Document Sample
DF IEF Powered By Docstoc
					Divergent flow isoelectric focusing

Karel Šlais

UIACh AV ČR v.v.i., Veveří 97, 602 00 Brno


Continuous flow isoelectric focusing (CF IEF) has a potential to be an important method in
proteome analysis.

Till now, the devices used do not fully use the advantages of IEF , since they do not take into
consideration all important IEF features including changes in background conductivity during
the focusing run.

We suggest to improve the peformance of CF IEF by optimizing the geometry and inner
space structure of the separation channel as well by management of electrolyte flow within
the channel.


Continuous flow isoelectric focusing (CF IEF) (A, A1, B,C,K) is a process wherein a sample stream
is continuously introduced into a carrier ampholyte solution flowing as thin film. By introducing an
electric field perpendicular to the flow direction, ampholytic sample components are separated by IEF
according to their differences in pI values. Finally, separated ampholytes are collected at the exit of the
separation space in collection vessels. Because of such arrangement of the electric field and buffer
flow, continuous separation of the sample can be achieved. To obtain reasonable sample throughput
and zone capacity, high voltage drop over reasonable channel width has to be applied. Nevertheless,
CF IEF has been examined also in microfluidic channels (P,Q,R,S). Fast separations are expected
while sacrifying separation efficiency and loadability. Also, connectorization is often a
problem in microfluidics.

Since the first applications of CF IEF for preparative protein fractionation (G,H,I,D) till the recent
sophisticated preparative devices (C,L,M,O) and           microchanels (P,Q,R,S) the parallelepipedic
separation space with parallel flow and constant perpendicular potential are used. It means that input
solution is subjected to the same transverse voltage drop as the solution at the channel output.
However, it is well known that in IEF with carriers the conductivity decreases by up to two orders of
magnitude during the focusing run. Thus, in IEF, the voltage has to be increased during focusing to
maintain stable wattage fed into area unit of the separation layer as mentored in textbooks of gel IEF.

In IEF as well as in CF IEF, the electrode solutions are inevitable in the creation of contacts between
the electrodes and the separation medium. They not only carry the electricity, they also make the
margins of pH gradient and feed the separation space with water ions. The interesting innovation in
CF IEF is to allow electrode electrolytes to flow within the separation chamber as sheath liquids (L).

Gel between the electrodes and the separation space ( )

Since, in different CF IEF instruments, the various electrode electrolytes are used, their electric
conductivity has to be considered to estimate the electric conditions within the channel.

        The solution of some of above compromises in continuous flow IEF is suggested here. It
is based on combination of the features of small channels with those of large ones. The basic
idea is planar divergent flow of medium subjected to transverse IEF. It is achieved by continuous
widening of flat channel in which the liquid flows from channel inputs toward the outputs. At
the same time, the measures are taken to maintain the small voltage at the channel input and high
voltage at the channel output. The features of divergent flow isoelectric focusing (DF IEF) can
be outlined as follows.
        Let us consider the situation close to the channel input where the channel is narrow and
the separation paths perpendicular to the flow stream are short. Due to suitable transverse voltage
the pH gradient is established soon as well as the analyzed ampholytes reach their positions
close to the local steady state. Under strongly simplified conditions often used for calculations in
IEF, the peak widths are determined by local field strength and pH gradient. The resulting
resolution of focused peaks as well as achievable peak capacity are low due to low applied
voltage drop over the local channel width. The low transverse voltage is necessary to prevent the
medium overheating.
        Now, for a moment, let us stop the electricity and allow the zones to travel toward the
broad place in channel solely by divergent flow. In this way, omitting the dispersion, the peak
resolution and conductivity of the medium remain the same. The geometrical distances between
the peak maxima increases proportionally to the increase in the channel width.
        Finally, without any flow, let us act the electricity keeping the same wattage per channel
area as at narrower position. After some time, the peaks are focused to the widths corresponding
to local pHand field strength. While the pH differences do not change during the voltage
application, the positions of peaks remain constant, however, they get narrower. At the broader
position of the channel the voltage drop is higher than that at narrower position both due to
longer distance and due to lower conductivity. Since the total pH difference remains the same,
the obtained peak capacity at the broader position is higher than that at the narrower one. The
above combination of divergent flow and focusing can be made in infinitely small steps or, in
other words, at the same time to obtain DF IEF.
        Thus it can be deduced that divergent flow together with suitable voltage management
reduce the amount of electricity needed for achieving some peak resolution in IEF. The final
issue of above model is that divergent flow isoelectric focusing should enable faster focusing
and/or higher performance in comparison with CF IEF in channel of constant width


       The device design

       The model above assumes to increase the transverse voltage drop over the increasing
channel width. The solution suggested here is to use electrode electrolytes at the margins of the
channel as electrolytic conduits. Such conduits, when connected to highest delivered voltage of
power supply close to the channel outlet act as the resistors which maintain the continuous
decrease of voltage from the channel outlet toward channel inlet. To fix the flow within the
conduits through the channel, the appropriate solutions should be continuously delivered at the
channel inlet close to the channel margins.
       The void closed channels are now mostly used in CF IEF experiments both in
miniaturized and preparative designs. Besides some advantages it also brings about the problems
associated with    the bubbles, leaking, limited shape flexibility, contacts with electrodes,
mounting of liquid inputs and outputs, and last but not least, the problems with minimizing the
channel height. The problem in the development of successful continuous flow devices for
electrophoretic   separation    is   also     the   control   of   undesired   convective    and
electrohydrodynamic flows which degrade the separation. In early attempts of CF IEF (D,I) as
well as in the recent designs (U) the anticonvective media were examined. In order to make
the demonstration of the idea of DF IEF as simple and straightforward as possible we have
examined here the non-woven fabrics for creating the separation space similarly as in paper
electrophoresis age. In comparison with paper and alike materials, modern nonwovens has some
advantages including mechanical and chemical stability and much lager permeability. The use of
thin porous sheet enables not only to create the porous non convective separation layer but also
to simply attach many liquid flow inputs and outputs as well as the electrode contacts. Fig. 1
shows the shape of hydrophilized spunbond polypropylene nonwoven web Pegatex S , 17 g m2
, nominal thickness 0.1 mm ( PEGAS Nonwowens, Znojmo, Czech Republic) used for
realization of whole flow manifold of divergent flow IEF device. The separation space, (A), has
a trapezoidal shape with input width 15 mm, output width 80 mm and length 150 mm which
gives the area 71.25 cm2. The input of the separation space is connected to three strips of 5 mm
in width and 50 mm in length for introduction of anolyte, (B) , background electrolyte mixed
with analytes (C) and catholyte ( D), respectively. Close to the output, at the margins of the
separation space, the strips for contacts with anode (E) and cathode (F) are realized. The output
of separation space transits seamless to outlet strips (G). In our case, 12 output strips are used to
match their number with microplate wells used for fraction collection.

       The view of whole device for examination of DF IEF is seen in photo, Fig. 2. The
combination of capillary elevation and hydrostatics is used here to continuously move the
working liquids through the porous layer without any need of pumps and capillaries. Also, the
open system simplifies the visual, photographic and electric inspection of the process. Thus, the
input strips B, C, D are dipped in respective solutions in Petri dishes, 80 mm in diameter, see Fig.
2, above. The electrode contacts E, F freely lies on horizontally positioned carbon rods 5 mm in
diameter and 7 cm in length serving as electrodes, they are fed by high voltage power supply (
VNZ         ,   ). The whole nonwoven layer A lies on white flexible PVC sheet Durofol , 0.25
mm thick, (FATRA, Napajedla, Czech Repubic) which enables the tips of output strips G to be
directed down by 80 mm below the horizontal plane of the separation space A. Thus, the output
tips hang just above the microplate 8 x 12 wells, 0.25 ml each ( ), see the Fig. 2, front bottom.


       The suggested model of DF IEF is foreseen for preparative purposes. Thus, the layer
dimensions will be optimized rather larger ones which also implies larger consumption of
electrolytes and chemicals.
        From many possible carrier electrolytes useful for generation of pH gradient we prefer
here the our mixtures of simple defined buffers similar to those described previously (C,M,N
my). ) and which is based on previous suggestions (      )
        The visual inspection of the CF IEF can conveniently be made with use of colored
ampholytes (D). Advantageously, we use here our synthetic low-molecular weight pI markers
( my ). The pI’s of our red pI markers used here are 3.3 , 4.75, 6.25, 7.6, and 11.
          The conductivity of the input background electrolyte together with our low-molecular
weight pI markers was 0.75 mS. cm-1

        The anolyte used was 0.5 Mol. l -1 H3PO4 , conductivity 5.2 (Tab 5.5 mS ).

        The catholyte was 0.05 Mol.l-1 NaOH 11, ( Tab. 0.125 M je 25 mS ).

Digital multimeter.

Results and discussion

Flow management

The nonwoven      web used is hydrophilic enough to keep the water soultions used and,
simultaneously, permeable enough to enable reasonable tangential flow under minimum
hydrostatic difference which is determined by the height difference between the horizontal
plane of the separation space A and ends of output strips. It amounts here 80 mm.
        The flow rates can conveniently be controlled by active length and width of both input
and output strips as well as by the height difference between the separation space and tips of
output strips.
        The PVC sheet under the separation space and the output strips is hydrophobic enough
to prevent the liquids to spread out from the porous layer. Thus, the output strips are mutualy
separated by few tenths of milimeter which is sufficient to prevent a mixing of the separated
The liquid flow velocity can be conveniently estimated by observing (and taking photo) of the
dye zone spiked on the separation area, by observing the flow pattern change after the voltage
change and by the evaluation of inlet liquids consumption. The flow rate during all the
experiments discussed below was adjusted to 4 ml/hour of the solution of carriers and 1 ml/hour
of electrode electrolytes each. By the measurement of the IEF pattern response to voltage
switching off, the holdup time of the separation area was found to be 10 min (10 min = 1/6 h)
which gives its holdup volume 1.0 ml which together with its area 71.25 cm2 =
1.0/71.25 = gives 0.14 mm effective thickness of liquid layer. Further, it gives the output flow
velocity (0.1 ml/ min /    (8 cm width x 0.014 cm height) = 0.1 / 0.112 = 0.9 cm/min which is
in agreement with the measurement of linear velocity of dye spot at the outlet. The density of
polypropylene 0.9 g / ml and specific area mass 0.0017 g / cm2 gives the specific volume of
solid phase 0.0017 / 0.9 = 0.0019 ml / cm2 = 0. 019 mm and together with the effective
thickness of liquid layer finally the effective porosity 14 / (14 + 1.9) = 0.88 . The loss of the
liquids by evaporation was measured by balance of inlet and outlet flows and it was estimated
to be about 10% under the running conditions with use of suitable device cover. The flow
pattern under no voltage applied can be seen in Fig. 3, up most left photo. The colorless
electrode electrolytes frame the solution of carriers and ampholytic dyes, the width of the middle
solution increases proportionally to the layer width, as expected. No considerable mixing
between the solutions is observed.


from switched off to switched on conditions in DF IEF is seen in Fig 3 where one minute interval
photo series of the separation area is collected. The constant power source giving 1 watt power is
switched on simultaneously with taking the up most left photo. The photo series shows that the
focusing takes place along the whole separation layer from the inlet corner toward the outlet
ramp. It indicates complex pattern of electric current which flows at any place. It has both a part
perpendicular to liquid flow streamlines which is manifested by the focusing process as well as
a part parallel to streamlines which enables the transfer of charge toward the inlet which is far
upstream from electrodes. The electric current directed toward the inlet is carried mainly by
electrode solutions which have their conductivity at least by one order of magnitude larger than
middle solution, Experimental. However, along the electrode solutions, the current is
continuously bifurcated toward the middle solution to enable IEF process in it. Due to such
bifurcation and due to the ohmic loss on the streamlines of electrode electrolytes the voltage drop
over the layer width decreases in direction from outlet to inlet. The conditions are changing with
time till the dynamic steady state is achieved under which the electric pattern is stable in time.
Under conditions used it is approached within 15 minutes.

The focusing pattern under such steady state is in Figure 4, left which is taken under the flow and
concentrations conditions mentioned above and under supply of 1 watt power. While the voltage
between the electrodes approaches 730 V, the voltage drop over the separation layer at its 4 cm
width is 500 V and voltage measured between the electrode electrolyte inputs is only 180 V.
Nevertheless, due to the conductivities of input solutions, the most of this voltage drop is loaded
over less than 1 cm wide stream of the central solution. Thus, the effective field strength of order
200V/cm can be expected over the input stream of solution of carriers and analytes. Since the
photo indicates that the pH gradient is formed on the path of about one centimeter and the linear
velocity is here about 5 cm . min-1 , it corresponds to about 12 seconds gradient forming time.
While the liquid flows toward the wider separation space the distance and resolution between
zones of gradient components as well as those of analytes proportionally increases which
consumes the electricity as discussed in the Introduction.

For comparison, we have arranged the CF IEF experiment in the rectangular separation space
with use of long parallel electrodes, see Fig. 4, right. For this purpose the same material, length
and output width of the separation space as for the DF IEF device at the left photo were used.
Also, the solutions, the output flow rate and electricity power load were the same. Under the
steady state, the voltage drop over the electrode achieved 380 V which indicates that most of the
electricity is spent at the channel input where the conductivity of inlet solution is high. It can be
clearly seen that the pH gradient is hardly formed before the solution leaves the separation space
and that some analytes are not focused to visible separated zones.


Modern nonwovens are suitable for design simple continuous flow electrophoretic devices which
enable us to setup the complicated flow and electricity manifolds. Such device can be disposable
and allow wide experimental variability and fast realization. Open design removes the problems
with bubbles at electrodes. The working with ( kalne) biological samples can be solved by
disposable design of the porous manifold. The wavy streamlines of focused dyes indicate some
heterogenity of nonwoven used.

The   location of electrodes downstream close to output of solution minimize the danger of
contact of analytes with them.

The design shown has potential for further optimization both in choice of material, scaling up
and down, as well as toward design of closed channels.

It was shown that DF IEF combines the advantageous features of both miniaturized as well as
preparative designs of CF IEF.

In comparison with current status we expect the improvement in the performance, speed and
capability of miniaturisation of free flow isoelectric focusing.


A – Righeti P.G., Castagna A., Herbert B., Reymond F., Rossier J.S., Proteomics , 2003,3,
1397-1407 Rev, + historie
B - Křivánková L., Boček P., Electrophoresis 1998,19, 1064-1074 , rev, fundamentals, ref. to
Octopus C.H.I.E.F. , Weber - margin buffers flowing though the electrophoresis chamber at
both sides

C - Canut H., Bauer J., Weber G., J. Chromatogr.B., 1999,722,121 -139, rev, all modes in single
instrument + Octolyt

D – Hedenskog G., J. Chromatogr., 107 (1975) 91 – 98, Sephadex, hydrostatic, 12 output paper
strips, photo - hemoglobin, , (albumin+ BrP blue) , + 1 M H3PO4, - 1m NaOH, 1% ampholines,
(celkem dosud citovane 4x !)

E, F - Konig, P., 1937 , Pocatky paper elektroforezy //Michl ( j.Ch. 31 (1967) 274)
G – Wagner H., Speer W., J Chromatogr 157 (1978) 259 -265 cont FF IEF

H – Hannig, K., Electrophoresis 1982, 3, 235 – 243 ( prvni FF IEF bez naplne ?)

I – Fawcet, J.S., Ann. N.Y.Acad.Sci. 1973 , 209, 112-126. vertical, density gradient, first
continuos flow

J – Kuhn R., Wagner H., J. Chromatogr. 1989, 481, 343- 350 popis Octopus

K – Roman M., Brown P.R., Anal. Chem., 66 , 2, 1994, 86A – 94A., rev,

L- Weber G., Bocek P., Electroporesis 17, 1996, 1906-1910, concepts CFE, new design
Octopus , for all modes , anolyte , catholyte as sheath liquids, in CF IEF ( fig 3A)

N / Bier M., Ostrem J. Marquez R.B., Electrophoresis 1993,14, 1011 – 1018. A new buffering
system, pilot experiments

O – Soulet, Roux-de Balmann H.,, Sanchez V., Electrophoresis, 1998,19, 1294-1299., major
problem in FFE hydrodynamics , single instrument for all electrophoretic modes.

U- Hagedorn, Electrophoresis, 26, 2495-2502, 2005, //, gel, micro fluid IEF

V - Albrecht JW, Jensen KF, Electrophoresis, 27, (24) 4960 – 4969 , micro, FF IEF gel mezi
eledama a sep. prostor

W Acevedo F., J.Cromaogr., 545, (1991) 391-196 – jednoduche pufry pro ITP proteinu.

T - PP nonwoven, smaceni vodou, cinani, 2006, polymer testing, 717-222

1. Šlais K., Friedl Z., Low-molecular weight pI markers for isoelectric focusing. J. Chromatogr.
A, 661 (1994) 249-256. .

2. Šlais K., Friedl Z., Ampholytic dyes for spectroscopic determination of pH in electrofocusing.
J.Chromatogr. A, 695 (1995) 113 - 122.

M - 1. Weber G, Bocek P
Recent developments in preparative free flow isoelectric focusing
ELECTROPHORESIS 19 (10): 1649-1653 JUL 1998 Octolyte consisting of defined
chemicals , ref. Bier, , Octopus for any CFE

A1 - 2. Prefractionation techniques in proteome analysis: The mining tools of the third
Author(s): Righetti PG, Castagna A, Antonioli P, Boschetti E
Source: ELECTROPHORESIS 26 (2): 297-319 JAN 2005

3. Šlais K., Transient electrophoretic processes in capillaries of non-uniform cross- section.
        J. Chromatogr. A, 730 (1996) 247 - 259.

4. Šťastná M., Šlais K., Dynamics of gel isoelectric focusing with ampholytic dyes monitored
by camera in real-time, J. Chromatogr. A 2003, 1008, 193 – 203. (I.F. 3.098)

5. Ivory

Cui, H., K. Horiuchi, P. Dutta and C. F. Ivory, ŇIsoelectric focusing in a
Poly(dimethylsiloxane) microfluidic chip," Submitted 2004.

Ivory, C. F., ŇPreparative Free-Flow Electrofocusing in a Vortex-Stabilized Annulus,Ó
Electrophoresis 25 360-374 (2004).

Tracy, N. and C. F. Ivory, ŇModeling Two-Component Isoelectric Focusing Buffers in a
Vortex-Stabilized Electrophoresis Apparatus,Ó Biotechnol. Progress 20(1) 193-199 (2004)

Tracy, N. and C. F. Ivory, ŇFocusing Proteins in Defined Isoelectric Focusing Buffers,Ó
Electrophoresis, Electrphoresis, 25(12) 1748-1757 (2004).

Ista, L. K., G. P. Lopez, C. F. Ivory, M. J. Ortiz, T. A. Schifani, C. D. Schwappach, and S. S.
Sibbett, ŇMicrochip Countercurrent Electroseparation,Ó Lab-on-a-Chip, 3 266-272 (2003)

Ross, D., C. F. Ivory, L. E. Locascio, and K. E. Van Cott, ŇPeak Compression and Resolution
for Electrophoretic Separations in Diverging Microchannels,Ó Anal. Chem., (2003)

Generation of natural pH gradients in microfluidic channels for use in isoelectric focusing,
Macounová, K., Cabrera, C.R., Holl, M.R. and Yager, P., Analytical Chemistry, 72(16), 3745-
3751 (2000).

Use of isoelectric focusing for sample preconditioning in a microfluidic electrochemical flow
cell, Cabrera, C.R., Macounová, K., Holl, M.R., and Yager, P., Proceedings of the IEEE 1st
Annual Conference on Microtechnology in Medicine and Biology, Lyon France, October,
2000, D. Beebe, Ed.,IEEE Press (2000)

Q -Formation of natural pH gradients in a microfluidic device under flow conditions: Model
and experimental validation, Cabrera, C.R., Finlayson, B.A., and Yager, P., Analytical
Chemistry, 73(3), 658-666 (2001)

Continuous concentration of bacteria in a microfluidic flow cell using electrokinetic
techniques, Cabrera, C.R. and Yager, P., Electrophoresis, 22(2), 355-362 (2001)
P - Concentration and separation of proteins in microfluidic channels on the basis of
transverse IEF, Macounová, K., Cabrera, C.R., and Yager, P., Analytical Chemistry, 73(7),
1627-1633 (2001)

 - R - Shinohara, Analytical Sciences,2001,
40 (w) x 48 (L) x 0.03 (D) mm
silicone, Pyrex glass,

S - 7. Manz

T – Nonwovens

1. Šlais K., Friedl Z., Low-molecular weight pI markers for isoelectric focusing. J. Chromatogr.
A, 661 (1994) 249-256. .

2. Šlais K., Friedl Z., Ampholytic dyes for spectroscopic determination of pH in electrofocusing.
J.Chromatogr. A, 695 (1995) 113 - 122.

 [8]   Šlais, K., Friedl, Z., J. Chromatogr. A 1994, 661, 249-256.

[9]    Šlais, K., Friedl, Z., J. Chromatogr. A 1995, 695, 113-122.

[10]   Štastná, M., Trávníček, M., Šlais, K., Electrophoresis 2005, 26, 53-59.

[11]   Štastná, M., Šlais, K., Anal. Bioanal. Chem. 2005, in press, Online First.

[12]   Štastná, M., Šlais, K., J. Chromatogr. A 2003, 1008, 193-203.
Fig. 1: The shape of nonwoven web used for DF IEF. A- trapezoidal separation area, input
width 1.5 cm, output width 8 cm, length 15 cm, B - input strip for anolyte , C - input strip for
solution of carriers and analytes; D - input strip for catholyte; E - contact for anode; E- contact
for cathode, G – 12 strips 10 cm in length for outlet of separated fractions.
Fig. 2.: Overall view of simple device for divergent flow IEF. The polypropylene nonwoven
web 0.1 mm thick shaped according to Fig. 1 lies on white polyvinylchloride flexible sheet
(Durofoil ), the respective strips are contacted as follows: Petri dishes 80 mm in diameter
containing: above left – anolyte - strip B; above middle - solution of carriers and pI markers -
strip C; above right – catholyte - strip D; middle left - the carbon rod 5 mm in diameter anode
– contact E ; middle right – cathode - contact F; bottom - microplate wells – contact G.
Traces of pI markers from left - 3.3 , 4.7, 6.2, 7.6, 11.0, respectively.
Fig. 3.: From left to right: One minute interval photo sequence of the separation area of device
in Fig. 2 showing the dynamics of early stages of transition from stand by flow with no
voltage loaded toward the running conditions. The supply of one watt electric power is
switched on simultaneously with taking of the first photo.
Fig. 4.: Comparison of DF IEF (left photo) with CF IEF in the rectangular space and parallel
voltage load (right photo). The separation space material, length and output width as well as
power load, solutions and output flow rate are the same in both configurations. Other details
see text.

Shared By: