李笃信, 张凌怡, 李 彤, 杜一平, 张维冰*
(1.华东理工大学,上海 200237;2.大连依利特分析仪器有限公司,辽宁大连 116023)
In recent years,the com p rehensive tw o-dim ensional liquid chrom atography(2D-LC)has gained m uch attention forthe separation ofcom p lex sam p les in various fields of research[1,2].Several2D-LC system s have been developed through the past over a decade[3-5].Colum n sw itching betw een the tw o dim ensions is a delicate operation to design and imp lem ent for its functions of connections,valving,storage and enrichm ent[6].
There are several kinds of interfaces[7,8] have been used in online2D-LC system,like samp le loop,parallel colum n and trap colum n interface.Sam p le loop interface is w idely used in2DLC system. How ever,sm all loop volum e lim its the flow rate and colum n dim ension of the first dim ension.The use of parallel colum n in the second dim ension colum ns can carry out several second dim ension separations sim ultaneously [9]. A lthough this im p roves the overall speed of the 2D-LC analysis,it requires m ore elaborate instrum entation,and causes difficulty in the data analysis[10,11].
The trap colum n interface[12]refers to packing suitable adsorbent in the sam p le loop to collect the fractions eluted from the first dim ension colum n and storing them for periodic re-injection into the second dim ension colum n.W hile one colum n traps the com ponents at its head,the other colum n back-flushes to induce the com ponents to the second dim ension separation[6,13].
Excep t for the various com binations of the tw o dim ensions,m ost system s em phasized the separation speed of the second d im ension[14,15]. Several kinds of the fast separation(fast gradient elution or/and short colum n)w ere used in the second dim ension in the online2D-LC system [16 -19]. How ever, the second d im ension could achieve high separation speed but lose efficiency w hen a shorter colum n w as used.In this paper,a longer trap colum n w as em p loyed to construct an interface that allow ed increased separation efficiency in the second dim ension w ithout affecting the valve sw itching frequency. The com ponents could be p re-separated through interface before entering to the second dim ension.The rat serum sam p le w as used to evaluate the perform ance of the2D-LC system.
A ll of the w ater used in the experim ent w as p repared using a Sartorius A rium 611 System (Sartorius, Germ any). Trifluoroacetic acid (99%)w as obtained from J&K Chem ical(B eijing,China).Acetic acid,sodium acetate and N aCl of analytical grade w ere purchased from Sinopharm Chem ical Reagent Co.,L td. (Shanghai,China).Acetonitrile(HPLC gradient grade) w as purchased from Xingke(Shanghai Xingke B iochem istry Co.,Ltd.(Shanghai,China).Rat serum sam p le w as kindly donated by Laboratory of System s B iology (Institutesof B iom edical Sciences,Fudan U niversity).Rat blood w as centrifuged at5 000r/m in for10m in;the upper layer w as taken as the sam p le.
For the first dim ension separation,a standard Elite230Series chrom atographic instrum ent controlled by EC2000Chem station softw are version v1.5 (EliteHPLC L td.,D alian, China)w as used.This instrum ent w as equipped w ith M odel 7725injection valve (Rheodyne,USA)w ith a hom e m ade300μL sam p le loop and abinary pum p ing system.A w eak anion exchange(WAX) colum n(AX250mm×4.6mm,5μm,B ioBasic) w as used as the first dim ension.A reversed-phase (RP)colum n(ODS-BP50mm ×4.6mm,5μm, SinoChrom)w as chosen as the second d im ension. Independent binary pump system identical to first dim ension w as used.Tw o RP colum ns(ODS-BP 35mm ×4.6mm,5μm,SinoChrom)w ere connected by10-port valve(M odel EV750-102,Rheodyne,USA)through0.01″i.d. PEEK tubing to serve as trap colum n.The UV detector w as set at the end of the second dim ension colum n.
In order to increase the speed of the second dim ensional separation,fast separation of second dim ension is often used in the2D-LC system s.A lthough it increases the valve sw itching the frequency,it also causes colum n efficiency loss. In our w ork,to acquire the enhanced colum n efficiency and m aintain the separation speed,longer trap colum ns w ereem p loyed and w erealw ays flushed forw ard w hatever in the loading status or elution status. The trap colum ns had tw o functions:one w as to trap the elution;the other w as to p re-separate the com ponents retained on the colum n.The p rincip les of im p roved trap colum n interface are show n in Fig.1.
F ig.1 Princip les of the in te rface in2D-LC system
Colum n1and2,served as thetrap colum ns and had the function of p re-separation.A t position1,colum n1w as connected to the first dim ension.The fractions from the p revious dim ension w ere cap tured by colum n1and retain at the head of the colum n.Solvent of the first dim ension w as drained to the w aste directly.W hen changed to position2,the colum n1w as connected to the second dim ension directly.Increased organic solvent gradient from the pum p system eluted the colum n1and second d im ension colum n in series. Com ponent enriched at the head of colum n1in position1got retention at colum n1and the second d im ension colum n,and w ere eluted by the organic solvent gradient gradually.A t position2, colum n1perform ed the function of p re-separation.Sim ultaneously,colum n2p lays the role of trap colum n.W hen the separation of the second dim ension com p leted,the valve w as sw itched to the position1,and colum n2w as eluted by the pum p system and the circulation p rocess started. The p re-separation w as carried out through the longer trap colum n before com ponents arrived the second d im ension.The sw itching frequency of the valve can increased the separation efficiency of the second dim ension.
F ig.2 O ne dim ensiona l sepa ra tions of a ra t se rum sam p le(a.IEX sepa ra tion;b.RP sep a ra tion)Conditions:a.B ioBasic AX colum n,250mm ×4.6mm,5 μm;m obile phases,(A)20mm ol/L Tris-HCl(pH7.5)and (B)20mm ol/L Tris-HCl,3m ol/L NaCl; flow rate,0.2 mL/m in;gradient elution,0-60m in,0%B-60%B;detection w avelength,280nm;b.SinoChrom ODS-BP colum n,250mm ×4.6mm,5μm;m obile phases,(A)w ater containing0.1% TFA and(B)acetonitrile containing0.1%TFA;flow rate,1 mL/m in;gradient elution,0-8m in,30%B-40%B;8-40 m in,40%B-60%B;detection w avelength,280nm.
Before the2D-LC analysis,the rat serum w as investigated by the one d im ensional m ode. The WAX colum n and the RP colum n w ere used and conditioned to get the op tim ized chrom atographic param eters. The chrom atogram s of the one dim ensional separation are show n in Fig.2.In Fig. 2a,only a few peaks are recognized in the ion exchange chrom atography(IEX)separation m ode, w hich did not give a pow erful resolution. The sam e situation exists in the RP separation in Fig. 2b.How ever,the one dim ensional separation w as restricted by its low er capacity.
W hen the35mm colum n w asused as the trap colum n to perform the p re-separation function, the separation pow er of the second d im ension, w hich w as50mm long,w ould theoretically increase by70%.B ecause of the volum e of the m ixer and the trap colum n,w hich generated a delay tim e,the t im e of m obile phase of the initial gradient reaching the separation colum n w aspostponed by2.5m in.As aresult, the valve w as sw itched w hen the gradient had passed2.5m in and data acquisition started at the sam e tim e.The m odulation tim e of the second dim ension w as5 m in. The chrom atogram s of rat serum sam p les separatedon this2D-LC system areshow n in Fig.3.
F ig.3 2D-LC chrom a togram s of aqueous ex tract of ra t se rum sam p lesConditions:first dim ension,B ioBasic AX colum n,250mm × 4.6mm,5μm;m obile phases,(A)20mm ol/L Tris-HCl(pH 7.5)and(B)3m ol/L NaCl,20mm ol/L Tris-HCl;flow rate: 0.2mL/m in;gradient elution,0-60m in,0%B-60%B;Second dim ension,SinoChrom ODS-BP colum n,50mm ×4.6 mm,5μm;m obile phases,(A)w ater containing0.1%TFA and (B) acetonitrile containing0.1%TFA; flow rate, 1 mL/m in;gradient elution,0-2.5m in,30%B-60%B;2.5-5 m in,0%B-30%B;detection w avelength,280nm.
A total of16fractions of the first dim ension w ere introduced to the second dim ension. The negative peak betw een0.8m in and1.6m in attributed to the UV absorbance of the m obile phase cap tured in the trap colum n.W hen com paring w ith the traditional trap colum n interface, the elution from0-0.8m in w as additional separation pow er p rovided by the forw ard-flushed longer trap colum n.
According to Shalliker’s report[20]that the viscous fingering(VF)phenom enon w as generated from the velocities difference of the m obile phases of the tw o dim ensions in the region they contacted directly.W hen the viscosity of injection p lug w as bigger than that of the m obile phase,VF generated at the rear of the injection p lug.O n the contrary,VF w ould be generated at the head of the injection p lug.Since the m obile phase in the first d im ension w as pure buffer,w hose viscosity w as bigger than that of m obile phase of the second d im ension,so the VF generated at around 1.6m in.M ass transfer in direct contact regions w as very slow.The intact region betw een the m obile phases of tw o dim ensions w as just like the m oving reaction boundary.So the desalting of the fraction and re-equilibrium of the colum n can be avoided.Fast gradient elution of the second dim ension can be carried out directly.
A WAX-RP2D-LC system w as constructed w ith im p roved trap colum n interface. This interface greatly increases the efficiency of the second dim ension colum n w ithout affecting the separation speed.W hen35mm trapcolum ns w ere used, com paring to a50mm second d im ensional colum n, the separation efficiency w as theoretically increased by70%. Prim ary separation of the rat serum sam p le w as used to analysis and evaluate the utility of the system. The VF phenom enon w as generated attributing to viscosity of the m obile phases of the tw o dim ensions.This interface w ould be an alternative selection to the traditional trap colum n interface in2D-LC system for the separation ofcom p lex m ixture in p roteom ics. Based on the im p roved interface, the WAX-RP 2D-LC system also could be app lied in the other areas such as the research of traditional Chinese m edicines.
[1] Dugo P,Cacciola F,Kumm T,et al.J Chrom atogr A, 2008,1184(1/2):353
[2] Issaq H J,Chan K C,Janini G M,et al.J Chrom atogr B, 2005,817(1):35
[3] Bushey M M,Jorgenson J W.Anal Chem,1990,62(10): 978
[4] M urphy R E,Schure M R,Foley J P.Anal Chem,1998,70 (8):1585
[5] Lan T,J iao F L,Tang T,et al.Chinese Journal of Chrom atography,2008,26(3):374
[6] Guiochon G,MarchettiN,M riziq K,et al.J Chrom atogr A, 2008,1189(1/2):109
[7] Shellie RA,Haddad P R.Anal B ioanal Chem,2006,386 (3):405
[8] W ang Z C,Zhang Q H,Zhao Z Y,et al.Chinese Journal of Analytical Chem istry,2005,33(5):722
[9] Tanaka N,Kim ura H,Tokuda D,et al.Anal Chem,2004, 76(5):1273.
[10] Felinger A,Kele M,Guiochon G.J Chrom atogr A,2001, 913(1/2):23
[11] W agner K,M iliotis T,M arko V G,et al.Anal Chem,2002, 74(4):809
[12] M aw uenyega K G,Kaji H,Yam auchi Y,et al.J Proteom e Res,2003,2(1):23
[13] Sw eeney A P,Shalliker R A.J Chrom atogr A,2002,968 (1/2):41
[14] StollD R,Carr P W.J Am Chem Soc,2005,127(14): 5034
[15] Stoll D R,Cohen J D,Carr P W.J Chrom atogr A,2006, 1122(1/2):123
[16] Cesla P,Hajek T,Jandera P.J Chrom atogr A,2009,1216 (16):3443
[17] Dugo P,Cacciola F,Herrero M,et al.J Sep Sci,2008,31 (19):3297
[18] Ikegam i T,Hara T,Kim ura H,et al.J Chrom atogr A, 2006,1106(1/2):112
[19] Im K,Park H W,Lee S,et al.J Chrom atogrA,2009,1216 (21):4606
[20] Shalliker R A,Catchpoole H J,Dennis G R,et al.J Chrom atogr A,2007,1142(1):48