赤霉素检测

Journal of Chromatography A,1416(2015)64–73Contents lists available at ScienceDirectJournal of ChromatographyAj o u r n a l h o m e p a g e :w w w.e l s e v i e r.c o m /l o c a t e /c h r o maOne-pot preparation of a mixed-mode organic-silica hybridmonolithic capillary column and its application in determination of endogenous gibberellins in plant tissuesZheng Zhang,Yan-Hong Hao,Jun Ding,Sheng-Nan Xu,Bi-Feng Yuan,Yu-Qi Feng ∗Key Laboratory of Analytical Chemistry for Biology and Medicine (Ministry of Education),Department of Chemistry,Wuhan University,Wuhan 430072,PR Chinaa r t i c l ei n f oArticle history:Received 4April 2015Received in revised form 12August 2015Accepted 30August 2015Available online 5September 2015Keywords:Organic-silica hybrid monolithic capillary columnPipette tip solid phase extractionCapillary liquid chromatography–mass spectrometry Derivatization Gibberellina b s t r a c tA newly improved one-pot method,based on “thiol-ene”click chemistry and sol–gel approach in microemulsion system,was developed for the preparation of C 8/PO(OH)2-silica hybrid monolithic cap-illary column.The prepared monolith possesses large specific surface area,narrow mesopore size distribution and high column efficiency.The monolithic column was demonstrated to have cation exchange/reversed-phase (CX/RP)mixed-mode retention for analytes on nano-liquid chromatography (nano-LC).On the basis of the developed nano-LC system with MS detector coupled to pipette tip solid phase extraction (PT-SPE)and derivatization process,we then realized simultaneous determination of 10gibberellins (GAs)with low limits of detection (LODs,0.003–0.025ng/mL).Furthermore,6endogenous GAs in only 5mg rice leaves (fresh weight)were successfully detected and quantified.The developed PT-SPE-nano-LC–MS strategy may offer promising applications in the determination of low abundant bioactive molecules from complex matrix.©2015Elsevier B.V.All rights reserved.1.IntroductionOrganic-silica hybrid monolithic stationary phases have been widely applied in capillary liquid chromatography (c LC)and capil-lary electrochromatography (CEC)due to their unique properties,such as fast mass transfer,good permeability and high column efficiency [1].On the other hand,mixed-mode stationary phases have recently attracted extensive attention in HPLC analysis of small molecules.Nowadays,mixed-mode packed columns have been commercial available and widely used,and better separation can be achieved compared to single-mode packed columns [2–4].However,the reported methods for the preparation of mixed-mode organic-silica hybrid monolithic capillary column often need fussy multistep.Though some one-pot strategies were recently developed [1,5,6],the organic-silica hybrid monolithic capillary column with strong hydrophobicity could not be prepared through these reported methods due to the incompatibility of hydropho-bic monomer in sol–gel system.Recently,Lin et al.developed a one-pot method for preparing organic-silica hybrid mono-lithic capillary column with glutathione as functional group [7].∗Corresponding author.E-mail address:yqfeng@ (Y.-Q.Feng).However,the application is limited due to less monomer.There-fore,significant challenges remain in the development of facile and universal methods for preparation of mixed-mode organic-silica hybrid monolithic capillary column.Gibberellins (GAs),a class of plant growth hormones,play essen-tial roles in the regulation of plant growth and developmental processes,including stem elongation,germination,flowering and fruit development [8,9].Natural GAs are a group of diterpenoid acids and exist in almost all green plants [10].Determination of endogenous GAs is critical for the in-depth understanding of their biological functions in plants.However,effective detection of GAs is normally difficult due to the low abundance of endogenous GAs present in plants as well as the serious effect of plant matrix [11,12].Therefore,development of sensitive and selective analytical meth-ods for the determination of GAs is highly needed.Liquid chromatography–mass spectrometry (LC–MS)has been demonstrated to be a powerful platform for sensitive determina-tion of phytohormones [13–15].However,the MS-based methods are still restricted to achieve satisfactory MS response for nega-tively charged carboxylated phytohormones such as GAs,mainly due to their low ionization efficiency [16],matrix effect [17],and co-elution of analytes [18].In this respect,derivatization is an effec-tive strategy and frequently used to improve the MS detection sensitivity of target compounds through labeling the compounds/10.1016/j.chroma.2015.08.0710021-9673/©2015Elsevier B.V.All rights reserved.Z.Zhang et al./J.Chromatogr.A1416(2015)64–7365with easily ionized group[19–23].For example,to enhance the ionization efficiency,a derivatization reagent(mass probe)with quaternary amine group was employed to label GAs[24].How-ever,the separation efficiencies of some GA derivatives,such as GA4,GA51,GA1and GA3are very poor on C18column.Similarly, GA derivatives generated from other derivatization reagents with tertiary amine group also have the same problem[23,25].Thus, effective separation of these GA derivatives in suitable separation mode is favorable.In the meantime,although conventional particulate packed columns were widely employed for GAs analysis,a relatively large plant amount was normally required.Nano-LC was demon-strated to be an effective approach to reduce the sample amount and improve detection sensitivity[26].Recently our group devel-oped a nano-LC–MS platform by using porous polymer monolithic capillary column as the separation medium for the sensitive determination of phytohormones in plant tissues[24],and5-methylcytosine and5-hydroxymethylcytosine in genomic DNA [27].In this respect,monolithic stationary phases are promising alternative for nano-LC due to the good permeability and broad selectivity.Considering for the chemical properties of mass probe-derived GAs and the advantages of monolithic stationary phase, a hydrophobic/cation exchange monolithic capillary column may provide good separation resolution for the target analytes.In this current study,a newly improved one-pot method,based on“thiol-ene”click chemistry and sol–gel approach in microemul-sion system,was developed for the preparation of C8/PO(OH)2-silica hybrid monolithic capillary column.The prepared monolith possesses large specific surface area,narrow mesopore size dis-tribution and high column efficiency.The monolithic column was demonstrated to have cation exchange/reversed-phase(CX/RP) mixed-mode retention for analytes on nano-LC.To develop a method for determination of endogenous GAs in plant tissues on the basis of nano-LC–MS with CX/RP monolithic capillary column, we also proposed a sample preparation strategy combining pipette tip solid phase extraction(PT-SPE)and derivatization process. Taken together,we achieved the simultaneous determination of10 GAs with low limits of detection(LODs,0.004–0.032ng/mL)using PT-SPE-nano-LC–MS system.Furthermore,the developed method was successfully applied to the determination of endogenous GAs in only5mg rice leaves(fresh weight).2.Materials and methods2.1.Chemicals2,2-Azobisisobutyronitrile(AIBN),urea(>95%,w/w),sodium dodecyl sulfonate(SDS,>95%,w/w),ammonia water(contain-ing25–28%NH3,w/w),and poly(ethylene glycol)with the molecular weight of10,000(PEG-10000)were all purchased from Shanghai Chemical Reagent Corporation(Shanghai,China). AIBN was purified by recrystallization from ethanol at40◦C. Tetramethoxysilane(TMOS,98%,w/w),n-octyltrimethoxysilane (C8-TMOS,98%,w/w),octadecyltrimethoxysilane(C18-TMOS,98%, w/w)and␥-mercaptopropyltrimethoxysilane(SH-TMOS,98%, w/w)were purchased from Wuhan University Silicone New Mate-rial(Wuhan,China).The fused-silica capillaries(75␮m i.d.,360␮m o.d.)were purchased from Yongnian Optic Fiber Plant(Hebei, China).Stable isotope-labeled compounds and standards,[2H2] GA1(internal tracers of GA1and GA3),[2H2]GA4(internal trac-ers of GA4and GA7),[2H2]GA9(internal tracers of GA9),[2H2]GA19 (internal tracers of GA19and GA44),[2H2]GA51(internal tracers of GA5and GA51),[2H2]GA53(internal tracers of GA53),GA1,GA3,GA4, GA5,GA7,GA9,GA19,GA44,GA51,GA53,Zeatin-9-glucoside(Z9G), zeatin-riboside(ZR),N6-isopentenyladenine-9-glucoside(iP9G),isopenteny-ladenine riboside(iPR),and zeatin(Z)were purchased from Olchemim Ltd.(Olomouc,Czech Republic).The structure of GAs can be seen in Fig.S1.Thiourea(>95%,w/w),toluene,ethyl-benzene(>95%,w/w),propylbenzene(>95%,w/w),ammonium formate(HCOONH4,98%,w/w),formic acid(FA,>88%,w/w),and triethylamine(TEA,98%,w/w)were purchased from Shanghai General Chemical Reagent Factory(Shanghai,China).2-Chloro-1-methyl-pyridinium iodide(CMPI,98%,w/w)and N,N-diethyl ethylenediamine(DEED)(98%,w/w)were purchased from Aladdin (Shanghai,China).HPLC-grade acetonitrile(ACN)were obtained from TEDIA Company Inc.(OH,USA).Milli-Q water(Millipore,Brad-ford,USA)was used in all experiments.HiCapt SAX SPE adsorbent, a strong anion exchanger containing quaternary ammonium group, was purchased from Weltech Company(Wuhan,China).The parti-cle size was labeled as200–300mesh.The pipette tips for PT-SPE were prepared by packing the SAX adsorbent into a200-␮L pipette tip,with a polyethylene frit for blocking the SAX sorbent.The frits(approximately1.0mm diameter,1.4mm thickness)were pur-chased from Biocomma(Shenzhen,China).All other reagents were of analytical reagent grade unless otherwise indicated.2.2.Preparation of the monolithic capillary columnsThe fused-silica capillaries were washed with1mol/L NaOH (2h),H2O(30min),1mol/L HCl(1h),H2O(30min),and methanol (30min)successively to activate the silanol groups.Then the capillaries were allowed to dry under nitrogenflow at160◦C for 5h.C8/PO(OH)2-silica hybrid monolithic capillary column was pre-pared using improved one-pot approach.Typically,H2O(500mg, 61.1%w/w total),PEG-10000(90mg,11.0%w/w total),ammo-nium hydroxide(7.5mg,0.9%w/w total)and vinylphosphonic acid (10mg,1.2%w/w total)were mixed to form solution A.TMOS (185mg,22.6%w/w total),SH-TMOS(10mg,1.2%w/w total),n-octyltrimethoxysilane(15mg,1.8%w/w total),SDS(1mg,0.1%w/w total)and AIBN(1mg,10%w/w SH-TMOS)was mixed to form solu-tion B.Subsequently,212mg of solution B was added to607mg of solution A and degassed by a10-min ultrasonication.The homoge-neous mixture was then manually introduced into the fused-silica capillary to an appropriate length by a syringe.After both ends of the capillary were sealed with two pieces of rubber,the mixture was incubated at40◦C for12h for simultaneous polycondensation and“thiol-ene”click reaction.The resulting monolith was com-pletelyflushed with water and ACN successively to remove the PEG and other residuals.For comparison,a PO(OH)2-silica hybrid monolithic capillary column(30cm-long,75␮m i.d.,360␮m o.d.)was also prepared. The preparation procedure is the same as that for the preparation of C8/PO(OH)2-silica hybrid monolithic capillary column but without the addition of C8-TMOS.C18-silica monolithic capillary column(30cm-long,75␮m i.d., 360␮m o.d.)used for control experiment was prepared accord-ing to previous report[28].Detail information can be seen in Supporting Information.2.3.Characterization of the monolithic capillary columnsThe surface area and pore size distribution were measured by a specific surface area and pore size distribution analyzer(Beijing JWGB Sci.&Tech.,Beijing,China).The microscopic morphology was examined by scanning electron microscope(SEM),using a Quanta 200scanning electron microscope(FEI Company,Holland).The ele-mental contents of the prepared monoliths were determined on Shimadzu EDX-720energy-dispersive X-ray analysis(EDX,Kyoto, Japan)by using Mg–Ka radiation as the excitation source.Per-meability measurements were performed by using a Shimadzu66Z.Zhang et al./J.Chromatogr.A1416(2015)64–73 LC-10AT pump(Kyoto,Japan)under the constantflow mode.ACNwas pumped through the prepared monolithic column(10-cm long,75␮m i.d.,360␮m o.d.)at aflow rate of2␮L min−1.The back pres-sure was recorded when the pressure stabilized.Permeability(K)was calculated according to Darcy’s Law by using Eq.(a),in whichu(ms−1)is the linear velocity of the mobile phase,Áis the viscosityof the mobile phase(0.38×10−3Pa s for ACN at20◦C in the currentstudy),L is the length of the monolithic column(m),and P is thepressure drop across the monolithic column(Pa).K=uÁLP(a)The nano-LC column evaluation experiments were performed on a Shimadzu nano-LC system containing one FVC valve of two positions with a splitter,a sample loop(5␮L),two Shimadzu LC-20A pumps(Tokyo,Japan)and one GL Sciences MU701UV–vis detector with a6nL detection cell(Tokyo,Japan).The actual flow rate in nano-LC system can be controlled by changing the splitting ratio and keeping the totalflow rate unchanged.The detection wavelength of UV detector was set at254nm for CKs and214nm for other analytes.Data acquisition and processing were performed using LabSolutions software(version5.53sp2, Shimadzu,Tokyo,Japan).After connecting to the nano-LC sys-tem,the organic-silica hybrid monolithic capillary columns were conditioned with the mobile phase for30min.For chromato-graphic separation,a30cm long monolithic column(75␮m i.d., 360␮m o.d.)was employed.Van-Deemter plots were obtained by plotting the height equivalent to a theoretical plate(HETP)of ace-tophenone,toluene and aniline,versus linear velocity of mobile phase.The effect of mobile phase on the retention of DEED-GAs was investigated on the nano-LC–MS system.Subsequently,the sepa-ration of DEED-GA standards,method validation and detection of endogenous GAs in rice leaves were performed on the same sys-tem,with an additional use of a C18on-line trapping column(5␮m, 5mm×0.3mm,Agilent Techologies,Inc.).Detailed nano-LC–MS conditions can be seen in Section2.7.2.4.Preparation of plant samplesPlant sample(rice)was grown in growth chambers in a16h light/8h dark photoperiod with80%humidity under28◦C.Light intensity wasfixed to120lx/m2/s.After approximate7days grow-ing,rice leaves were collected,weighted,immediately frozen in liquid nitrogen,and then stored at−80◦C.To extract phytohor-mones,plant samples(5mg)were frozen in liquid nitrogen and finely grounded followed by extraction with50␮L ACN at4◦C for12h.[2H2]GA1(2.00ng/g),[2H2]GA4(2.00ng/g),[2H2]GA9 (2.00ng/g),[2H2]GA19(2.00ng/g),[2H2]GA51(2.00ng/g),and[2H2] GA53(2.00ng/g)were added to plant samples as internal standards(I.S.)prior to grinding.2.5.Purification of GAs by PT-SPEAfter extraction of plant samples,the supernatants(50␮L) were collected by centrifugation at10,000rpm for5min.Then the supernatants were alkalized with0.5␮L ammonia water(25–28%, NH3,w/w),and pipetted up and down20times to allow analytes adsorbed on the SAX adsorbents.(The PT-SPE was pre-conditioned with50␮L1%FA in ACN,50␮L H2O,and50␮L ACN for three times before use).After sample loading,the PT-SPE was rinsed with50␮L ACN,50␮L ACN/H2O(v/v,9/1),50␮L ACN/H2O(v/v,1/9)and50␮L ACN for three times.After that,200␮L ACN with1%FA(v/v)was applied to elute the adsorbed GAs.2.6.Derivatization of GAsThe eluted GAs solution from PT-SPE was dried under nitrogen gas and reconstituted in100␮L ACN.To the resultant solution,2␮L triethylamine(TEA)(20␮mol/mL)and2␮L CMPI(20␮mol/mL) were added and vortexed for5min.DEED(2␮L,40␮mol/mL)was subsequently added and the reaction solution was vortexed for1h at35◦C.The reaction mixture was then evaporated under nitrogen stream followed by re-dissolving in20␮L H2O/ACN/FA(950/49/1, v/v/v)for further analysis.2.7.Nano-LC–MS conditionsAnalysis of samples was performed on the nano-LC–MS sys-tem consisting of an AB4500QTRAP mass spectrometer(Applied Biosystems,Foster City,CA,USA)with an electrospray ionization source(Turbo Ionspray),an Eksgent nano-flow liquid chromatog-raphy system including a binary solvent manager and sample manager.Data acquisition and processing were performed using AB SCIEX Analyst1.5Software(Applied Biosystems,Foster City,CA, USA).H2O/ACN/FA(950/49/1,v/v/v,solvent A)and H2O/ACN/FA (49/950/1,v/v/v,solvent B)were employed as mobile phase for the separation of DEED-GAs.To enhance the capacity of the analytical system,a C18column(5␮m,5mm×0.3mm,Agilent Techologies, Inc.)was employed for online trapping.Five microliter of sample solution was automatically injected into the trapping column at a flow rate of5␮L min−1for10min with mobile phase A.The trapped DEED-GAs were then separated on the C8/PO(OH)2-silica hybrid monolithic capillary column with a gradient elution at aflow rate of500nL/min.The gradient condition was set as0–15min5–40% B,15–40min40–55%B,40–50min55–95%B and50–60min95%B.The MS analysis was performed under MRM mode(positive mode),and the parameters were set as follows:sheath gas(nitro-gen)flow rate,50arb;aux/sweep gas(nitrogen)flow rate,10arb; spray voltage,1.8kV;capillary temperature,150◦C;capillary volt-age,−4V;tube lens offset,−5V.Helium was used as the collision gas in the ion trap.Optimal parameters for analytes are listed in Table S1in Supporting Information.3.Results and discussion3.1.Preparation of the C8/PO(OH)2-silica hybrid monolithic capillary columnThe schematic procedure for the preparation of C8/PO(OH)2-silica hybrid monolithic capillary column is shown in Fig.1As the hydrophobic C8-TMOS could not dissolve or disperse uniformly in polymerization mixture,SDS was adopted to form a homogeneous microemulsion environment.Two major processes were involved in the preparation of C8/PO(OH)2-silica hybrid monolithic capil-lary column:(1)hydrolysis and condensation of TMOS,C8-TMOS and SH-TMOS;(2)“thiol-ene”click reaction between vinylphos-phonic acid and SH-end silica monolithic matrix.In the one-pot approach,a mixed-mode capillary monolithic column with homo-geneous structure can be achieved.The morphology of monolithic columns can befine-tuned by changing conditions such as the composition of the polymeriza-tion mixture.Herein,several parameters affecting the formation of C8/PO(OH)2-silica hybrid monolithic capillary column were evalu-ated(Table1).The content of PEG will affect the rate of phase separation,thus resulting in a significant change on the porosity of silica monoliths [28].As shown in Table1,decreased values of permeability were observed as the content of PEG increased(Column A,B and C),Z.Zhang et al./J.Chromatogr.A1416(2015)64–7367Fig.1.The scheme for the one-pot preparation of C8/PO(OH)2-silica hybrid monolithic capillary column.indicating decreased skeleton and pore sizes of monoliths.The effects of C8-TMOS and SH-TMOS content on the hybrid mono-liths were also investigated.Although the condensation rate may be affected by the content of C8-TMOS,no apparent changes of col-umn permeability(RSD13.7%)and optical microscope image were observed as the content change of C8-TMOS(Column B,D and E).As shown in Table1,the permeability of the as-prepared hybrid mono-liths was dependent on the content of SH-TMOS.With decreasing the content of SH-TMOS,no apparent change in the permeability of Column G and B was observed;while the permeability of Col-umn F greatly decreased with the decrease of SH-TMOS.We reason that pH of the polymerization mixture will increase due to the click reaction between SH-TMOS and vinylphosphonic acid(the p K a of phosphate groups will increase from1.5to2.7after coupling to SH-TMOS),resulting in the change of condensation rate.Therefore, when the content of SH-TMOS was reduced to some degree,the pH of reaction system may become relatively low,thus resulting to a relatively fast condensation rate and more compact structure. Considering an appropriate permeability and homogeneous mor-phology,Column B was employed for further experiments.3.2.Characterization of the C8/PO(OH)2-silica hybrid monolithic capillary columnTo account for the effective incorporation of C8groups into the C8/PO(OH)2-silica hybrid monolithic capillary column,a com-parison of C8/PO(OH)2-silica hybrid monolithic capillary column and PO(OH)2-silica hybrid monolithic capillary column for the separation of alkylbenzenes was performed(Fig.2A).It can be seen that alkylbenzenes exhibited stronger retention on the C8/PO(OH)2-silica hybrid monolithic capillary column than on the PO(OH)2-silica hybrid monolithic capillary column,indicating that C8/PO(OH)2-silica hybrid monolithic capillary column had higher hydrophobility due to the introduction of C8group.The percentage of O,Si,S,and P(w/w)of the C8/PO(OH)2-silica hybrid mono-lithic capillary column were determined by EDX to be28.7%, 63.45%,4.08%,and3.77%,respectively(Fig.2B),indicating that phosphate and mercapto groups were successfully incorporated into the monolith.The atom ratio of P to S was1.12(the calculated value was1),suggesting a complete consume of mercapto group.The porous structure of the C8/PO(OH)2-silica hybrid monolithic capillary column was measured by nitrogen adsorption–desorption measurement,and SEM.The specific surface area of the monolith was found to be230m2/g with a narrow mesopore distribution (Fig.2C).The adsorption/desorption isotherm indicated a meso-porous skeleton of prepared monolith(Fig.2D).Continuous silica monolithic network was observed on the C8/PO(OH)2-silica hybrid monolith and no shrink was found in the capillary(Fig.2E and F),indicating a stable and homogeneous network structure.Taken together,the results demonstrated the C8/PO(OH)2-silica hybrid monolithic capillary column had homogeneous porous structure and large specific surface area.3.3.Chromatographic evaluation of the C8/PO(OH)2-silica hybrid monolithic capillary columnThe lowest plate heights of approximate7.6␮m (∼132,000plates/m),7.1␮m(∼141,000plates/m)and10.8␮mTable1Optimization of the preparation conditions of C8/PO(OH)2-silica hybrid monolithic capillary columns.Column SH-TMOS(mg)C8-TMOS(mg)PEG-10000(mg)Permeability(×10−14m2)Optical microscope imageA13158015.3B1315905.4C1315100Hard topumpD1310906.3E1320904.8F81590Hard topumpG1815906.3The amount of H2O,ammonium hydroxide,vinylphosphonic acid,TMOS,SDS and AIBN wasfixed and can be seen in Section2.2.68Z.Zhang et al./J.Chromatogr.A 1416(2015)64–73Fig.2.(A)Comparison of the separation of alkylbenzenes on C 8/PO(OH)2-silica hybrid monolithic capillary column and PO(OH)2-silica hybrid monolithic capillary column (15cm-long,75␮m i.d.,360␮m o.d.).(B)EDX patterns of C 8/PO(OH)2-silica hybrid monolithic capillary column.(C)Specific surface area and pore distribution of C 8/PO(OH)2-silica hybrid monolithic capillary column.(D)The N 2adsorption/desorption isotherm of C 8/PO(OH)2-silica hybrid monolithic capillary column.(E,F)SEM images of the cross section of the C 8/PO(OH)2-silica hybrid monolithic capillary column.Scale bars,(E)50␮m;(F)5␮m.Experimental condition for (A):flow rate,400nL/min;mobile phase,H 2O/ACN,70/30(v/v);detection wavelength,214nm;temperature,25◦C.Analytes:1,thiourea;2,toluene;3,ethylbenezene;4,propylbenezene.All of analytes were dissolved with mobile phase and the concentrations of them were 10mg/L.The injection volume was 5␮L with splitting.(∼93,000plates/m)were obtained on the C 8/PO(OH)2-silica hybrid monolithic capillary column,using acetophenone,toluene,and aniline as probes,respectively (Fig.3).The high column efficiency indicated a uniform structure of the prepared monolith.To explore the retention mechanism of the C 8/PO(OH)2-silica hybrid monolithic capillary column,alkylbenzenes and cytokinins were chosen as probes.As shown in Fig.4A,a linear relationship of ln k versus ACN content can be observed for alkylbenzenes,indicating the RP retention character of the monolithic column.On the other hand,a linear relationship of k versus 1/[HCOONH 4]can be observed (Fig.4B),indicating the ion-exchange retention mechanism.According to the two-site models of mixed-mode retention mechanisms,we can confirm the contribution of reversed phase mechanism to separation of CKs,for the inter-cepts of fitting equation in Fig.4B were positive.Subsequently,the C 8/PO(OH)2-silica hybrid monolithic capillary column wasZ.Zhang et al./J.Chromatogr.A1416(2015)64–7369Fig.3.Van Deemter plot of the height equivalent to a theoretical plate as a func-tion of linear velocity.Experimental condition:analytical column,C8/PO(OH)2-silica hybrid monolithic capillary column(15cm-long,75␮m i.d.,360␮m o.d.).Mobile phase,H2O/ACN,75/25(v/v).Detection wavelength,214nm;temperature,25◦C. All of analytes were dissolved with mobile phase and the concentrations of them were10mg/L.The injection volume was5␮L with splitting.employed to explore the retention behavior toward DEED-GAs.As shown in Fig.5A,the retention time decreased dramatically with the increase of ACN content.In addition,a dramatic decrease in the retention as the increase of salt concentration from10to25mM was observed(Fig.5B).Taken together,the results demonstrated the cation exchange/hydrophobic retention mechanism of the C8/PO(OH)2-silica hybrid monolithic capillary column.In order to confirm that CX/RP stationary phase can improve the separation for DEED-GAs,we made a comparison for separa-tion of them on the C8/PO(OH)2-silica hybrid monolithic capillary column and C18-silica monolithic capillary column.As shown in Fig.S2,co-elutions for GA1and GA3and much lower resolution could be observed on C18-silica monolithic capillary column The CX/RP monolithic column exhibited better separation selectivity and stronger retention for DEED-GAs.Therefore,the C8/PO(OH)2-silica hybrid monolithic capillary column monolithic column will be further investigated.To investigate the reproducibility of the C8/PO(OH)2-silica hybrid monolithic capillary column,two additional batches were prepared.The reproducibility of three batches was chromato-graphically evaluated on the nano-LC–MS platform,using3GA derivatives(DEED-GA3,DEED-GA9and DEED-GA51)as probes.As shown in Fig.S3,the batch-to-batch reproducibility of the retention factor k values was obtained with RSD being1.5%,2.3%and1.4%,for GA9,GA3,GA51,respectively,indicating an excellent reproducibility of the preparation strategy for the monoliths.3.4.Development of PT-SPE method and derivatizationGA analysis is commonly hampered by the complicated sam-ple matrix,low endogenous content,and poor MS responses.To improve the MS response of GAs,chemical derivatization of GAs was performed according to our previous work[25],through label-ing with a tertiary amine group.The derivatization was completed through a reaction between the amino group of DEED and the car-boxyl group of GAs,with the help of catalysts(TEA and CMPI).After derivatization,the detection mode was changed into positivemode, Fig.4.(A)The plot of log k versus ACN content(%)for the separation of alkylbenzenes.(B)The plot of k versus1/[HCOONH4mM]for the separation of CKs.Experimental conditions:analytical column,C8/PO(OH)2-silica hybrid monolithic capillary column(30cm-long,75␮m i.d.,360␮m o.d.).Flow rate,500nL/min.Mobile phase:H2O/ACN (v/v)for A,HCOONH4(pH4.5)/ACN,(80/20,v/v)for B.Detection wavelength,214nm for A and254nm for B;temperature,25◦C.All of analytes were dissolved with mobile phase and the concentrations of them were10mg/L.The injection volume was5␮L withsplitting.Fig.5.(A)The effect of ACN content on the retention of DEED-GAs.(B)The effect of HCOONH4content on the retention of DEED-GAs.Experimental conditions:analytical column,C8/PO(OH)2-silica hybrid monolithic capillary column(30cm-long,75␮m i.d.,360␮m o.d.).Flow rate,500nL/min.Mobile phase:10mM HCOONH4(pH4)/ACN for A and HCOONH4(pH4)/ACN(70/30,v/v).Temperature,25◦C.The analytes were dissolved with mobile phase and the concentrations of them were10ng/mL.The injection volume was100nL.。