A PCR BASED HIGH THROUGHPUT METHOD FOR CONSTRUCTION OF FULL SITES SMALL INTERFERING RNA (SiRNA) POLYNUCLEOTIDES AND

RELATED COMPOSITIONS

FIELD OF THE INVENTION

This invention relates to biotechnology and kits for constructing an siRNA polynucleotide pool from a sample and the selected polynucleotide pools produced thereby, more particularly for preparing an siRNA polynucleotide pool using type III restriction/modification enzymes and the corresponding linkers.

DESCRIPTION OF RELATED ARTS

An understanding of the biological role of any gene comes only after observing the phenotypic consequences of altering the function of that gene in a living cell or organism. RNA interference (RNAi) is a well-established experimental technology for silencing gene expression both in cultured eukaryotic cells and living organisms. Currently, RNAi also opens a way for gene therapy, such as for defense against viral infection, cancer as well as vascular diseases etc. RNAi induces the sequence-specific degradation of a single mRNA species by short interfering RNA (siRNA, a double-stranded small interference RNA), which is believed to be processed through the highly conserved Dicer family of RNase III enzymes in vivo. The process includes: 1) delivery of homologous double-stranded RNAs (dsRNAs) to the cytoplasm of a cell. 2) dsRNA cleavage by the RNase Ill-like enzyme, Dicer, to 21-23 bp siRNAs. 3) siRNA incorporating into a protein complex, the RNA-induced silence complex (RISC). 4) the antisense strand of the duplex siRNA guiding the RISC to the homologous mRNA, where the RISC-associated endoribonuclease cleaves the target mRNA, resulting in silencing of the target gene.

The siRNA molecule of interest can be synthesized in vitro by chemical and enzymatic (Dicer) methods. They can also be synthesized in vivo. When a synthetic oligonucleotide is cloned into siRNA expression vectors with a RNA polymerase III promoter (including U6, human Hl, and tRNA promoters), or a polymerase II promoter with a minimal poly(A) signal sequence, siRNA can be transcribed in vivo. Typically a single promoter is used to express a short hairpin (shRNA)

sequence, although two tandem polymerase III promoters have also been used to transcribe the sense and antisense siRNA sequences. In addition to plasmid-based systems, PCR-derived siRNA expression cassettes based on the single-promoter system is an alternative format for suppressing transfected gene activity.

The siRNA-mediated gene silencing efficiency is affected by many parameters. An important limiting fact is that only about 25% of selected target siRNA sequences are functional due to some factors, such as secondary structures, non-gene-specific reactions and some unknown ones. Thus several synthetic siRNAs need to be generated and tested for every target gene. This is much expensive and time consuming to identify a suitable construct. Furthermore, to find the best siRNA binding sites (RNAi drug targets) is more challenging. To resolve this off-target phenomenon, extensive studies have been done on selecting potential target sequences for siRNA. Using algorithms based on sequences-efficacy correlations is current practice for designing effective siRNAs. Although these criteria significantly increase chance of success for achieving gene silencing, there are many highly effective siRNA sequences do not anticipate the current algorithms because different genes have different sequence preference. To ensure that the best siRNAs are identified, an siRNA library constructed from cDNA or DNA offers a better alternative way to search for sequences that have the best potential silencing effect. Moreover, such an siRNA library can be a useful research tool to functional genomics, more preferably, in a high throughput way on effective screening for RNAi therapeutic targets.

Recently siRNA library approaches (whole genomic or gene-specific or domain-specific siRNA libraries) have been widely used as a powerful tool for RNAi therapeutic targets screening. In all those approaches, efforts to generate siRNA sequences with appropriate length have been employing Mmel- a type II restriction enzyme, by which a maximum 20 nucleotides can be generated. However, a maximum 20bp siRNA in length can not completely mimic the cleavage (21-23bp) by an RNase Ill-like enzyme- Dicer. This can cause the best siRNA target sites underpresentation. For all these approaches, the double-stranded (ds) cDNA is randomly cleaved into small fragments by DNase I (some use restriction enzymes for fragmentation but the representation is an issue ) and subsequently ligated to an artificial loop-anchor which contains Mmel- a type II restriction site, then by Mmel restriction enzyme to cut 18-20 nucleotides away from the recognized site. Through complex and multiple steps of manipulations for the second

anchor ligation, loop extension and PAGE purifications, the ds-cDNA is then converted into a 20-nt palindromic structure with a loop (shRNA) and finally clone into the siRNA expression vector with an RNA polymerase III promoter. In all these approaches, efforts to generate siRNA sequences with appropriate length have been employing Mmel restriction enzyme, by which only a maximum 20 nucleotides can be generated. This is the longest length can be generated among the type II restriction enzymes. However, there are existed three major drawbacks in this approach: 1) shRNA library cannot be generated by PCR due to a palindromic structure. The complicated steps together with heavily cDNA loss in multiple manipulations make this approach practical difficulty and impossible to become a high throughput tool for functional genomics and towards to siRNA therapeutic target screening. 2) The palindromic structure is unstable during cloning in E. coll This can lead to reduction in library complexity and potential loss of the best therapeutic target sites. 3) A maximum 20bp siRNA in length cannot completely mimic the cleavage by the RNase Ill-like enzyme, Dicer, from 21 to 23bp. This may be another limitation for the best RNAi therapeutic target sites screening by using this kind of library.

An attempt employing PCR to construct siRNA library from cDNA has currently been reported. In this system, the dsRNAs corresponding to the cDNA of interest are prepared by T7 RNA polymerase mediated transcription from DNA templates flanked by T7 RNA promoter and subsequent annealing. The dsRNAs are then digested with cloned human Dicer in vitro, yielding 21-23 bp siRNAs (modified bacterial RNase III can be used to replace Dicer, but the generated siRNA is 20-25 bp). Cleavage products were denatured, purified by PAGE and dephosphorylated. RNA adapters were attached subsequently to the 3'- and 5'- ends of the cleavage products by T4 RNA ligase. RNAs are subsequently converted into dsDNA by RT-PCR using primers complementary to the adapters. After digestion with appropriate restriction enzymes, the 21 - 23 bp siRNAs corresponding to the cDNA fragments are ligated into the siRNA expression vector with dual RNA polymerase III promoters: U6 and Hl. Taking advantage of PCR, this approach can tolerate the staring material heavily loss during multiple manipulations and generate enough molecules for cloning. Another advantage is that RNA fragmentation with Dicer, a distinct random 21 - 23 bp siRNA in length can be generated. However, the manipulations on cDNA-RNA-cDNA conversion are obviously a complex procedure, even more complicated than shRNA library construction described above. Furthermore, RNA degradation during multiple manipulations (e.g.,

T7 DNA polymerase-mediated DNA to RNA transcription, Dicer digestion, RNA PAGE purification, dephosphate and anchor ligation as well as RT-PCR) is unavoidable, which may heavily loose some best siRNA target sites in the library.

Another attempt on siRNA library construction is based on DNase I digestion only. In this approach, dscDNA is partially digested with DNase I, followed by PAGE gel purification for 20-3 Obp. These fragments are either directly blunt-end cloned into siRNA expression vector or attached a PCR anchor by ligation, PCR amplification and subsequently cloning into siRNA expression vector. It sounds much simpler and straightforward. However cutting 20-3 Obp from gel is very challenging. Contamination from smaller (e.g., <16 bp) and larger (e.g. >30 bp) cannot avoided. Too short siRNA (e.g., <16 bp) is not efficiency. For > 30bp larger siRNA, it is back to the original RNAi technology in past decades, which predominantly employed dsRNA > 3 Obp and was proved to be not effective in mammalian cells because their introduction activates an interferon and protein kinase R (PKR) pathways, resulting in nonspecific gene silencing and apoptosis. Such an siRNA library may contain a high frequency of undesirable ("junky") clones which may not only drastically impair the overall efficiency of the approach, but also seriously compromise the integrity of the data that are generated. Thus this approach is not ideal on effective screening for the best siRNA sequence site of functional genomics and RNAi therapeutics.

An ideal siRNA library, especially gene-specific library, should contains every site represented by multiple overlapping sequences, and individual sequence should have widely accepted rational length: 19-23 bp, and should easily and simply be amplified by PCR to meet a high throughput library construction format, really accelerating the screenings for the best siRNA sequence site of functional genomics and RNAi therapeutics.

Current appearance of a type III restriction enzyme- Eco?\5\ can cleave maximal 25-27bp of DNA outside of their recognition site. So far there are no any related arts employing any type III restriction /modification enzymes for siRNA library construction yet, mainly due to two technical difficulties: a.) the cleavage 25-27bp is out of scale of 19-23bp which is a widely accepted rational length of siRNA; b.) two inversely oriented recognition sites of type III restriction /modification enzymes is required for effective cleavage (using EcoP15I as an example):

£eoP15I ►

5'CAGCA(N) _{25 27} (N)GTGCTG3'

3'GTCGT(N) _{27 25} (N)CACGAC5'

-4

Eco?15l

In the art of present invention bellow, the addressed two technical difficulties above can be overcome by an EcoP15I artificial loop phosphate linker(s).

SUMMARY OF THE PRESENT INVENTION

A main object of the present invention is to provide a type III restriction/modification enzyme mediated PCR high-throughput method for preparing an siRNA polynucleotide pool from a DNA sample (cDNA or genomic DNA and so on) for RNAi therapeutic targets screening. The resulting siRN A polynucleotide pool length can distinctly distributed from 16-27 bp, more preferably, from 21-23 bp which completely mimics the length of siRNA naturally generated by Dicer enzyme in living cells, overcoming the bottleneck and underpresentation by all conventional siRNA polynucleotide pool construction approaches mediated by a type II restriction enzyme-Mmel, in which the longest siRNA generated is 18-20bp in length.

Another object of the present invention is to provide an artificial oligonucleotide shaped in a loop after self-annealing (Loop-1 linkers) containing the recognition sites of a type III restriction/modification enzyme and a type II restriction enzyme.

The general structure formula of Loop-1 linker(s) is:

5'pC(n)TTTTN(n) Type IIIase site N(n) Type IIase site-PCR anchor-loop 3' G(n)AAAAN(n) Type IIIase site N(n) Type IIase site-PCR anchor-loop

Where

P: phosphate group;

N: any bases of G, C, T, A; n: the number of base (form 0 to 20)

Type IIIase: type III restriction/modification enzyme;

Type IIase: type II restriction enzyme.

Another object of the present invention is to provide another artificial oligonucleotide shaped in a loop after self-annealing (Loop-2 linker) containing a type II restriction enzyme recognition sequences.

The general structure formula of Loop-2 linker is:

5'pC(n)TTTTN(n) Type IIase site-PCR anchor-loop 3' G(n)AAAAN(n) Type IIase site-PCR anchor-loop

Where

P: phosphate group;

N: any bases of G, C, T, A; n: the number of base (form 0 to 20)

Type IIase: type II restriction enzyme.

Another object of the present invention is to provide a protocol for a PCR based high- throughput method in preparing an siRNA polynucleotide pool from a DNA sample, which is: a.) Partial digestion of cDNA or genomic DNA with DNaseI in the present of Mn ²⁺ . The resulting products are blunt-ended. b.) Loop-1 linker(s) blunt-end ligation; c.) PCR amplification with a single primer that is a portion of homolog sequences to the antisense strand of the Loop-1 linker(s) (the strand with a poly A stretch). The resulting PCR products contain double type III restriction/modification enzyme sites (e.g., EcoP15I) in inversed orientation at the DNA both ends, which are cleavable. d.) A type III restriction/modification enzyme site (e.g., Eco?l5ϊ) cleavage. The enzyme cleaves a maximal 25-27bp of DNA outside of their recognition sites. The resulting siRNA polynucleotide pool distributed froml9 to 23bp in length by adjusting the adjacent number of poly (AJT) sequences in the Loop-1 linker(s); e.) Blunt-ending by filling in with a DNA polymerase in the present of dNTPs; f.) Loop-2 linker ligation. A type II restriction site (e.g., Fokl) is included in the Loop-2; g.) Second PCR amplification with 5' Loop-1 and 3' Loop-2 primers; h.) A type II restriction enzyme (e.g., Fokl in both loops) digestion to generate over A ₄ hands (cloning sits) at both 5' ends; i.) Cloning to a pre-prepared siRNA expression vector with over T ₄ hands at both 3 ' ends, flanked by two tandem RNA polymerase III promoters such as U6 and Hl, to complete an siRNA polynucleotide pool construction. Ploy (A/T)s act as the initial and termination signals for RNA polymerase III promoters after cloning.

BRIEF DESCRIPTION OF THE DRAWINGS

FIG.l A Schematic Diagram of Technical Flow-Chart and the Structure of Loop- 1 Linker(s) (using

EcoP15I as an example):

FIG.2 Schematic Diagram of an siRNA Expression Vector (pU6Hl-GFP) Map and MCS

Sequences Before and After Cloning.

FIG.3 Electrophoresis of SARS coronavirus membrane protein cDNA (666bp) on 1% Agarose gel.

FIG.4 A partial digestion pattern (Lanes 1-3) of SARS coronavirus membrane protein cDNA on

1% Agarose gel, along with a lkb plus DNA ladder (Lane M; NEB).

FIG.5 (A) PCR-I products (Lanes: 1-5: 19-; 20-; 21-; 22- and 23 bp Loop-1 anchored, respectively) on 1% Agarose gel along with a lkb plus DNA ladder (Lane M). (B) After phenol/chloroform extraction and ethanol precipitation.

FIG.6 A PAGE analysis for the 66bp DNA fragments cleaved by EcoP15I, along with a Low

Molecular Weight DNA Ladder (NEB).

FIG.7 Analysis for PCR-2 product on 20% PAGE, along with a Low Molecular Weight DNA

Ladder (LMW; NEB).

FIG.8 Analysis for Fokl digestion patterns on 20% PAGE, along with a Low Molecular Weight

DNA Ladder (LMW; NEB).

FIG.9 Analysis of PCR insert screening products on 1% Agarose gel, along with a Low Molecular

Weight DNA Ladder (NEB; the arrows indicated).

FIG.10 Analysis of the Sfil digested PCR insert screening products (1% Agarose gel; the arrow indicated is the Sfil digested product).

FIG.ll Plasmid sample check before sequencing on 1% Agarose gel.

THE DETAILED DESCRIPTION OF THE PREFERED EMBODIMENT

In the Loop-1 linker(s) of present invention (see FIG.l and "The general structure formula" described in the SUMMARY OF THE PRESENT I]SfVENTION), the poly(T/A) sequences between £cσP15I and DNA inserts play three key roles: a.) The resulting DNA insert length after Ecό?\5l cleavage can be controlled from 19- 23bp by addition or deletion of poly(T/A) sequences between EcoPlSl in Loop-1 and DNA substrates. If using a Loop-1 linker only with a distinct poly(T/A) number, the length of DNA insert can also be fixed at a certain length to construct an siRNA polynucleotide pool at a distinct length, or mix all

Loop-1 linkers together to generate an siRNA polynucleotide pool in 19-23bp size distribution, according to the needs. It can also expand to construct an siRNA polynucleotide pool in other length distribution by addition or deletion of the A/T sequences in Loop-1 linker(s) (e.g., 16-18bp; 24-27bp polynucleotide pool); b.) Generation of a cloning site- over A ₄ hands at both 5' ends by Fokϊ, a type II restriction enzyme, conducting a poly T/A cloning. For 24-27bp siRNA polynucleotide pool construction, other type restriction enzymes can be employed to create the corresponding cloning sites; c.) Generating the initiation and termination signals of RNA polymerase III promoters. After the siRNA polynucleotide inserts are cloned into an siRNA expression vector, the "AAAAA" (an initiation signal) and "χχχ7T" ( _a termination signal) of the promoters are generated. There are no additional cloning sequences between RNA polymerase III (U6 and Hl) promoters and DNA inserts, which can increase specificity of RNAi therapeutic targets screening.

The number of "N" between Eco?l5l and Fokl maintains the A ₄ adhesive ends (over A ₄ hands at both 5' ends) producing. As soon as ligation into an siRNA expression vector which has T5/A1 adhesive ends, the length of initiation signal (AAAAA) and termination signal (TTTTT) for RNA polymerase III promoters (U6 and Hl) are created.

Addition of the "G" before the initiation signal "AAAAA" of RNA polymerase III promoters is to improve an siRNA transcription efficiency. The present invention features that all Loop-1 linkers contain "G" before PoIy(A), except for construction of 23bp siRNA polynucleotide pool. It is widely accepted that placing "G" before the initial signal "AAAAA" enables to enhance RNA polymerase III promoters' activity, especially for U6 promoter. However, "G" has to be omitted to maintain siRNA polynucleotide 23bp in length by Loop-1 linker and can be added to an siRNA expression vector alternatively. If needed, for 24-27bp siRNA polynucleotide pool construction, it has to place the initiation/termination signals together with its enhancing base "G" into an siRNA expression vector. In this case an appropriate cloning site has to be selected in both Loop-1 linker(s) and vector to conduct an efficient cloning. The present invention provides unlimited formats, especially with the future appearance of other type III restriction/modification enzymes that have longer cleavage outside of their recognition sequences than EcoP15I.

PCR anchor and a loop structure designed in Loop-1 linker(s) is to select a right orientation for type III restriction/modification enzyme cleavage. As mentioned above, the type III

restriction/modification enzymes require two 5' sequences in inversed orientations within the same DNA molecule to accomplish cleavage. Single primer PCR enables to select such molecules. For a successful single primer-based PCR selection, a loop shaped structure is helpful. Unlike a tandem linker, a looped linker has no primer activity, thus ensures the single primer PCR successful.

In the Technical Flow Chart of the present invention:

DNA is randomly and partially digested to the blunt-ended fragments by DNase I in the presence OfMn ²⁺ :

DNase I, (RNase-free) is an endonuclease that nonspecifically cleaves DNA to release products with 5 ^' -phosphorylated and 3 '-hydroxylated end. The concentration for DNase I partial digestion is 0.01-0.3U per 1 g DNA depending on the length and purity of DNA as well as the source of the enzyme. Due to a nonspecific fragmentation, the resulting products are random fragments that can be a representative distribution of the siRNA polynucleotide pool generated. In the presence of Mn ²⁺ (a final concentration is 1 niM in a buffer, e.g., NEB buffer-2), the partially digested DNA are blunt-ended. The preferable length for downstream Loop-1 linker(s) ligation and siRNA polynucleotide pool construction is 100-300bp.

2.) Loop-1 linker(s) ligation:

As shown in FIG. 1, the present invention has designed various kinds of corresponding Loop -1 linker(s). They can be used in a mixture to generate 19-, 20-, 21-, 22-, and 23bp siRNA polynucleotide pool in length simultaneously, or separately use if one has specially interest in a distinct length polynucleotide of siRNA. As described in the Loop-1 phosphate linker(s) structure, a blunt-end ligation with the partially digested DNA can be performed by any DNA ligases, more preferably, a T4 DNA ligase in the presence of ATP. The molar ratio of Loop-1 linker(s) and DNA is 10:1.

3.) Single Primer PCR Amplification:

Efficient cleavage by type III restriction enzymes requires the presence of two inversely oriented substrate sites as above described. A head to head configuration in inverse orientation is required. Use EcoV\5l as an example:

(A):

£coP15I ^

5' -CAGCA(N) ₂₅ (polyA) (polyT) ₂₇ (N)GTGCTG-3'

3'-GTCGT(N) ₂₇ (polyT) (polyA) ₂₅ (N)CACGAC-5'

^M EcoV15I

However, the blunt-end ligation of Loop-1 linker(s) and DNA not only can generate the cleavable molecules (A), also generate other uncleavable molecules (B and C) as well:

(B):

5'-GTCGTC(N) ₂₇ (PoIyT) ( polyT) ₂₇ (N)CTGCTG~3'

3 '-CAGCAG(N) ₂₅ (PoIyA) (polyA) _2S (N)GACGAC~5'

(C):

5'-CAGCAG(N) ₂₅ (poIyA) (polyA) ₂₅ (N)GACGAC-3'

3 '-GTCGTC(N) ₂₇ (PoIyT) (polyT) ₂₇ (N)CTGCTG~5'

The present invention provides a single primer PCR amplification for selection of molecule (A) that can be cleavable.

This single primer is a portion of sequences homolog to the strand with poly(A) stretch in the Loop-1 linker(s). The single primer PCR amplification completes a selection for molecule (A) that has two inversely orientated 5 '-ends. The resulting PCR products are EcoP 151 cleavable. The unclevable molecules (B) and (C) are removed during the PCR process:

5' ^3' (same primer)

5'— CAGCAG(N) ₂₅ (polyA) (polyT) ₂₇ (N)CTGCTG— 3'

3 '..-GTCGTC(N) ₂₇ (polyT) (polyA) ₂₅ (N)GACGAC— 5'

(Same primer) 3' ^ 5'

3.) EcoP 151 Digestion:

The type III restriction-modification enzyme- EcoP 151, consisting of two modification (Mod) subunits and two restriction (Res) subunits, requires the interaction of two unmethylated, inversely oriented recognition sites 5'-CAGCAG in head to head configuration to allow an efficient DNA cleavage. ATP is required for the cleavage. Cleavage efficiency is also affected by the distance between the two sites. EcoP 151 can efficiently recognize the sites in distance up to 1.7 kb.

The Loop-1 linkers contain a type III restriction/modification enzyme site-EcoP15I and a type II restriction site-Fo/cl adjacent to a PCR anchor,. EcoP 151 cleaves 25-27bp outside of their

recognition sequences, thus adjusting the number of the poly A/T sequences between EcoP15I recognition site in Loop- 1 linker(s) and DNA inserts, the resulting siRNA polynucleotide pool has a distinct siRNA distribution in functional length (e.g., 19-23bp)..

4.) Blunt-ending and Loop-2 linker ligation:

The DNA is cohesive-ended after EcoP15I cleavage (sense strand 25bp and antisense strand 27bp), forming an over 5' two bases hand and can be filled-in using a DNA polymerase in the presence of dNTPs at a final concentration is 0.02 mM.

After a gel purification, the ligation of blunt-ended DNA and a Loop-2 linker is catalyzed by T4 DNA ligase in the presence of ATP. The ligation products are template for the second PCR amplification.

5.) Second PCR amplification:

The blunt-end ligation of EcoP15I digested siRNAs and Loop-2 linker generates the following two molecules due to different orientations: (A): Fok I (in Loop-2 ) siRNA Fok I ( in Loop-1)

5'.—GGATG—(polyA) (polyA) GTAGG 3'

3'-—CCTAC—(polyT) (polyT) CATCC-—5'

(B):

Fok I (in Loop-1 ) siRNA Fok I ( in Loop-2)

5'.—GGATG—(polyA) (polyA) CATCC -—3'

3'-—CCTAC—(polyT) (polyT) GTAGG 5'

After Fok I cleavage, only molecule (B) can generate the right cloning sites as below: siRNA insert

IU A A A \ '

The second PCR amplification selects molecule (B) by 5' Loop-1 and 3' Loop-2 primers:

5' Loop-1 primer ^ siRNA

5' GGATG (polyA) (polyT) — CATCC - — 3'

3' GCTAC (polyT) (polyA) — GTAGG - — 5'

3'Loop-2 primer

6.) Fokl digestion:

Loop-2 linker plays a role in creating a PCR anchor and a cloning site. As a type II restriction &nzymQ-Fokl can cleave any neighboring 9- 13bp sequences, after digestion, the DNA inserts generate over A ₄ hands at both 5' ends as described in (5.).

7.) Cloning into an siRNA expression vector:

After a gel purification, the Fokl digested products can be cloned into a pre-prepared siRNA expression vector with over T ₄ hands at both 3' ends (dephosphorylated), flanked by two tandem RNA polymerase III promoters such as U 6 and Hl (FI G.2).

Ploy(A/T) ₅ act as the initial and termination signals for RNA polymerase III promoters as described previously.

After E. coli transformation, the resulting siRNA polynucleotide pool has a size distribution same as that the Loop-1 linkers restricted (e.g., 19-23bp).

EXAMPLE- 1: OIigo Sequences and Preparation

(A) Loop Phosphate Linkers: (synthesized by Sigma-aldrich) Loop-1 linker(s): 19LP:

5 'PCTTTTTTTCTGCTG CATCCCTGAACTGGGATCCGTTGGATGTGT

3 'GAAAAAAA GACGAC GTAGG GACTTGACCCTAGGCAACCT ACACA

£coP 151 F ok I (P= phosphate group)

0LP:

-IT +1G 'PCTTTTTTCTGCTGGCATCCCTGAACTGGGATCCGTTGGATGTGT

3'GAAAAAAGACGACCGTAGGGACTTGACCCTAGGCAACCTACACA 1LP: -2T +2G

5'PCTTTTTCTGCTGGGCATCCCTGAACTGGGATCCGTTGGATGTGT 'GAAAAAGACGACCCGTAGGGACTTGACCCTAGGCAACCTACACA

22LP:

-3T +3G

5'PCTTTT CTGCTGGGGCATCCCTGAACTGGGATCCGTTGGATGTGT k λ

3 'GAAAA GACGACCCCGTAGGGACTTGACCCTAGGCAACCTACACAC/

Eco?l5l Fok l

23LP:

-1C; -3T +4G

5'PTTTTCTGCTGGGGGCATCCCTGAACTGGGATCCGTTGGATGTGT

3'AAAAGACGACCCCCGTAGGGACTTGACCCTAGGCAACCTACACA

Loop linker-2 :

5'PCTTTTTGAGACCGACATCCCTCTGCAGACGATCCATCAGAGTCAG

3 'GAAAAACTCTGGC TGTAGGGAGACGTCTGCTAGGTAGTCTCAGTC

Fok l

Each (10OmM) phosphate linker listed above was denatured at 95 ⁰ C in a PCR thermocycler for 1 min and then self annealed to a loop by cooling down to room temperature. After gel purification, diluted each looped linker with 1 X TE buffer to 5OmM and stored at -70 ⁰ C until use.

(B) PCR primers (the following PCR primers synthesized by Invitrogen)

PCR-I primer:

BHl : 5' ACACATCCA ACGGATCCCAGTTCAG 3' PCR-2 primers:

BHl : 5' ACACATCCA ACGGATCCCAGTTCAG 3'

LG: 5 ' GACTCTGATGGATCGTCTGCAGAG 3 ' (C) PCR quality control primers: 5' U6: 5' AAGGTCGGGCAGGAAGAGGGC 3' 3' Hl :5' TATTTGCATGTCGCTATGTGTTCT 3' Diluted each one with 1 X TE buffer to 10 mM and stored at -7O ⁰ C until use.

EXAMPLE 2: DNA Partial digestion by DNaseI

(1) Starting DNA: SARS coronavirus membrane protein (NCBI accession No.: AY536759) full-length cDNA (666bp; obtained from an academic source) as show in FIG 3.

(2) OneQjgof SARS coronavirus membrane protein full-length cDNA is partially digested into 100-300bp blunt-end fragments by DNaseI (Roche Biosystem; 0.01-0.03U) in a Mn ²⁺ buffer at a final concentration of 1OmM Tris-HCl (pH9.0); 2 mM MgSO ₄ ; 1OmM KCl; 8mM (NH4)) ₂ SO ₄ and 1 mM MnCl ₂ . The reaction is performed on ice for 1 min following by heating inactivation for 20 min at 70 °C. The resulting products were analyzed on 1% Agarose gel (FI G.4).

EXAMPLE 3: Loop-1 Linkers Ligation and PCR-I Amplification Loop-1 Linkers Ligation

The partially digested products were ligated to Loop-1 phosphate linker(s) (19-; 20-; 21-; 22- and 23-bp Loop-1 linker) respectively. 5μl of ligation reaction mix contains: 2.5 μl partially digested DNA; 0.5 μl lOxligation reaction buffer (500 mM Tris-HCl (pH 7.5, 25 ⁰ C), 100 mM MgCl ₂ , 100 mM DTT, 25 μg/μl BSA); 1 μl (10 mM) Loop-1 linker; 0.5 μl 10 mM ATP, 0.5 μl T4 DNA Ligase (NEB). The reaction was performed at 16 ⁰ C overnight. PCR-I Amplification

The reaction mixture (50μl) contains: 0.5 μl each Loop-1 ligated product; 2μl single primer BHl (20 μM), 1 μl dNTP (1OmM), 5 μl 1OX PCR buffer (200 mM Tris-HCl (pH 8.4), 500 mM KCl, 15 mM MgCl ₂ ), 0.5 μl Taq DNA polymerase and 41 μl PCR H ₂ O. The thermal cycling is as follows:

preheating at 95 ⁰ C for 1 min; 28 cycles of 95 ⁰ C for 15s; 68 °C for lmin. Three μl of each PCR products were analyzed on 1% Agarose gel (FIG.5).

EXAMPLE 4: Eco¥15l Cleavage. Blunt-ending and PAGE Purification ϋcøP15I Cleavage

Mix 5 ul of each PCR-I products generated by five (19-; 20-; 21-; 22- and 23 bp) Loop-1 linkers. An EcoP15I digestion solution (lOOμl) contains: 10 μl PCR-I product mixture, 10 μl ATP (10 mM), lOul 10xNEBuffer-3, 1 μl BSA (10Ox; 10mg/ml), 10 μl (100U) Eco?l5ϊ (NEB), 59 μl D-H ₂ O. Place in a 37 ⁰ C water bath for an overnight incubation. Blunt-ending

After phenol-chloroform extraction and ethanol precipitation, dissolve the pellet with 11 μl D-H ₂ O, add 1.5μl (4.5U) T4 DNA polymerase (NEB), 1.5 μl NEB Buffer-2 and 2 μl dNTP (ImM) to a total 15 μl reaction volume. Incubate at 37 "C for 15 min in a PCR thermal cycler. Inactivate the reaction by heating at 68 °C for 20 min. PAGE Purification

Add 1.5 μl DNA Loading Buffer (30 mM EDTA; 36% (V/V) glycerol; 0.05% (WfV) BPB, pH=7.0) into the 15 μl reaction. Load 6.5 μl/well for 20% TBE polyacrylamide gel Electrophoresis (at 200V for 90 min). As shown in FIG.6, the specific 66bp DNA fragments can be observed after 10% EtBr staining (20min) under a UV lamp. Cut the gel pieces containing 66bp fragments, put together in a fresh tube containing 150 μl gel diffusion buffer (0.5M NH ₄ AC, 10 mM Mg(Ac) ₂ , 1 mM EDTA; pH 8.0). Incubate the tube at 55 ⁰ C for overnight. Extract DNA with QIAEX II Gel Extraction Kit and elute DNA with D-H ₂ O.

EXAMPLE 5: Loop-2 Linker Ligation and PCR-2 Amplification Loop-2 Linker Ligation

Five μl ligation reaction contains: 2.5 μl DNA isolated by PAGE, 0.5 μl lOxligation reaction buffer (500 mM Tris-HCl (pH 7.5, 25 ⁰ C), 100 mM MgCl ₂ , 100 mM DTT, 25 μg/μl BSA), 1 μl (20 mM) Loop-2 linker, 0.5 μl ATP (1OmM), 0.5 μl T4 DNA Ligase (NEB) and 0.5 μl D-H ₂ O. Incubate the reaction at 16 ⁰ C overnight in a PCR thermal cycler. PCR-2 Amplification

Dilute 5 μl ligation-2 product with 45 μl D-H ₂ O and take out 0.5μl as a template for PCR-2 amplification. A 50 μl reaction contains: 0.5μl diluted ligation-2 template, 5μl 1OX PCR buffer (200 mM Tris-HCl (pH 8.4), 500 mM KCl, 15 niM MgCl ₂ ), 1 μl LG primer (lOμM), 1 μl BHl(I OμM), l μl dNTP (1OmM), 0.5 μl Taq DNA polymerase, 41 μl PCR H ₂ O. The thermal cycling is as follows: preheating at 95 ⁰ C for lmin; 28 cycles of 95 ⁰ C for 15s; 68 ⁰ C for 1 min. After amplification, 10 μl PCR product was analyzed on a 20%TBE polyacrylamide gel. As shown in FIG.7, a specific 109 bp amplicon can be observed after 10% EtBr staining (20min) under a UV lamp. After phenol: chloroform extraction, the ethanol precipitation was performed at -20 °C at least for 2 hours. Pellet DNA and dissolve in 20 μl D-H ₂ O.

EXAMPLE 6: Fokl Digestion

A 100 μl reaction contains: 20 μl above PCR-2 product, 2 μl Fokl (8U; NEB), 10 μl NEBuffer-3 and 72 μl D-H ₂ O. Incubate the reaction at 37 ⁰ C for 2 hrs. After digestion, analyze 10 μl Fokl digested DNA on a 20% TBE polyacrylamide gel. As shown in FIG.8, the target bands are distributed from 29 to 33bp in length (as an arrow indicated), which is corresponding to siRNA in 19-23 bp plus lObp cloning sites ("AAAAG" at both 5' ends). Three other 3 bands from top to bottom are: an undigested 109bp band, a partial digested band (siRNA with a loop at one side), a Loop-1 and Loop-2 overlapping band (owing to Loop-1 and Loop-2 overlap each other, the signal is strong). PAGE purification for the DNA fragments of interest is as described in EXAMPLE 4.

EXAMPLE 7: Vector Ligation and Completion of siRNA Library Ligation into an siRNA Expression Vector

As shown in FIG.10, DNA fragments after Fokl digestion were ligated into an siRNA expression vector: pU6Hl-GFP (NT Omics, USA) with double promoters of U6 and Hl (FIG.2).

A 15μl ligation mixture contains: 1 μl (lOOng) pU6Hl-GFP, 2 μl Fokl digested DNA, 10 μl 1.5xPlasmid Ligation Buffer (90 mM Tris (PH8.5-9.0); 12 mM DTT; 6OmM MgCl ₂ ; 40% PEG8000), 0.5 μl ATP (1OmM), 0.25 μl T4 DNA Ligase (NEB), 2.25 μl D-H ₂ O. The reaction was incubated at 16°C for 30 min. Add 85 ul D-H ₂ O to the ligation mixture and proceed to an ethanol precipitation (-7O ⁰ C , 30min). Pellet DNA and dissolve in 5 μl D-H ₂ O. Electrotransformation

Thaw electro-competent cells, MegaX DH10B™ (Invitrogen) on ice. Mix 5μl ligation product in 40 μl electro-competent cells and stand it on ice for lmin. Transfer the mixture into a cuvette (BioRad) to perform electro transformation in a MicroPulser (BioRad). After transformation, transfer to a fresh tube containing 150 μl SOC medium (2% (W/V) ) Tryptone, 0.5% (W/V) Yeast, 0.05% (W/V) NaCl, 2.5 mM KCl, 1OmM MgCl ₂ , 2OmM glucose, pH=7.0). Incubation was performed at 37°C with a continuous shaking for 40~60 min at 220 rpm, followed by plating the bacteria culture on a solid medium (1% (W/V) Tryptone, 0.5% (W/V) NaCl, 1.5% (W/V) Agar, pH=7.0; 50μg/ml Kanamycine) and incubation at 37°C in an air incubator overnight. One ligation reached IX lO ³ independent clones (colonies).

EXAMPLE 8: siRNA Library Validation Sample Inoculation

Randomly pick up the well-isolated colonies and inoculate separately into each 1.5 ml eppendorf tube (or 48-well plate), each containing 400ul LB medium (1% (W/V) Tryptone, 0.5% (W/V) Yeast, 1% (W/V) NaCl, pH=7.0). Incubation was performed at 37 ⁰ C with a continuous shaking for 2 hours at 220rpm. PCR-based Insert Screening

The reaction mixture (30 μl) contains: 2μl above bacteria culture, 0.5 μl 5' U6 primer (10 μM), 0.5 μl 5' Hl primer (10 μM), 0.5μl dNTP (10 mM), 3μl 1OX PCR buffer (200 mM Tris-HCl (pH 8.4), 500 mM KCl, 15 mM MgCl ₂ ), 0.3 μl Taq DNA polymerase and 23.2μl PCR H ₂ O. The thermal cycling is as follows: preheating at 95°C for 5min; 25 cycles of 95°C for 15s; 62 ⁰ C for 30s; 72°C for 40s; holding at 72 °C for 7min for last extension. Five μl of each PCR products were analyzed on 1% Agarose gel. As shown in FIG. 9, the positive bands of PCR products are ~400bp in length, the omitted bands are PCR negative due to failures in colony picking-up or bacteria inoculating. The positive rate for sampling is 91.6% (44/48). Sfil Digestion of PCR Products

The siRNA is only distributed 19-23bp in length. By 400bp PCR products only, it is very difficult to distinguish siRNA positive clones from the negative ones (just a vacant vector without siRNA insert).

In order to further validate siRNA positive clones, Sfil restriction enzyme was added to PCR

products. The Sfil restriction site is originally existed in the MCS (multiple cloning sites) region of pU6/Hl-GFP vector used. A positive digestion indicates siRNA insert negative (Gust a vacant vector).

Sfil digestion mixture contains: 4μl above PCR products, 0.25 μl (5U) Sfil (NEB), 0.75 μl NEBuffer-2. Digestion was incubated at 50°C for 1.5hr in a PCR thermal cycler. Five μl Sfil digestion mixture was analyzed on 1% Agarose gel. As shown in FIG.10, out of the 43 testing samples, only 1 can be digested by Sfil, indicating that the recombination rate of the experimental SARS siRNA library is 97.6% (43/44). Plasmid Mini Preparation and Sequencing

Inoculate lOOμl bacteria culture with PCR/S/zI tested positive into a culture tube containing 4 ml LB medium (1% (W/V) Tryptone, 0.5% (W/V) Yeast, 1% (W/V) NaCl, pH=7.0; 50 μl/ml Kanamycine). An overnight incubation was performed at 37 °C with a continuous shaking for 2 hours at 220 rpm. Take 3.2 ml of it for mini sale plasmid extraction with TIAN Prep Mini plasmid Kit ((TIANGEN), store the remaining in 20% glycerol at -20 °C for further use. One μl (150-200 ng) purified plasmid of each sample was analyzed on 1% Agarose gel. FIG.ll demonstrates a group of sample check before sequencing reaction (the expect size of the plasmid is 5.5kb).

The plasmids were proceeded for sequencing (Shanghai Invitrogen). Align the obtained sequences with SARS full-length cDNA sequence by an alignment program software. The results were listed in TABLE I. The length of siRNA clones is distributed from 19 to 23bp with a random binding sites distribution. The results confirm that the experimental SRAR siRNA library is representative both in length and binding sites.

TABLE I. Sequencing Results for Clones Randomly Selected from the Experimental SARS siRNA Library

(* Repeated clones)