401685

7/29/2019 401685

http://slidepdf.com/reader/full/401685 1/6

Talanta 52 (2000) 1105–1110

Effect of temperature on DNA fractionation in slalomchromatography

Eric Peyrin a,*, Yves C. Guillaume c, Catherine Garrel b, Anne Ravel a,Annick Villet a, Catherine Grosset a, Josette Alary a, Alain Favier b

a Laboratoire de Chimie Analytique, UFR de Pharmacie, Domaine de la Merci , 38700 La Tronche, Franceb LBSO, UFR de Pharmacie, Domaine de la Merci , 38700 La Tronche, France

c Laboratoire de Chimie Analytique, UFR de Medecine et Pharmacie, Place Saint-Jacques, 25030 Besancon Cedex, France

Received 20 March 2000; received in revised form 26 May 2000; accepted 31 May 2000

Abstract

Slalom chromatography is an alternative chromatographic procedure for the analysis of relatively large double-

stranded DNA molecules and is based on a hydrodynamic principle. The retardation of the DNA fragments from the

cleavage of the Lambda DNA by the KpnI restriction enzyme was studied using an acetonitrile-phosphate buffer as

a mobile phase (flow rate equal to 0.3 ml/min) and a C1 column as a stationary phase at various temperatures. It was

shown that the temperature constituted an important parameter for the separation of the DNA fragments in slalom

chromatography. The DNA hydrodynamic behavior with the temperature was related to the variation in the fluidviscosity and the modification of the elastic properties of the biopolyrner. © 2000 Elsevier Science B.V. All rights

reserved.

Keywords: Slalom chromatography; Column temperature; DNA; Relative retention time

www.elsevier.com /locate/talanta

1. Introduction

The slalom chromatography mode constitutes a

new research approach to the analysis and study

of double-stranded DNA molecules. The principleis based on a hydrodynamically driven mechanism

where relatively large DNA polymers (greater

than several kilobases) are retained in the gel

permeation column without interacting with the

stationary phase [1,2]. Recently, a model has been

proposed to describe the mechanistic aspects of

the fractionation [3]. The progression of DNA

molecules in the column was modeled taking into

account the direction changes of the macro-

molecule in response to frequent changes in the

flow direction through the interstitial spaces. The

major advantage of the slalom chromatography

mode is the rapidity and the simplicity of the

experimental procedure. For example, three frag-

ments of approximately 4, 9 and 23 kilobases (kb)

can easily be separated in less than 2 min with a

conventional chromatographic system using a gel

* Corresponding author. Tel.: +33-4-76637145; fax: +33-

4-76518667.

E -mail address: [email protected] (E. Peyrin).

0039-9140/00/$ - see front matter © 2000 Elsevier Science B.V. All rights reserved.

PII: S 0 0 3 9 - 9 1 4 0 ( 0 0 ) 0 0 4 8 2 - 3

7/29/2019 401685


E . Peyrin et al . / Talanta 52 (2000) 1105–1110 1106

permeation column [4]. As well, it has been shown

that the flow rate increase is associated with a

decrease in the analysis time and the enhancement

of the apparent separation factor between non

retained and retained DNA fragments [5]. This

fact is a rather novel concept and gives a real

advantage to the hydrodynamic principle of the

slalom chromatography over the equilibrium prin-

ciple of the classical chromatographic modes.

However, the main limitation of this technique is

represented by the resolution ability which is less

than that of slab or capillary gel electrophoresis.

Thus, it was of interest to study the possibilities of

improving the fractionation. capacities of the

slalom mode. On the basis of the model which has

been previously proposed and the experimental

data reported by Hirabayashi et al. [6], it was

expected that column temperature was one of the

main parameters which could influence DNA sep-

aration. In order to gain further insight into the

behavior of DNA in a hydrodynamic flux and

enhance the efficiency of the technique, the reten-

tion of DNA fragments on a C1 stationary phase

was analyzed over a wide range of column tem-

perature (3–60°C). The effects of this parameter

are discussed in relation to the model previously

established.

2. Materials and methods

2 .1. Apparatus

The HPLC system consisted of a Shimadzu

pump LC 10 AT VP (Touzart et Matignon,

Courtaboeuf, France), an Interchim Rheodyne in-

jection model 7125 (Montlugon, France) fitted

with a 20 ml sample loop and a Merck L 4500

diode array detector. A C1 Kromasil column

(particle size: 5 mm, column size: 150×4.6 mm)

Fig. 1. Theoretical T dependence (Eqs. (2) and (3)) on relative retention time (RRT) for various DNA fragments (17.05, 24, 29.95

and 35 kb) using a C1 column with a particle diameter of 5 mm and an acetonitrile-phosphate buffer as a mobile phase (—).

Experimental data obtained for the 17.05 and 29.95 kb DNA fragments ( and ).

7/29/2019 401685



Fig. 2. Experimental variations of apparent separation factor happ () in relation to T (°C) for the two DNA fragments 17.05 and

29.95 kb using a C1 column with a particle diameter of 5 mm and an acetonitrile-phosphate buffer as a mobile phase.

supplied by Interchim, was used with controlled

temperature in an Interchim Crococil oven TM

N° 701.

2 .2 . Reagents

Lambda DNA (48.50 kb) and restriction en-

zyme KpnI were supplied by New England Bio-

labs (Gagny, France). Ethanol, EDTA,

acetonitrile, glycerol, sodium hydrogen phosphate

and sodium dihydrogen phosphate were pur-

chased from, Prolabo (Paris, France). Water was

obtained from an Elgastat option water purifica-

tion system (Odil, Talant, France) fitted with a

reverse osmosis cartridge.

2 .3 . Digestion of lambda DNA

Restriction enzyme KpnI was used for the

cleavage of the lambda DNA into three fragments

of different sizes: 29.95, 17.05 and 1.50 kb. The

lambda DNA (2 mg) was treated with 3 U of KpnI

in 15 ml of the reaction mixture at 37°C for 3 h,

precipitated by ethanol dissolved in 20 ml of water

and stored at −20°C until use.

2 .4 . Chromatographic conditions

The mobile phase consisted of a sodium phos-

phate salt 0.01 M-EDTA 0.001 M mixture at pH6.8. The column temperature varied from 3 to

60°C. A volume of 20 ml of DNA solution was

injected and the retention times were measured at

a flow rate value equal to 0.3 ml/min. The reten-

tion time tNR corresponding to the void fraction

was obtained using the 1.5 kb fragment which was

not retained [7].

7/29/2019 401685



2 .5 . Model for the temperature dependence on

DNA beha6ior

By introducing the notion of the reorientation

time of a polymer [8], two equations for the

relative retention time RRT (ratio between the

retention time tR of a retained molecule and the

retention time of a non retained one tNR) have

been given in relation to the DNA length [3]. As

well, to estimate the temperature dependence of

the acetonitrile-water mobile phase viscosity, the

empirical relationship reported by Ghrist et al. [9]

was used:

p=p25298

T

6(1)

where p25 is the viscosity al 25°C. Thus, the RRT

values can be linked to the column temperature

by rearranging the model equations [3] with Eq.

(1):

RRT(1)L,6=

1−hL'

1+uL

T 7−1

−1

(2)

RRT(1)L,6=

h %L'

1+uL

T 7−1

(eh¦L/( 1+(uL/T

7)−1)−1)

−1

(3)

where the hL and uL are constants dependent on

the DNA length and the particle diameter. These

two equations can be used to provide an expres-

sion of the RRT value in relation to the column

temperature T . As well, an apparent separation

factor happ

defined as the ratio tR1/tR2

for two

adjacent retained DNA fragments 2 and 1 was

determined to characterize the separation.

3. Results and discussion

3 .1. Model 6alidation

The retention time values for the 17.05 and

29.95 kb fragments (tR) and for the 1.50 kb

fragment which corresponded to the void volume

marker (tNR) were obtained at various column

temperatures. From the tR and tNR values, the

experimental RRT were calculated for the differ-

ent chromatographic conditions. All the experi-

ments were repeated three times. The variation

coefficients of the RRT values were less than 4%

in most cases, indicating a high reproducibility

and good stability for the chromatographic sys-

Fig. 3. Chromatograms of the three DNA fragments (1.5 (1),

17.05 (2) and 29.95 (3) kb) at the optimal conditions (T =

50°C).

7/29/2019 401685



tem. With a non linear regression procedure

which was used in earlier chromatographic studies

[10,11], the data obtained at various flow rates

and viscosity were fitted to Eqs. (2) and (3). After

the non linear regression procedure, the calculated

hL and uL constants were used to estimate the

RRT values with the measured values for the two

DNA fragments at the different T values. The

correlation between all the predicted and experi-

mental RRT values exhibited slopes equal to 0.97

with r2\0.96. This good correlation between the

predicted and experimental values can be consid-

ered to be adequate to verify the model.

3 .2 . Temperature dependence on conformation and

separation of DNA

It has been previously demonstrated that

columns developed for reversed-phase chromatog-

raphy (such as the C1 column used in this study)

[12] are useful for slalom chromatography. In

order to eliminate the eventual hydrophobic inter-

action which could interfere with the hydrody-

namic principle, an aqueous mobile phase

containing 5 –20% of organic modifier such as

acetonitrile was used [12]. It was found that the

hydrophobic interaction was negligible in such

conditions. Thus, in this study, the experiments

were carried out with a large proportion of aceto-

nitrile in the mobile phase (20%). The fact that no

significant difference in the tNR values was ob-

tained at a constant flow rate for various acetoni-

trile proportion was consistent with a ‘pure’

hydrodynamic mechanism. To represent the sepa-

ration between non retained (1.5 kb fragment)

and retained molecules, the theoretical and ob-

served RRT values were plotted against the

column temperature T . Fig. 1 shows the curves

obtained from all the data for the T dependence

on RRT values for the 17.05 and 29.95 kb frag-

ments. As well, the theoretical RRT values for

various DNA fragments were added to the graph

of Fig. 1. As described by the model, the RRT

values increased when T decreased. This result

confirmed that the temperature acted on the DNA

behavior via two effects: the increase in the hydrodynamic force gener-

ated by the mobile phase due to the tempera-

ture dependence on the liquid viscosity

the enhancement of the DNA steady-state ex-

tension when the temperature decreased.

For the separation between two retained polymers

(17.05 and 29.95 kb), happ was calculated and

plotted against T in Fig. 2. These experiments

showed that temperature was a very important

factor in slalom chromatography. The RRT value

varied with the column temperature (T 7 function)

more strongly that with the linear velocity (6

function) which had previously been defined as

the main important parameter of this chromato-

graphic mode [3]. The lower the column tempera-

ture, the greater separation between non retained

and retained molecules. Thus, the optimal condi-

tions for the best separations between the void

DNA fraction and other DNA fragments were

represented by the lowest value of the column

temperature at a constant flow rate which was

compatible with a practicable back pressure and

the prevention of the physical degradation of

DNA fragments. In the case of the separation of

the two 17.05 and 29.95 kb large retained frag-

ments, the optimal apparent selectivity was at-

tained for T above 50°C over the temperature

range and at 0.3 ml/min. Fig. 3 shows the chro-

matogram for the separation of the three frag-

ments analyzed at this optimal apparent

selectivity.

In summary, this paper demonstrated that tem-

perature was a parameter which governed the

retention in slalom chromatography. The model

previously established was adequate to describe

the variation in the relative retention time of

DNA molecules with T and predict the hydrody-

namic behavior in slalom chromatography.

References

[1] J. Hirabayashi, K. Kasai, Nucleic Acid Res. Symp. Ser.

20 (1988) 67.

[2] J. Hirabayashi, K. Kasai, Anal. Biochem. 178 (1989) 336.

[3] E. Peyrin, Y.C. Gufflaume, A. Vinet, A. Favier, Anal.

Chem. 72 (2000) 853.

[4] K. Kasai, J. Chromatogr. 618 (1993) 203.

[5] J. Hirabayashi, N. Ito, K. Noguchi, K. Kasai, Biochem-

istry 29 (1990) 9515.

7/29/2019 401685



[6] J. Hirabayashi, K. Kasai, in: T.T. Ngo (Ed.), Molecular

Interactions in Bioseparations, Plenum Press, New York,

1993, p. 69.

[7] B.E. Boyes, D.G. Walker, P.L. McGeer, Anal. Biochem.

170 (1988) 127.

[8] G.W. Shater, J. Noolandi, Biopolymers 25 (1986) 431.

[9] B.F. Ghrist, M.A. Stadalius, L.W Spyder, J. Chromatogr.

387 (1987) 1.

[10] E. Peyrin, Y.C. Guillaume, C. Guinchard, Biophys. J. 77

(1999) 1206.

[11] Y.C. Guillaume, E. Peyrin, Anal. Chem. 71 (1999) 1326.

[12] J. Hirabayashi, K. Kasai, J. Chromatogr. A 722 (1996) 135.

401685

Documents

Transcript of 401685