Analytical Biotechnology - ACS Publications - American Chemical

Chapter 4

On-Column Radioisotope Detection for Capillary Electrophoresis 1,3

1

2

Stephen L. Pentoney, Jr. , Richard N. Zare , and Jeff F. Quint Downloaded by NANYANG TECHNOLOGICAL UNIV on June 15, 2016 | http://pubs.acs.org Publication Date: July 17, 1990 | doi: 10.1021/bk-1990-0434.ch004

1

Department of Chemistry, Stanford University, Stanford, CA 94305 Advanced Development Unit, Beckman Instruments, Inc., Fullerton, CA 92634 2

Three on-line radioactivity detection schemes for capillary electrophoresis are described. The first detector system utilizes a commercially available semiconductor device positioned external to the separation channel and responding directly to impinging γ or high-energy β radiation. The second detector system utilizes a commercially available plastic scintillator material and a cooled photomultiplier tube operated in the photon counting mode. The third detector system utilizes a plastic scintillator material and two room-temperature photomultiplier tubes operated in the coincidence counting mode. The system performance and detector efficiency are evaluated for each of the detection schemes using synthetic mixtures of P-labeled sample molecules. The detection limits are determined to be in the low nanocurie range for separations performed under standard conditions (an injected sample quantity of 1 nanocurie corresponds to 110 x 10 moles of P). The lower limit of detection is extended to the sub-nanocurie level by using flow (voltage) programming to increase the residence time of labeled sample components in the detection volume. The separation of P-labeled oligonucleotide mixtures using polyacrylamide gel-filled capillaries and on-line radioisotope detection is also presented. When desired, the residence time can be made almost arbitrarily long by freezing the contents of the capillary, permitting autoradiograms to be recorded. This last technique is applied to gel-filled capillaries and provides a detection sensitivity of a few DPM per separated component, corresponding to subattomole amounts of radiolabel. Current address: Advanced Development Unit, Beckman Instruments, Inc., Fullerton, CA 92634 32

-18

32

32

3

0097-6156/90/0434-0060$08.50A) © 1990 American Chemical Society

Horváth and Nikelly; Analytical Biotechnology ACS Symposium Series; American Chemical Society: Washington, DC, 1990.

4. PENTONEY ET AL.

On-Column Radioisotope Detection

The highly e f f i c i e n t separations afforded by c a p i l l a r y electrophoresis (CE) are a d i r e c t r e s u l t of employing extremely narrow separation channels. Effective d i s s i p a t i o n of heat generated by the passage of e l e c t r i c a l current through the separation medium occurs only when the r a t i o of c a p i l l a r y inner surface area to i n t e r n a l volume i s s u f f i c i e n t l y large ( t y p i c a l l y 10 to 10 m" ). For this reason c a p i l l a r y tubes with i n t e r n a l diameters ranging from 10 μπι (1) to 200 μτη have been selected for CE separations. Early i n the development of c a p i l l a r y electrophoresis, i t was noted that the successful detection of separated sample components present within the narrow confines of these c a p i l l a r y tubes posed a major challenge (2). In response to this challenge, much research has been directed toward the development of sensitive and selective detectors for c a p i l l a r y electrophoresis. CE detector technology has been largely borrowed from the f i e l d of high-performance l i q u i d chromatography (HPLC), especially from micro-column HPLC. Successful extension of the various HPLC detection schemes to c a p i l l a r y electrophoresis has generally involved miniaturizing e x i s t i n g technology while s t r i v i n g for improved s e n s i t i v i t y . Radioisotope detection i s used widely i n HPLC but has received l i t t l e attention i n c a p i l l a r y electrophoresis applications (3-6). The a v a i l a b i l i t y of an on-line radioisotope detector for CE i s especially appealing for several reasons. F i r s t , state-of-the-art radiation detection technology offers extremely high s e n s i t i v i t y . Second, radioisotope detection affords unrivaled s e l e c t i v i t y because only radiolabeled sample components y i e l d a response at the detector. T h i r d , the radiolabeled molecule possesses the same chemical properties as the un-labeled molecule, thereby permitting tracer studies. Fourth, radioisotope detection can be d i r e c t l y c a l i b r a t e d to provide a measurement of absolute concentration of the labeled species. F i n a l l y , a c a p i l l a r y electrophoresis system i n which r a d i o a c t i v i t y detection i s coupled with more conventional detectors adds extra v e r s a t i l i t y to this new separation technique. Radioisotope detection of P, C , and " T c was reported by Kaniansky et al. (7,8) for isotachophoresis. In t h e i r work, isotachophoretic separations were performed using fluorinated ethylene-propylene copolymer c a p i l l a r y tubing (300 μιη i n t e r n a l diameter) and either a Geiger-Mueller tube or a p l a s t i c s c i n t i l l a t o r / p h o t o m u l t i p l i e r tube combination to detect emitted β particles. One of t h e i r reported detection schemes involved passing the radiolabeled sample components d i r e c t l y through a p l a s t i c scintillator. Detector efficiency for C - l a b e l e d molecules was reported to be 13-15%, and a minimum detection l i m i t of 0.44 nCi was reported for a 212 nL c e l l volume. A l t r i a et al. reported the CE separation and detection of radiopharmaceuticals containing T c , a 7 emitter with a 6-hour h a l f - l i f e (9, see also JjO). Their design involved passing a c a p i l l a r y tube ( « 2 cm long) through a s o l i d block of s c i n t i l l a t o r material and detecting the l i g h t emitted as technetium-labeled sample zones traversed the detection volume. Unfortunately, detection l i m i t s and detector efficiency were not reported. 4

Downloaded by NANYANG TECHNOLOGICAL UNIV on June 15, 2016 | http://pubs.acs.org Publication Date: July 17, 1990 | doi: 10.1021/bk-1990-0434.ch004

61

S 2

s

1 4

l4

9 e m


1

62

ANALYTICAL BIOTECHNOLOGY

We report here the design and characterization of three simple, on-line radioisotope detectors for c a p i l l a r y electrophoresis. The f i r s t detector u t i l i z e s a commercially available semiconductor device responding d i r e c t l y to 7 rays or β p a r t i c l e s that pass through the walls of the fused s i l i c a separation channel. A s i m i l a r semiconductor detector for 7-emitting radiopharmaceuticals separated by HPLC was reported by Needham and Delaney ( H ) . The second detector u t i l i z e s a commercially available p l a s t i c s c i n t i l l a t o r material that completely surrounds (360 ) the detection region of the separation channel. Light emitted by the p l a s t i c s c i n t i l l a t o r i s collected and focused onto the photocathode of a cooled photomultiplier tube. A l t e r n a t i v e l y , a t h i r d detection scheme u t i l i z e s a disk fashioned from commercially available p l a s t i c s c i n t i l l a t o r material positioned between two-room temperature photomultiplier tubes operated i n the coincidence counting mode. This t h i r d scheme maintains favorable c o l l e c t i o n geometry ( 3 6 0 ° ) while minimizing detector background noise by e l e c t r o n i c a l l y r e j e c t i n g non-coincident photomultiplier pulses. The detectors described i n the present work are applicable to both high-energy β emitters and 7 emitters. We report here on their a p p l i c a t i o n to the detection of P - l a b e l e d molecules separated by c a p i l l a r y electro phoresis .


e

S2

Experimental Section On-Line Radioactivity Detectors. Our f i r s t on-line r a d i o a c t i v i t y detector (see Figure 1) consisted of a model S103.1/P4 spectroscopic-grade cadmium t e l l u r i d e semiconductor detector and a model CTC-4B radiation counting system (Radiation Monitoring Devices, I n c . , Watertown, MA). The cadmium t e l l u r i d e detector probe consisted of a 2-mm cube of CdTe set i n a thermoplastic and positioned behind a thin film of aluminized mylar at a distance of approximately 1.5 mm from the face of an aluminum housing (see Figure 1). A Pb aperture (2 mm wide, 0.008 inch thick) shielded the CdTe detector element from radiation o r i g i n a t i n g from regions of the c a p i l l a r y adjacent to the detection volume. The aluminum housing incorporated a BNC-type connector that f a c i l i t a t e d both physical and e l e c t r i c a l connection to a miniature, charge-sensitive preamplifier. The CdTe probe and preamplifier assembly were mounted on an x-y t r a n s l a t i o n stage and the face of the aluminum housing was brought into d i r e c t contact with the polyimide-clad f u s e d - s i l i c a c a p i l l a r y / P b aperture assembly. The CdTe detector was operated at the manufacturer's suggested bias voltage of 60 V and the detector signal (the creation of electron-hole pairs produced as β p a r t i c l e s were decelerated within the semiconductor material) was amplified by the charge-sensitive preamplifier and sent through a six-foot cable to the counting and display electronics of the CTC-4B counting u n i t . Although the CTC-4B i s capable of room-temperature energy discrimination, a l l experiments reported here were performed with a r e l a t i v e l y large energy window. The upper energy discriminator setting was 1 MeV (the maximum setting for the CTC-4B) and the low energy setting was 0.01 MeV. The CdTe r a d i o a c t i v i t y detector was computer interfaced to a laboratory microcomputer (IBM PC-XT) by using the open c o l l e c t o r



PENTONEY ET A L


32P

Radioisotope

Detector

Data Acquisition Detector

Capillary Inlet

Capillary Outlet

r U

Electrolyte . Buffer

Electrolyte Buffer ^

Reservoir

Reservoir HV

Figure 1. Experimental setup of the c a p i l l a r y electrophoresis/radioisotope detector system. The inset shows the positioning of the CdTe probe with respect to the c a p i l l a r y tubing. The 2-mm Pb aperture i s not shown i n this illustration.



64


output of the CTC-4B counting u n i t . The open c o l l e c t o r output was t i e d high by way of a 1-ΚΩ pull-up r e s i s t o r so that the unit provided a negative-going TTL pulse for each count measured. This TTL signal was sent to a photon counter (model 1109, EG&G Princeton Applied Research, Princeton, NJ) and counting intervals ( t y p i c a l l y 1 second) for run-time data acquisition were preselected by way of front-panel thumbwheel switches on the photon counter. The binary coded decimal (BCD) output of the photon counter was read at the end of a preset counting i n t e r v a l (strobe sent by the 1109 counter) by a laboratory microcomputer (IBM PC-XT) using a 32-bit d i g i t a l I/O board (Model DT2817, Data Translation, I n c . , Marlboro, MA). Data a c q u i s i t i o n and storage were accomplished using software written i n house (BASIC). Migration times and peak widths reported here were determined manually from scale-expanded portions of the recorded electropherograms. Our second on-line r a d i o a c t i v i t y detector consisted of a p l a s t i c s c i n t i l l a t o r material (BC-400, Bicron Corp., Newbury, OH) that was machined from 1-inch-diameter rod stock into a 5/8-inchdiameter (front face) s o l i d parabola (see Figure 2). A special rotating holder was constructed for the p l a s t i c s c i n t i l l a t o r and the curved outer surfaces were coated by vacuum deposition with a thin film of aluminum i n order to r e f l e c t the emitted l i g h t toward the front face of the s c i n t i l l a t o r . A detection length of 2 mm was defined within the parabola by aluminum mounting rods (0.250 inch outer diameter) that were p r e s s - f i t (coaxial to the separation c a p i l l a r y ) i n the sides of the s c i n t i l l a t o r , as i l l u s t r a t e d i n Figure 2. As radiolabeled sample passed through the detection region, the s c i n t i l l a t o r emitted l i g h t , which was collected and focused onto the photocathode of a cooled photomultiplier tube (Centronic # 4283) by a condenser lens combination (Physitec, # 06-3010, focal length 16 mm). Photon counting was accomplished using a Model 1121A discriminator control unit and a Model 1109 photon counter (EG&G Princeton Applied Research). The background count rate observed under t y p i c a l operating conditions for t h i s system was approximately 12 counts per second. Our t h i r d on-line r a d i o a c t i v i t y detector consisted of a modified HPLC radioisotope detector ( Model 171 Radioisotope Detector, Beckman Instruments, I n c . , Palo A l t o , CA). The standard HPLC flow c e l l was removed and the unit was modified for use as a c a p i l l a r y electrophoresis detector, as i l l u s t r a t e d i n Figure 3. A 2-mm-wide disk of the p l a s t i c s c i n t i l l a t o r material surrounded the detection region of the separation c a p i l l a r y and was positioned between two photomultiplier tubes operated i n the coincidence counting mode. For a l l experiments reported here, the coincidence gate was 20 nanoseconds. In this manner, most of the random background pulses associated with the two room temperature photomultiplier tubes were rejected. A large number of photons were emitted i s o t r o p i c a l l y for each β p a r t i c l e decelerated within the s c i n t i l l a t o r material and a burst of l i g h t was thereby sensed at both of the photomultiplier tubes within the gated time i n t e r v a l . The background count rate of the modified coincidence detector used i n t h i s work was approximately 30 counts per minute. Data from this detector was acquired and analyzed using the Chromatographics



PENTONEY ET AJL On-Column Radioisotope Detection

Figure 2. Exploded diagram of the p l a s t i c s c i n t i l l a t o r radioisotope detector. The fused s i l i c a c a p i l l a r y i s exposed to a 2-mm section of the p l a s t i c s c i n t i l l a t o r located between the p r e s s - f i t aluminum mounting rods.

Figure 3. Exploded diagram showing the design of the coincidence radioisotope detection scheme. The f u s e d - s i l i c a c a p i l l a r y i s exposed to a 2-mm length of p l a s t i c s c i n t i l l a t o r material located between two photomultiplier tubes operated i n the coincidence counting mode.


65

66



software package (Beckman Instruments, I n c . ) . Both of the o p t i c a l detection systems yielded a large response (at the onset and completion of each separation), which was associated with the application or termination of high voltage. The magnitude of this response was dependent upon the length of time that the system was held at zero p o t e n t i a l between runs, with longer zero potential periods corresponding to larger responses at the onset of voltage. Although the cause of this response i s not yet f u l l y understood, the signal was observed to damp out within the f i r s t 1-2 minutes of a run, and therefore posed no interference to the separations reported here. Apparatus. The experimental setup of the home-built c a p i l l a r y electrophoresis system was s i m i l a r to that described previously (11.12) and i s i l l u s t r a t e d i n Figure 1. For the semiconductor detector, a 2-mm section of a f u s e d - s i l i c a c a p i l l a r y tube (100 μιη internal diameter, 365 μιη outer diameter, 100 cm long, TSP 100/365, Polymicro Technologies, Inc.) was exposed to the CdTe semiconductor by placing the Pb f o i l aperture d i r e c t l y between the face of the detector housing and the fused s i l i c a c a p i l l a r y at a distance of 75 cm from the i n l e t end of the c a p i l l a r y tube. For the parabolic p l a s t i c s c i n t i l l a t o r detector, a 2-mm section of a f u s e d - s i l i c a c a p i l l a r y (100 μιη i n t e r n a l diameter, 140 cm long) was exposed to the s c i n t i l l a t o r by passing the c a p i l l a r y through a 400-μιη hole d r i l l e d through the aluminum mounting rods and the central 2-mm portion of the parabola at a distance of 75 cm from the c a p i l l a r y i n l e t . For the coincidence detection scheme a 2-mm section of f u s e d - s i l i c a c a p i l l a r y (100 μιη i n t e r n a l diameter, 100 cm long) was exposed to the p l a s t i c s c i n t i l l a t o r disk by passing the c a p i l l a r y through a 400-μπι hole d r i l l e d through both the aluminum shields and the disk i t s e l f . The length of c a p i l l a r y extending from the i n l e t to the detection region was 55 cm i n the efficiency experiments reported here. This resulted i n a detection volume of approximately 15 nL for each of the three detectors. Each end of the c a p i l l a r y tubing was dipped into a 4-mL glass v i a l containing approximately 3 mL of e l e c t r o l y t e - b u f f e r s o l u t i o n . A s t r i p of P t - f o i l submersed i n each of the buffer reservoirs provided connection to high voltage. Plexiglass shielding (0.25 i n thick) was placed around the i n l e t buffer reservoir because the top of this v i a l was quickly contaminated by sample solution c a r r i e d on the outside of the c a p i l l a r y tube during the sample i n j e c t i o n procedure. This contamination, i f unshielded, lead to unnecessary operator exposure to r a d i a t i o n . The current through the system was monitored as a potential drop across a 1-ΚΩ r e s i s t o r i n the c i r c u i t . The c a p i l l a r y system and detector were enclosed i n a Plexiglass box to prevent operator exposure to high voltage. Electroosmotic flow rates for freesolution separations reported here were measured i n a manner s i m i l a r to that described by Huang et al. (14). The c a p i l l a r y tube was f i l l e d with running buffer d i l u t e d by 10%, the buffer reservoirs were f i l l e d with running buffer, and the current was monitored as one tubing volume was displaced by supporting buffer under the influence of the applied p o t e n t i a l . Sample introduction i n a l l f r e e - s o l u t i o n separations reported here was accomplished by


4. PENTONEY ET AL.


67

hydrostatic pressure. E l e c t r o k i n e t i c sample introduction was used for a l l c a p i l l a r y gel separations. The high-voltage power supply (Model MG30N100, Glassmann High Voltage, Whitehouse Station, NJ) was continuously programmable from 0 to -30 kV by means of an external 0-to-10-volt DC s i g n a l . The flow-programming experiments reported here were accomplished by manually reducing the program voltage to the high-voltage supply. Reagents. Aqueous ethanol solutions of the triethylammonium s a l t s of adenosine-5'-[a- P]triphosphate ( a - P A T P ) , adenosine-5'-[7P]triphosphate ( 7 - P - A T P ) , thymidine-5'-[a- P]triphosphate (QP - T T P ) , c y t i d i n e - 5 ' - [ a - P ] t r i p h o s p h a t e ( a - P - C T P ) , and guanosine-5'-[a- P]triphosphate (a- P-GTP) were purchased from Amersham (Arlington Heights, I L ) . Radioactive sample concentrations reported for detector efficiency determination were adjusted from the manufacturer's specifications after subjecting several d i l u t e d aliquots of the stock solution to l i q u i d s c i n t i l l a t i o n counting. The concentration was further corrected for radiochemical purity according to the manufacturer's specifications because l i q u i d s c i n t i l l a t i o n counting measures the t o t a l sample a c t i v i t y and does not account for the presence of radiolabeled impurities. Stock solutions were stored at -15*C or -20 C i n order to minimize sample loss due to hydrolysis. Injected sample solutions were prepared i n 0.25 mL p l a s t i c v i a l s by d i l u t i n g stock solutions with buffer or deionized water and were also stored at -15 C or - 2 0 ° C. The supporting electrolyte for a l l free-solution separations reported here was a borate buffer (pH 8.1 or 8.26, 0.10 M or 0.20 Μ), prepared from reagent-grade sodium borate decahydrate and boric acid ( J . Τ Baker). For the preparation of g e l - f i l l e d c a p i l l a r i e s , a solution consisting of 50 mM t r i s , 50 mM boric a c i d , and 7 M urea (pH - 8.3) was used both to prepare the gel and as the running buffer. For some of the g e l - f i l l e d c a p i l l a r i e s used i n this work, the buffer also contained 3% PEG 20,000 (15) (Fluka Chemical Corp., Ronkonkoma, NY). The f u s e d - s i l i c a c a p i l l a r i e s were rinsed with 1 N HC1, IN NaOH, and then methanol. A 1:1 mixture of methacryloxypropyltrimethoxysilane and methanol was introduced into the c a p i l l a r y (100 or 75 μιη internal diameter) by syringe and allowed to s i t at room temperature for at least 3 hours. The acrylamide/crosslinker (N, N' methylenebisacrylamide) solution was prepared i n 10 ml of running buffer to y i e l d a gel composition of either 7% T, 3% C, or 6% T, 5% C. To i n i t i a t e polymerization, 2-5 ul* aliquots of 10% solutions of both ammonium persulfate (APS) and Ν , Ν , Ν ' , Ν ' - t e t r a m e t h y l e t h y l e n e d i a m i n e (TEMED) were added to 1 mL of the monomer solution. The monomer solution was then quickly introduced into the c a p i l l a r y by syringe and l e f t at room temperature overnight for complete polymerization. Once polymerization was complete, care was taken to keep the c a p i l l a r y ends submerged i n buffer at a l l times i n order to prevent the gel from drying out. The oligonucleotide samples described here were labeled at the 5' end according to the following procedure. Approximately 0.5 OD of oligomer sample was combined with 5 μ ί of 7 - P ATP (10 mCi/ml, 5000 ci/mmole, Amersham), 5 μΐ* of lOx kinase reaction buffer 32

32

3 2


3 2

32

32

32

32

32

32

e

e

3 2


68


(Amersham) , and 2 /iL of kinase. The t o t a l solution volume was adjusted to 50 /*L by the addition of HPLC grade water. The solution was gently mixed and allowed to react at 37* C for one hour. At the end of this period, the labeled oligomer sample was isolated from unincorporated 7 - P ATP by either ethanol p r e c i p i t a t i o n or by size exclusion chromatography using Sephadex G-25. The labeled oligomer was then l y o p h i l i z e d and dissolved i n an appropriate volume of deionized water before being e l e c t r o k i n e t i c a l l y injected into the gel-filled capillaries. S 2


Results and Discussion The process of β decay for 32

S 2

P can be written as

(1£),

32 Ρ

-•

S

15

+

β~

+

ν

(1)

16

where β' represents the negatively charged β p a r t i c l e and ν i s the antineutrino. P i s an example of a "pure β emitter" that populates only the ground state of the product nucleus. Each βdecay t r a n s i t i o n i s characterized by a fixed decay energy shared between the β p a r t i c l e and the antineutrino. As a r e s u l t , the β p a r t i c l e i s emitted with an energy that varies from decay to decay and ranges from zero to the β end-point energy," which i s numerically equal to the t r a n s i t i o n decay energy. A ^-energy spectrum for P shows a maximum p a r t i c l e energy of 1.7 MeV and an average p a r t i c l e energy of approximately 0.57 MeV. The penetrating a b i l i t y of β p a r t i c l e s through various media may be obtained from l i t e r a t u r e range-energy plots i n which the product of p a r t i c l e range and medium density ("mass thickness") i s plotted against p a r t i c l e energy. Such plots are especially useful because they may be used to predict the penetration length at a given p a r t i c l e energy i n media other than that used to obtain the o r i g i n a l p l o t (16). From such p l o t s , one would predict that the average β p a r t i c l e energy ( » 0.57 MeV) produced by decay of P would correspond to a range of approximately 2000 μιη i n water and approximately 950 μπι i n fused silica. Thus, P decay would be detectable by a sensitive device positioned external to the fused s i l i c a c a p i l l a r y tubing (of the dimensions normally selected for c a p i l l a r y electrophoresis separations). Successful detection of P - l a b e l e d molecules separated by c a p i l l a r y electrophoresis using the above detection schemes, i n which a sensor was positioned external to the separation channel, was made possible by several factors. These included (1) the large energy associated with β decay of P (1.7 MeV), (2) the high s e n s i t i v i t y and small size of commercially available semiconductor detectors, (3) the a v a i l a b i l i t y of e f f i c i e n t s o l i d s c i n t i l l a t o r materials and sensitive photomultiplier tubes, (4) the short lengths of fused s i l i c a ( c a p i l l a r y wall thickness) and aqueous e l e c t r o l y t e through which the radiation must pass before s t r i k i n g the detector, and (5) the r e l a t i v e l y short h a l f - l i f e of P (14.3 days). Because the CdTe detector was not v i s i b l e through the aluminized mylar f i l m , i t was necessary to check for proper S 2

η

S 2

S 2

S 2

32

3 2

3 2


4. PENTONEY ET A L


69

alignment of the c a p i l l a r y tube with respect to the CdTe cube. This was accomplished by f i l l i n g the detection volume with radioactive material and monitoring the signal l e v e l as the detector was translated with respect to the c a p i l l a r y . The observed signal was not very sensitive to positioning when the c a p i l l a r y was offset over a range of ± 1.5 mm from the center of the aluminum housing but dropped o f f rapidly at greater distances. A l l experiments reported here were performed with the c a p i l l a r y positioned at the center of the aluminum housing, as indicated i n Figure 1. Detector E f f i c i e n c y . In order to calculate the e f f i c i e n c y of the on-line r a d i o a c t i v i t y detectors for P , i t was necessary to determine the volume of sample injected onto the c a p i l l a r y tube by the gravity-flow technique. The volume of sample introduced by hydrostatic pressure was determined as follows: A plug of P labeled ATP was introduced onto the c a p i l l a r y by r a i s i n g the sample v i a l above the high-voltage reservoir for a c a r e f u l l y timed i n t e r v a l . The end of the c a p i l l a r y was then returned to the anode reservoir and electrophoresis was performed for 5 minutes at high voltage. This 5-minute high-voltage period served to transfer the sample plug toward the detector and away from the i n j e c t i o n end of the c a p i l l a r y as i f an actual separation were being performed. At the end of the 5-minute period the voltage was switched off and the e l e c t r o l y t e within the c a p i l l a r y tube was driven, v i a syringe, into a l i q u i d s c i n t i l l a t i o n v i a l located beneath the c a p i l l a r y o u t l e t . This process was continued u n t i l approximately 8 c a p i l l a r y volumes of e l e c t r o l y t e were c o l l e c t e d . The collected sample plugs were mixed with s c i n t i l l a t i o n c o c k t a i l and subjected to l i q u i d s c i n t i l l a t i o n counting. The i n j e c t i o n volumes were determined by r e l a t i n g the a c t i v i t y of the sample plugs to that of the injected sample s o l u t i o n . The i n j e c t i o n volumes and r e p e a t a b i l i t y of the manually performed hydrostatic injections for the three detector e f f i c i e n c y determinations are shown i n Tables I, I I , and I I I .


S 2

3 2

Table I .

Injection Data for CdTe Radioisotope Detector CE System

Injection No. 1 2 3 4 5 6 7 8 9 10 Average Std. Dev. % RSD Injection Volume

DPM 101885 110449 101884 111375 103512 113018 109432 104581 106740 107357 107023 3996 3.7 84 nL


70



Table I I .

Injection Data for P l a s t i c S c i n t i l l a t o r Radioisotope Detector CE System

Injection No. 1 2 3 4 5 6 7 8 9 10 Average Std. Dev. % RSD Injection Volume

DPM 89732 95220 90353 94939 90564 90039 94648 96628 95212 93150 93049 2621 2.8 72 nL

Table I I I . Injection Data for Coincidence Radioisotope Detector CE System Injection No. 1 2 3 4 5 6 7 8 9 10 Average Std. Dev. % RSD Injection Volume

DPM 37167 38220 40071 39010 32237 36784 35845 36032 35606 36767 36773 2146 5.8 60 nL

Replicate c a p i l l a r y electrophoresis runs were made i n which a standard solution of P - l a b e l e d ATP was injected into the capillary. The results are shown i n Tables IV, V, and V I . These tables l i s t the migration time, peak area, residence time, and detector e f f i c i e n c y . Representative electropherograms corresponding to the three detector efficiency determinations and the conditions under which the separations were performed are shown i n Figures 4, 5, and 6. The e f f i c i e n c i e s reported i n the tables were calculated using the following equation: 32

NOC -

(DPM

peak

){Residence Time){Efficiency}

{DPMp ){Detector Length/Zone Velocity){Efficiency}, eak


(2)


PENTONEY ET AL.

0


10

20

Time (min) F i g u r e 4. C a p i l l a r y e l e c t r o p h e r o g r a m o f adenosine-5'-[αP ] t r i p h o s p h a t e o b t a i n e d by i n j e c t i n g a p p r o x i m a t e l y 51 n C i (7 χ 1 0 " M s o l u t i o n ) onto t h e c a p i l l a r y and a p p l y i n g a c o n s t a n t p o t e n t i a l o f -20 kV. The s e p a r a t i o n was c o n t i n u o u s l y m o n i t o r e d u s i n g t h e CdTe s e m i c o n d u c t o r radioisotope detector. Data were s u b j e c t e d t o a 5 - p o i n t s l i d i n g smooth. E l e c t r o l y t e was 0.2 M b o r a t e b u f f e r , pH 8.1. S 2

8




15

20

Time (inin)

Figure 5. C a p i l l a r y electropherogram of adenosine-5[aP ] triphosphate obtained by i n j e c t i n g approximately 38 nCi (6 χ 10" M solution) onto the c a p i l l a r y and applying a constant potential of -25 kV. The separation was monitored using the parabolic p l a s t i c s c i n t i l l a t o r radioisotope detector. Data were subjected to a 5-point s l i d i n g smooth. The e l e c t r o l y t e was the same as i n Figure 4. 3 2

8

333 ATP

υ (Ο 3 Ο

Ο

4.5

7 Time

9.5

12

15

(min)

Figure 6. C a p i l l a r y electropherogram of adenosine-5'-[7P ] triphosphate obtained by i n j e c t i n g approximately 25 nCi (6 χ 10" M solution) onto the c a p i l l a r y and applying a constant potential of -25 kV. The separation was monitored using the coincidence radioisotope detector. S 2

8


4. PENTONEY ET AL.

73


where "NOC" represents the number of observed counts integrated over a peak, "DPMpeak" represents the number of radioactive transformations occurring each minute i n the injected sample plug, "Residence Time" i s the amount of time ( i n minutes) a radioactive molecule within a given sample zone spends i n the detection volume, and "Efficiency" i s the f r a c t i o n a l number of events sensed by the detector. The e f f i c i e n c i e s for the on-line detectors described here Table IV.


Run 1 2 3 4 5 6 7 8 9 10 11 av sd %RSD

E l u t i o n Time, mins 18.33 17.96 18.30 18.07 17.98 17.48 17.76 17.32 17.91 17.37 17.77 17.84 0.34 1.9 Table V.

Run 1 2 3 4 5 6 7 8 9 10 av sd %RSD

CdTe Radioisotope Detector E f f i c i e n c y Data Peak Area, counts 1519 1372 1461 1511 1407 1601 1197 1195 1392 1266 1359 1389 132 9.5

Residence Time, mins 0.049 0.048 0.049 0.048 0.048 0.047 0.047 0.046 0.048 0.046 0.047

Efficiency 27.6% 25.4% 26.5% 28.0% 26.1% 30.3% 22.7% 23.1% 25.8% 24.5% 25.7% 26.0% 2.2% 8.4%

Parabolic Radioisotope Detector E f f i c i e n c y Data

Elution Time, mins 20. 65 20. 62 20. 55 20. 54 20.,69 20.,84 21. 22 21..00 20..84 20..88 20,.78 0,.22 1 .06

Peak Area, counts 3375 3088 3119 3091 2953 2881 2883 2952 3213 3122 3068 156 5.07

Residence Time, mins 0.055 0.055 0.055 0.055 0.055 0.056 0.057 0.056 0.056 0.056

Efficiency 72. 7% 66.,5% 67.,2% 66.,5% 63..6% 60..9% 59..9% 62,.4% 67,.9% 66,.0% 65,.4% 3,.8% 5 .8%

are largely a function of detector c o l l e c t i o n geometry, i . e . , positioning of the CdTe probe or p l a s t i c s c i n t i l l a t o r with respect to the c a p i l l a r y . Note that the residence time within the detector must be determined for each component i n a mixture because separated sample zones t r a v e l with different v e l o c i t i e s according to t h e i r individual electrophoretic m o b i l i t i e s . This i s i n sharp contrast to radio-HPLC detection, i n which the residence time for each sample component i s the same and i s given by the r a t i o of the detector c e l l


74


volume to the mobile phase flow rate. The residence time for a p a r t i c u l a r sample component separated by c a p i l l a r y electrophoresis i s e a s i l y obtained from i t s migration time and from the length of c a p i l l a r y to which the detector i s exposed. Table V I .


Run 1 2 3 4 5 6 7 8 9 10 11 12 av sd

Coincidence Radioisotope Detector E f f i c i e n c y Data

Elution Time, mins 11. 52 11..58 11..87 11..65 11. 58 11..45 11..18 10,.75 10..92 11,.05 11,.77 11,.02

Peak Area, counts 2236 2196 2031 2277 2237 1966 1936 2038 2182 2311 2072 2200

Residence Time, mins 0.042 0.042 0.043 0.042 0.042 0.042 0.041 0.039 0.040 0.040 0.043 0.040

Efficiency 103% 101% 92% 105% 103% 91% 91% 101% 112% 101% 93% 106% 100% 6.97 6.97%

%RSD

Results obtained for the replicate runs shown i n Tables IV, V, and VI indicate that the measured efficiency of P detection for the on-line CdTe r a d i o a c t i v i t y detector i s approximately 26%, while the efficiency of the p l a s t i c s c i n t i l l a t o r r a d i o a c t i v i t y detectors is approximately 65% (parabola) and 100% (coincidence u n i t ) , r e f l e c t i n g the improved geometry of the l a t t e r two devices. The background noise l e v e l of the CdTe detector system i s a function of the low energy discriminator setting. The value of 0.01 MeV selected for a l l experiments reported here gave a background count rate of approximately 10 counts per minute while leaving a wide energy window open for detection. Comparison of signal-to-noise r a t i o s i n the three electropherograms indicates that the three detectors exhibit quite s i m i l a r s e n s i t i v i t i e s despite the fact that the e f f i c i e n c y of the p l a s t i c s c i n t i l l a t o r detectors i s considerably greater than that of the semiconductor detector. This difference i n s e n s i t i v i t y i s caused by the extremely low background noise l e v e l of the CdTe device compared with a photomultiplier tube. The large gain i n s e n s i t i v i t y afforded by on-line radioisotope detection i n comparison with the more commonly used UV-absorbance detector i s i l l u s t r a t e d i n Figure 7. In this example, a UVabsorbance detector, monitoring at 254 nm, was positioned 8.5 cm downstream from a CdTe radioisotope detector, and P - l a b e l e d ATP was injected at a concentration of approximately 5 χ 10" M. Under these conditions, ATP i s detected with an excellent signal-to-noise r a t i o by the radioisotope detector but i s completely undetectable by UV absorbance. 3 2

32

8



4. PENTONEY ET A L


Time (min) Figure 7. Electropherograms showing (A) CdTe radioisotope detector response and (Β) UV absorbance detector response. The injected sample was 5 χ 1 0 ' · M P - l a b e l e d ATP. The UV absorbance detector was located 8 cm downstream from the radioisotope detector, and absorbance was monitored at 254 nm. S2


75


76


Flow Programming. Equation 2 suggests that the number of counts measured (the detector s e n s i t i v i t y ) over a sample peak may be increased by lengthening the residence time of the sample i n the detection volume. This i s equivalent to increasing the counting time on a l i q u i d s c i n t i l l a t i o n counter and this concept has been recognized i n radio-HPLC applications (12) and applied to radioisotope detection i n isotachophoresis (2). In c a p i l l a r y electrophoresis, the v e l o c i t y of a sample zone may be reduced and i t s residence time increased by simply reducing the applied potential as the zone passes through the detection volume. The most e f f i c i e n t implementation of this flow-programming concept would involve reducing the zone v e l o c i t i e s only while the labeled sample was present within the detection volume and operating at a r e l a t i v e l y high potential at a l l other times. To our knowledge, this type of flow programming has not previously been explored i n c a p i l l a r y electrophoresis. Although i t i s demonstrated only for radioisotope detection here, this methodology should be applicable to other modes of sample detection i n CE. The flow-programming concept is demonstrated i n Table V I I , which l i s t s the peak width and peak area for six c a p i l l a r y Table VII. Run

Peak Width, mins 0.38 0.34 0.43 0.38 0.48 0.35

Peak Area, counts 985 1098 1236 1016 1078 968

Average -

0.39

1064

7 8 9 10 11 12

0.87 0.79 0.71 0.81 1.12 0.80

2404 2705 2081 2448 2695 2673

0.85

2501

1 2 3 4 5 6

Elution Time, mins 18.00 18.34 19.36 18.04 18.75 18.14

Flow-Programmed Runs

18.75 18.45 17.30 17.65 16.82 19.02

Average Peak Area Ratio Current Ratio Voltage Ratio

Voltage Program 20 kv constant

20 kv i n i t i a l 10 kv during detection period

2.4 2.4 2.0

electrophoresis separations performed with and without flow programming. Separations 1 through 6 were performed at a constant potential of -20 kV using the CdTe radioisotope detector, while i n runs 7 through 12 the potential was reduced to -10 kV as soon as signal was detected above the detector background l e v e l . Because the zone v e l o c i t y i s d i r e c t l y proportional to the applied f i e l d strength, the average temporal peak width and area (number of counts observed) for the six flow-programmed runs were approximately



4. PENTONEY ET A L


77

doubled. This improvement i n s e n s i t i v i t y i s , however, accompanied by an increase i n analysis time as well as a small loss i n resolution due to zone broadening. The magnitude of the resolution losses incurred during flow programming w i l l be strongly dependent upon the amount of sample injected and the additional run time associated with the flow programming process. For injected sample plug lengths several times larger than the length associated with d i f f u s i o n a l broadening ( t y p i c a l operating conditions), the resolution loss w i l l not be s i g n i f i c a n t . In the l i m i t of injected sample plugs with no i n i t i a l width (6 function), the additional peak variance increases l i n e a r l y with programming time (ignoring analytewall interactions) and the resolution loss w i l l become quite significant. A s t r i k i n g example of increased s e n s i t i v i t y gained through the application of flow programming is i l l u s t r a t e d i n Figures 8 and 9. In Figure 8, a synthetic mixture of thymidine-5'-[a- P]triphosphate ( P - T T P ) , c y t i d i n e - 5 ' - [ a - P ] t r i p h o s p h a t e ( P - C T P ) , and adenosine s' - [ Q - P ] t r i p h o s p h a t e ( P - A T P ) , with each component present at a concentration of approximately 3 χ 1 0 " · M ( « 19 nCi injected), was injected using hydrostatic pressure and separated under the influence of a constant -20 kV applied p o t e n t i a l . In Figure 9, the sample solution and i n j e c t i o n volume were the same as i n Figure 8, but the residence time of each component was increased by reducing the applied p o t e n t i a l from -20 kV to -2 kV as radioactive sample was passing through the detection volume. At the same time, the counting i n t e r v a l was increased proportionately (from 1 second to 10 seconds) and the detector signal was plotted as a function of e l e c t r o l y t e volume displaced through the c a p i l l a r y tube. Note that this results i n a time-compressed abscissa over the flow programmed period of the electropherogram (the entire separation required about 70 minutes i n this case). It i s important to point out that the lower l i m i t of radioisotope detection refers to the lowest sample a c t i v i t y contained within a peak that can be detected and accurately quantified. From the data presented i n Table VII and Figures 4-9 i t is apparent that the lower l i m i t of detection for this system is greatly dependent upon the conditions under which the analysis i s performed, and that detector s e n s i t i v i t y may be extended by an order of magnitude or more using flow programming. The s e n s i t i v i t y gain afforded by this flow-programming methodology w i l l ultimately be limited by p r a c t i c a l considerations of analysis time and resolution losses caused by d i f f u s i o n a l broadening of the sample zones. S i m p l i c i t y and consideration of analysis time, however, s t i l l make flow-counting detection for c a p i l l a r y electrophoresis an a t t r a c t i v e alternative to the quantitatively superior batch-counting approach i n which fractions are c o l l e c t e d and subjected to conventional counting techniques (12). The batch-counting approach, provided that s u f f i c i e n t l y small fractions may be c o l l e c t e d , does offer the advantage of decoupling separation considerations from measurement time. Considering only the l i m i t a t i o n imposed by d i f f u s i o n a l spreading of sample zones during the flow-programmed portion of a run, i t i s possible to predict the extent to which detector s e n s i t i v i t y may be improved by flow programming. For an i n j e c t i o n volume of 84 nL, as used i n this example, and a maximum allowable increase i n zone variance defined S2

S2

32

S2

82

32




90 CTP ATP

701

TTP

50H

ϋ 30

10

lulu*. τ

0

•

1

5

•

1

10

•

1

'

15 Time

1

20

'

1

•

25

1

30

(min)

Figure 8. C a p i l l a r y electropherogram of t h y m i d i n e - 5 [ a P ] triphosphate, c y t i d i n e - 5 ' - [ a - P ] triphosphate, and a d e n o s i n e - 5 ' - [ Q - P ] triphosphate obtained by i n j e c t i n g approximately 19 nCi (2 χ 1 0 " · M solution) of each component onto the c a p i l l a r y and applying a constant p o t e n t i a l of -20 kV. S 2

S 2

S2


4. PENTONEY ET A L



790

590 cd > CD CU

390 λ

E c D Ο

ϋ

190

0

5 Electrolyte

10

15

Volume Displaced (/xL)

Figure 9. Flow-programmed c a p i l l a r y electropherogram of thymidine-5'-[a- P] triphosphate, c y t i d i n e - 5 ' - [ a - P ] triphosphate, and adenosine-5'-[a- P] triphosphate obtained by i n j e c t i n g approximately 19 nCi (2 χ 1 0 " · M solution) of each component onto the c a p i l l a r y . The separation was flow programmed by applying a constant potential of -20 kV u n t i l radiolabeled sample approached the detection volume and then reducing the potential to -2 kV as the sample zones traversed the detection region. Note that the detector signal i s plotted as a function of electrolyte volume displaced, r e s u l t i n g i n a time-compressed abscissa over the flow programmed region of the electropherogram. The operating current was 38 μΑ at -20 kV and 3.8 μΑ at -2 kV. The data were subjected to a 5-point s l i d i n g smooth. S2

S2

S2


79

80


to be 10%, approximately 84 minutes of flow programming would be permitted (this c a l c u l a t i o n assumes a rectangular i n j e c t i o n p r o f i l e and a solute d i f f u s i o n coefficient of 10" cm /sec, and neglects both d i f f u s i o n a l broadening p r i o r to flow programming and v e l o c i t y dependent analyte-wall i n t e r a c t i o n s ) . This 10% increase i n v a r i a n c e would be accompanied by a 250-fold increase i n the number of counts observed over a peak. Because the s e n s i t i v i t y of radioisotope detection i s governed by counting s t a t i s t i c s , a 16-fold increase i n the signal-to-noise r a t i o ( N 0 C / ( N 0 C ) w o u l d r e s u l t . Thus, a lower l i m i t of detection of about 10" M would seem to be a conservative extrapolation. Obviously, the limitations imposed by d i f f u s i o n a l broadening would become more severe i f the i n i t i a l i n j e c t i o n volume were reduced. In an automated implementation of the flow-programming methodology, that i s , with the high-voltage power supply under computer c o n t r o l , there i s a further l i m i t a t i o n imposed upon achievable s e n s i t i v i t y gains. There must be enough sample present to generate a signal s u f f i c i e n t l y large to exceed the detector background l e v e l under standard (non-flow-programmed) conditions, in order to i n i t i a t e the flow-programming procedure. In c e r t a i n instances, however, p r i o r knowledge of e l u t i o n times for the compounds of interest would permit this l i m i t a t i o n to be overcome. 6

2


11

Application to C a p i l l a r y Gel Electrophoresis. Recently, Karger and co-workers demonstrated the use of polyacrylamide g e l - f i l l e d c a p i l l a r i e s to separate peptide/protein (SDS PAGE) (18) and oligonucleotide mixtures (12.2Û) by c a p i l l a r y electrophoresis. This mode of CE operation may prove to couple well with on-line radioisotope detection. The results of several preliminary c a p i l l a r y electrophoresis separations using g e l - f i l l e d c a p i l l a r i e s and on-line radioisotope detection using the coincidence unit described here are presented below. The c a p i l l a r y gel electrophoresis separation of a threecomponent, P - l a b e l e d , nucleotide mixture i s i l l u s t r a t e d i n Figure 10. It i s interesting to note that the migration order of ATP and CTP i s reversed with respect to the free solution separations presented i n Figures 8 and 9. This i s caused by the absence of electroosmotic flow i n the g e l - f i l l e d c a p i l l a r y . Figure 11 i l l u s t r a t e s the CE separation of synthetic polythymidylic oligomers. The c a p i l l a r y gel electrophoresis separation of this sample has previously been described by Paulus and Ohms (21) using UV-absorbance detection. The polythymidylic 50mer sample was synthesized with the reaction conditions purposely adjusted to increase the f a i l u r e rate at every f i f t h base, beginning with the 15-mer. Figures 12 and 13 i l l u s t r a t e the CE separation of P - l a b e l e d 29- and 30-base heterooligomers, respectively. In the two electropherograms, the major component i s n i c e l y resolved from several f a i l u r e sequences that were also phosphorylated i n the l a b e l i n g procedure of the 5' end. Figure 14 i l l u s t r a t e s the c a p i l l a r y gel separation of a mixture containing these two heteropolymers. These two polymers d i f f e r only by the absence or presence of a 3' terminal thymidine residue. 32

32


PENTONEY ET A L


190ATP 170150130-


110GTP

90CTP

ο ϋ

70503010-

U ι

22

•

1

•

24

1

>

26

1

— •

28

— ι

30

1

1

32

Time (min)

Figure 10. Capillary gel electrophoresis separation of a simple three-component nucleotide mixture with on-line radioisotope detection using the coincidence u n i t .

Ί 2

1 12

1 22

1 33.6

Time (min) Figure 11. C a p i l l a r y gel electrophoresis separation a poly (T) oligomer sample P - l a b e l e d at the 5' end. Detection was accomplished using the coincidence radioisotope detector. S2




10090-

32-P labeled 29-mer

805' GCCCACCACCACAAGCTTGT ATTCTGTCA 3*

7060ο φ

50-

w c 3 Ο ϋ

40302010-

^ ^ ^ ^

o-1020

—ι— 30

40

—ι— 50

60

Time (min) Figure 12. Capillary gel electrophoresis separation of heteropolymer P - l a b e l e d at the 5' end. The oligomer was 29 units i n length with the sequence shown i n the Figure. Detection was accomplished using the coincidence radioisotope detector. 32



PENTONEY ET A L


-

32-P labeled 30-mer

5' GCCCACCACCACAAGCTTGT ATTCTGTCAT 3'

20

30

40

50

60

Time (min) Figure 13. C a p i l l a r y gel electrophoresis separation of heteropolymer - l a b e l e d at the 5' end. The oligomer was 30 units i n length with the sequence shown i n the f i g u r e . Detection was accomplished using the coincidence radioisotope detector. 3 2




60 29-mer

|

50 30-mer 40 H ο ο

Q > ço c 3 Ο ϋ

30

ι

20 Η

10Η

35

40

45

50

55

Time (min) Figure 14. C a p i l l a r y gel electrophoresis separation of 29and 30-mer, P - l a b e l e d at the 5' end. Detection was accomplished using the coincidence radioisotope detector. S2


4. PENTONEY ET A L


85

I t i s worth noting that the s e n s i t i v i t y advantage of radioisotope detection i n comparison with UV absorbance detection i s not so large for the biopolymers studied here as i t i s for small molecules such as the nucleotide triphosphates. This i s because the labeling scheme at the 5' end transfers only one P atom per oligomer molecule while the same oligomer, containing many chromophores, absorbs quite strongly. Hence the s e n s i t i v i t y advantage of radioisotope detection w i l l continue to decline as the length of the oligomer increases. Obviously, a l a b e l i n g scheme that resulted i n the transfer of more radiolabels per molecule would increase the s e n s i t i v i t y difference of the two detectors. Perhaps a more s i g n i f i c a n t advantage of radioisotope detection i s that the separation medium need not be o p t i c a l l y transparent. This opens up a broader range of buffer components and matrix s t a b i l i z e r s from which the separation medium may be formulated. The application of flow programming to the analysis of samples using g e l - f i l l e d c a p i l l a r i e s also results i n a s e n s i t i v i t y gain. This i s i l l u s t r a t e d i n Figures 15 and 16. Figure 15 shows the CE separation of a polydeoxyadenosine 40-60 mer sample using a polyacrylamide-filled c a p i l l a r y under constant voltage. The electropherogram presented i n Figure 16 was obtained using the same sample and column but under flow-programmed conditions. The enhancement i n s e n s i t i v i t y i s marked.


S 2

Autoradiography. Of course, residence time can be a r b i t r a r i l y increased by removing the applied f i e l d . However, there i s an obvious trade-off between measurement time and resolution loss caused by d i f f u s i o n a l broadening. The l a t t e r may be s i g n i f i c a n t l y reduced by freezing the c a p i l l a r y contents. This allows an autoradiographic view of the separation channel by exposing d i r e c t l y the frozen c a p i l l a r y to x-ray film (Kodak Diagnostic XAR-5, Rochester, NY). Autoradiography i s a well-established technique (26). This method i s not limited to high energy β emitters; i t can be applied to detect radiolabels that have less penetrating r a d i a t i o n , provided that the c a p i l l a r y contents are formulated to scintillate. Figure 17 shows the autoradiographic detection of the plabeled polydeoxyadenosine homopolymer sample, which i s the same sample as shown i n Figures 15 and 16. The gain i n s e n s i t i v i t y i s several orders of magnitude. Indeed, i t i s even possible to discern P - l a b e l e d material "trapped" i n the gel matrix between peaks. Clearly this method i s the method of choice when s e n s i t i v i t y i s an issue. The same method can also be applied to f r e e - s o l u t i o n c a p i l l a r y electrophoresis. 3 2

S2

Conclusion Three simple, on-line radioisotope detectors for c a p i l l a r y electrophoresis were described and characterized for the analysis of P - l a b e l e d analytes. The minimum l i m i t of detection for these systems was shown to be strongly dependent upon the conditions under which the analysis i s performed. For standard CE separations performed at a r e l a t i v e l y high (constant) voltage, the minimum l i m i t of detection was found to be i n the low nanocurie (injected sample 32



86


poly d(A) 4 0 - 6 0 , 5' e n d - l a b e l e d , inject 15 s e c , 5 kV; run 15 kV constant, gel capillary column 7 8 / 9 8 cm, 7 5 micron i . d .

400

c

ε (0

+—

c

D Ο

ϋ

mm

il

11

1

\\\

80

101 Time (min)

Figure 15. Electropherogram i l l u s t r a t i n g the c a p i l l a r y electrophoresis separation of poly d(A) 40-60 mer sample, P - l a b e l e d at the 5' end. Detection was accomplished using the coincidence detector. The separation was accomplished using a polyacrylamide g e l - f i l l e d c a p i l l a r y and a constant p o t e n t i a l of 15 kV. The sample a c t i v i t y i n this example was approximately 4800 DPM/nL. 82


PENTONEY ET A L


4οομ

ίΛ/iiir


poly d(A) 4 0 - 6 0 , 5' e n d - l a b e l e d , inject 15 s e c , 5 k V ; run 15 kV, flow program by reducing potential to 1.5 kV at 81.3 min. gel capillary column 7 8 / 9 8 cm, 75 micron i.d.

Ik u

J ι) \J * 81

275 Time (min)

Figure 16. Electropherogram i l l u s t r a t i n g the flowprogrammed separation of the same ply d(A) 40-60 mer sample presented i n Figure 15. The sample was separated at 15 kV and the potential was reduced to 1.5 kV as radiolabeled sample reached the detector. The s e n s i t i v i t y improvement afforded by flow programming is readily apparent i n this figure.

Figure 17. Autoradiogram showing the separated poly d(A) 40-60 mer sample located within a polyacrylamide g e l - f i l l e d capillary. The autoradiogram was measured by placing the c a p i l l a r y on a piece of x-ray film and freezing the c a p i l l a r y - f i l m combination at -20 C for 15 hours. The leading end of the sample is on the right i n this photograph. e


88


quantity) range, corresponding to an analyte concentration of about 10* M. The lover l i m i t of detection for this type of detection system was extended to the sub-nanocurie l e v e l ( » 10" M) by application of flow programming methodology which served to increase the residence time of labeled sample components within the detection volume. Further large gains have been demonstrated by freezing the contents of the c a p i l l a r y after separation and exposing the frozen c a p i l l a r y to f i l m (autoradiography). Thus radioisotope detection, when applicable, has a s e n s i t i v i t y superior to most other detection schemes, comparable to electrochemical detection (22.21) laserinduced- fluorescence detection (12.24.25). One improvement to the current systems would involve automation of the flow-programming methodology, and such efforts are currently underway. A second improvement over the current semiconductor system would involve optimizing the detector geometry by capturing a larger s o l i d angle with the CdTe detector. The performance of the parabolic p l a s t i c s c i n t i l l a t o r detector would be greatly improved by reducing the background noise l e v e l through the use of a quieter photomultiplier tube. In c e r t a i n instances i t would be desirable to reduce the effective detection volume of these systems i n order to increase resolution. This could be accomplished by i n s t a l l i n g a narrower aperture i n the semiconductor detector or machining a smaller detection region from the p l a s t i c s c i n t i l l a t o r materials. In either case, detector s e n s i t i v i t y would be reduced, because the detection volume and effective residence time would be decreased. Hence, there is once again a p r a c t i c a l trade-off between detector s e n s i t i v i t y and resolution. On-line radioisotope detection has been demonstrated to be a p r a c t i c a l alternative to UV absorbance detection when g e l - f i l l e d c a p i l l a r i e s are used for CE separations. Significant improvement i n detection l i m i t s i s r e a l i z e d with radioisotope detection. The greatest improvement i s r e a l i z e d for small molecules and i s roughly one to two orders of magnitude (for runs i n which the residence time i s not enhanced). Future work i n this area w i l l focus on the extension of this detection scheme to include other radioactive isotopes. The present systems are applicable to high-energy β and 7 emitters. Peptide and protein samples labeled with ΐ or I should prove to be an interesting a p p l i c a t i o n . Weaker β emitters w i l l require that the sensing device be placed i n d i r e c t contact with the e l e c t r o l y t e solution and that the sensing device be compatible with changing f i e l d gradients. 9

10


a

1 2 δ

n

d

1 3 1

Acknowledgments S . L . P . wishes to thank David J . Rakestraw, Patrick H. Vaccaro, W. Howard Whitted and James Burns for many helpful conversations pertaining to this work. The assistance of Aaron Paulus, Andras Guttman, and Sushma Rampai i n the preparation of g e l - f i l l e d c a p i l l a r i e s and Karen Wert i n the efficiency determinations is also gratefully acknowledged. Credit Support for this work by Beckman Instruments, acknowledged.

Inc. i s

gratefully



4. PENTONEY ET A L


89

Literature Cited 1. Nickerson, Β.; Jorgenson, J. W. J. High Resolut. Chromatogr. Chromatogr. Commun. 1988, 11(7), 533-534. 2. Jorgenson, J. W.; Lukacs, K. D. Science 1983, 222, 266-272. 3. Kessler, M. J. In Analytical and Chromatographic Techniques in Radiopharmaceutical Chemistry;Wieland,D. M.; Tobes, M. C.; Mangner, T. J., Eds.; Springer-Verlag: New York, 1987; Ch. 5-7. 4. Roberts, T. R. Radiochromatography, Journal of Chromatography Library; Elsevier: Amsterdam, 1978, Ch. 6. 5. Kessler, M. J. Am. Lab. 1988, 20(6), 86-95. 6. Kessler, M. J. Am. Lab. 1988, 20(8), 76-81. 7. Kaniansky, D.; Rajec, P.; Švec, Α.; Havaši, P.; Macášek, F. J. Chromatogr. 1983, 258, 238-243. 8. Kaniansky, D.; Rajec, P.; Švec, Α.; Marák, J.; Koval, M.; Lúcka, M.; Franko, Š.; Sabanoš, G. J. Radioanal. Nucl. Chem. 1989, 129(2), 305-325. 9. Altria, K.D.; Simpson, C.F.; Bharij, Α.; Theobald, A.E. Paper presented at the 1988 Pittsburgh Conference and Exposition, abstract no. 642, New Orleans, February 1988. 10. Berry, V. LC/GC 1988, 6, 484-491. 11. Needham, R.E.; Delaney, M.F. Anal. Chem. 1983, 55, 148-150. 12. Gassmann, E.; Kuo, J.E.; Zare, R.N. Science 1985, 230, 813814. 13. Gordon, M. J.; Huang, X.; Pentoney, S.L., J r . ; Zare, R.N. Science 1988, 242, 224-228. 14. Huang, X.; Gordon, M.J.; Zare, R.N. Anal. Chem. 1988, 60, 1837-1838. 15. Karger, B. L.; Cohen, A. U. S. Patent #4865707, 1989. 16. Knoll, G.F. Radiation Detection and Measurement; Wiley: New York, 1979. 17. See, for example, Markl, P. In Instrumentation for High Performance Liquid Chromatography; Journal of Chromatography Library, Volume 13; Elsevier: Amsterdam, 1978, pp. 151-161. 18. Cohen, A. S.; Karger, B. L. J. Chromatogr. 1987, 397, 409417. 19. Guttman, Α.; Paulus, Α.; Cohen, A. S.; Karger, B. L. Electrophoresis '88, Proc. Int. Electrophoresis Society Meeting, 6th, Copenhagen, 1988, pp. 151-159. 20. Cohen, A. S.; Najarian, D. R.; Paulus, Α.; Guttman, Α.; Smith, J. Α.; Karger, B. L. Proc. Natl. Acad. Sci. 1988, 85, 9660-9663. 21. Paulus, Α.; Ohms, J. J. J. Chromatogr. 1989, accepted for publication. 22. Wallingford, R. Α.; Ewing, A. G. Anal. Chem. 1988, 60, 258263. 23. Wallingford, R. Α.; Ewing, A. G. Anal. Chem. 1987, 59, 1762-1766. 24. Gozel, P.; Gassmann, E.; Michelsen, H.; Zare, R. N. Anal. Chem. 1987, 59, 44-49. 25. Dovichi, N. J. Paper presented at the 41st ACS Summer Symposium on Analytical Chemistry, Stanford University, 26 to 29 June 1988. 26. Stryer, L. Biochemistry; Freeman and Company, 1981, Ch. 24. RECEIVED December 20, 1989


Analytical Biotechnology - ACS Publications - American Chemical

Recommend Documents