TWI232867B - Apparatus and methods for massively parallel oligonucleotide synthesis - Google Patents

Abstract

Apparatus and method are described for parallel oligonucleotide synthesis by the number of hundreds of different sequences and lengths at a time. Standard phosphoramitide chemistry is employed and the synthesis reactions take place in the container compliant with the industrial standard microplate format to allow using the readily available automated instruments for subsequent processing. Key parameters in reducing synthesis volume in small reaction wells are discussed. Solutions to the difficulties in low volume and large number synthesis reactions are provided in this invention.

Description

1232867 A7 B7 五、發明説明（Ί ) 本發明部分由中華民國台灣國家科學委員會及研究院贊 助及支持。 發明簡單敘述 本發明係關於一種以多管道合成大量寡核芬酸之裝置及 方法，且特別是關於同時固相合成數百或數千種寡核甞酸 之裝置及方法。 發明背景 由於寡核甞酸之化學合成之發展，合成之生物聚合物 (biopolymer)表許多領域已演舉足輕重之角色，如分子生物 學、法醫科學、及醫學診斷。近年來，由於基因研究之許 多大規模計劃，如基因多形性鑑定及記錄（scoring)、STS含 量圖譜、II射雜交圖譜、DNA定序法中之引子移動 (walking)、人類基因組計劃（Human Genome Project)、及丨其他 等需要各種不同寡核答酸，因此對現有寡核甞酸之要求有 明顯提昇。由於出現多管道分析技術如DNA微陣列，因此 發展出對許多不同寡核芸酸的新需求。 各種不同寡核甞酸合成之成本中，以現有合成方法顯得 過高。降低共轭或未改質核酸之試劑消耗及成本有助於各 種不同寡核甞酸之應用。例如，寡核甞酸微陣列應用中， 各寡核甞酸之5’-端以間隔子及使寡核苷酸固定在固體基 材上之胺基連結子改質。胺基連結子及間隔子之成本均遠 高於核甞亞胺磷酸鹽。再者，合成小於40奈莫耳之現有商 業製劑之規模已足以應用於大部分應用。PCR擴增通常需 要微微莫耳（picomole)量之寡核替酸引子。對各探針而言， 本紙張尺度適用中國國家標準(CNS) A4規格(210X 297公$) 1232867 A71232867 A7 B7 V. Description of the invention (Ί) This invention is partly supported and supported by the National Science Council and Research Institute of Taiwan, Republic of China. Brief description of the invention The present invention relates to a device and method for synthesizing a large number of oligonucleotide acids by multiple channels, and more particularly to a device and method for synthesizing hundreds or thousands of oligonucleotide acids in a solid phase at the same time. BACKGROUND OF THE INVENTION Due to the development of chemical synthesis of oligonucleotide, synthetic biopolymers have played an important role in many fields, such as molecular biology, forensic science, and medical diagnostics. In recent years, due to many large-scale projects in genetic research, such as gene polymorphism identification and recording (scoring), STS content mapping, II-ray hybridization mapping, primer walking in DNA sequencing, human genome project (Human Genome Project), and others require a variety of different oligonucleotides, so the requirements for existing oligonucleotides have been significantly increased. Due to the emergence of multi-channel analysis technologies such as DNA microarrays, new requirements for many different oligonucleotic acids have been developed. Among the various synthetic oligonucleotide costs, the existing synthetic methods appear to be too high. Reducing the consumption and cost of reagents for conjugated or unmodified nucleic acids is helpful for the application of various oligonucleotides. For example, in the application of oligonucleotide microarrays, the 5'-end of each oligonucleotide is modified with a spacer and an amine linker that immobilizes the oligonucleotide on a solid substrate. The cost of amine linkers and spacers is much higher than that of riboimine phosphate. Furthermore, the scale of existing commercial formulations that are less than 40 nanomoles is sufficient for most applications. PCR amplification usually requires a picoole amount of oligonucleotide primers. For each probe, this paper size applies Chinese National Standard (CNS) A4 specification (210X 297 $) 1232867 A7

______ B7 五、發明説明（2 ) —' ^ 需要小於1微微莫耳之寡核菩酸微陣列。高產量合成之衝 擊在基因生物技術應用具重要性，其需要數十萬之寡核菩 酸。 在尋求合成各種寡核甞酸同時，現有通常使用兩類多管 道系統，以一次合成至少數十種寡核甞酸。第一類利用具 有8個螺f閥及輸送一試劑至96孔微盤8排中之射出噴嘴之 1 -次元移動控制系統。其實例為Am〇s系統，其係由 Lashkad及其同事發展之9卜管道合成儀（pr〇c 八㈣％ USA· I995，$2，7912_7915)。同時，AM〇s 系統中需要 8S個螺 管閥，以注入其他試劑如改質之基劑（不容易操作）。當需 要其他試劑如標記試劑或改質之基劑時，需要另加8個閥 及8個噴嘴。AMOS系統中之惰性反應氣體係藉由使鐵氟龍 片充氣密封密封物而維持。其試劑之最小注入容積約丨為1〇 械升。4.5小時之產出為20-聚物（2〇-mer)之96寡核：¾:酸。 AMOS設計之優點為··（1)高產出效能，同時使用8個噴嘴輸 送試劑至8個孔；（2)由於反應室容積小，因此氬氣消耗 低；（3)簡單之一次元移動設計。AM〇s系統之缺點為·· 〇) 對各試劑需要8個閥及噴嘴，而不易使用其他試劑如改質 之基劑；（2)僅可使用96-孔。為了使用384_孔，因此需排 列噴嘴位置，且螺管閥及噴嘴數目需加倍至超過一百，使 得系統操作及維護相當不易。 第二類為具有一個閥及用以利輸送各試劑至％•孔微盤 射出喷嘴之二次元移動控制系統。96孔再藉由供接收試劑 之射出噴嘴下方之移動控制系統光柵掃描。此類型之實例 1232867 A7 B7 _____ 五、發明説明（3 ) 為由Rayner及其同事發展之MerMade合成儀（Genome Res·， 1998, 8, 741-747)。MerMade系統利用二次元移動控制系統以 及各試劑一個噴嘴。使用二個96-孔盤，使產出在17小時内 提升為192個寡核甞酸。系統之最小反應容積約為60微升。 優點為：（1)可使用不同格式之盤；（2)藉由加入各其他試 劑之一個閥及一個喷嘴，易於將其他試劑加到系統中。缺 點為：（1)對於使用384-動盤（見於下）之合成，實際之注入 容積太大。（2) 2-盤設計佔用大空間且機器尺寸為3 X 3 X 6 呎。 * 許多多管道寡核苷酸合成儀為市售，ABI 3900可同時合 成48個寡核甞酸，且可經數次轉換，在9小時内合成288個 寡核答酸，GfeneMachines之Polyplex可同時合成96個寡核嘗 酸，且BioAutomation銷售之LCDR/MerMade可同時合成丨96x2 個寡核甞酸。寡核甞酸合成之方法及裝置為主之兩個專利 已公告為習知技藝。AMOS系統為USP 5,529,765，及 MerMade系統為USP 5,368,823及5,541，314號。所有寡核苷酸 合成儀均以原來或由 Beaucage 等人（Tetrahedron I.ett. 1981，22, 1859-1862)及 Adams 等人（J. Am· Chem· Sac.，1983, 105, 661-663 ) 發展之改質亞胺磷酸鹽化學品為準。以亞胺磷酸鹽化學品 為主之專利（USP 4,415,732)於1983年授與Caruthers等人，且 C-I-P專利（USP 4,973,679)則在1990年授與同一發明人。相 關專利在1992年授與Caruthers等人（USP 5，132,418)。使用執 行固相亞胺磷酸鹽化學品之控制孔隙玻璃（CPG)為主之專 利於 1985年授與 Urdea等人（USP 4,517,338)。 本紙張尺度適用中國國家標準(CNS) A4規格(210X297公$) A7 B7 1232867 五、發明説明（ ） 4 就完成模型有機體之基因組以及人類基因計劃之定序而 言，需要大量基因專一探針以及PCR引子，以擴增有機體 中所有基因，以研究其功能、細胞角色、及基因其他特 性。例如，me/awogas/er果繩約有 14,000 個基因， 且C. e/eM似約有18,000個基因。為了擴增基因組中之所有 蜂蠅基因，需要約28,000個PCR引子。先前技藝 需要約6個月以完成上述有機體之所有基因專一 PCR引子之 合成。評估100,000個人類基因，以習知技藝需要超過一年 之時間。_ 改善合成產量之一方式為增加管道合成數至高於先前技 藝所述。為了增加管道合成數，可使用尺寸大於標準微盤 格式之盤或反'應槽，或使用具有較高密度孔如384-孔之標 準微盤格式。使用大於標準微盤格式之盤可克服後合成加 工中之困難度，如產量定量、產品純化等。對於大量寡核 甞酸之後合成加工需要自動化儀器。幾乎所有市售自動化 儀器均設計成使用標準微盤。未使用後合成加工用之自動 化液體處理或移液儀器，無法同時執行數百個多管道寡核 甞酸合成。 對標準微盤而言，96-孔中反應孔直徑約為7毫米且孔距 為9毫米。反應孔直徑降低至3.5毫米且孔距降低至4.5毫米 之384孔排列在合成上出現許多困難。1536-孔格式之孔間 隔尺寸進一步降低至2.25毫米。經常發生合成問題如相鄰 反應孔交叉污染、孔間產量大為改變、反應劑及合成支撐 物之飛濺等。因此，藉由使用標準384-孔或1536-孔微盤增 本紙張尺度適用中國國家標準(CNS) Α4規格(210 X 297公釐) 1232867 A7 B7 五、發明説明（5 ) 加管道合成數，以增加合成之產量並非全然沒價值。先前 技藝均未教示藉由同時使用數百個直徑小於96-孔盤格式 之反應孔而獲得高產量寡核苷酸之方法及裝置。 前述先前技藝出現許多問題。第一種為反應過程中合成 支撐物（控制之孔隙玻璃、CPG)脫離。脫離發生以約2 %比 例隨機分佈在整個盤上（參照AMOS)。當孔中發生支撐物 脫離時，將招致有害作用。不希望之結果可包含（l)CPG珠 粒因脫離而流失之孔為零產物產率，（2)由於脫離之CPG珠 粒造成真空f線隨後阻塞，及（3)發生CPG珠粒脫離之孔中 真空過度抽氣，導致其他孔中排放失效。此等不期望作用 可減低合成儀效能，或進一步使整個盤合成失敗。當孔密 度增加時，該作用變得更明顯，如本文討論之384-孔盤。 第二種問題為各反應孔中試劑容積微小變化或廢棄物排 放速率可能對排放速率比可用於整個盤合成之其他者快速 之某個孔產生毀滅性之作用。先前技藝之另一問題為多管 道合成儀之試劑注入最小容積對384-孔或更高密度合成而 言太大。最小注入容積對AMOS系統為10微升，而對 MerMade系統為20微升。該容積對於反應期間適當混合試 劑而言將太大，係由於孔直徑限制在比標準384-孔微盤小 3.5毫米之故。 本發明操作性能優於先前技藝者如下：（1)合成失敗率 最小；（2)有機廢棄物量較低；（3)合成規模較低；（4)可彈 性地利用標準微盤，如384-孔及96-孔；（5)反應中合成產 量差異較小。實驗結果顯示並未發現單一合成失敗。使用 本紙張尺度適用中國國家標準(CNS) A4規格(210 X 297公g 1232867 A7______ B7 V. Description of the invention (2) — '^ An oligonuclear acid microarray with less than 1 picomolar is required. The impact of high-yield synthesis is of great importance in the application of genetic biotechnology, which requires hundreds of thousands of oligonucleotides. While seeking to synthesize various oligonucleotides, two types of multi-channel systems are currently used to synthesize at least dozens of oligonucleotides at one time. The first type uses a 1-dimensional movement control system with 8 screw f valves and injection nozzles that deliver a reagent to 8 rows of 96-well microdisks. An example of this is the Am0s system, which is a 9b pipe synthesizer developed by Lashkad and colleagues (pr0c ㈣% USA · I995, $ 2,7912_7915). At the same time, 8S solenoid valves are needed in the AM0s system to inject other reagents such as modified bases (not easy to operate). When other reagents such as labeling reagents or modified bases are needed, additional 8 valves and 8 nozzles are required. The inert reaction gas system in the AMOS system is maintained by inflating the Teflon sheet with a hermetic seal. The minimum injection volume of the reagent is about 10 mechanical liters. The output in 4.5 hours was a 20-mer (20-mer) 96 oligo: ¾: acid. The advantages of AMOS design are: (1) high output efficiency, using 8 nozzles to deliver reagents to 8 holes at the same time; (2) low argon consumption due to the small volume of the reaction chamber; (3) simple one-time movement design. The shortcomings of the AM〇s system are: ○) 8 valves and nozzles are required for each reagent, and it is not easy to use other reagents such as modified bases; (2) Only 96-well can be used. In order to use 384 holes, it is necessary to arrange the nozzle positions, and the number of solenoid valves and nozzles must be doubled to more than one hundred, which makes the system operation and maintenance quite difficult. The second type is a two-dimensional movement control system with a valve and a micro-disc injection nozzle that facilitates the delivery of reagents to the pores. The 96 wells are raster scanned by a movement control system below the ejection nozzle for receiving reagents. Examples of this type 1232867 A7 B7 _____ 5. Description of the invention (3) is a MerMade synthesizer developed by Rayner and colleagues (Genome Res ·, 1998, 8, 741-747). The MerMade system uses a two-dimensional movement control system and a nozzle for each reagent. Using two 96-well plates, the yield was increased to 192 oligonucleotides in 17 hours. The minimum reaction volume of the system is approximately 60 microliters. The advantages are: (1) disks of different formats can be used; (2) it is easy to add other reagents to the system by adding a valve and a nozzle for each other reagent. The disadvantages are: (1) For synthesis using a 384-moving disc (see below), the actual injection volume is too large. (2) The 2-disk design takes up a lot of space and the machine size is 3 X 3 X 6 feet. * Many multi-channel oligonucleotide synthesizers are commercially available. The ABI 3900 can synthesize 48 oligonucleotides at the same time, and can be converted several times to synthesize 288 oligonucleotides within 9 hours. GfeneMachines' Polyplex can be simultaneously 96 oligonucleotides are synthesized, and LCDR / MerMade sold by BioAutomation can synthesize 96x2 oligonucleosides simultaneously. Oligonucleotide synthesis method and device-based two patents have been announced as conventional techniques. The AMOS system is USP 5,529,765, and the MerMade system is USP 5,368,823 and 5,541,314. All oligosynthesizers are based on original or by Beaucage et al. (Tetrahedron I.ett. 1981, 22, 1859-1862) and Adams et al. (J. Am. Chem. Sac., 1983, 105, 661-663 ) Development of modified imine phosphate chemicals shall prevail. A patent based on imine phosphate chemicals (USP 4,415,732) was granted to Caruthers et al. In 1983, and a C-I-P patent (USP 4,973,679) was granted to the same inventor in 1990. Related patents were granted to Caruthers et al. (1992, USP 5,132,418). Urdea et al. (USP 4,517,338) was the main patent on the use of controlled pore glass (CPG) that performs solid-phase imine phosphate chemicals. This paper scale applies Chinese National Standard (CNS) A4 specification (210X297 public dollars) A7 B7 1232867 V. Description of invention () 4 In order to complete the genome of the model organism and the sequencing of the human gene plan, a large number of gene-specific probes and PCR primers to amplify all genes in the organism to study their functions, cellular roles, and other characteristics of genes. For example, me / awogas / er fruit rope has about 14,000 genes, and C. e / eM seems to have about 18,000 genes. To amplify all the bee fly genes in the genome, about 28,000 PCR primers are required. The previous technique required about 6 months to complete the synthesis of specific PCR primers for all genes of the above organisms. Assessing 100,000 human genes to acquire skills requires more than a year. _ One way to improve the yield of synthesis is to increase the number of pipeline synthesis more than described in the previous technology. In order to increase the number of pipeline synthesis, a disk or a reaction slot larger than the standard microdisk format can be used, or a standard microdisk format with higher density holes such as 384-wells can be used. Using disks larger than the standard microdisk format can overcome difficulties in post-synthesis processing, such as yield quantification and product purification. For large quantities of oligonucleotide, post-synthetic processing requires automated instruments. Almost all commercially available automation instruments are designed to use standard microdisks. Unused automated liquid processing or pipetting equipment for post-synthesis processing cannot perform hundreds of multi-channel oligonucleotide synthesis at the same time. For a standard microdisk, the diameter of the reaction wells in a 96-well is about 7 mm and the pitch is 9 mm. The array of 384 wells with the diameter of the reaction well reduced to 3.5 mm and the pitch of the well reduced to 4.5 mm presents many difficulties in synthesis. The 1536-hole format hole size was further reduced to 2.25 mm. Synthesis problems often occur, such as cross-contamination of adjacent reaction wells, large changes in yield between wells, splashing of reactants and synthetic supports, and so on. Therefore, by using standard 384-well or 1536-well microplates to increase the paper size, the Chinese National Standard (CNS) A4 specification (210 X 297 mm) 1232867 A7 B7 is used. 5. Description of the invention (5) plus the number of pipeline synthesis, To increase the yield of synthesis is not entirely worthless. Previous techniques have not taught methods and devices for obtaining high-yield oligonucleotides by using hundreds of reaction wells with a diameter less than 96-well disk format simultaneously. Many problems arose with the aforementioned prior art. The first is the release of synthetic supports (controlled pore glass, CPG) during the reaction. The detachment occurred randomly distributed over the entire disk at a rate of about 2% (see AMOS). Harmful effects can occur when the support is detached from the hole. Unwanted results may include (1) zero yield of pores lost by CPG beads due to detachment, (2) subsequent blockage of the vacuum f-line due to detached CPG beads, and (3) detachment of CPG beads Excessive vacuum in the holes caused exhaust in other holes to fail. These undesired effects can reduce the performance of the synthesizer, or further make the entire disc synthesis fail. This effect becomes more pronounced as the pore density increases, as discussed in the 384-well plate. The second problem is that a small change in reagent volume or waste discharge rate in each reaction well may have a devastating effect on a well that is faster than the others that can be used for the entire disc synthesis. Another problem with the prior art is that the minimum volume of reagent injection for multi-channel synthesizers is too large for 384-well or higher density synthesis. The minimum injection volume is 10 μl for the AMOS system and 20 μl for the MerMade system. This volume will be too large for proper mixing of reagents during the reaction because the pore diameter is limited to 3.5 mm smaller than a standard 384-well microdisk. The operating performance of the present invention is superior to the previous artist as follows: (1) the minimum synthesis failure rate; (2) the amount of organic waste is low; (3) the synthesis scale is low; (4) the standard microdisk can be flexibly used, such as 384- Wells and 96-wells; (5) The difference in synthetic yield during the reaction is small. Experimental results show that no single synthesis failure was found. Use This paper size applies to China National Standard (CNS) A4 (210 X 297 g g 1232867 A7)

過滤器尖端作為反應孔之具體例可有效地支撐合成用之 CPG珠粒’且不會發生cPG珠粒流失。沒有合成失敗使系 、先更可非’且使後續合成加工步驟簡化。由於反應孔内徑 小’因此對各孔中合成基劑產生之廢棄物為約0.33亳升。 只驗結果顯示合成規模小於10奈莫耳可在高品質下進行 (圖7) ’且各基劑僅需要〇36毫克之亞胺磷酸鹽。本發明已 達到新穎之低化學品用量與低廢棄物產生量。試劑消耗減 f對於合成中需要昂貴之亞胺磷酸鹽之合成如封端標示 劑、改質基劑、RNA亞胺磷酸鹽及分隔基等相當重要。此 等試劑之成本可能高達一般DNA核酸亞胺磷酸鹽之數十倍 以上。降低試劑消耗對於降低總成本有莫大效益。 ' 發明目的及概要 本發明之一目的係提供一種以多管道格式同時合成數百 種寡核菩酸之裝置及方法。 本發明另一目的係提供一種將微量試劑送入微盤格式之 數百個反應孔之裝置及方法。 本發明又一目的係教示一種將寡核甞酸合成試劑留在微 型盤格式之數百個反應孔中之裝置及方法。 本發明再另一目的係教示一種使微盤格式之多管道合成 孔-對-孔之產出差異最小之裝置及方法。 本發明又另一目的係教示一種避免合成支撐物在合成過 程濺出之方法。 本發明之前述及其他目的可由本文所述用於多管道合成 數百個寡核菩酸之裝置及方法而達成。該裝置包含於標準 張尺度適财_家標準(CNS) A_4規格(21()χ 297公^ " 1232867 A7 B7 五、發明説明（7 ) 微盤格式中形成數百個反應孔之設備，及在試劑噴嘴下方 之具有使固體反應珠粒濺出為最小之傳送微盤用之設備， 及使孔-對·孔之合成產量差異最小之設備。 附圖簡單敘述 本發明之前述及其他目的可藉由下列敘述及配合附圖而 更易瞭解，其中： 圖1為說明本裝置基本組件之概略圖。 圖2A為由具有熔料或薄膜以支撐CPG珠粒之過濾器尖端 構成之反應'孔具體例。 圖2B為由冠狀盤、維持CPG珠粒之介質（薄膜），及基劑 盤之高密度反應盤另一具體例。 圖3為顯示流體流速、氣體流速及排放速率之相對大小 對反應孔中試劑排放完全性之影響總圖。 ！ 圖4為說明反應孔剖面及其對CPG珠粒濺出及反應孔中 氣泡形成之影響簡圖之總概略圖。 圖5為顯示CPG珠粒濺出高度及試劑射出容積之閥射出 壓力函數之圖，該圖亦說明如何決定最佳射出壓力及客 積。 圖6為顯示將附著在反應孔壁上之CPG珠粒洗滌下來所 需乙腈容積之閥射出壓力函數之圖。 圖7為以不同規模合成寡核甞酸之HPLC層析之總圖。 較佳具體例詳細敘述 參照圖1，裝置係由5個主要組件組成： (1)反應盤配件（assembly)及真空配件：特製384-微型版組件 本紙張尺度適用中國國家標準(CNS) A4規格(210X 297公 1232867 A7 B7 五、發明説明（8 作為反應匣，且其中各孔均為獨立反應槽。真空配件在 各合成步驟後，以真空泵浦將試劑排出。 ⑺試劑㈣及管：儲存試劑且將試劑自瓶送到螺管間及射 出噴嘴。 ⑺試劑閥配件：一組螺管閥及快速反應驅動電路，以將射 出容積控制在每次射出低於2微升。 (4) 射出噴嘴及移動系統：電腦控制2 •次元試劑輪送***； 使喷嘴下方之微型版移動’以收集自噴嘴注入各反應孔 中之試劑。 (5) 控制軟體：自主檔讀取寡核甞酸結果且控制合成程序之 軟體程式。A specific example of the filter tip as a reaction hole can effectively support CPG beads for synthesis without cPG bead loss. The absence of synthesis failures makes the system more negligible and simplifies the subsequent synthesis processing steps. Since the inside diameter of the reaction holes is small, the waste generated by the synthetic base in each hole is about 0.33 liters. The results showed that the synthesis scale of less than 10 nmoles can be performed with high quality (Fig. 7) 'and each base agent only requires 036 mg of imine phosphate. The invention has achieved novel low chemical usage and low waste generation. Reduction of reagent consumption f is very important for the synthesis of expensive imine phosphates such as capping markers, modifying bases, RNA imine phosphates and separators. The cost of these reagents may be several tens of times higher than the average DNA nucleic acid imine phosphate. Reducing reagent consumption is of great benefit in reducing overall costs. '' Object and Summary of the Invention One object of the present invention is to provide a device and a method for synthesizing hundreds of oligonuclear acids simultaneously in a multi-channel format. Another object of the present invention is to provide a device and method for feeding a small amount of reagents into hundreds of reaction wells in a microdisk format. Yet another object of the present invention is to teach a device and method for leaving oligonucleotide synthesis reagents in hundreds of wells in a microdisk format. Still another object of the present invention is to teach a device and method for minimizing the difference in output of hole-to-hole in a multi-channel synthesis of micro-disk format. Yet another object of the present invention is to teach a method for preventing the synthetic support from splashing during the synthesis process. The foregoing and other objects of the present invention can be achieved by the devices and methods described herein for multi-channel synthesis of hundreds of oligonuclear acids. This device is included in the standard Zhangjiazhuang Standard (CNS) A_4 specification (21 () χ 297 public ^ " 1232867 A7 B7 V. Description of the invention (7) Equipment for forming hundreds of reaction holes in the microdisk format, And a device for transporting microdisks with a minimum of solid reaction beads splashing under the reagent nozzle, and a device for minimizing the difference in the yield of well-to-well synthesis. The drawings briefly describe the foregoing and other objects of the present invention. It can be more easily understood through the following description and accompanying drawings, in which: Figure 1 is a schematic diagram illustrating the basic components of the device. Figure 2A is a reaction 'hole formed by a filter tip having a melt or a film to support CPG beads Specific example. Figure 2B is another specific example of a high-density reaction plate consisting of a crown plate, a medium (film) that maintains CPG beads, and a base plate. Figure 3 shows the relative size of the fluid flow rate, gas flow rate, and discharge rate. The general diagram of the effect of the completeness of the reagent discharge in the reaction well. Figure 4 is a general schematic diagram illustrating the cross section of the reaction well and its effect on the splash of CPG beads and the formation of bubbles in the reaction well. Figure 5 shows the CPG beads Splash height A diagram of the valve injection pressure function of the reagent injection volume, which also illustrates how to determine the optimal injection pressure and guest product. Figure 6 shows the valve injection pressure of the acetonitrile volume required to wash the CPG beads attached to the wall of the reaction well Function diagram. Figure 7 is a general diagram of HPLC chromatography for the synthesis of oligonucleotides on different scales. The specific example is described in detail with reference to Figure 1. The device consists of 5 main components: (1) reaction plate assembly (assembly) ) And vacuum accessories: special 384-mini version components This paper size is applicable to China National Standard (CNS) A4 specifications (210X 297 male 1232867 A7 B7) V. Description of the invention (8 as a reaction box, and each hole is an independent reaction tank. After each synthesis step, the vacuum accessories discharge the reagents with a vacuum pump. ⑺Reagents and tubes: store reagents and send the reagents from the bottle to the spiral tube and the ejection nozzle. ⑺Reagent valve accessories: a set of solenoid valves and fast response Drive circuit to control the injection volume to less than 2 microliters per injection. (4) Injection nozzle and moving system: computer-controlled 2 • dimension reagent rotation system; the micro plate under the nozzle is moved 'Are collected from the wells in each injection nozzle of reagent (5) Control software: autonomous read oligonucleotide Chang acid profile results and control software programs of the synthesizing program.

反應版配件、、射出噴嘴及轉譯階段均置於充填有氣體如 氬氣或氮氣以提供合成用惰性氣體之密閉室中。 I 參照圖1，合成反應發生在内部建構之384_孔反應盤配 件中，且反應盤為標準微盤格式。所有試劑噴嘴均依序排 列，且一起組裝在噴嘴座上。該座係設計成支撐2〇個不同 喷嘴且固定在轉譯階段之—上。喷嘴座以_次元，藉由移 動轉譯階段掃描該微盤。反應盤及真空配件架設在二次元 移動系統之另一轉譯階段上。此二轉譯階段一起構成整個 反應盤配件上噴嘴之X - γ向掃描。 為確保成分之射出容積約為2微升，因此重點為快速反 應閥之操作。反應時間快亦將導致較高之射出速率。對裝 置實際應用而言，需要高於20Hz之射出速率。 圖2A顯示384-孔反應盤配件之第一具體例。該384•孔反 本紙張尺度適用中國國家標準(CNS) A4規格(210 X 21 1232867 A7 B7 五、發明説明（9 ) 應盤配件係由用以獨立支撐384個配置有過濾器玻璃熔料 之變尖移液管尖端之承板所構成。移液管尖端内徑約為2.2 毫米，尖端距離為4.5毫米。承板係以圖1中所示之真空配 件固定。承板與真空配件間之墊片可確保真空密封。真空 配件連接至亦可作為廢棄物容器之真空壓縮物。試劑由各 孔頂端注入且抽真空排出。承板之製造係使移液管尖端緊 貼在各承板上，使合成過程中未發現尖端脫離或真空洩 漏，因此確保可靠且一致之產率。當寡核甞酸合成數目低 於384時，可有利地使用具有不需合成之承板中之隱藏尖 端之承板，因而不會發生真空流失及浪費試劑。 反應盤配件之第二具體例示於圖2B中。該配件係由三個 部份，冠狀盤'、維持CPG珠粒之介質及基劑盤所組成。該 冠狀板包含底部直徑約2毫米、深度約15毫米之圓錐型 孔。適當之維持CPG珠粒之介質需符合氣體流速低於排放 用施加抽真空口之流體流速之流動特性要求。親水性膜之 起泡點通常高於排放用之真空，且對於本目的較理想。基 劑盤之形狀為具有各孔之收縮開口，以協助收集產物而不 會使孔間交互污染。 圖3說明排放過程中，適當維持CPG珠粒之介質及選用 之泵浦速率之重要性。該實例中說明二反應孔，但多孔之 原理相同。2個反應孔間，使孔1具有Π升/分鐘之較高廢 棄物排放（經過熔料或薄膜之流體流速），且孔2具有F2升 /分鐘之較低排出速率。亦即 Fl> F2 本紙張尺度適用中國國家標準(CNS) A4規格(210 X 297公_爱^ 1232867 A7 B7 五、發明説明（1Q ) 使排放用之泵浦速率為P1升/分鐘，且孔1、孔2之氣體 流速分別為A1升/分鐘及A2升/分鐘。試劑送入後，孔1及 孔2均假詨收到等量之試劑。排放製成期間，廢棄物分別 以F1及F2之流速流經各孔。由於流速差異，因此孔1先放 空，接著在A1速率下使空氣流經孔1。 在此階段，不同之流速條件會產生不同結果： (1) 若Al < F1，則空的孔1不會低於孔2之抽氣口，且孔2可 排放而無問題。 (2) 若Al > F1且P > A1，則由於大量氣體以A1流速流經孔 1，使得孔2上之抽氣口降低。不過，高的泵浦速率可 維持孔2上之弱但易抽氣之口，且孔仍可在犧牲較長排 放時間及較大量之高成本氬氣下排放。 (3) 若Al > F1，且P与A1，則孔2上之抽氣口接近零，且孔2 之排放時間長到無法忍受。 上述實例說明排放速率差異會使合成效率及產率改變。 針對具有多反應孔之盤，若部分孔在添加下一合成試劑之 前未完全排出，則殘留之廢棄物會干擾合成化學，導致合 成產率低或無產率。 參照圖4，其敘述適當孔輪廓之重要性。圖4之上層板顯 示淺的反應孔，其中CPG珠粒在試劑輸送過程中易被沖 掉。中層板顯示深-窄之反應孔，氣泡易於其中形成，且 因表面張力而包在其中。氣泡受試劑流動之干擾，導致產 率變化或合成失敗。圖4之低層板顯示對有效維持CPG珠粒 輸送不含氣泡之試劑之適當輪廓。反應孔之適當内徑對於 本紙張尺度適用中國國家標準(CNS) A4規格(210 X 297公衰^ 1232867 A7 B7 五、發明説明（11 ) 高合成品質相當重要。内徑較小有助於降低試劑消耗量， 但亦容易因表面張力而產生氣泡。許多市售過濾器尖端之 平滑表面良好至足以避免形成氣泡。適當孔輪廓之決定說 明於圖7。 原型裝置中’所有試劑均裝在以氬氣加壓之玻璃瓶中， 且經過内徑〇·03英吋或更低之管輸送。將試劑自瓶送到小 型螺管閥（Lee公司之LFVX〇5〇84〇A)使用鐵氟龍管。由瓶至 閥至射出噴嘴之輸送線孔孔容積藉由使用儘可能最短之管 長而僅可旎維持在最低。各試劑之射出容積係以各螺管閥 、打]時間k制。具有快速反應時間之螺管閥驅動電路為 必要，以確保低射出容積。 384-孔盤比孔盤更需要低射出容積。過量之試劑未必 可仵軚问產率，尤其是偶合借段。在偶合階段處，三成分 (亦即口成支撐物、亞胺磷酸鹽及四唑(偶合反應之活化 劑需完全混合，而可有效偶合。過量試劑會造成適當混 、用之液把量過同。本發明人曾檢視射出特性，且發現總 容積需低於15微升，徒之可古上冬、日人 入 更 了有效化合。此會使亞胺磷酸鹽 容積低於5微升。 射出喷嘴為具有2G達因/公分低表面張力之越切割鐵氟 龍管。噴嘴需使用低表面張力材料，使每次射出後累積在 =嘴端上之殘留賴量為最小。下㈣明使試㈣積在喷 角大端（另-因素。μ嘴尖端切割後未經去毛邊，則备 造成試劑累積在喷嘴尖端處。為避免此問㈣嘴尖端： 藉由將鐵氟龍管尖端處加熱去毛邊。去毛邊後，、 端The reaction plate accessories, injection nozzles and translation stages are placed in a closed chamber filled with a gas such as argon or nitrogen to provide an inert gas for synthesis. I Referring to Figure 1, the synthesis reaction takes place in an internally constructed 384-well reaction disk assembly, and the reaction disk is a standard microdisk format. All reagent nozzles are arranged in sequence and assembled together on the nozzle holder. The seat is designed to support 20 different nozzles and is fixed to the translation stage. The nozzle holder scans the microdisk in the _ dimension by moving the translation stage. The reaction plate and vacuum accessories are erected on another translation stage of the two-dimensional mobile system. These two translation stages together constitute the X-γ scanning of the nozzles on the entire reaction plate assembly. To ensure that the injection volume of the components is about 2 microliters, the focus is on the operation of the quick response valve. Faster reaction times will also result in higher injection rates. For the actual application of the device, an emission rate higher than 20Hz is required. FIG. 2A shows a first specific example of a 384-well reaction plate accessory. The size of the 384-hole anti-paper paper is in accordance with Chinese National Standard (CNS) A4 specifications (210 X 21 1232867 A7 B7. V. Description of the invention (9) The tray accessories are used to independently support 384 glass frits equipped with filters. It consists of a carrier plate with a pointed pipette tip. The inner diameter of the pipette tip is approximately 2.2 mm and the tip distance is 4.5 mm. The holder plate is fixed with a vacuum fitting as shown in Figure 1. Between the holder and the vacuum fitting Gaskets ensure vacuum tightness. Vacuum fittings are connected to vacuum compacts that can also be used as waste containers. Reagents are injected from the top of each hole and evacuated. The carrier plate is manufactured by placing the pipette tip against each carrier plate. No tip detachment or vacuum leak was found during the synthesis, thus ensuring a reliable and consistent yield. When the number of oligonucleotides synthesized is less than 384, it can be advantageous to use a tip with a hidden tip in a substrate that does not require synthesis. Support plate, so no vacuum loss and waste of reagents occur. The second specific example of the reaction plate accessory is shown in Figure 2B. The accessory is composed of three parts, the crown plate, the medium to maintain the CPG beads, and the base plate. Make up. The The coronal plate includes a conical hole with a diameter of about 2 mm at the bottom and a depth of about 15 mm. The medium to properly maintain CPG beads must meet the flow characteristics of the gas flow rate lower than the flow velocity of the fluid applied to the vacuum port for discharge. The hydrophilic membrane The bubble point is usually higher than the vacuum used for the discharge, and is ideal for this purpose. The base plate is shaped as a constricted opening with holes to help collect the product without contaminating the holes. Figure 3 illustrates the discharge process. The importance of properly maintaining the medium of the CPG beads and the selected pumping speed. In this example, two reaction holes are described, but the principle of porosity is the same. Between the two reaction holes, the hole 1 has a high waste rate of Π liters / minute. Material discharge (fluid flow rate through the melt or film), and hole 2 has a lower discharge rate of F2 liters / minute. That is Fl & F2 This paper size applies to China National Standard (CNS) A4 specifications (210 X 297 mm_ Love ^ 1232867 A7 B7 V. Description of the Invention (1Q) The pumping rate for discharge is P1 liter / minute, and the gas flow rate of hole 1 and hole 2 is A1 liter / minute and A2 liter / minute. After the reagent is sent in , Hole 1 and hole 2 are both詨 Received the same amount of reagent. During the discharge process, the waste flows through the holes at the flow rate of F1 and F2. Due to the difference in flow rate, the hole 1 is first emptied, and then air is passed through the hole 1 at the rate of A1. At this stage, different flow conditions will produce different results: (1) If Al < F1, the empty hole 1 will not be lower than the suction port of hole 2, and the hole 2 can be discharged without problems. (2) If Al > F1 and P > A1, because a large amount of gas flows through hole 1 at a flow rate of A1, the suction port on hole 2 is reduced. However, a high pumping rate can maintain a weak but easy-to-pump port on hole 2. And, the holes can still be discharged at the expense of longer discharge time and a large amount of high cost argon. (3) If Al > F1, and P and A1, the suction port on hole 2 is close to zero, and the discharge time of hole 2 is unbearable. The above example shows that the difference in emission rate will change the synthesis efficiency and yield. For disks with multiple reaction wells, if part of the wells are not completely discharged before the next synthesis reagent is added, the remaining waste will interfere with the synthesis chemistry, resulting in low or no yield. Referring to Figure 4, the importance of proper hole contouring is described. The upper plate in Figure 4 shows shallow reaction wells, in which CPG beads are easily washed away during reagent delivery. The mid-layer plate shows deep-narrow reaction pores, bubbles are easily formed therein, and are enclosed by surface tension. Bubbles are disturbed by reagent flow, which can lead to yield changes or synthesis failures. The lower plate of Figure 4 shows the appropriate profile for effectively maintaining CPG beads to deliver reagents that do not contain air bubbles. The appropriate inner diameter of the reaction hole is applicable to the Chinese National Standard (CNS) A4 specification (210 X 297 public attenuation ^ 1232867 A7 B7) for this paper size. 5. Description of the invention (11) High synthesis quality is very important. A smaller inner diameter helps reduce Consumption of reagents, but also easy to produce bubbles due to surface tension. The smooth surface of the tip of many commercially available filters is good enough to avoid the formation of bubbles. The decision of the appropriate pore profile is illustrated in Figure 7. In the prototype device, 'all reagents are packed in Argon-pressurized glass bottles are transported through a tube with an inner diameter of 0.03 inches or less. Reagents are sent from the bottle to a small solenoid valve (LFVX 050884A from Lee) using iron fluoride. Long tube. The volume of the hole of the conveying line from the bottle to the valve to the injection nozzle can be kept to a minimum by using the shortest possible tube length. The injection volume of each reagent is made by each solenoid valve and the time k. A solenoid valve drive circuit with fast response time is necessary to ensure a low injection volume. A 384-well disk requires a lower injection volume than a well disk. Excess reagents may not be questionable for yield, especially the coupling borrow section. Coupling stage The three components (that is, oral support, imine phosphate, and tetrazole (the activator of the coupling reaction need to be completely mixed, and can be effectively coupled. Excessive reagents will cause proper mixing and use the same amount of liquid.) The present invention People have inspected the injection characteristics and found that the total volume needs to be less than 15 microliters, which can effectively combine the ancient winter and Japanese people. This will make the imine phosphate volume less than 5 microliters. The injection nozzle is provided with 2G dyne / cm low Teflon cutting Teflon tube. The nozzle needs to use a low surface tension material to minimize the residual residual amount accumulated on the mouth of the nozzle after each shot. Spray angle big end (other-factor. Μ tip is not deburred after cutting, it will cause the reagent to accumulate at the tip of the nozzle. To avoid this, pinch the tip: Remove the burr by heating the tip of the Teflon tube. After deburring, end

1232867 A7 B7 五、發明説明（12 噴角大鳊疋平滑表面及形狀使試劑射出 過程中累積之試劑為# I ^上 凡1為瑕小，因而提供一致之射出容積，且 使反應孔間之相互冷染為最小。 再者^缸射出逯度對於合成之成功率亦相當重要。射 、速度.一喷角尖端處之累積愈少。另-方面，射出速 度愈向會使CPG珠粒錢出反應孔。因此，需決定最佳之射 出速度，使噴嘴尖端處之試劑累積及CPG珠粒之濺出二者 得到均衡。 取佳射出速度〈選擇述於圖5。CPG珠粒錢出之傾向與 射出速度成正比，射出速度係由射出壓力p及喷嘴内径決 足。射出速度愈低’ CPG珠粒錢出愈少，但試劑消耗愈 夕累積《試劑會留在噴嘴尖端，直到試劑成長到表面張 力無法再支撐該液滴之特定尺寸為止。當射出速度增加 時，射出容積之變異係數（CV)曲線達到最小，但因此合使 CPG珠粒濺出高度増加。 曰 隹乂可使用較深之反應孔，以留住錢出之cpG珠粒，避 免冲出孔之外’深的反應孔對使合成產率變化最小化產生 另一=難點。此係由於心之㈣珠粒（雖留在反應孔内部) 緊t著在反應孔内壁上，且在下一合成步驟前，需將此等 CPG珠粒洗滌到孔底冑’以確保適當之反應效率。濺出高 度愈高意指需要較大之洗務容積，以將此等⑽珠粒： 下此作用造成試劑消耗較大，試劑輸送時間較長，且排 出時間較長。需考慮此等因素，以決定適當之射出速度。 圖6顯示將射出速度增加時所增加之濺出cpG珠粒洗滌 从边川_參標半(CNS) A4規格(2ΐ〇χ297,『 1232867 A7 -—_______________ B7 五、發明説明（C "-—--- 下來所須之CH3CN容積。據此，會增加廢棄物之排出時間 及總合成時間。為了決定最佳射出速度，以射出速度之函 數測量f入容積之cv。由於射出速度為許多參數（如射出 壓力、管I.D.、及試劑黏度）之複雜函數，但對各試劑需決 疋其取小可接受之射出速度，且最小之射出壓力可能隨不 同試劑而變。所有試劑之總體最佳射出速度接著藉由選擇 對所有試劑可接受之射出壓力而決定。 、 至於實例，圖5證明三氯乙酸（TCA)注入容積之cv係以 射出壓力之函數測量。圖5中，注入容積之cv突然增加至 低於導致試劑累積之壓力。cv曲線符合下列之指數函數： y = A * exp[.(x.x〇)/t] 程式！ 其中 、 y為注入容積之cv 4 A = 2.2，振幅 xo= 1，分支 t = 0.53，曲線之衰減常數。 因此，為了達到0.1之CV，因此射出壓力計算為 (2.53* t) + x〇 〜=2.5 (psi) 設定此值為a。射出壓力a為TCA達到注入容積CV為0.1之 最佳射出壓力。同樣的，可測定各試劑最佳射出壓力及總 體最佳射出速度。 一旦測定出射出壓力，則可如下列般測定反應孔深度。 CPG珠粒濺出之高度係使用所需之射出壓力測量。實例 本紙張尺度適用中國國家標準(CNS) A4規格(210 X 297公表^ 1232867 A7 ΓΤί明説明（“ )~--—-- 中，TCA係在射出壓力為主入，如圖5中所示，且平均 減出高度係藉由重複數次射出而測量。在4.5 PS1射出壓力 下，99%信賴度下所得之平均賤出高度為48±16毫米。此 例中，反應孔深度至少需4.8毫米+ 16毫米=64毫米。 原型裝置中’易使用之程式係以Visual Basic(微軟）寫 成。以使用纟輸入《合成參數為$，該程式會計算且對轉 譯階段發出移動指令，對螺管閥發出射出指令。使用者提 供簡易樓案中之序列資訊。電腦程式讀取資訊，且依全自 動方式進行合成製程。當操作中384反應孔未被完全使用 時’程式會略過未使用孔之試劑釋出，以節省試劑。使用 者可輕易地加入合成參數’如試劑注入容積、反應時間、 及改質基劑之'添加。此等參數可儲存在構案中以利後續使 用。1232867 A7 B7 V. Description of the invention (12 The spraying angle is large and the surface is smooth and the shape makes the reagent accumulated during the injection of the reagent is # I ^ 上 凡 1 is small, so it provides a consistent injection volume, and the distance between the reaction holes Cold dyeing with each other is the smallest. Furthermore, the injection degree of the cylinder is also very important for the success rate of the synthesis. The shot, the speed, the less the accumulation at the tip of a spray angle. On the other hand, the more the injection speed, the more CPG beads will be made. Out of the reaction hole. Therefore, it is necessary to determine the optimal injection speed to achieve a balance between the accumulation of reagents at the tip of the nozzle and the splashing of CPG beads. Take the best injection speed (selection is shown in Figure 5. The CPG beads are from the money) The tendency is directly proportional to the injection speed. The injection speed is determined by the injection pressure p and the inner diameter of the nozzle. The lower the injection speed, the lower the CPG beads, but the less the reagent consumption accumulates, the "reagent will stay at the nozzle tip until the reagent grows Until the surface tension can no longer support the specific size of the droplet. When the injection speed increases, the coefficient of variation (CV) curve of the injection volume reaches a minimum, but therefore the CPG bead splash height increases. Deeper reaction holes can be used to retain the cpG beads from the money and avoid punching out of the holes. 'Deep reaction holes' have another effect on minimizing the change in synthetic yield. This is due to the bead of the heart. The particles (although staying inside the reaction well) are tightly attached to the inner wall of the reaction well, and before the next synthesis step, these CPG beads need to be washed to the bottom of the well to ensure proper reaction efficiency. The higher the splash height It means that a larger washing volume is required to make these spheroids beads: This effect causes a large consumption of reagents, a longer reagent delivery time, and a longer discharge time. These factors need to be considered to determine the appropriate injection Figure 6 shows the increased splashing cpG beads when the injection speed is increased from the edge of the river _ reference standard half (CNS) A4 specifications (2ΐ〇χ297, "1232867 A7 -----_______________ B7 V. Description of the invention (C " ------ The volume of CH3CN required to go down. According to this, the waste discharge time and total synthesis time will be increased. In order to determine the optimal injection speed, the cv of the f-in volume is measured as a function of the injection speed. Because of the injection speed For many parameters (such as injection pressure, ID, and reagent viscosity), but for each reagent it is necessary to choose a small acceptable injection speed, and the minimum injection pressure may vary with different reagents. The overall optimal injection speed for all reagents is then selected by The acceptable injection pressure for all reagents is determined. As an example, Figure 5 demonstrates that the cv of the trichloroacetic acid (TCA) injection volume is measured as a function of the injection pressure. In Figure 5, the cv of the injection volume suddenly increased to below The cumulative pressure of the reagent. The cv curve conforms to the following exponential function: y = A * exp [. (Xx〇) / t] Formula! Where y is the injection volume of cv 4 A = 2.2, amplitude xo = 1, branch t = 0.53, the attenuation constant of the curve. Therefore, in order to achieve a CV of 0.1, the injection pressure is calculated as (2.53 * t) + x〇 ~ = 2.5 (psi). Set this value to a. The injection pressure a is the optimal injection pressure at which the TCA reaches the injection volume CV of 0.1. Similarly, the optimal injection pressure and overall injection speed of each reagent can be measured. Once the injection pressure is measured, the reaction hole depth can be measured as follows. CPG bead splatter height is measured using the required injection pressure. Example The paper size is applicable to Chinese National Standard (CNS) A4 specification (210 X 297 public form ^ 1232867 A7) Note (“) ~ ----- In the TCA system, the injection pressure is the main input, as shown in Figure 5. , And the average subtracted height is measured by repeating several injections. At 4.5 PS1 injection pressure, the average low output height obtained at 99% confidence is 48 ± 16 mm. In this example, the depth of the reaction hole needs to be at least 4.8 Mm + 16 mm = 64 mm. The 'easy-to-use' program in the prototype device was written in Visual Basic (Microsoft). Using 纟 to enter "Synthetic parameters are $, the program will calculate and issue movement instructions for the translation stage, and for the solenoid The valve issues an injection instruction. The user provides the sequence information in the simple building case. The computer program reads the information and performs the synthesis process in a fully automatic manner. When the 384 reaction wells are not fully used in operation, the program will skip the unused wells The reagents are released to save reagents. The user can easily add synthetic parameters 'such as reagent injection volume, reaction time, and modified base agents'. These parameters can be stored in the plan for subsequent use.

I 實例1 2發明者以上述原型裝置執行不同順序及長度之384寡 核甘酸&合成。合成支撐物（控制孔隙之玻璃I Example 12 The inventors performed the 384 oligonucleotide & synthesis in different sequences and lengths using the prototype device described above. Synthetic support

Research或Chemgenes))係在以移液管送到反應孔前，以 1 . 1 (v/v)(比例懸浮在氯仿/二溴甲垸中。合成前，藉由對 反=孔抽真空，排出溶劑。亞胺墙酸鹽（Chemg或綱 係落於無水乙腈（Aldrich，含水量〇〇〇1%)中且含亞胺磷 酸鹽之瓶與合成儀相連。合成儀室在進行合成前先以氮氣 沖洗30分鐘。 亞胺磷酸鹽合成化學品包含四個步驟，亦即去封阻、偶 合、封端及氧化。該實例中之合成參數如下：對去封阻而Research or Chemgenes)) is suspended in chloroform / dibromomethane at a ratio of 1.1 (v / v) (ratio of 1.1 (v / v)) before being pipetted to the reaction wells. Drain the solvent. The imine wall acid salt (Chemg or class is in anhydrous acetonitrile (Aldrich, 0.001% water content) and the bottle containing the imine phosphate is connected to the synthesizer. The synthesizer room is first synthesized. Rinse with nitrogen for 30 minutes. The imine phosphate synthesis chemical contains four steps, namely deblocking, coupling, capping and oxidation. The synthesis parameters in this example are as follows:

1232867 A7 B7 五、發明説明（15 ) 舌’係將40微升之TCA注入各孔中，接著以丨1〇微升cH3CN 洗滌。對於偶合階段，為確保最高之可達到偶合效率，偶 合反應每次以4微升亞胺磷酸鹽及8微升四唑進行兩次。各 反應均在試劑排出前進行60秒。將全部8微升之亞胺磷酸 鹽及16微升之活化劑注入各反應孔中，接著將$微升封端 齊J注入各孔中。反應在試劑排出前進行2〇秒。接著將5微 升氧化劑送到各反應孔中。反應在試劑排出前進行3〇秒。 由於反應孔之直徑（如前述約為2.2亳米），因此5微升試劑 已足以覆盍所有珠粒，使各階段之反應完全。氧化階段 後’以CNSCN洗滌反應孔，且再由去封阻階段開始重複合 成反應。 合成完成後，以乙腈（CH3CN)充分洗滌反應孔，以移除 反應孔中之殘留試劑。寡核甞酸藉由注入60微升鹼,溶液 (32 %氫氧化銨或40%甲胺，依所用之CPG珠粒而定），自 固態支撐物分離到各孔中。反應在室溫進行5至60分鐘。 再將反應板配件架在聚丙烯收集盤上，及將該二盤置於離 心機上，沉澱出粗產物。收集盤接著密封且在55培育8 小時，以移除寡核茹酸分子上之保護基。該盤在移除密封 前再度冷卻至室溫。粗產物於真空室中烘乾，再溶於去離 子水中且備用。寡核甞酸可以另一方法乾燥。正丁醇萃取 顯示可產出對PCR、定序等方面具充分品質之寡核甞酸。 反應盤配件可重複使用數次，而不會降低合成產物品 質。所用之反應盤配件先浸在丙酮中，且以超音波震動30 分鐘。各孔依序以400微升乙腈、60微升去離子水、乙腈及 本紙張尺度適用中國國家標準(CNS) A4規格(210 X 297公复j)8 1232867 A7 B7 五、發明説明（16 ) 40%甲胺洗滌。在2 60奈米下測量各孔溶離液之UV吸收度 約為0.01，約與溶劑相同。因此，上述製程可用以回收該 反應盤配件。 為測量合成寡核甞酸之純度，係以逆向C8管柱進行高效 能液相層析（HPLC)。HPLC系統包含二個島津LC-6A泵浦、 控制泵浦用之SCL-6A及當作偵測器之光電二極管陣列。使 用個人電腦控制HPLC系統，且得到層析圖。即時監測分析 之UV-Vis吸收光譜，以助於尖峰辨識。HPLC分離之移動相 係由試劑A 0.1M三乙基乙酸（TEAA)及試劑B :乙腈所組 成。試劑B之溶離液成分在0、24、34分鐘時分別為8 %、 20%、40%。全長之寡核甞酸產物在約30分鐘下溶離，且 不合格之序列在14-18分鐘間溶離。 為測量合成產率，將384-孔格式中合成之寡核甞酸移到 四個96-孔盤（Corning Coaster No.3635)中，且各寡核嘗酸之 吸收度以盤讀取機（SpectraMax Plus，分子裝置公司）讀取。 對各合成產物測量200奈米至400奈米之吸收光譜。使用260 奈米、280奈米及320奈米之吸收度評估最終產物量。結果 顯示盤中各孔均含有寡核甞酸。合成過程中並未發生CPG 珠粒流失，且未發現合成失敗。5奈莫耳公稱規格合成之 384寡核甞酸實際之產物產量平均約為6奈莫耳。孔-對-孔 產量變化之變異係數（CV)經測量為23%。當以相同裝置承 載96-孔反應盤時，發現15%之較小產率CV。經由比較， 先前技藝在96-孔格式中合成之寡核苷酸CV為23%至25%。 該等發現首先說明使反應孔密度增加4倍，亦即由96-孔增 本紙張尺度適用中國國家標準(CNS) A4規格(210X 297公袭^ 1232867 五、發明説明（17 力口 至 3 8 4 - L -V' 昨 、°式’且未犧牲產率之變異並非全然無價值。 第 就本發明所述方法而言，可達到比先前技藝更低之 孔-對-孔產量差異。 口成之v驟產率係以全長產物產率對總產物產率之比而 田使用瓜之cpG珠粒時，長度An核酸之寡核芬酸 平均步騾產率係依據下式計算： #[NF/NT]1/n 程式2 、 中丁及叫刀別為全部及全長產物之產率。其量可藉由 逆向HPLC測定。為了測量步騾之產率，爪序列之撕寡核 、、7:5莫耳至20奈莫耳之規模合成。總產物係隨機選 自數個孔。全部及全長度序列之相對產率係以HPLC測定。 圖8顯示不同孔產物之數種Ηριχ層析。各孔之平均步驟產 量係依程式2計算，且以各層析圖顯示。步騾產率f大於 99 /ί>係對合成規模不大於丨〇奈莫耳所達成。可忽略步驟產 率中孔-對-孔之差異。由於該裝置對多樣但小量之寡核苷 酸合成為最佳，因此對20奈莫耳之合成規模得到較低之步 驟產率並不意外。結果證明對於不同規模之合成需使合成 參數最佳化。 實例2 增加合成儀之產量可使用僅有一個合成盤轉譯控制系統 之多重試劑輸送系統。1536-管道（384x4 )合成儀係藉由使 用一組試劑儲槽、一個轉譯控制系統、四組合成儀配件、 及四組如實例1中所述384-管道合成儀之試劑輸送用之螺 管閥及噴嘴進行。為了增加12 -試劑合成製程用之額外合 本紙張尺度適用中國國家標準(CNS) Α4規格(210X 297公^j 1232867 A7 B7 五、發明説明（18 ) 成儀配件，需要12個額外之試劑輸送螺管閥及3個額外之 試劑分佈閥。與先前技藝如AMOS系統比較，其中一個額 外合成糢組需要約100個額外螺管閥。本發明在提昇產量 而增加額外之合成模組上明顯地更有彈性。 本紙張尺度適用中國國家標準(CNS) A4規格(210 X 297公g1232867 A7 B7 V. Description of the invention (15) The tongue 'is filled with 40 microliters of TCA into each well, and then washed with 10 microliters of cH3CN. For the coupling stage, to ensure the highest achievable coupling efficiency, the coupling reaction was performed twice with 4 microliters of imine phosphate and 8 microliters of tetrazole each time. Each reaction was performed for 60 seconds before the reagent was discharged. A total of 8 microliters of imine phosphate and 16 microliters of activator were injected into each reaction well, followed by injection of $ microliters of end caps into each well. The reaction was performed for 20 seconds before the reagent was discharged. Then 5 microliters of oxidant was sent to each well. The reaction was performed for 30 seconds before the reagent was discharged. Due to the diameter of the reaction well (about 2.2mm as mentioned above), 5 microliters of reagent is enough to cover all beads and complete the reaction in each stage. After the oxidation stage, the reaction wells were washed with CNSCN, and the reaction was recombined from the deblocking stage. After the synthesis is completed, the reaction wells are thoroughly washed with acetonitrile (CH3CN) to remove the residual reagents in the reaction wells. Oligonucleotide was separated from the solid support into each well by injecting 60 microliters of base, solution (32% ammonium hydroxide or 40% methylamine, depending on the CPG beads used). The reaction is performed at room temperature for 5 to 60 minutes. The reaction plate assembly was mounted on a polypropylene collection tray, and the two trays were placed on a centrifuge to precipitate a crude product. The collection dish was then sealed and incubated for 8 hours at 55 to remove the protecting groups on the oligonucleotide molecule. The tray was cooled to room temperature again before the seal was removed. The crude product is dried in a vacuum chamber, redissolved in deionized water and set aside. Oligonucleotides can be dried in another way. The n-butanol extraction shows that it can produce oligonucleotides with sufficient quality for PCR, sequencing, etc. The tray accessories can be reused several times without degrading the quality of the synthetic product. The reaction plate accessories used were first immersed in acetone and shaken with ultrasound for 30 minutes. Each well is in order of 400 microliters of acetonitrile, 60 microliters of deionized water, acetonitrile, and the size of this paper. Applicable to China National Standard (CNS) A4 specifications (210 X 297 public compound j) 8 1232867 A7 B7 V. Description of the invention (16) Washed with 40% methylamine. The UV absorbance of the eluate of each well measured at 2 60 nm is about 0.01, which is about the same as that of the solvent. Therefore, the above process can be used to recover the reaction plate assembly. To measure the purity of synthetic oligonucleotides, high performance liquid chromatography (HPLC) was performed on a reversed C8 column. The HPLC system includes two Shimadzu LC-6A pumps, SCL-6A for controlling the pump, and a photodiode array as a detector. A personal computer was used to control the HPLC system and a chromatogram was obtained. Real-time monitoring and analysis of UV-Vis absorption spectra to help identify spikes. The mobile phase separated by HPLC consisted of reagent A 0.1M triethylacetic acid (TEAA) and reagent B: acetonitrile. The eluent composition of Reagent B was 8%, 20%, and 40% at 0, 24, and 34 minutes, respectively. The full-length oligonucleotide product dissolves in about 30 minutes, and the unqualified sequence dissolves in 14-18 minutes. To measure the synthetic yield, the oligonucleotide synthesized in the 384-well format was moved to four 96-well plates (Corning Coaster No. 3635), and the absorbance of each oligonucleotide was measured by a disk reader ( SpectraMax Plus, Molecular Device Corporation). The absorption spectrum of 200 nm to 400 nm was measured for each synthesized product. The final product amount was evaluated using the absorbances of 260 nm, 280 nm, and 320 nm. The results showed that each well in the dish contained oligonucleotides. No CPG bead loss occurred during the synthesis, and no synthesis failure was found. The actual yield of 384 oligonucleotide synthesized at 5 nmole nominal specifications is about 6 nmole. The coefficient of variation (CV) of the hole-to-hole yield change was measured to be 23%. When a 96-well reaction disk was loaded on the same device, a smaller yield of CV was found to be 15%. By comparison, the CV of the oligonucleotide synthesized in the 96-well format by previous techniques is 23% to 25%. These findings first explained that the density of the reaction pores was increased by a factor of four, that is, the paper size of the 96-well sterilization paper was adapted to the Chinese National Standard (CNS) A4 specification (210X 297 public attack ^ 1232867 V. Description of the invention (17 Likou to 3 8 4-L -V 'Yesterday, °' without sacrificing yield variation is not entirely worthless. For the method described in the present invention, it is possible to achieve lower hole-to-hole yield differences than previous techniques. Mouth The yield rate of V is based on the ratio of the full-length product yield to the total product yield. When Tian uses cpG beads of melons, the average step yield of oligonucleotide of length An nucleic acid is calculated according to the following formula: # [ NF / NT] 1 / n Formula 2, Zhong Ding and Dao Ding are the yields of all and full-length products. The amount can be determined by reverse HPLC. In order to measure the yield of the step, the claw sequence oligonucleotide ,, 7: 5 mole to 20 nanomole synthesis. The total product was randomly selected from several wells. The relative yields of all and full-length sequences were determined by HPLC. Figure 8 shows several Ηρχχ chromatography of products of different wells. The average step yield of each well is calculated according to formula 2 and shown in each chromatogram. The step yield f is greater than 99 / ί > Achieved when the synthesis scale is not larger than 〇〇Νmol. The pore-to-pore difference in the step yield can be ignored. Since the device is optimal for the synthesis of a variety of small amounts of oligonucleotides, It is not surprising that the yield of the 20 nanomole synthesis step is low. The results prove that the synthesis parameters need to be optimized for different scale synthesis. Example 2 Increase the output of the synthesizer can use only one synthesis disk translation control system Multiple reagent delivery system. The 1536-channel (384x4) synthesizer uses a set of reagent storage tanks, a translation control system, four combination instrument accessories, and four sets of 384-channel synthesizers as described in Example 1. Solenoid valve and nozzle for reagent delivery. In order to increase 12-reagent synthesis process, additional paper size is applicable to Chinese National Standard (CNS) A4 specification (210X 297 public ^ j 1232867 A7 B7 V. Description of the invention (18) For the instrument accessories, 12 additional reagent delivery solenoid valves and 3 additional reagent distribution valves are required. Compared with previous technologies such as the AMOS system, one additional synthesis module requires about 100 additional solenoid valves. The invention increases significantly increase production more flexible on the synthesis of additional modules. This paper scales applicable Chinese National Standard (CNS) A4 size (210 X 297 male g

裝 訂Binding

線line

Claims (1)

88號專利申請案 •利範圍替換本(94年！月）Patent Application No. 88 • Replacement of the scope of interest (1994! Month) 利範圍 A8 B8 C8 D8 i丨 m3 一種藉由平行運作方式或更多種寡核苷酸之裝 置，該裝置包含： 一反應盤配件’其包含3 8 4或更多個反應孔及支撐該 反應孔之承板’其中該反應孔尖端直徑為3 · 5毫米 或更小且及各反應孔之孔距為4 · 5毫米或更小； 一真空配件’其包含一或多個真空泵浦，其中該真空 配件與該承板連接以固定該承板； 一移動系統’其包含電腦控制之試劑輸送系統及一或 多個試劑喷嘴’其中該試劑噴嘴位於該反應盤上方 及該移動系統與反應盤連接並使反應盤移動以收集 自試劑噴嘴注入各反應孔中之試劑； 一試劑閥配件，其包含一或多個螺管閥及螺管閥驅動 電路’其中該螺管閥驅動電路連接該螺管閥及該移 動系統； 一或多個試劑儲槽及管，其中該管連接該試劑儲槽、 該螺管閥及該試劑噴嘴，及由該管將試劑自試劑儲 槽遞送到螺管閥及該試劑噴嘴；及 一電腦系統，其包含控制軟體以控制寡核芬酸合成程 序並讀取寡核甞酸結果。 2.如申請專利範圍第丨項之裝置，其中該反應孔由冠狀 板、維持合成支撐物之介質及基劑板所構成，其中冠狀 板包含底部直徑約2毫米及深度約1 5毫米之圓錐形孔。 3·如申請專利範圍第2項之裝置，其中該維持合成支據物 之介質之起泡點大於真空配件壓差。 69922-940131.DOC 本紙張尺度適用中國國家標準(CNS) A4規格(210 X 297公釐) 1232867 A8 B8 C8 D8 申請專利範圍 4·如申請專利範圍第2項之裝置，其中該合成支撐物為控 制孔隙玻璃（C P G )。 5·如申請專利範圍第丨項之裝置，其中該真空泵浦之速率 大於通過介質之氣體流動速率。 6·如申叫專利範圍第1項之裝置，其中該試劑噴嘴具有低 表面張力。 7·如申請專利範圍第1項之裝置，其中該試劑噴嘴為具有 2 0達因/公分低表面張力之鈍切割鐵氟龍管。 8.如申請專利範圍第〗項之裝置，其中該試劑噴嘴之試劑 噴出速度須使噴嘴口處之試劑累積及合成支撐物濺出為 最小。 9·如申請專利範圍第1項之裝置，其中該試劑之射出容積 約為2微升。 ' 10· —種使用如申請專利範圍第丨項之裝置合成數百或更多 種寡核棼酸之方法，其中該方法包括下列步驟： 以該裝置之電腦系統控制寡核苷酸之合成程序，將 合成支撐物、核苷酸反應物及合成反應試劑導入該 裝置之反應孔；及 使核芬酸反應物進行去封阻、偶合、封端及氧化反 應以合成暴核菩酸。 11_如申請專利範圍第1 〇項之方法，其中該合成支搏物為 控制孔隙玻璃（c P G )。 12·如申請專利範圍第i 〇項之方法，其中該合成反應試劑 為亞胺磷酸鹽。 69922-940131.DOC · 2 · 本紙張尺度適用中國國家標準(CNS) A4規格(210 X 297公釐） 8 8 8 8 A B c D 1232867 六、申請專利範圍 13.如申請專利範圍第1 0項之方法，其中該合成寡核甞酸 反應所需的最佳試劑喷注壓力參數係由如申請專利範圍 第3、5或8項之裝置測試而得。 69922-940131.DOC ~ 3 ~ 本紙張尺度適用中國國家標準(CNS) A4規格(210 X 297公釐)A8 B8 C8 D8 i 丨 m3 A device that operates in parallel or more oligonucleotides, the device includes: a reaction plate accessory 'which contains 3 8 4 or more reaction holes and supports the reaction Hole-bearing plate 'where the diameter of the tip of the reaction hole is 3.5 mm or less and the distance between each reaction hole is 4.5 mm or less; a vacuum accessory' which contains one or more vacuum pumps, where The vacuum accessory is connected to the carrier plate to fix the carrier plate; a mobile system 'which includes a computer-controlled reagent delivery system and one or more reagent nozzles', wherein the reagent nozzle is located above the reaction disk and the mobile system and the reaction disk Connect and move the reaction disc to collect the reagents injected into the reaction holes from the reagent nozzles; a reagent valve accessory, which includes one or more solenoid valves and solenoid valve drive circuits, wherein the solenoid valve drive circuit is connected to the solenoid Tube valve and the moving system; one or more reagent storage tanks and tubes, wherein the tube is connected to the reagent storage tank, the solenoid valve and the reagent nozzle, and the reagent is delivered from the reagent storage tank to the spiral tube by the tube And said reagent nozzle; and a computer system, comprising software for controlling the controlled oligonucleotide synthesis procedures diclofenac oligonucleotide Chang acid and reads the results. 2. The device according to the scope of the patent application, wherein the reaction hole is composed of a crown plate, a medium to maintain a synthetic support, and a base plate, wherein the crown plate includes a cone with a diameter of about 2 mm at the bottom and a depth of about 15 mm. Shaped hole. 3. The device according to item 2 of the scope of patent application, wherein the bubble point of the medium for maintaining synthetic receipts is greater than the pressure difference of the vacuum fittings. 69922-940131.DOC This paper size is applicable to Chinese National Standard (CNS) A4 specification (210 X 297 mm) 1232867 A8 B8 C8 D8 Patent application scope 4. If the device of the second scope of the patent application is applied, the synthetic support is Controlled Pore Glass (CPG). 5. The device according to item 丨 of the patent application range, wherein the rate of the vacuum pump is greater than the gas flow rate through the medium. 6. The device as claimed in claim 1 in which the reagent nozzle has a low surface tension. 7. The device according to item 1 of the patent application range, wherein the reagent nozzle is a blunt-cut Teflon tube having a low surface tension of 20 dyne / cm. 8. For the device in the scope of the patent application, the reagent ejection speed of the reagent nozzle must minimize the accumulation of reagents at the mouth of the nozzle and the spillage of synthetic support. 9. The device according to item 1 of the scope of patent application, wherein the ejection volume of the reagent is about 2 microliters. '10 · —A method for synthesizing hundreds or more oligonucleotides using a device such as the one described in the patent application, wherein the method includes the following steps: The computer program of the device controls the synthesis of oligonucleotides , Introducing a synthetic support, a nucleotide reactant, and a synthetic reaction reagent into the reaction wells of the device; and deblocking, coupling, capping, and oxidizing the nuclear fenamic acid reactant to synthesize nucleate. 11_ The method of claim 10 in the scope of patent application, wherein the synthetic beat substance is a controlled porosity glass (c P G). 12. The method according to item i 0 of the patent application range, wherein the synthetic reaction reagent is imine phosphate. 69922-940131.DOC · 2 · This paper size is applicable to Chinese National Standard (CNS) A4 specification (210 X 297 mm) 8 8 8 8 AB c D 1232867 6. Scope of patent application 13. If the scope of patent application is the 10th item The method, wherein the optimal reagent injection pressure parameter required for the synthetic oligonucleotide reaction is obtained by testing the device such as the scope of patent application No. 3, 5 or 8. 69922-940131.DOC ~ 3 ~ This paper size applies to China National Standard (CNS) A4 (210 X 297 mm)