Created: 1st September 2000, last updated: 30th October 2000, © 2000 ABRF

METHODS AND REVIEWS

 

10-nmol Oligonucleotide Synthesis for the ABI Model 394 DNA Synthesizer

Douglas A. Bintzler and Catherine E. Terrell

University of Cincinnati DNA Core Facility, Cincinnati, OH

Sequencing and gene contig assembly generally involve primer walking, in which an oligonucleotide primer is used once and then discarded. Because the smallest commonly available scale for commercial oligonucleotide synthesizers is the 40-nmol scale, a large excess of product is produced. This results in reagent waste, excess cost to the researcher, and increased production time. A logical solution would be to develop a smaller scale for the commercial synthesizers. A 10-nmol scale was developed for the Applied Biosystems ABI model 394 DNA synthesizer. To develop the 10-nmol scale, a column support was first developed from commonly available materials. Then a 10-nmol cycle was developed from the systematic reduction of the reagents. The final product of the 10-nmol cycle was tested for quality by monitoring the coupling efficiency, polyacrylamide gel electrophoresis, and automated DNA sequencing. The final cycle reduces the cost of the oligonucleotide and decreases the time required to produce it. (J Biomol Tech 2000;11:122-134)

Key Words: DNA synthesis, oligonucleotide, primer.

Address correspondence and reprint requests to: Douglas A. Bintzler, 231 Bethesda Avenue, Cincinnati, OH 45267-0524 (email: bintzlda@email.uc.edu).

 

DNA oligonucleotide synthesis produces a single-stranded oligonucleotide product. The oligonucleotide is synthesized on a support filled with control pore glass (CPG) or polystyrene or is synthesized using a membrane. The oligonucleotide synthesis cycle is depicted in Figure 1.1 The DNA is synthesized from 3' to 5' with the first base on the support prior to oligonucleotide synthesis.1 Each base thereafter is delivered in the form of a phosphoramidite, which is protected on the 5' end by a hydroxyl group with dimethoxytrityl (DMT).1 After a base is added, the DMT is removed by exposure to trichloroacetic acid (TCA); this removal is apparent because of the bright orange color of DMT while going through the support.1 Because only bases that successfully attach to the growing oligonucleotide chain are present during the removal of DMT, this is used as a means of determining the efficiency of base coupling during each cycle. The next step is activating the 5' hydroxyl by treatment with a weak acid, tetrazole, which is also referred to as the activator.1 The addition of the activator in conjunction with the next base ensures attachment of the next base. Combining methylimidazole and acetic anhydride in the support forms an acetylating reagent. The acetylating reagent caps the end of failure sequences and unincorporated bases.1 The newly attached base is treated with iodine, an oxidizer, to become a phosphotriester, which is a more stable product; it is then ready to receive the next base.1 After synthesis of the oligonucleotide is complete, it is cleaved from the support using 1.25 mL of 30% ammonium hydroxide and deprotected in a heat bath at 80°C for 45 minutes under pressure in a tightly sealed vial.

FIGURE 1. Schematic diagram of the oligonucleotide synthesis cycle. Oligonucleotide synthesis is initiated by removal of the dimethoxytrityl detritylation from the first base attached to the support. Detritylation is followed by coupling the next base to the oligonucleotide chain on support. Unincorporated bases and failure sequences are capped with an acetylating agent. The newly added base is oxidized for preparation of the next base.

 

The process of measuring the DMT efficiency was first performed manually by spectrophotometric absorbencies.1 Currently, the DNA synthesizers have automated DMT monitoring using absorbency or conductivity. The ABI model 394 DNA synthesizer (Applied Biosystems, Foster City, CA) was fitted with conductivity flow cells for each column. The conductivity flow cells monitor the DMT cation, or signal value, as it is cleaved from the oligonucleotide chain.2 The background, or noise value, is measured after the DMT has gone through the conductivity flow cell.2 The difference between signal value and noise value is a measurement of the DNA base most recently attached to the oligonucleotide and is converted to percentage as coupling efficiency.1,2

For the purpose of developing a 10-nmol cycle, it was necessary to first develop a support with a minimum void volume to allow for the reduction of reagents for oligonucleotide synthesis. Next, an attempt was made to reduce the amount of deblock, activator, oxidizer, and acetonitrile wash necessary for successful oligonucleotide synthesis.

 

MATERIALS AND METHODS

10-nmol Support

The support for 10-nmol DNA synthesis is a circular membrane that is 0.4 cm in diameter. Each membrane is cut from a PerSeptive Biosystems 0.2-µmol MemSyn (Framingham, MA), a DNA membrane support with the 3' oligonucleotide base attached. The 0.2-µmol membrane is cut using a 0.4-cm hole punch purchased from McMaster Carr (Cleveland, OH). Two membranes are pushed inside a PerSeptive Biosystems column union support and pressed between two porcelain filters called frits. The frit filters are acquired by removing the frit from a PerSeptive Biosystems post column filter, a filter device placed on the PerSeptive Biosystems M.O.S.S. Expedite DNA synthesizers. A diagram of the 10-nmol DNA synthesis support is shown in Figure 2. The column union supports and frit filters are cleansed by sonication and rinsed with 100% ethanol after DNA synthesis and reused.

FIGURE 2. Schematic diagram of the 10-nmol DNA synthesis support. The DNA synthesis support consists of three components: (1) the union column support, which provides containment for the DNA synthesis support membrane; (2) the frit filters, which hold two DNA synthesis membranes in the center of the union column support; and (3) the DNA synthesis support membranes.

 

The size of the membrane inside the 10-nmol DNA synthesis support is equivalent to 5% of the surface area of the membrane of the 0.2-µmol MemSyn. However, the circular shape will allow only 10 10-nmol DNA synthesis supports to be made from each 0.2-µmol MemSyn. A preliminary study to determine the amount of synthesized oligonucleotide cleaved from different areas of the 0.2-µmol MemSyn show negligible variance; therefore, the entire membrane inside the 0.2-µmol MemSyn can be used.

10-nmol Cycle Development

The cycle for 10-nmol oligonucleotide synthesis was developed from a modified Applied Biosystems LV40cycle.3 Cycle optimization was achieved by reducing the amount of each reagent individually to accommodate for the low void volume (essentially zero) of the 10-nmol support. For each step that a reagent was reduced, the cycle was tested by synthesizing an m13(-21) primer (TGTAAAACGACGGCCAGT) and an m13rev primer (CAGGAAACAGCTATGACC). Coupling efficiency was determined for each base synthesized by recording DMT values, signal values, and noise levels.

The two universal primers were synthesized on the Applied Biosystems LV40 column with an in-house modified LV40 cycle to function as standards. Two universal primers were also synthesized with the 10-nmol support using the LV40 cycle to function as the controls. This was designated test cycle 1.The oligonucleotides synthesized for each cycle during the optimization of the 10-nmol cycle were compared with the standard and control oligonucleotides for all quality testing.

The DMT is removed from the base once it is attached to the oligonucleotide chain in steps 79 through 99 of the LV40 cycle. Steps 79 through 90 are active only when the instrument is programmed to measure DMT during its removal. This requires methylene chloride as an additional solvent. As a means of control, DMT is normally read every fourth base. However, for this experiment the DMT was read for every base. Steps 91 through 99 remove the DMT from the base without DMT measurement. By decreasing the acetonitrile wash and flush steps, reagent waste was further reduced.

In test cycles 2, 3, and 4, deblock reagent was gradually reduced. The changes made to the cycle during deblock reduction are shown in Table 1. The changes made for cycle 2 were retained and used in the final synthesis modifications.

 

TABLE 1
Reagent Reduction of Deblock Solution in 10-nmol Cycle Development
Step Control Test Cycle 2 Test Cycle 3 Test Cycle 4
Number     Step Action     Time (s)     Time (s)     Time (s)     Time (s)
79 If monitoring
80 19 to column 18.0 12.0 10.0 12.0
81 14 to column 3.0 3.0 2.0 2.0
82 Monitor trityls
83 14 to column 18.0 12.0 10.0 12.0
84 Monitor noise
85 14 to column 10.0 4.0 4.0 0.0
86 Stop monitoring
87 18 to column 10.0 5.0 5.0 5.0
88 Reverse flush 8.0 4.0 4.0 4.0
89 If not monitoring  
90 14 to column 6.0 6.0 6.0 6.0
96 18 to column 10.0 6.0 6.0 6.0
97 Trityl flush 8.0 5.0 5.0 5.0
98 14 to column
98 End monitoring
99 Wait
99 18 to column 8.0 6.0 6.0 6.0

Chemical reduction occurs in steps 80, 81, 83, 85, 87, and 88 during the dimethoxytrityl (DMT) monitoring phase in the cycle. Chemical reduction occurs in steps 96, 97, and 99 during DMT removal steps without DMT monitoring. Test cycle 2 maintained the best coupling efficiency and was kept for activator reduction. All changes made to the cycle are shown in bold.

 

The effects of activator reduction were measured in test cycles 5, 6, and 7. The proper volume of activator was critical in development of the 10-nmol cycle for two reasons. First, the correct volume of activator is necessary to carry the phosphoramidite into the column during base coupling. If too much activator is used, the base will pass through the column without attaching. Second, the distance the activator travels in the valve block increases for each column on the four-column instrument; thus, the time must also increase. Although columns 1 and 2 were tested during the initial development of the cycle, columns 1 through 4 were tested during the quality-control phase. The activator reductions for cycles 5 through 7 are given in Table 2.

 

TABLE 2
Reagent Reduction of Activator Solution in 10-nmol Cycle Development
Step Control Test Cycle 5 Test Cycle 6 Test Cycle 7
Number     Step Action     Time (s)     Time (s)     Time (s)     Time (s)
8 Block vent 2.0 2.0 2.0 2.0
9 TET to column 0.5 0.5 0.4 0.4
10 B+TET to column 0.2 0.2 0.2 0.2
11 TET to column 1.3 1.1 0.8 0.7
12 Flush to waste 0.6 0.6 0.6 0.6
13 18 to waste 4.0 4.0 4.0 4.0
14 Block flush 3.0 3.0 3.0 3.0
15 Column 1 off
16 Column 2 on
17 Block vent 2.0 2.0 2.0 2.0
18 TET to column 0.5 0.5 0.4 0.4
19 B+TET to column 0.2 0.2 0.2 0.2
20 TET to column 1.6 1.5 1.1 1.0
21 Flush to waste 0.6 0.6 0.6 0.6
22 18 to waste 4.0 4.0 4.0 4.0
23 Block flush 3.0 3.0 3.0 3.0

Chemical reduction occurs in steps 9, 11, 18, and 20 during the base coupling phase in the cycle. The amount of activator solution (TET) is important for pushing the phosphoramidite (B) into the support during the coupling phase. Test cycle 6 maintained the best coupling efficiency determined by dimethoxytrityl monitoring and was kept for oxidizer reduction. All changes made to the cycle are shown in bold.

 

The final reagent tested for reduction was the oxidizer. Reagent reductions from test cycle 6 were carried over, and the effects of the oxidizer reductions were measured in cycles 8 and 9. The cycle modifications are shown in Table 3.

 

TABLE 3
Reagent Reduction of Oxidizer Solution in 10-nmol Cycle Development
Step Control Test Cycle 8 Test Cycle 9
Number     Step Action     Time (s)     Time (s)     Time (s)
64 Block flush 3.0 3.0 3.0
65 15 to column 4.0 2.0 3.0
66 18 to waste 4.0 4.0 4.0
67 Block flush 3.0 3.0 3.0
68 Wait 15.0 15.0 15.0

Chemical reduction of oxidizer occurs in step 65. The initial reduction reduces the oxidizer by 50% in cycle 8 but results in a significant decrease in coupling efficiency. Cycle 9 reduces the oxidizer by 25% and also shows a reduction in coupling efficiency. Volume of oxidizer was not reduced for the 10-nmol cycle used. All changes made to the cycle are shown in bold.

 

Quality Control

All oligonucleotides synthesized from cycles 2 through 9, including both the standard and control, were tested by 15% polyacrylamide gel electrophoresis (PAGE). The samples were electrophoresed at 300 V for 1.25 hours. The oligonucleotide bands were visualized by staining with 1% methylene blue for 30 minutes. Excess stain was removed by destaining in 1% acetic acid in deionized water for 15 minutes.

Each oligonucleotide was diluted to a final concentration of 20 ng/µL in high-performance liquid chromatography (HPLC)-grade water, and a 3-µL aliquot was used as a primer for automated DNA sequencing using the Applied Biosystems BigDye terminator chemistry and following the protocol described. The primers were not desalted before sequencing. The pGem standard included in the BigDye chemistry kit was the template for all reactions. The ability of the oligonucleotides to function as DNA sequencing primers was determined visually from the quality of the sequence, including the read-length, signal strength, and number of incorrectly identified bases.

The final test of coupling efficiency was to synthesize oligonucleotides with longer base lengths. A 50-base, 75-base, and 100-base sequence was selected from the pGem-3Zf(+) vector sequence as a model. The sequences were synthesized on each of the four columns using test cycle 6. The final products were tested by 15% PAGE (300 V for 1.25 hours) and visualized with 1% methylene blue.

These longer oligonucleotides were amplified by polymerase chain reaction (PCR) to generate a double-stranded product using m13(-21) and m13rev tailed primers. The final products were purified through Spin-20 columns (Princeton Separations, Adelphia, NJ) and then sequenced on an Applied Biosystems Model 377 automated DNA sequencer. The results were compared with the original sequence that was synthesized.

 

RESULTS AND DISCUSSION

The size and shape of the 10-nmol DNA synthesis support greatly reduced the total volume; therefore, a systematic reduction of reagents was possible. DMT data for the standard and control oligonucleotides show the signal strength values and noise values (Table 4). The coupling efficiency for the standards and controls are similar. However, the signal value for the control oligonucleotides is less than one half that of the standard oligonucleotides. Because the signal values are lower, a decrease in DMT, and thus calculated coupling efficiency, is more sensitive to a decrease in the signal. Therefore, coupling efficiency values below 98% are considered normal for 10-nmol oligonucleotide synthesis.

 

TABLE 4
Dimethoxytrityl Measurement for Standard and Control Oligonucleotides
m13rev m13(-21)
Base     Trityl (%)     Signal     Noise     Base     Trityl (%)     Signal     Noise
Standards
C 100.0 626 87 G 100.0 666 88
A 100.0 633 90 A 100.0 686 93
G 95.9 577 98 C 99.1 687 101
T 96.9 646 108 C 99.4 695 113
A 97.5 632 129 G 98.6 688 134
T 97.9 650 158 G 97.8 692 173
C 98.0 678 206 C 98.1 764 217
G 96.0 632 241 A 97.9 758 257
A 96.4 691 233 G 98.0 764 268
C 96.8 700 245 C 98.2 760 268
A 97.1 712 242 A 98.1 748 270
A 97.3 688 258 A 98.2 776 283
A 97.5 683 256 A 97.9 735 284
G 97.7 664 266 A 98.1 739 288
G 97.8 665 255 T 98.2 736 285
A 98.0 685 262 G 98.0 720 291
C 98.1 685 260 T 98.1 713 263
Controls
C 100.0 266 139 G 100.0 287 135
A 96.8 261 142 A 99.0 290 141
G 97.9 262 138 C 99.3 288 139
T 96.9 260 148 C 97.4 284 147
A 95.3 258 158 G 96.5 287 160
T 96.1 265 162 G 97.0 298 163
C 95.6 260 167 C 97.4 292 166
G 96.2 257 161 A 97.7 291 165
A 95.4 249 166 G 97.2 283 165
C 95.8 254 167 C 96.8 273 163
A 95.8 244 165 A 96.6 266 162
A 95.5 240 167 A 96.3 258 161
A 95.4 233 164 A 96.6 260 163
G 95.7 238 163 A 96.0 249 163
G 95.9 231 163 T 96.3 255 163
A 96.0 230 164 G 96.4 242 158
C 95.7 224 164 T 96.6 245 159

m13(-21) And m13rev were synthesized by the Applied Biosystems LV40 cycle before the 10-nmol cycle was developed by reagent reduction. The standards were synthesized on the LV40 support with the Applied Biosystems LV40 cycle. Dimethoxytrityl (DMT) measurements for the standards show a 98% coupling efficiency, which is typical for LV40 oligonucleotide synthesis. Controls were synthesized on the 10-nmol support using the Applied Biosystems LV40 cycle. DMT measurements show a stable value for the noise and a significant reduction in the conductivity signal. The signal strength value for the control samples was approximately 50% of the value for the standard samples. The lower signal value for the controls increases the sensitivity to DMT measurement. The coupling efficiency for 10-nmol oligonucleotide synthesis is typically calculated below 98%.

 

DMT values for reducing deblock are shown in Table 5. The reduction of the reagents in steps 79 through 99 maintained a better oligonucleotide synthesis during test cycle 2. This cycle included a 40% reduction in reagent use for steps related to removing DMT. The DMT values for steps associated with delivering the phosphoramidite and activator mix to the 10-nmol support are shown in Table 6. Test cycle 6 appeared to have the most efficient delivery of activator and phosphoramidite to the support. This cycle included a 0.6-second decrease in activator. The volume of phosphoramidite was not changed for the 10-nmol cycle, because it is sufficiently reduced by the Applied Biosystems LV40 cycle.

 

TABLE 5
Dimethoxytrityl Measurement During Reduction of Deblock in 10-nmol Cycle Development
Test Cycle 2 Test Cycle 3 Test Cycle 4
Base     Trityl (%)     Signal     Noise     Trityl (%)     Signal     Noise     Trityl (%)     Signal     Noise
m13rev
C 100.0 314 141 100.0 316 177 100.0 355 172
A 96.2 308 148 96.3 309 180 100.0 359 168
G 97.4 312 146 100.0 310 165 100.0 357 161
T 98.1 316 156 94.2 302 188 97.3 355 179
A 96.5 299 154 92.1 284 188 96.6 345 180
T 97.1 305 158 93.4 300 190 97.2 357 180
C 97.5 308 162 94.3 295 180 94.2 324 195
G 97.3 296 157 95.0 293 177 94.9 336 169
A 97.6 292 151 95.5 294 182 95.5 325 182
C 97.0 285 158 96.0 284 184 95.9 327 184
A 97.2 287 154 95.9 273 182 96.3 311 182
A 97.5 284 152 94.8 258 182 96.6 314 178
A 97.2 266 147 94.9 255 182 96.5 303 179
G 97.1 271 157 95.2 259 173 96.8 310 178
G 97.3 277 151 95.5 266 177 97.0 301 172
A 97.4 276 156 95.8 261 176 96.7 296 182
C 96.5 258 163 96.0 256 183 96.6 295 186
m13(-21)
G 100.0 371 149 100.0 351 177 100.0 344 163
A 99.5 373 153 84.1 323 200 96.1 341 174
C 100.0 377 152 89.1 320 190 96.2 338 177
C 97.6 367 163 91.7 319 195 97.1 348 178
G 98.1 372 162 93.3 328 192 97.7 341 178
G 98.4 372 166 93.6 315 198 97.3 333 179
C 98.3 379 179 93.9 311 199 96.6 341 199
A 97.9 364 174 93.2 294 195 97.0 335 188
G 98.1 352 163 93.9 309 191 97.3 332 186
C 97.3 341 169 94.5 299 196 97.2 331 195
A 97.6 343 171 95.0 295 193 96.8 319 193
A 97.5 334 168 95.0 284 190 96.5 305 187
A 97.6 327 163 94.9 274 186 96.6 303 188
A 97.2 321 169 94.4 270 192 96.3 297 191
T 97.4 321 167 94.0 268 199 95.7 290 196
G 97.1 308 168 94.4 282 184 95.6 259 171
T 97.2 332 168 94.7 268 193 95.8 268 173

The volume of reagents used during the phase in oligonucleotide synthesis associated with removing dimethoxytrityl (DMT) from an attached base were reduced in cycles 2, 3, and 4. The effect of reagent reduction was determined by synthesizing an m13rev and an m13(-21) while measuring DMT. The DMT values for cycle 2 maintained a coupling efficiency average equal to 97% and was kept for the activator reduction phase of the research.

 

 

TABLE 6
Dimethoxytrityl Measurement During Reduction of Activator in 10-nmol Cycle Development
Test Cycle 5 Test Cycle 6 Test Cycle 7
Base     Trityl (%)     Signal     Noise     Trityl (%)     Signal     Noise     Trityl (%)     Signal     Noise
m13rev
C 100.0 338 141 100.0 299 142 100.0 246 153
A 100.0 337 0 99.7 307 151 100.0 244 150
G 100.0 351 0 97.4 298 153 100.0 256 150
T 74.7 350 241 98.0 303 147 95.7 246 157
A 79.1 340 0 97.6 301 162 96.6 252 160
T 80.1 338 245 98.0 305 164 95.6 253 172
C 82.7 337 0 97.7 294 161 96.2 250 167
G 84.7 339 0 97.9 296 164 96.7 254 165
A 86.3 325 0 97.6 294 168 96.9 244 164
C 87.6 316 0 97.0 283 167 96.7 241 165
A 87.6 322 240 97.1 281 167 97.0 249 168
A 88.6 306 0 97.1 277 167 96.4 234 166
A 88.5 296 224 97.3 280 166 96.5 233 166
G 89.3 306 0 97.5 278 167 96.8 235 159
G 89.6 299 231 97.4 272 167 97.0 241 166
A 90.3 292 0 97.3 271 169 96.9 231 167
C 90.8 286 0 97.1 268 172 97.1 237 167
m13(-21)
G 100.0 315 0 100.0 337 140 100.0 325 144
A 80.1 308 106 100.0 352 155 100.0 328 143
C 49.0 306 269 99.8 350 154 95.3 311 151
C 58.5 291 132 100.0 350 144 96.4 332 158
G 65.2 307 246 98.2 353 165 97.1 331 161
G 70.0 297 209 98.5 362 166 97.6 340 168
C 73.6 296 216 98.7 356 163 100.0 357 172
A 76.5 292 25 98.1 346 169 96.8 308 165
G 78.8 301 168 98.3 353 173 97.2 322 168
C 62.8 286 283 97.8 334 169 97.5 321 168
A 65.5 282 253 98.0 345 170 97.7 335 175
A 67.9 280 198 98.2 334 167 97.9 313 164
A 69.9 276 184 98.3 331 167 97.0 291 167
A 71.7 265 251 97.8 322 172 97.2 326 174
T 73.3 270 256 97.9 328 174 97.4 294 168
G 74.8 277 227 98.0 328 171 97.5 299 172
T 76.1 263 168 98.1 323 174 97.5 290 170

The reagents associated with coupling the base to the oligonucleotide chain were reduced in cycles 5, 6, and 7. A specific volume of activator is necessary to carry the base phosphoramidite into the support during base coupling. For each level of activator reduction, an m13rev (6a) and an m13(-21) were synthesized, and coupling efficiency was determined by dimethoxytrityl (DMT) measurement. Low DMT values may reflect a low coupling efficiency caused by too much activator pushing the base past the support before coupling. Cycle 6 maintained the most efficient oligonucleotide synthesis, and base coupling was approximately 98%. The reagent reduction in test cycle 6 was maintained for the oxidizer reduction phase of the research.

 

Table 7 shows the DMT values for oxidizer reduction. Test cycle 8 is a 50% reduction. This was equivalent to a 2-second decrease in the cycle. Although the presence of oxidizer was visualized in the support, there was a significant decrease in DMT values. Test cycle 9 had only a 25% reduction in oxidizer and also showed a slight decrease in DMT values. Therefore, the 10-nmol cycle did not include oxidizer reduction.

 

TABLE 7
Dimethoxytrityl Measurement During Reduction of Oxidizer in 10-nmol Cycle Development
Test Cycle 8 Test Cycle 9
Base     Trityl (%)     Signal     Noise     Trityl (%)     Signal     Noise
m13rev
C 100.0 271 148 100.0 259 150
A 100.0 283 151 98.6 257 151
G 97.7 274 147 94.8 253 160
T 100.0 285 152 94.2 249 163
A 98.4 279 155 95.4 258 167
T 96.4 277 170 95.8 252 168
C 96.9 271 164 96.2 254 171
G 97.3 271 164 96.4 247 166
A 96.0 260 168 96.8 243 162
C 96.4 265 160 97.0 246 166
A 96.7 265 166 96.5 237 163
A 97.0 254 158 96.1 232 164
A 96.3 243 161 96.4 236 162
G 96.6 242 160 96.7 235 166
G 96.8 255 159 96.9 242 164
A 96.9 245 165 97.1 239 166
C 97.1 247 163 97.3 231 163
m13(-21)
G 100.0 331 151 100.0 311 146
A 92.8 305 150 99.4 316 153
C 95.1 305 148 97.5 312 159
C 96.3 317 159 95.6 304 166
G 97.1 319 160 96.5 320 166
G 97.5 345 183 100.0 328 162
C 97.9 335 180 96.8 308 176
A 97.7 332 182 97.2 310 165
G 97.8 325 178 97.5 314 164
C 98.0 327 175 96.7 291 172
A 98.2 323 172 97.0 298 167
A 97.6 304 169 97.3 292 170
A 96.9 286 166 97.0 282 170
A 96.8 287 173 96.8 277 172
T 96.1 274 175 97.0 291 170
G 96.3 271 169 97.2 285 171
T 96.5 270 169 97.3 284 172

Cycles 8 and 9 were attempts to reduce the oxidizer, step 65, in the 10-nmol cycle. For both cycles 8 and 9, an m13rev and an m13(-21) were synthesized, and oxidizer was reduced while determining the coupling efficiency through dimethoxytrityl (DMT) measurement. The 50% oxidizer reduction in cycle 8 resulted in a decrease in base coupling. A 25% reduction of the oxidizer in cycle 9 also resulted in lower coupling efficiency. The 10-nmol cycle currently used does not include a reduction of the oxidizer reagent.

 

The quality of the oligonucleotides synthesized by the test cycles were tested by 15% PAGE, as shown in Figure 3. A nonpurified oligonucleotide typically shows a single strong band of the desired oligonucleotide product following a ladder of bands representing the failure sequences. The PAGE test of oligonucleotide quality shows a strong band for each oligonucleotide, but there was little difference in the quality assessment of each band. The PAGE test as an individual test did not prove the quality of each oligonucleotide as conclusive. However, combined with the other quality tests, an assessment of the oligonucleotide synthesis can be made.

FIGURE 3. Fifteen percent polyacrylamide gel electrophoresis testing m13rev oligonucleotide synthesized by each cycle during reagent reduction. The standards (S1, S2) were synthesized on the LV40 support using the Applied Biosystems LV40 cycle. The control oligonucleotide (C), was synthesized with the LV40 cycle on the 10-nmol support. The oligonucleotides that were synthesized during deblock reduction (2, 3, and 4), activator reduction (5, 6, and 7) and oxidizer reduction (9 and 10) are also shown. Sample 10 was synthesized after the oxidizer reagent was returned to the initial level of the LV40 cycle and is the same as cycle 6. This appeared to synthesize oligonucleotides with the highest coupling efficiency.

 

The second test used to determine the quality of the oligonucleotides synthesized by the test cycles was by automated DNA sequencing of the pGem standard. The primary purpose of the 10-nmol cycle was to synthesize an oligonucleotide that would function well as a DNA sequencing primer. Therefore, DNA sequencing was an important measurement of oligonucleotide quality. Results from automated DNA sequencing were assessed by length of base read and quality of peaks seen in the electropherogram. The m13(-21) and m13rev oligonucleotides synthesized in test cycle 6 both produced sequence results equivalent to that of the pGem sample used as a standard on every DNA sequencing gel, and the total DNA sequence base read was not shortened or compromised.

The final test of the 10-nmol cycle was to determine the maximum oligonucleotide length that could be synthesized. The 50-base, 75-base and 100-base oligonucleotides that were synthesized were electrophoresed by PAGE. The 50-base and 75-base oligonucleotides successfully synthesized well across all four columns on the ABI model 394 DNA synthesizer and produced strong bands when tested by PAGE. The 100-base oligonucleotide initially produced a lower DMT value, which was the direct result of the low conductivity signal strength for the 10-nmol oligonucleotide synthesis. Improvement was made to the synthesis of a 100-base oligonucleotide by making a 10-nmol column with four membranes instead of two. Although this was not technically a 10-nmol oligonucleotide synthesis, the added cost to make this column was negligible. The sequences of the longer oligonucleotides were confirmed by PCR followed by automated DNA sequencing. The sequencing results are shown in Figure 4. The sequences for the three different-length oligonucleotides were confirmed, as were the m13(-21) and m13rev primer sequences that were attached as tails to the products.

FIGURE 4. Fifty-base, 75-base, and 100-base oligonucleotides were synthesized on the 10-nmol support with 10-nmol cycle 6 on each ABI model 394 column position. To determine whether the oligonucleotides synthesized correctly, the oligonucleotides were amplified by polymerase chain reaction to generate a forward and complimentary strand and then sequenced on Applied Biosystems 377 automated sequencer with m13(-21) and m13rev. The results were assembled and compared with the original sequence. The underlined bases show m13(-21) and m13rev tailed primers used to amplify the products of 10-nmol oligonucleotide synthesis.

 

CONCLUSIONS

The complete 10-nmol cycle is shown in Table 8. The design of the 10-nmol DNA synthesis support eliminated excess volume and reduced the reagent waste. The 10-nmol cycle that was developed increased the rate of oligonucleotide synthesis from 10 bases/hour for the initial LV40 cycle to 14 bases/hour. The current cost of phosphoramidites added to the reduced reagent volume used in the LV40 cycle was small to produce a 40-nmol oligonucleotide. However, the cost of the 40-nmol support was not significantly reduced and equaled 40% of the overall cost to produce an average-length 40-nmol oligonucleotide. Producing 10 10-nmol DNA synthesis supports from one 0.2 MemSyn reduced the cost of the DNA synthesis support to 5% of the overall cost to produce an average-length oligonucleotide. By reducing the cost of the DNA synthesis support and reducing reagent waste, the cost of producing a 10-nmol oligonucleotide is one half that of producing a 40-nmol oligonucleotide. The cost of labor is also minimal. The 10-nmol DNA synthesis supports are produced in large numbers, and 30 or more columns can be made in a 15-minute period.

 

TABLE 8
The 10-nmol Cycle
Step     Function     Cycle     |     Step     Function     Cycle     |     Step     Function     Step
1 Begin | 40 18 to waste 4.0 | 79 If monitoring
2 18 to waste 3.0 | 41 Block flush 3.0 | 80 19 to column 12.0
3 18 to column 5.0 | 42 Column 4 off | 81 14 to column 3.0
4 Reverse flush 4.0 | 43 Flush to waste 0.9 | 82 Monitor trityls
5 Block flush 4.0 | 44 Wait 1.0 | 83 14 to column 12.0
6 Phos prep 3.0 | 45 Reverse flush 0.4 | 84 Monitor noise
7 Column 1 on | 46 Wait 1.0 | 85 14 to column 4.0
8 Block vent 2.0 | 47 Flush to waste 0.7 | 86 Stop monitoring
9 TET to column 0.4 | 48 Wait 1.0 | 87 18 to column 5.0
10 B+TET to column 0.2 | 49 Reverse flush 0.4 | 88 Reverse flush 4.0
11 TET to column 0.9 | 50 Wait 1.0 | 89 If not monitoring
12 Flush to waste 0.6 | 51 Flush to waste 0.7 | 90 14 to column 6.0
13 18 to waste 4.0 | 52 Wait 1.0 | 91 Trityl flush
14 Block flush 3.0 | 53 Reverse flush 0.4 | 91 Wait 5.0
15 Column 1 off | 54 Wait 1.0 | 92 Trityl flush 5.0
16 Column 2 on | 55 Flush to waste 0.7 | 93 Wait
17 Block vent 2.0 | 56 Cap prep 3.0 | 93 14 to column 6.0
18 TET to column 0.4 | 57 18 to waste 4.0 | 94 Trityl flush
19 B+TET to column 0.2 | 58 Reverse flush 4.0 | 94 Wait 5.0
20 TET to column 1.2 | 59 Block flush 3.0 | 95 14 to column
21 Flush to waste 0.6 | 60 Cap to column 3.0 | 95 Trityl flush 5.0
22 18 to waste 4.0 | 61 Wait 5.0 | 96 Wait
23 Block flush 3.0 | 62 18 to waste 4.0 | 96 18 to column 6.0
24 Column 2 off | 63 Reverse flush 4.0 | 97 Trityl flush 5.0
25 Column 3 on | 64 Block flush 3.0 | 98 14 to column
26 Block vent 2.0 | 65 15 to column 4.0 | 98 End monitoring
27 TET to column 0.4 | 66 18 to waste 4.0 | 99 Wait
28 B+TET to column 0.2 | 67 Block flush 3.0 | 99 18 to column 6.0
29 TET to column 1.3 | 68 Wait 15.0 | 100 Trityl flush
30 Flush to waste 0.6 | 69 18 to column 10.0 | 100 Reverse flush 5.0
31 18 to waste 4.0 | 70 Flush to waste 4.0 | 101 18 to column
32 Block flush 3.0 | 71 18 to column 6.0 | 101 Block flush 4.0
33 Column 3 off | 72 Reverse flush 4.0 | 102 Trityl flush
34 Column 4 on | 73 Block flush 3.0 | 102 End
35 Block vent 2.0 | 74 Start detrityl | 103 End monitoring
36 TET to column 0.4 | 75 18 to waste 4.0 | 104 18 to column 8.0
37 B+TET to column 0.2 | 76 18 to column 6.0 | 105 Reverse flush 5.0
38 TET to column 1.6 | 77 Reverse flush 4.0 | 106 Block flush 4.0
39 Flush to waste 0.6 | 78 Block flush 3.0 | 107 End

The 10-nmol cycle currently used is cycle 6, which provided the best coupling efficiency. The tests for oligonucleotide quality and efficiency support the ability of this cycle to produce an oligonucleotide at 14 bases/hour. The function titles are as given in the Applied Biosystems LV40 cycle.3

TET, tetrazole; Phos and B, phosphoramidite; Cap, acetic anhydride and methylimidazole.

 

The coupling efficiency was tested up to a 100-base oligonucleotide synthesis across the four columns. Although a 100-base oligonucleotide was barely visible by 15% PAGE, the oligonucleotide was amplified by PCR and confirmed by automated DNA sequencing. This facility currently maintains a 75-base limit for 10-nmol oligonucleotide synthesis. Cycle development is an ongoing process. The future direction of this research is the development of oligonucleotide synthesis with a coupling efficiency capable to produce a 150-base 10-nmol oligonucleotide at a rate of 18 bases/hour.

 

ACKNOWLEDGMENTS

We would like to express our appreciation to Brent Baldwin, Yonggen Song, and Michael Jordan of University of Cincinnati DNA Core Facility for assisting with automated DNA sequencing.

 

REFERENCES

1. Chemistry for automated DNA/RNA synthesis. In: Models 392 and 394 DNA/RNA Synthesizers User's Manual, version 2. Foster City, CA: Applied Biosystems, 1992:6.2-6.19.

2. Kaufman J, Le M, Ross G, et al. Trityl monitoring of automated DNA synthesizer operation by conductivity: a new method of real time analysis. BioTechniques 1993;14:834-839.

3. Fast, Economical Oligonucleotide Synthesis With the ABI LV40 Columns, User Bulletin 88. Foster City, CA: Applied Biosystems, 1996.

 

Return to table of contents