o
he calib a ion cu e wi h hose ob ained om he s anda d addi ion
calib a ion cu e in in he selec ed g oup o samples. Mo eo e , o
assu e he applicabili y o he de eloped me hod o he di e en g oup
o samples a ma ix ex ension s udy was pe o med o powe ollow-up,
liquid s a e and liquid ollow-up o mulae a h ee le els 250, 500 and
100 ng/mL in quin uplica ed.
3. Resul s and discussion
3.1. Pa ame e s op imiza ion o head-space solid-phase mic oex ac ion
me hodology
The choice o a sui able de i a iza ion eagen is c ucial o he
me hod’s e ec i eness. I ’s essen ial ha his eagen o ms a s able
p oduc when eac ing wi h he analy e and is compa ible wi h he
chosen sepa a ion and de ec ion echniques. The eac ion be ween he
eagen and analy e should also occu apidly, which is pa icula ly
impo an in GC analysis due o he need o undecomposed e apo a ion
o de i a i es. Common de i a iza ion eac ions o ca bonyl com-
pounds in GC in ol e o ming oximes and hyd azones [27,32,35,36]. In
ou s udy, we e alua ed h ee hyd azine eagen s (DNPH, PFPH, and
PH) o hei abili y o c ea e hyd azones wi h speci ic ca bonyl com-
pounds (MDA, GO, MGO, DMGO, and ACOd6) in iplica e. As shown in
Fig. 1SA (Supplemen a y ma e ial), he highes signal, in e ms o
ela i e ch oma og aphic a ea, was achie ed when PFPH was used as
he de i a i e eagen . Consequen ly, PFPH was chosen o u he
analysis. The key di e ence be ween PFPH, DNPH, and PH lies in he
na u e o he subs i uen s on he phenyl ing (Fig. 2S). PFPH con ains
luo ine a oms on he phenyl ing, which can enhance s abili y and
eac i i y. The inc eased pola i y o hese de i a i es can lead o
imp o ed esolu ion in ch oma og ams [35]. In con as , DNPH, cha -
ac e ized by a phenyl ing wi h ni o g oups, is ecognized o i s eac-
ion wi h ca bonyl compounds, yielding colo ed de i a i es u ilized in
UV/ is o luo escence spec oscopy applica ions ollowing liquid
ch oma og aphic sepa a ion; ne e heless, ecen inno a ions ha e
in oduced gas ch oma og aphic quan i ica ion me hods o DNPH de-
i a i es [25,26].
The concen a ion o de i a i e eagen was hen s udied a 1, 3 and
5 g/L based on li e a u e (Fig. 1S.B). Fo his applica ion he highes
signal in e ms o ela i e ch oma og aphic a ea was achie ed using 5 g/
L; he e o e, his concen a ion o PFPH was selec ed. Hyd azine
de i a iza ion eac ions equi e acidic condi ions [25,26,35,38], and
di e en acids we e es ed o achie e a pH o 5.0 du ing he de i a i-
za ion p ocedu e. In Fig. 1S.C, i is e iden ha he highes ela i e
ch oma og aphic a eas a e ob ained when HCl is used. This di e ence
a ises om he ac ha s ong acids almos comple ely dona e hei
p o ons o wa e molecules, whe eas weak acids es ablish an equilib-
ium be ween undissocia ed acid molecules and ions in solu ion. This
beha io is quan i ied by he acid dissocia ion cons an (K
a
). The e o e,
s ong acids exhibi a highe deg ee o ioniza ion compa ed o weak
acids a he same pH, which a ec s hei eac i i y and beha io in
chemical eac ions. Di e en ammonium sal s—ammonium sul a e
(AS), ammonium o ma e (AF), and ammonium ace a e (AA)—we e
igo ously es ed in iplica e as sal ing-ou agen s (Fig. 3SA). The esul s
indica ed ha AS yielded he mos p omising ou comes in e ms o
ela i e ch oma og aphic a ea, wi h AS consis en ly su passing 80%
ela i e a eas ac oss mos analy es. No ably, while MGO showed highe
a eas wi h AA, AS s ill achie ed ela i e a eas exceeding 80%. Thus, AS
was chosen as he p ima y sal ing-ou agen o subsequen esea ch
endea o s. Fu he mo e, he impac o agi a ion me hods—ul asound
e sus magne ic s i ing—was in es iga ed (Fig. 3B). The indings
e ealed ha magne ic s i ing consis en ly esul ed in supe io ela i e
ch oma og aphic a eas o all analy es. As a esul , he decision was
made o conduc u he expe imen s u ilizing magne ic s i ing as he
p e e ed agi a ion me hod.
We conduc ed es s using a ious SPME ibe s, each coa ed wi h
di e en ma e ials, ollowing PFPH de i a iza ion unde he condi ions
ou lined abo e. This de i a iza ion was ca ied ou o 15 min in he
headspace o he ials. The coa ings es ed included PDMS, PA, CAR/
PDMS, DVB/CAR/PDMS, and PDMS/DVB. These es s aimed o assess
hei pe o mance in ex ac ing he MDA-PFPH, GO-PFPH, MGO-PFPH
and DMGO-PFPH de i a i es in iplica e (Fig. 4S). DVB/CAR/PDMS
and PDMS/DVB coa ings demons a ed he highes ela i e a eas,
showcasing hei e ec i eness in e aining hese compounds. Addi-
ionally, we in es iga ed he kine ics o simul aneous de i a iza ion and
ex ac ion using he DVB/CAR/PDMS ibe (Fig. 5S). I was obse ed
ha equilib ium is achie ed a e 10 min o he eac ion o hese ana-
ly es unde he gi en condi ions. The e o e, hese condi ions we e
selec ed o u he expe imen s o examine he key pa ame e s in lu-
encing he pe o mance o HS-SPME du ing he simul aneous ex ac ion
and PFPH de i a iza ion o FCHO, MDA, ACE, ACEd4, ACOd6, GO,
MGO, DMGO, and PhCHO. This e alua ion was ca ied ou using an
asymme ic sc eening design, conside ing he ollowing ac o s: b
1
,
J.A. Cus odio-Mendoza e al.
Talan a 273 (2024) 125816
5
empe a u e; b
2
, SPME coa ing ma e ial; b
3
, concen a ion o he sal ing-
ou agen ; and b
4
, pH. To minimize andom e o s, all expe imen s we e
conduc ed andomly. The esul s we e analyzed using o al e ec and
del a weigh plo s [24], which depic he di e en ial e ec s o each
ac o when conside ing pai wise combina ions o wo di e en le els.
In del a plo s, he leng h o he ba s e lec s he ela i e magni ude o he
e ec s. Posi i e alues indica e e ec s a o ing highe le els, while
nega i e alues indica e e ec s a o ing lowe le els. Do ed lines on he
plo s ep esen he s a is ical signi icance le els, de e mined using he
Leng h’s me hod a a 95% con idence le el. An e ec is conside ed
s a is ically signi ican when i su passes his e e ence line. By consid-
e ing he sign and alue o he e ec s, we we e able o selec he op imal
le el o each expe imen al pa ame e . In o al e ec plo s, he leng h o
he ba s di ec ly co ela es wi h he e ec o each ac o on he analy -
ical esponse. Fig. 2 illus a es he indings om he del a plo s e ealing
ha none o he pa ame e s we e s a is ically signi ican o malon-
dialdehyde (Fig. 2B), glyoxal (Fig. 2F) and dica bonyl compounds All o
a iables exhibi ed s a is ical signi icance o aldehydes as deu e a ed
ace one (Fig. 2D). Rema kably, all analy es displayed simila beha io
ega ding hese ac o s. Despi e hese a ied esponses, he o al e ec
cha (Fig. 2A, C, E) indica ed ha he bes esul s we e achie ed when
conduc ing he HS-SPME a 60 ◦C, u ilizing a DVB/CAR/PDMS coa ing
on he SPME ibe , and employing a 20% w/w concen a ion o he
sal ing-ou agen . Acco ding o he pa e o cha s, a pH o 3.0 was
op imal o aldehydes (Fig. 2A–C), and o dica bonyl compounds, pH
7.0 yielded he highes esponse, ollowed by pH 3.0 (Fig. 2E). Since he
selec ion o pH was s a is ically signi ican o aldehydes (Fig. 2D), i
was decided o u ilize a sample solu ion wi h a pH o 3.0.
3.2. Analy ical alida ion o he HS-SPME-GC-MS me hodology
Following FDA guidelines [22,23], we conduc ed a comp ehensi e
e alua ion o he HS-SPME p ocedu e’s analy ical ea u es. Ini ially, we
assessed he selec i i y and speci ici y by selec ing one quan i ie and
wo quali ie ions wi h speci ic e en ion imes o FCHO, MDA, ACE,
ACEd4, ACOd6, GO, MGO, DMGO, and PhCHO. I is wo hwhile
poin ing ou ha all he a ge analy es and IS exhibi ed dis inc i e
e en ion imes, and no ma ix in e e ences we e obse ed in he GC-MS
ch oma og am (Fig. 3).
The me hod’s cha ac e is ics, including de e mina ion limi s, line-
a i y, and sensi i i y, a e summa ized in Table 1. ACEd4 ac ed as he IS
o FCHO, MDA, ACE, and PhCHO, while ACOd6 was employed o GO,
MGO, and DMGO. By employing s anda d addi ion wi h IS calib a ion
plo s o each analy e, we co e ed a concen a ion ange om 50 o
1500 ng/mL, wi h he IS concen a ion ixed a 500 ng/mL. This
app oach showcased ema kable linea i y (
2
≥0.9990). Fu he mo e,
when assessing sensi i i y a he lowe limi o quan i ica ion (LLOQ),
we achie ed a ela i e s anda d de ia ion (RSD) o ≤7.6%. We used QC
samples o e alua e he me hod’s accu acy and p ecision, as ou lined in
Table 2. Ou accu acy esul s ell wi hin he ange o 84.8%–111.5% o
eco e ies, and bo h in aday and in e day p ecision exhibi ed excellen
ou comes, wi h RSD alues o ≤13.8% and ≤12.1%, espec i ely,
mee ing he es ablished c i e ia o accep abili y.
Table 3 summa izes he esul s ob ained om he assessmen o he
Ma ix E ec and Ma ix Ex ension S udy. No ably, a signi ican ma ix
e ec is e iden o all he analy es when e alua ed in powde ed s a e
in an o mulae, as compa ed o ollow-up powde ed in an o mula and
liquid in an o mulae. This obse ed ma ix e ec aligns wi h p io
indings in he con ex o SPME and is likely a ibu ed o he p esence o
in e e ing compounds which may compe e wi h he a ge analy es o
binding si es on he SPME ibe o in luence he chemical equilib ium
du ing he ex ac ion p ocess, as epo ed in p e ious s udies [16–18,20,
21].
Gi en ha such in e e ence can in oduce a ia ions in ex ac ion
e iciency and po en ially lead o inaccu a e quan i ica ion o he ana-
ly es, i becomes impe a i e o employ speci ic calib a ion cu es o
each dis inc g oup o samples wi hin he se [23]. On he o he hand,
he Ma ix Ex ension S udy, as p esen ed in Table 3, demons a es
accep able accu acy (80–116% eco e y) and p ecision (0.4–19.0 %
RSD) ac oss he di e en sample g oups. This ou come unde sco es he
applicabili y o he HS-SPME-GC-MS me hod o he en i e se o samples
[23].
Compa a i ely, p e iously published a icles (see Table 1S in he
supplemen a y ma e ial) on ch oma og aphic me hods o ca bonyl
Fig. 2. To al e ec s and del a weigh plo s o he asymme ical 2232//9 sc eening design o malondialdehyde (A, B), deu e a ed ace one (C, D), glyoxal (E, F).
J.A. Cus odio-Mendoza e al.
Talan a 273 (2024) 125816
6
Fig. 3. Ch oma og am o o maldehyde (FCHO), malondialdehyde (MDA), ace aldehyde (ACE), deu e a ed ace aldehyde (ACEd4), deu e a ed ace one (ACOd6),
glyoxal (GO), me hylglyoxal (MGO), dime hylglyoxal (DMGO), benzaldehyde (PhCHO) a 1
μ
/mL o powde s a e in an o mula.
Fig. 4. A box and whiske plo o he a iabili y in ca bonyl compounds concen a ions among posi i e samples.
J.A. Cus odio-Mendoza e al.
Talan a 273 (2024) 125816
7
compound analysis epo ed quan i ica ion limi s ei he simila o o
highe han hose p esen ed he e [7,25–32]. No ably, Paiano e al. [33]
achie ed lowe ins umen al de e mina ion limi s han ou s udy in he
single analysis o ACE in alcoholic and non-alcoholic be e ages using he
HS-SPME-GCMS me hod p e iously de eloped by Wang e al. [34],
which also exhibi ed simila linea i y bu highe a ia ion coe icien s.
Simila ly, Wang e al. [35] epo ed lowe limi s in he single GC-MS
analysis o MDA in mea using hyd olysis and PFPH de i a iza ion,
wi h eco e ies and p ecision on pa wi h ou indings. Finally, Lim &
Shim [36] epo ed lowe de e mina ion limi s han ou s udy in he
simul aneous GC-MS analysis o GO and MGO in alcoholic be e ages ia
HS-SPME. Pa allelly, he u iliza ion o pen a luo o benzyl hyd oxyl-
amine de i a iza ion has demons a ed imp o emen in he ex ac ion
o ca bonyl compounds using SPME om alcoholic be e ages. Yu e al.
[37] epo ed lowe quan i ica ion limi s o GO, MGO, DMGO, and
PhCHO in hei a ge ed me abolomics analyses o Huangjiu, a adi-
ional Chinese alcoholic be e age. Pie gio anni e al. [38] epo ed
lowe limi s in he analysis o PhCHO in wines. Mo ei a e al. [39] also
documen ed lowe quan i ica ion limi s o GO, MGO, DMGO, and
PhCHO in po wines. Addi ionally, He nandes e al. [40] epo ed lowe
limi s o ACE and FCHO in c a bee .
Ou HS-SPME-GC-MS p ocedu e dis inguishes i sel by achie ing
lowe de ec ion limi s, highe accu acy (app oaching 100% eco e y),
and g ea e p ecision (wi h an RSD o ≤14.7%) when compa ed o p io
in es iga ions in ol ing he de e mina ion o MDA and
α
-dica bonyl
compounds in in an o mula and baby ood [28,29,41–43]. I is no e-
wo hy ha ou app oach demons a es he capabili y o iden i y and
quan i y up o se en ca bonyl and dica bonyl compounds, showcasing
Table 1
Speci ici y, sensibili y, linea i y and de e mina ion limi s o he HS-SPME me hod o simul aneous GC-MS de e mina ion o ca bonyl and
α
-dica bonyl compounds in
in an o mulae.
Analy e RT Quan i ie ion Quali ie ions R
2
LOD LLOQ ULOQ Sensi i i y
min m/z m/z m/z ng/mL ng/mL ng/mL % RSD
FCHO 4.65 210 211 182 0.9990 16 50 1500 6.5
MDA 5.32 234 233 215 0.9999 20 50 1500 7.6
ACE 5.77 224 182 196 0.9993 50 150 1500 6.9
ACEd4
a
5.73 228 227 229 – – – – –
ACOd6
a
6.55 244 242 245 – – – – –
GO 6.55 281 182 155 0.9995 15 50 1500 2.1
MGO 7.74 252 253 182 0.9992 15 50 1500 2.5
DMGO 8.97 266 267 223 0.9996 16 50 1500 2.1
PhCHO 11.15 286 287 285 0.9998 50 150 1500 2.7
a
Used as in e nal s anda d; RT, e en ion ime; LOD, limi o de ec ion; LLOQ, lowe limi o quan i ica ion; ULOQ, uppe limi o quan i ica ion; RSD, ela i e
s anda d de ia ion; FCHO, o maldehyde; MDA, malondialdehyde; ACE, ace aldehyde; ACEd4, deu e a ed ace aldehyde; ACOd6, deu e a ed ace one; GO, glyoxal;
MGO, me hylglyoxal; DMGO, dime hylglyoxal; PhCHO, benzaldehyde.
Table 2
Accu acy and P ecision o he HS-SPME Me hod o Simul aneous GC-MS De e mina ion o Ca bonyl and
α
-Dica bonyl Compounds in In an Fo mulae.
Analy e Accu acy P ecision
(n =3)
% Reco e y
In aday (n =5)
% RSD
In e day (n =5)
% RSD
250 ng/mL 500 ng/mL 1000 ng/mL 250 ng/mL 500 ng/mL 1000 ng/mL 250 ng/mL 500 ng/mL 1000 ng/mL
FCHO 86.8 93.1 97.9 14.7 7.7 12.8 9.6 5.2 3.2
MDA 105.8 110.8 110.1 13.7 6.6 13.8 8.3 9.1 12.1
ACE 94.4 100.5 98.5 13.1 11.5 4.3 5.7 4.9 4.6
GO 89.5 101.3 98.6 7.8 8.2 9.6 5.2 3.7 4.5
MGO 84.8 104.5 105.5 13.9 7.2 9.0 4.1 5.2 12.1
DMGO 108.2 111.3 101.1 13.4 9.4 4.9 6.6 11.1 14.1
PhCHO 105.1 102.6 111.5 9.6 7.6 11.9 5.1 3.5 11.8
RSD, ela i e s anda d de ia ion; FCHO, o maldehyde; MDA, malondialdehyde; ACE, ace aldehyde; GO, glyoxal; MGO, me hylglyoxal; DMGO, dime hylglyoxal;
PhCHO, benzaldehyde.
Table 3
In an Fo mulae’s ma ix e ec and ma ix ex ension s udy.
Analy e Powe ollow-up IF Liquid s a e IF Liquid ollow-up IF
Ma ix e ec
a
Accu acy
b
P ecision
c
Ma ix e ec
a
Accu acy
b
P ecision
c
Ma ix e ec
a
Accu acy
b
P ecision
c
% % Reco e y % RSD % % Reco e y % RSD % % Reco e y % RSD
FCHO 43.7 85–104 0.7–4.0 25.5 80–103 0.7–1.1 49.8 83–113 1.5–8.7
MDA 123.7 80–112 4.6–11.2 21.6 83–107 1.7–5.0 9.3 90–107 1.7–3.8
ACE 57.2 91–102 2.0–4.3 48.7 92–104 3.8–6.4 26.1 85–111 3.9–6.7
GO 19.6 102–113 4.7–12.7 21.5 88–106 2.3–9.2 14.8 89–108 1.8–3.8
MGO 1.63 92–115 0.7–5.1 7.13 88–103 0.5–1.5 3.8 98–118 0.4–6.9
DMGO 6.69 84–104 2.1–19.0 1.57 92–109 0.8–3.7 3.9 90–104 1.2–12.4
PhCHO 149 91–116 2.7–5.4 15.1 96–108 0.5–5.2 20.9 91–111 4.4–5.2
IF, in an o mula; RSD, ela i e s anda d de ia ion; FCHO, o maldehyde; MDA, malondialdehyde; ACE, ace aldehyde; GO, glyoxal; MGO, me hylglyoxal; DMGO,
dime hylglyoxal; PhCHO, benzaldehyde;
a
de e mined h ough a compa ison o slope alues wi h hose ob ained om he s anda d addi ion calib a ion cu e in
powde s a e in an o mula;
b
assessed a h ee di e en concen a ion le els (250, 500 and 100 ng/mL) in iplica e;
c
assessed in in a- and in e -day assays a h ee
le els o concen a ion (250, 500 and 100 ng/mL) in quin uplica e.
J.A. Cus odio-Mendoza e al.
Talan a 273 (2024) 125816
8
supe io analy ical pe o mance in compa ison o p e iously published
me hods (Table 1S).
3.3. Analy ical g eenness me ic o HS-SPME-GC-MS
To assess he en i onmen al impac o ou sample p epa a ion
me hod, we u ilized he AGREEp ep assessmen h ough he AGREEp ep
open access so wa e [44]. This ool helps pinpoin he s eng hs and
weaknesses o me hods, he eby acili a ing he de elopmen o mo e
eco- iendly sample p epa a ion p ocedu es. The AGREEp ep assess-
men in ol es en s eps, each assigning sco es be ween 0 and 1, whe e
0 ep esen s he poo es pe o mance and 1 he bes . These sco es, wi h
de aul weigh ings o each c i e ion, a e combined o gene a e an
o e all sco e, also anging om 0 o 1, whe e 1 signi ies op imal
pe o mance.
The en i onmen al impac compa ison be ween sample p epa a ion
me hods o ca bonyl and dica bonyl compound analysis in in an o -
mula and baby ood is ou lined in Table 2S (supplemen a y ma e ial).
Ou de eloped HS-SPME-GC-MS p ocedu e ob ained an o e all sco e o
0.62. This sco e mi o s i s pe o mance, in luenced by key ac o s such
as educed haza dous ma e ial olume (sco e: 0.67), signi ican ma e ial
eusabili y (sco e: 0.75), he capabili y o p ocess 6 samples simul a-
neously wi hin an hou , and semiau oma ic sys em in eg a ion (sco e:
0.50). These collec i e indings unde sco e ha he p esen ed me hod
p o ides a mo e en i onmen ally iendly al e na i e o ex ac ing
ca bonyl and dica bonyl compounds om in an o mulae and baby
ood compa ed o exis ing me hods [28,29,40].
3.4. Occu ence o ca bonyl and dica bonyl compounds in in an
o mulae
The inno a i e HS-SPME-GC-MS me hod was u ilized o de ec se en
ca bonyl and dica bonyl compounds in a ious IF samples, including
bo h powde ed and liquid s a e and ollow-up IF. The indings a e
de ailed in Table 4.
Fig. 4 plo s he a ia ion in ca bonyl compound concen a ions
among he posi i e samples. All samples es ed posi i e o MGO, wi h
concen a ions anging om 203.3 o 1409.6 ng/mL. No ably, he
lowes and highes MGO concen a ions we e ound in PSIF samples. A
signi ican 96% o he samples also es ed posi i e o FCHO and GO.
The highes FCHO concen a ion was obse ed in PFIF a 1385.7 ng/mL,
while he lowes was in PSIF a 78.9 ng/mL. Fo GO, he highes con-
cen a ion was in LFIF a 1464.4 ng/mL, and he lowes in PSIF a 61.4
ng/mL.
Among he samples, 68% es ed posi i e o MDA, wi h he highes
concen a ion a 1491.5 ng/mL ound in LFIF and he lowes a 54.1 ng/
mL in LSIF. Rema kably, MDA le els in PFIF samples we e non-
quan i iable. DMGO and PhCHO we e ound a quan i iable le els
only in he liquid samples, ep esen ing 24% o he samples. The highes
DMGO concen a ion, 894.9 ng/mL, was ound in LFIF, while he lowes ,
171.2 ng/mL, was in LSIF. The highes PhCHO concen a ion was
obse ed in LSIF a 651.4 ng/mL, and he lowes in LFIF a 137.5 ng/mL.
ACE was de ec ed in only 20% o he samples, wi h he lowes concen-
a ion, 155.1 ng/mL, ound in one PSIF sample, and he highes , 893.0
ng/mL, in LFIF. Simila esul s we e epo ed by Pozzo e al. [42] and
Cesa [41] ega ding MDA con en in IF. Wang e al. [43] also epo ed
simila GO con en in IF. Simila ly, Akıllıo˘
glu e al. [29] and Kocada˘
glı
e al. [28] epo ed compa able MGO esul s in he analysis o IF.
Fu he mo e, an ANOVA analysis (Table 3S) e ealed p- alues below
0.05 o he occu ence o FCHO, MDA, GO, and PhCHO, indica ing a
s a is ically signi ican di e ence among sample g oups based on he
concen a ion o hese analy es. Howe e , no s a is ically signi ican
di e ences we e ound o ACE, MGO, and DMGO. Fig. 6S p esen s a
dend og am o simila i ies o ca bonyl con en among IF samples using
he wa d-linkage me hod. This HCA classi ied in an o mulas in o h ee
dis inc ca ego ies based on he dissimila i ies o hei ca bonyl con en .
Class 1 comp ised all samples o LSIF and PS1F 2, Class 2 included PSIF
1, and PSIF samples 3 o 7, while Class 3 encompassed all ollow-up
o mulas. Simila ly, he ba g aph o clus e ing o IF in Fig. 7S p o-
ides in o ma ion abou he same h ee clus e s o med using he K-
means clus e ing me hod. In clus e 1, he in an o mulas exhibi ed high
posi i e alues, pa icula ly o ace aldehyde, glyoxal, and dime hyl-
glyoxal. Clus e 2 samples sco ed posi i ely o FCHO, ACE, MDA, GO,
and MGO, bu showed nega i e alues o DMGO and PhCHO. On he
Table 4
Ca bonyl compound occu ence in in an o mulae.
IF Fo maldehyde Malondialdehyde Ace aldehyde Glyoxal Me hylglyoxal Dime hylglyoxal Benzaldehyde C
ng/mL ±SD ng/mL ±SD ng/mL ±SD ng/mL ±SD ng/mL ±SD ng/mL ±SD ng/mL ±SD
PSIF 1 413.4 86.0 1432.8 23.1 ND – 70.2 49.3 499.1 61.2 ND – ND – 2
PSIF 2 812.4 38.1 1030.8 36.0 155.1 6.6 841.8 56.4 1409.6 93.6 ND – ND – 1
PSIF 3 318.8 39.3 1290.5 29.4 ND – 316.6 24.8 976.2 83.5 ND – ND – 2
PSIF 4 165.8 28.4 1011.4 64.1 ND – ND – 455.6 59.9 ND – ND – 2
PSIF 5 78.9 15.6 1258.3 14.7 ND – 61.4 32.1 257.9 73.7 ND – ND – 2
PSIF 6 329.6 26.5 1156.1 28.2 ND – 149.2 88.0 435.1 60.3 ND – ND – 2
PSIF 7 ND – 113.8 88.3 ND – 196.6 51.0 203.3 49.0 ND – ND – 2
PFIF 1 468.4 27.1 ND – ND – 196.0 16.91 223.9 33.1 ND – ND – 3
PFIF 2 1132.2 47.8 ND – ND – 118.0 69.89 431.2 25.3 ND – ND – 3
PFIF 3 1115.8 17.5 ND – ND – 131.0 34.73 484.7 13.9 ND – ND – 3
PFIF 4 441.3 28.7 ND – ND – 144.0 76.39 546.7 72.5 ND – ND – 3
PFIF 5 1385.7 42.9 ND – ND – 100.0 75.61 387.5 42.2 ND – ND – 3
PFIF 6 562.2 38.2 ND – ND – 154.0 43.4 432.5 10.5 ND – ND – 3
PFIF 7 861.7 36.8 ND – ND – 125.0 18.84 543.3 61.5 ND – ND – 3
LSIF 1 559.2 21.2 ND – ND – 152.9 42.1 614.6 39.8 257.2 85.3 ND – 1
LSIF 2 558.6 11.4 82.5 12.7 ND – 136.8 43.3 589.0 93.6 236.3 80.5 651.4 51.3 1
LSIF 3 1327.6 77.4 54.1 51.6 ND – 1014.9 31.2 761.3 76.7 264.6 75.5 169.4 54.4 1
LSIF 4 791.3 32.9 104.3 96.5 ND – 754.5 75.2 706.3 52.7 171.2 19.2 287.5 37.5 1
LFIF 1 187.6 12.7 298.2 47.7 ND – 436.5 21.8 210.3 21.2 ND – ND – 3
LFIF 2 1001.3 62.7 571.1 78.4 ND – 1182.1 24.6 1162.1 92.5 ND – ND – 3
LFIF 3 1006.3 57.5 621.5 83.7 893.0 34.6 619.4 29.9 1076.9 54.8 ND – ND – 3
LFIF 4 303.6 31.1 362.0 22.9 ND – 213.5 38.9 315.0 85.5 ND – 206.9 95.5 3
LFIF 5 360.9 47.4 933.5 47.4 160.5 19.4 1323.2 42.3 398.8 55.5 894.9 53.8 139.7 65.5 3
LFIF 6 1107.5 19.6 1453.1 17.0 176.9 40.9 1464.4 77.4 769.7 27.1 358.2 57.8 137.5 17.1 3
LFIF 7 986.5 44.5 1491.5 46.2 242.6 65.1 174.6 15.8 584.1 19.1 ND – ND – 3
IF, in an o mula sample; C, clus e ; PSIF, powde ed s a e IF; PFIF, powde ed ollow-up IF; LSIF, liquid s a e IF; LFIF, liquid ollow-up IF; ND, non-de e mined
(concen a ion below LLOQ).
J.A. Cus odio-Mendoza e al.
Talan a 273 (2024) 125816
9
o he hand, clus e 3 samples displayed nega i e alues o FCHO, ACE,
MDA, GO, MGO, and DMGO, wi h sligh ly posi i e alues o PhCHO.
This analysis, ca ego izing di e en in an o mulas based on hei
ca bonyl con en , s ongly suppo s he use o ca bonyl compounds as
ma ke s o lipid pe oxida ion and ood quali y. The clus e s highligh
how ca bonyl con en a ia ions se e as quali y indica o s, wi h lowe
le els indica ing eshness and p oduc quali y, and highe le els sug-
ges ing possible de e io a ion o s o age issues.
4. Conclusions
The de elopmen and op imiza ion o he headspace solid-phase
mic oex ac ion (HS-SPME) me hodology o analyzing ca bonyl and
dica bonyl compounds in in an o mulae has been a me iculous and
ui ul p ocess. Se e al key pa ame e s we e in es iga ed and op i-
mized o ensu e he me hod’s e ec i eness and eliabili y.
Fi s and o emos , selec ing he igh de i a iza ion eagen was
c ucial, and PFPH p o ed highly e ec i e, enhancing s abili y, eac-
i i y, and ch oma og aphic esolu ion. The op imized PFPH concen-
a ion o 5 g/L, along wi h he use o hyd ochlo ic acid o pH con ol,
ensu ed op imal condi ions. DVB/CAR/PDMS and PDMS/DVB coa ings
on SPME ibe s excelled in ex ac ing PFPH de i a i es, eaching equi-
lib ium in 10 min. An asymme ic sc eening design iden i ied HS-SPME
a 60 ◦C, DVB/CAR/PDMS coa ing, and a 20% w/w sal ing-ou agen as
op imal o simul aneous ex ac ion and de i a iza ion o FCHO, MDA,
ACE, GO, MGO, DMGO, and PhCHO. pH a ia ions we e conside ed o
speci ic analy es o op imize me hod pe o mance.
The analy ical alida ion o he HS-SPME-GC-MS me hodology ol-
lowed FDA guidelines, demons a ing excellen selec i i y, speci ici y,
linea i y, and sensi i i y. The de e mina ion limi s and p ecision me
es ablished c i e ia, wi h eco e ies anging om 84.8% o 111.5%.
Impo an ly, he me hod’s pe o mance su passed o equaled ha o
p e iously published me hods o ca bonyl compound analysis, making
i a obus and eliable ool o quan i ying ca bonyl and dica bonyl
compounds in in an o mulae.
The assessmen o he en i onmen al impac using he AGREEp ep
ool has p o ided aluable insigh s in o he eco- iendliness o he
de eloped HS-SPME-GC-MS p ocedu e o ca bonyl and dica bonyl
compound analysis in in an o mulae. No able s eng hs include
educed haza dous ma e ial usage, high ma e ial eusabili y, e icien
sample p ocessing, and semi-au oma ion. The me hod p esen s a g eene
al e na i e compa ed o exis ing app oaches, emphasizing i s po en ial
o sus ainable analy ical p ac ices in ood analysis.
The alida ed HS-SPME-GC-MS me hod was employed o de ec
ca bonyl and dica bonyl compounds in di e se in an o mula samples,
including powde ed and liquid a ian s. The analysis e ealed he
widesp ead p esence o compounds like FCHO, GO, and MGO, wi h
a ying concen a ions, while o he s like MDA, DMGO, and PhCHO
showed mo e selec i e occu ence. Signi ican di e ences in compound
concen a ions we e ound among sample g oups, as e idenced by
ANOVA esul s. Fu he mo e, clus e ing analyses ca ego ized he in an
o mula samples in o dis inc g oups based on hei ca bonyl con en ,
unde sco ing he po en ial o hese compounds as ma ke s o assessing
lipid pe oxida ion, and ood quali y.
In conclusion, he p esen ed me hod s ands as a obus ool ha can
be employed in u u e esea ch endea o s o in es iga e he a ia ions in
ca bonyl con en o in an o mulae unde di e se condi ions, p o iding
insigh s in o hei associa ion wi h eshness, p oduc quali y, and po-
en ial de e io a ion. Addi ionally, he p esence and quan i y o hese
compounds ha e signi ican nu i ional and heal h implica ions,
pa icula ly in he con ex o in an o mulas, making his analysis a
aluable ool o e alua ing hei o e all quali y, sa e y, and nu i ional
impac .
CRediT au ho ship con ibu ion s a emen
Jo ge A. Cus odio-Mendoza: W i ing – e iew & edi ing, W i ing –
o iginal d a , Visualiza ion, Valida ion, In es iga ion, Fo mal analysis,
Concep ualiza ion. Ana Lopez Blanco: Valida ion, In es iga ion. Ana
M. A es-Fuen es: W i ing – e iew & edi ing, W i ing – o iginal d a ,
In es iga ion. An onia M. Ca o Díaz: W i ing – e iew & edi ing,
W i ing – o iginal d a , Supe ision, Me hodology, Fo mal analysis,
Da a cu a ion, Concep ualiza 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
This wo k was suppo ed by he Spanish Minis y o Science, Inno-
a ion, and Uni e si ies (P ojec RTI2018-096450-B-I00) along wi h
FEDER ounds, and by he Polish Na ional Science Cen e (P ojec 2023/
07/X/NZ9/01113) as pa o he MINIATURA-7 p og am.
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. alan a.2024.125816.
Re e ences
[1] Wo ld Heal h O ganiza ion, Global S a egy o In an and Young Child Feeding,
2003. h ps://apps.who.in /gb/a chi e/pd _ iles/WHA54/ea54id4.pd . X Oc obe
2023.
[2] D.M. DiMaggio, N. Du, C. Sche e , S. B odlie, V. Shabano a, P. Belama ich, A.
F. Po o, Compa ison o impo ed Eu opean and US in an o mulas: labeling,
nu ien and sa e y conce ns, J. Pedia . Gas oen e ol. Nu . 69 (2019) 480–486,
h ps://doi.o g/10.1097/mpg.0000000000002395.
[3] I. Bi louez-A agon, M. Pische s iede , J. Lecle e, F.J. Mo ales, K. Hasenkop ,
R. Kien sch-Engel, C.J. Ducauze, D. Ru ledge, Assessmen o p o ein glyca ion
ma ke s in in an o mulas, Food Chem. 87 (2004) 253–259, h ps://doi.o g/
10.1016/j. oodchem.2003.11.019.
[4] B.M. Meh a, H.C. Dee h, Blocked lysine in dai y p oduc s: o ma ion, occu ence,
analysis, and nu i ional implica ions, Comp . Re . Food Sci. Food Sa . 15 (2016)
206–218, h ps://doi.o g/10.1111/1541-4337.12178.
[5] M. Li, M. Shen, J. Lu, J. Yang, Y. Huang, L. Liu, H. Fan, J. Xie, M. Xie, Mailla d
eac ion ha m ul p oduc s in dai y p oduc s: o ma ion, occu ence, analysis, and
mi iga ion s a egies, Food Res. In . 151 (2022) 110839, h ps://doi.o g/10.1016/
j. ood es.2021.110839.
[6] A. Kishimo o, S. Nomu a, K. Tanaka, Chemical sensing o ac olein-amine
conjuga es o ood quali y con ol: a case s udy o milk p oduc s, Bull. Chem. Soc.
Jpn. 92 (2019) 1018–1023, h ps://doi.o g/10.1246/bcsj.20190010.
[7] J.A. Cus odio-Mendoza, C. Caama˜
no-Fe nandez, M.A. Lage, P.J. Almeida, R.
A. Lo enzo, A.M. Ca o, GC–MS de e mina ion o malondialdehyde, ac olein, and
4-hyd oxy-2-nonenal by ul asound-assis ed dispe si e liquid-liquid
mic oex ac ion in be e ages, Food Chem. 384 (2022) 132530, h ps://doi.o g/
10.1016/j. oodchem.2022.132530.
[8] In e na ional Agency o Resea ch on Cance , Malonaldehyde, IARC Monog aphs
on he E alua ion o Ca cinogenic Risks o Humans, 1999. h ps://monog aphs.ia
c. /wp-con en /uploads/2018/08/14-002.pd . (Accessed 1 Augus 2023).
[9] In e na ional Agency o Resea ch on Cance , Me hylglyoxal, IARC Monog aphs on
he E alua ion o Ca cinogenic Risks o Humans, Co ee, Tea, Ma e,
Me hylxan hines and Me hylglyoxal, IARC, Lyon-F ance, 1991, pp. 443–458. h p
s://publica ions.ia c. /Book-And-Repo -Se ies/Ia c-Monog aphs-On-The-Iden i i
ca ion-O -Ca cinogenic-Haza ds-To-Humans/Co ee-Tea-Ma e-Me hylxan hines-An
d-Me hylglyoxal-1991.
[10] Eu opean Food Sa e y Au ho i y (EFSA), Opinion o he Scien i ic Panel on ood
addi i es, la ou ings, p ocessing aids and ma e ials in con ac wi h ood (AFC)
ela ed o use o o maldehyde as a p ese a i e du ing he manu ac u e and
p epa a ion o ood addi i es, EFSA J. 5 (2007) 415, h ps://doi.o g/10.2903/j.
e sa.2007.415.
[11] A. Papas e giadis, A. Fa ouh, L. Jacxsens, C. Lacha , K. Sh es ha, J. Daelman,
P. Kols e en, H. Van Langenho e, B. De Meulenae , Exposu e assessmen o
J.A. Cus odio-Mendoza e al.
