Compositions & methods for unlinked architect oligo-mediated DNA synthesis

Abstract

Reductions in costs to read and write DNA have driven advances in our ability to manipulate and engineer biology. However, writing DNA remains a limiting step in testing new hypotheses and engineering novel biological properties. Current synthesis methods lack the speed, low cost, and accessibility to match our current experimental outputs, such as high throughput screening and next generation sequencing. Ideally, DNA synthesis should become as ubiquitous and affordable as PCR. Such a step-change in affordability and accessibility would therefore lead to dramatic improvements in our ability to study and engineer biology. Towards this goal, we have developed a next generation DNA synthesis technology that relies on sequence-filtered, multiplexed ligations to enable template-independent, exponential synthesis of gene- or genome-length DNA. This approach is amenable to cost-effective automation and thus will enable cost-effective DNA "printers" that are as affordable as PCR machines.

Description

COMPOSITIONS & METHODS FOR UNLINKED ARCHITECT OLIGO-MEDIATEDDNA SYNTHESISSTATEMENT AS TO FEDERALLY SPONSORED RESEARCH

[0001] This invention was made with Government support under Federal Grant No. 5R21- HG011675-02 awarded by NIH / NHGRI. The Government has certain rights to this invention.REFERENCE TO A SEQUENCE LISTING

[0002] The instant application contains a Sequence Listing which has been filed electronically in ASCII format as 47381-69. xml created on June 1, 2023 and is 37,802 bytes in size and is hereby incorporated by reference in its entirety.FIELD OF INVENTION

[0003] This invention relates to methods of DNA synthesis. Using a limited set of “architect oligos”, DNA synthesis is accomplished by iterative rounds of oligonucleotide directed extension, ligation and cleavage. These cycles can be multiplexed for efficient synthesis.BACKGROUND

[0004] According to BCC Research, the current synthetic biology market will soon exceed $18 Billion USD annually. (Bergin 2019b) This growth is in large part driven by key advances in technologies to both read and write DNA. The market for DNA or gene synthesis products alone is expected to exceed $7 Billion USD by 2024. (Bergin 2019a) The cost of synthesis has lagged significantly behind the reductions seen in the cost of DNA sequencing and on a per base pair level synthesis is still 5 orders of magnitude higher than that of DNA sequencing. At current best prices for DNA synthesis (of ~$O.O5-$O.15 / bp) the synthesis of a relatively simple bacterial genomes, such as E. coli (~ 5Mbp) would still cost ~ $350,000, which is intractable for routine experimentation. (Kosuri and Church 2014) Additionally, the lowest reported costs per base pair are often not realized in practice. From recent purchases, the cost of a 4kbp “gene” ranges anywhere from $675.00 (~ $0.14 / bp) for a sequence verified clone to $575.00 for linear DNA fragments which need to be assembled and cloned. This corresponds to over $0.16 / bp. In addition to the fact that the actual costs for longer sequences are higher than the lowest price points, many additional “difficult” to manufacture sequences cannot be obtained from DNA synthesis providers. For example GC or AT rich sequences as well as sequences with repetitive elements need to be cloned with more traditional methodology. For the field of synthetic biology to realize its truepotential, the cost of writing DNA needs to be reduced by 100- to 1000-fold to make routine DNA synthesis (of even large or difficult sequences) a feasible tool for routine systematic experimentation even in academic labs. (Carr and Church 2009) Ideally, to be game changing, DNA synthesis technologies should be as simple and as affordable as PCR.SUMMARY

[0005] The Summary is provided to introduce a selection of concepts that are further described below in the Detailed Description. This Summary is not intended to identify key or essential features of the claimed subject matter, nor is it intended to be used as an aid in limiting the scope of the claimed subject matter.

[0006] Other methods, features and / or advantages is, or will become, apparent upon examination of the following FIGs and detailed description. It is intended that all such additional methods, features, and advantages be included within this description and are protected by the accompanying claims.BRIEF DESCRIPTION OF DRAWINGS

[0007] The novel features of the invention are set forth with particularity in the claims. A better understanding of the features and advantages of the present invention will be obtained by reference to the following detailed description that sets forth illustrative aspects, in which the principles of the invention are used, and the accompanying drawings of which:

[0008] Figure 1A-1B: Overview of Architect-Directed DNA Synthesis. 1A) Architect- directed DNA Synthesis involves 4 repeating steps: Extend, Ligate, Amplify, and Cleave, i) The first cycle starts with Donor and Architect oligos. Donor oligos encode an Architect sequence at their 3’ end (red or blue) and a “donor” sequence at their 5’ end that becomes part of the final synthesized sequence (grey). The 5’ end of the Donor oligo is phosphorylated. There is also a Cast 2a PAM site encoded in the Architect sequence or an additional Cut Site sequence encoding a Type IIS target sequence such that the DNA can be cut between the Architect and Donor sequences (not shown). The Architect oligos in this case contain a 5’ hairpin structure and encode a unique Architect that is complementary to the Architect sequence of the Donor oligo, ii) These oligos anneal in a sequence-specific manner and then (iii) a polymerase is used to “Extend” the double-stranded DNA (dsDNA), copying the “donated” sequence, (iv) These double-stranded molecules are then ligated together and (v) amplified in a PCR reaction using primers that bind to the Architect sequences, (vi) Following amplification, the product is cleavedsuch that either the blue or red Architect sequence is removed, creating a Donor for the next cycle. The cleaved reaction is then mixed with a (vii) parallel reaction to (viii) begin the next cycle. IB) Each cycle of Architect-directed DNA Synthesis doubles the length of DNA created. If starting with 2bp of “donated” DNA in Donor oligos, only 12 cycles of ADS would be required to synthesize 8,192bp of DNA.

[0009] Figure 2A-2C: Architect sequences enable directional control of ligations.2A) Donor oligos consist of an Architect sequence (1 : red and 2: blue) on the 3’ end and donor sequences (1 : dark grey and 2: light grey) on the 5’ end along with a 5’ phosphate. Architect oligos have hairpin structures on their 5’ ends (grey) and Architect sequences on their 3’ ends (1 : red and 2: blue) that are complementary to the Donor oligo Architect sequences. 2B) First, the Architect and Donor oligos anneal to each other and a polymerase is used to fill in the overhang that is created. The double-stranded products are ligated together, creating multiple ligation products including ligations between the same sequences. PCR amplification is performed using Q5 polymerase with primers that bind to the Architect sequences to select the products with Architect 1 and 2 on either end. 2C) As shown, amplification products were sequenced and all products consisted of either (i) 100% sequence correct product (Seq Correct) or (ii) products that identifiably contain the correct Architect 1 and 2 on either end with some portion of Donor sequence 1 and 2 in between (Struct Correct) or (iii) sequences that did not easily fall into one of these classifications due to too many errors or more than 2 Architect sequences. (Other). Importantly, none of these sequences showed self-ligation of, for example, Architect 1 and Donor 1 to Architect 1 and Donor 1.

[0010] Figure 3A-3E: Optimizing Architect Directed Synthesis (ADS) to prevent carryover of products between cycles. 3A) In this example of ADS, Architect 1 and 2 are used in both first cycle reactions. After the ligation and amplification in the first cycle, Architect 2 and Architect 1 are then cleaved from each reaction with Bsal and SapI, respectively, and the products’ overhangs are filled in and then the blunted products are ligated together and amplified with Architect 1 and 2. 3B) In this example of ADS, Architect 1 and 2 and Architect 3 and 4 are used in the first cycle reactions, respectively. Architect 2 and 3 are cleaved in their respective reactions (using Bsal and SapI, respectively), the overhangs are filled in with a polymerase, and the products are ligated together. This ligation product is amplified using primers that bind to Architect 1 and 4. Architect oligos can either (3C) contain no 5’ modifications or (3D) containhairpins at their 5’ ends, which help to prevent extraneous ligations. 3E) 2 cycles of ADS were performed using the conditions described in (a) and (c), (a) and (d), (b) and (c), or (b) and (d). PCR amplification of the second cycle was performed with primers the bind the expected Architect sequences. Switching Architect sequences and using 5’ hairpins on the Architect oligos was shown to produce the expected Cycle 2 product.

[0011] Figure 4A-4C: Demonstration of 3 cycles of Architect-directed DNA Synthesis. 4A) Using Donor oligos containing 30bp of “donated” DNA (Donors 1-8) and 6 unique Architect sequences (Architects 1-6), 3 Cycles of ADS were performed as illustrated. Architect oligos contained 5’ hairpin sequences to block unwanted ligations. Cleavage steps were performed using Casl2a with crRNA specific to each Architect sequence. The final sequence should contain 240bp of donated DNA (Donors 1-8) flanked by Architects 1 and 6. 4B) Gel image showing products of each reaction across all three cycles. 4C) Long read sequencing of Cycle 3 product (n=l) showed that over half the sequencing reads showed “structurally correct” sequences. In this case, unoptimized Casl2a wobble cleavage prevents sequence correct products from being formed.

[0012] Figure 5A-5B: An example of a minimal set of donor oligos could be used to make any sequence. Using six Architect sequences to enable Architect switching every 3 cycles, we show the complete set of 5A) Donor oligos and 5B) Architect oligos that would be needed to synthesize any sequence. These oligos could be purchased in bulk using lower throughput but high fidelity methods.

[0013] Figure 6A-6H: Examples of different types of starting Donor and Architect oligos and their impact on the first cycle of ADS. In all examples, Donor oligos contain a 5’ phosphate to facilitate ligation. 6A) In this example, Donor and Architect oligos are both single stranded with no modifications and 6B) the first cycle until the amplification step would consist of an Annealing step, an Extend step, a Ligate step, and then the Amplification step. 6C) Similar to (a) but with a 5’ hairpin on the Architect oligo to prevent unwanted ligations. 6D) The first cycle steps are the same as in (b). 6E) Oligos can be designed as a single sequence encoding both strands of the Architect and Donor and connected by a hairpin at one end. 6F) In this example, the Anneal and Extend steps would be skipped. 6G) A clonal population can be used as a template for generating starting material using PCR with primers that bind the Architect sequences and contain 5’ hairpins. 6H) In this example, the first cycle would start with acleavage step that would cleave one side or the other depending which Architect was required. After the cleavage step, the cycle would proceed with an Extend step, Ligation step, and then the Amplification step.

[0014] Figure 7A-7B: Altering Casl2a “wobble” cleavage with phosphorothioate(PS) bonds. Using LbCasl2a and a 19bp crRNA, target sites were cleaved that contained 7A) no modifications or 7B) contained phosphorothioate (PS) bonds connecting the backbones of nucleotides 13-19 from the PAM (shown in red with asterisks indicating the PS bonds between nucleotides). PAM sites are shown in orange. Shown are the cleavage events at each site along the target site as a percentage of all cleavage events (n=l for each condition).

[0015] Figure 8A-8C: Incorporating error correction into ADS enables sustained levels of fidelity. 8A) In traditional linear synthesis, very high levels of cycle efficiency (a.k.a. Coupling efficiency) are required. The percent of correct sequences can be calculated with the formula: (cycle efficiency)leng,h°fDNA~1. In this case, 95% is a relatively low coupling efficiency and by the time a 60bp oligo is synthesized, it is expected that only -5.1% of molecules are the correct sequence (0.9560~]* 100 = -5.1%). 8B) In ADS, the calculation for error propagation is represented as (xi * xt) * Y - Xi+i where z is the cycle number, x is the starting fraction correct in a given cycle, and Y is the cycle efficiency. Functionally, this means that to make the same length DNA with the same cycle efficiency, the same percent of sequences would be correct for ADS or traditional linear synthesis. However, ADS enables repeated application of error correction which is calculated as: [(xt * xt) * YJ + z * (l-[(xt * xt) * Y]) = x,+i, where z is the fraction of errors corrected with each application of error correction. 8C) Using this formula, even limited amounts of error correction each cycle (z) enable sustained fidelity at a certain threshold.DETAILED DESCRIPTION

[0016] Toward this goal, this invention is a next generation DNA synthesis technology. The process, illustrated in Fig. 1, has the potential to overcome many of the challenges associated with current methods of DNA synthesis and as a result also has the potential to enable extremely low costs for DNA synthesis and assembly. Traditional methodologies all still rely on the synthesis of oligonucleotides and the use of DNAs double stranded nature and enzymes to build larger dsDNA fragments. (Kosuri and Church 2014) While the cost of oligonucleotide synthesis has dropped significantly, a key limitation is the ability to assemble oligonucleotides into larger genes. (Carr and Church 2009) Assembly has mostly utilized double-stranded DNA ligases, (Auet al. 1998) polymerase cycling assembly (PCA), (Stemmer et al. 1995) or variations therein. These methods are limited by the inherent error rates in oligo synthesis (coupling efficiencies) and high complexity present in multiplexed-gene synthesis (providing an upper limit to the cost savings of oligo pool gene assemblies). (Kosuri and Church 2014) Thus the cost of larger DNA fragments hasn’t decreased with the drop in cost of oligonucleotide synthesis and sequencing. (Carr and Church 2009) This approach overcomes many of these challenges and enables template independent, exponential DNA synthesis, for example 2 bp to 4 bp to 8bp to 16 bp, etc,. One non-limiting example of this process is illustrated in Fig 1., Exponential growth enables the potential to synthesize DNA fragments of up to 10 kilobases in less than 14 cycles (which theoretically, can be achieved in an overnight “PCR type” reaction and is less cycles than a 15 bp oligo), reducing cycle number and compounding errors associated with oligo building technologies.I, Definitions

[0017] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention pertains. In case of conflict, the present specification, including definitions, will control.

[0018] Unless otherwise specified, “a,” “an,” “the,” “one or more of,” and “at least one” are used interchangeably. The singular forms “a”, “an,” and “the” are inclusive of their plural forms.

[0019] The recitations of numerical ranges by endpoints include all numbers subsumed within that range (e.g., 0.5 to 5 includes 1, 1.5, 2, 2.75, 3, 3.80, 4, 5, etc.).

[0020] The term “about,” when referring to a value or to an amount of mass, weight, time, volume, concentration, or percentage is meant to encompass variations of ±1% from the specified amount. The terms “comprising” and “including” are intended to be equivalent and open-ended. The phrase “consisting essentially of’ means that the composition or method may include additional ingredients and / or steps, but only if the additional ingredients and / or steps do not materially alter the basic and novel characteristics of the claimed composition or method. The phrase “selected from the group consisting of’ is meant to include mixtures of the listed group.

[0021] Moreover, the present disclosure also contemplates that in some aspects, any feature or combination of features set forth herein can be excluded or omitted. To illustrate, if the specification states that a complex comprises components A, B and C, it is specifically intendedthat any of A, B or C, or a combination thereof, can be omitted and disclaimed singularly or in any combination.

[0022] The term “heterologous DNA,” “heterologous nucleic acid sequence,” and the like as used herein refers to a nucleic acid sequence wherein at least one of the following is true: (a) the sequence of nucleic acids is foreign to (i.e., not naturally found in) a given host microorganism; (b) the sequence may be naturally found in a given host microorganism, but in an unnatural (e.g., greater than expected) amount; or (c) the sequence of nucleic acids comprises two or more subsequences that are not found in the same relationship to each other in nature. For example, regarding instance (c), a heterologous nucleic acid sequence that is recombinantly produced will have two or more sequences from unrelated genes arranged to make a new functional nucleic acid, such as a nonnative promoter driving gene expression. The term “heterologous” is intended to include the term “exogenous” as the latter term is generally used in the art. With reference to the host microorganism's genome prior to the introduction of a heterologous nucleic acid sequence, the nucleic acid sequence that codes for the enzyme is heterologous (whether or not the heterologous nucleic acid sequence is introduced into that genome). As used herein, chromosomal and native and endogenous refer to genetic material of the host microorganism.

[0023] As used herein, the term “gene disruption,” or grammatical equivalents thereof (and including “to disrupt enzymatic function,” “disruption of enzymatic function,” and the like), is intended to mean a genetic modification to a microorganism that renders the encoded gene product as having a reduced polypeptide activity compared with polypeptide activity in or from a microorganism cell not so modified. The genetic modification can be, for example, deletion of the entire gene, deletion or other modification of a regulatory sequence required for transcription or translation, deletion of a portion of the gene which results in a truncated gene product (e.g., enzyme) or by any of various mutation strategies that reduces activity (including to no detectable activity level) the encoded gene product. A disruption may broadly include a deletion of all or part of the nucleic acid sequence encoding the enzyme, and also includes, but is not limited to other types of genetic modifications, e.g., introduction of stop codons, frame shift mutations, introduction or removal of portions of the gene, and introduction of a degradation signal, those genetic modifications affecting mRNA transcription levels and / or stability, and altering the promoter or repressor upstream of the gene encoding the enzyme.

[0001] The terms partially and completely complementary and partially and completely hybridize or hybrid are used to describe the interaction between any oligonucleotides, polynucleotides, subsequence, or nucleic acid fragments of any length that are at least partially complimentary. The purpose of providing complementary sequences is to obtain a double stranded sequence recognizable by an endonuclease. That is to say that the hybridization between two complementary sequences needs to be sufficient to form an endonuclease recognition site but may not need to be completely perfectly hybridized or complementary to each other. There may be gaps or partially single stranded segments within a double stranded recognition sequence, yet not impede binding and cleavage by an endonuclease.

[0002] Any contiguous nucleotide sequence of a target polynucleotide is generally formed of nucleotides from the group consisting of: A, G, T, or C. Likewise, the donor and acceptor oligonucleotides are also generally formed of nucleotides A, G, T, or C. It is appreciated though that variants or structural equivalents or mimics or non-natural nucleotides may also be used in the oligonucleotides of the invention and in the target polynucleotide that is synthesized by the methods described. For example, uracil, inosine, isoguanine, xanthine (5-(2,2 diamino pyrimidine), 8-azaguanine, 5 or 6-azauridine, 6-azacytidine, 4-hydroxypyrazolopyrimidine, allopurinol, arabinosyl cytosine, azathioprine, aminoallyl nucleotide, 5-bromouracil, any isomer of any natural or non-natural nucleotide, thiouridine, queuosine, wyosine, methyl-substituted phenyl analogs, purine or pyrimide mimics may be used.

[0024] When the genetic modification of a gene product, i.e., an enzyme, is referred to herein, including the claims, it is understood that the genetic modification is of a nucleic acid sequence, such as or including the gene, that normally encodes the stated gene product, i.e., the enzyme.

[0025] Enzymes are listed here within, with reference to a UniProt identification number, which would be well known to one skilled in the art. The UniProt database can be accessed at http: / / www.UniProt.org / . When the genetic modification of a gene product, i.e., an enzyme, is referred to herein, including the claims, it is understood that the genetic modification is of a nucleic acid sequence, such as or including the gene, that normally encodes the stated gene product, i.e., the enzyme.

[0026] Where methods and steps described herein indicate certain events occurring in certain order, those of ordinary skill in the art will recognize that the ordering of certain steps maybe modified and that such modifications are in accordance with the variations of the invention. Additionally, certain steps may be performed concurrently in a parallel process when possible, as well as performed sequentially.

[0027] The meaning of abbreviations is as follows: “C” means Celsius or degrees Celsius, as is clear from its usage, DCW means dry cell weight, “s” means second(s), “min” means minute(s), “h,” “hr,” or “hrs” means hour(s), “psi” means pounds per square inch, “nm” means nanometers, “d” means day(s), “pL” or “uL” or “ul” means microliter(s), “mb” means milliliter(s), “L” means liter(s), “mm” means millimeter(s), “nm” means nanometers, “mM” means millimolar, “pM” or “uM” means micromolar, “M” means molar, “mmol” means millimole(s), “pmol” or “uMol” means micromole(s)”, “g” means gram(s), “pg” or “ug” means microgram(s) and “ng” means nanogram(s), “PCR” means polymerase chain reaction, “kDa” means kilodaltons, “g” means the gravitation constant, “bp” means base pair(s), “kbp” means kilobase pair(s), “% w / v” means weight / volume percent, “% v / v” means volume / volume percent, “rpm” means revolutions per minute, “HPLC” means high performance liquid chromatography, “GC” means gas chromatography, and “oligo” refers to an oligonucleotide comprising a series of contiguous nucleotides of any length.EXAMPLES

[0028] For the purposes of promoting an understanding of the principles of the present disclosure, reference will now be made to preferred aspects and specific language will be used to describe the same. It will nevertheless be understood that no limitation of the scope of the disclosure is thereby intended, such alteration and further modifications of the disclosure as illustrated herein, being contemplated as would normally occur to one skilled in the art to which the disclosure relates.

[0029] Common Methods

[0030] DNA and RNA sequences: All sequences were ordered from IDT (Coralville, IA). In one embodiment, Architect-Directed DNA Synthesis (ADS) relies on Architect oligos and Donor oligos. For the examples in this patent, Architect oligos consist of a unique 25bp sequence with no modifications or with a 5’ hairpin sequence (Table 1: EM01-12). All Architect oligos are designed with a Cast 2a PAM site (“TTTV” where “V” represents A, C, or G) that starts 8bp from the 5’ end of the Architect sequence. crRNA sequences for use with Casl2a were designed to target each Architect sequence with the PAM distal portion of the crRNA binding to Cut Site 1 or2 (Table 1 : EMI 2-18). Donor oligos (Table 1 : EM20-29) consist of a 5’ phosphate, a 30 bp Donor sequence at their 5’ end that is “donated” to the final synthesized sequence, a recognition sequence for Bsal or SapI (Cut Site 1 and 2, respectively) that is orientated to cleave between the Cut Site and Donor sequence and generate 4 bp overhangs within the Donor sequence, and a sequence that is complementary to one of the Architect sequences at their 3’ end.

[0031] “Extend” reaction in first cycle: Stocks of Donor and Architect oligos were prepared at O.luM concentrations. Extend reactions were either performed with an individual Donor and Architect oligo or with the two Donor and Architect oligos that would be ligated together in the ligate step. Architect and Donor oligos were mixed together at equal volumes to make 5 uL and then mixed with Q5® High-Fidelity 2X Master Mix (NEB Catalog # M0492S). In a thermocycler, after an initial denaturation step (98C for 10 sec), five cycles of denaturing (98C for 5 sec), annealing (65C for 5 sec) and extension (68C or 72C for 10 sec) were performed, followed by an additional extension (68C or 72C for 30 sec). If using Architect oligos with 5’ hairpins, the lower temperature extension was used, otherwise the normal extension temperature was used. Following the reaction, samples were diluted by adding 190 uL of water.

[0032] “Ligate” reaction: 1 uL of “Extend” reaction was added to 9 uL of ligation mix (1 uL of T4 DNA ligase (NEB Catalog # M0202L), 1 uL of 10X T4 DNA ligase buffer, and 7 uL of nuclease-free water) and incubated at 37C for 10 minutes, followed by a 10 minute denaturing step at 65C.

[0033] “Amplify” reaction: 1 uL of “Ligate” reaction was added to 24 uL of PCR mix containing: 12.5 uL of Q5® High-Fidelity 2X Master Mix (NEB Catalog # M0492S), 2.5 uL of each 10 uM Architect oligo (with or without 5’ hairpins), and 6.5 uL of water. The samples were placed in a thermocycler. Following an initial denaturation step (98C for 30 sec), 35 cycles of denaturing (98C for 5 sec), annealing (65C for 5 sec) and extension (68C or 72C for 10 sec) were performed, followed by an additional extension (68C or 72C for 30 sec). For ADS third cycle products, 15 second extension time was used. PCR products were confirmed on 2% agarose gel.

[0034] “Cleave” reaction using Type IIS enzymes: 3 uL of “Amplify” reaction were added to 27 uL of restriction enzyme mix containing: 1 uL of either BsaI-HF®v2 (NEB catalog R3733S) or SapI (NEB catalog R0569S), 3 uL of rCutSmart™ Buffer, and 23 uL of nuclease- free water. Reactions were placed into a thermocycler and kept at 37C for 1 hour, followed by a 20 minute denaturing step at 85C.

[0035] “Cleave” reaction using LbCasl2a: LbCasl2a (NEB Catalog M0653T) was diluted to 2uM using EnGen® Lba Casl2a Diluent provided with LbCasl2a. 3.1 uL of 2 uM Casl2a was mixed with 3 uL ofNEBuffer™ r3.1, 1.5 uL of 0.5 uM crRNA, 0.75 uL of 200 mM DTT, and 20.15 uL of nuclease-free water. This was incubated at room temperature for 10 minutes before adding 1.5 uL of “Amplify” product and placing at 37C for 10 minutes, followed by a denaturation step (10 minutes at 65C).

[0036] “Extend” reaction for cycle 2 onwards: 5 uL of “Cleave” reaction products were mixed with 5 uL of Q5® High-Fidelity 2X Master Mix (NEB Catalog # M0492S) and placed at 72C for 5 minutes.

[0037] Sequencing: All sequencing was performed with Premium PCR service provided by Primordium. Sequencing primers (Table 1 : EM30-41) bind to Architect sequences and contain a 5’ barcode sequence if multiplexing within the sequencing reaction. Otherwise, a normal Architect oligo was used. Custom Python code was used to analyze the data. Briefly, Phred quality scores were converted to base-calling error probabilities and averaged for the entire sequence. Sequences with an average score above 0.03 were excluded. To identify “structurally correct” reads that may contain mutations, a 9 bp sliding window was used to identify each Architect or Donor sequence and the order in which they appear. For “sequence correct” reads, the sequences were matched 100% with the reference sequence containing the expected Architects, Cut Sites, and Donors. To identify cut sites in Figure 7, all combinations of possible cut sites were matched to the reads and identified.

[0038] Unless otherwise stated, all materials and reagents were of the highest grade possible.two molecules of DNA. In this embodiment, an Architect sequence is a 25 bp unique sequence denoted by a number (e.g. Architect 1 is a 25 bp unique sequence and Architect 2 is a different unique 25 bp sequence). In Figure la, Architect oligos composed entirely of Architect sequences bind to complementary regions on Donor oligos. The Donor oligos consist of unique Architect sequences at their 3’ end, a unique endonuclease cut site (hereafter referred to as Cut Site 1 or Cut Site 2) in the middle of the oligo, and a 5’ sequence that is “donated” to the final sequence. As shown in Figure la, one Donor oligo encodes a complementary sequence to one Architect sequence while the other Donor oligo encodes a complementary sequence to another Architect sequence. In this embodiment, Cut Site 1 is 7bp and Cut Site 2 is 8 bp and they are cleaved by Bsal and SapI, respectively. The Donor DNA used in this embodiment is 30 bp and has a 5’ phosphate group. In Figure lb, a pair of Architect oligonucleotides (referred to as Architect 1 and Architect 2) each bind to Donor oligonucleotides (referred to as Donor 1 and Donor 2), generating 5’ overhangs that are 37 and 38 bp, respectively. A polymerase is then used to fill in these 5’overhangs, such that the Donor DNA sequence is now double-stranded DNA (dsDNA). This step is hereafter referred to as the “Extend” step. A ligase is then used to join Donor 1 and Donor 2 such that the desired synthetic DNA (referred to as Donor 1-2) is created. To confirm this, Donor 1-2 is amplified by primers that bind to the Architect sequences and long-read sequencing was used to evaluate results. As shown in Figure Id, 47.1+ / -1.8% of sequences were correct with another 51.9+ / - 1.5% containing mutations or truncations from oligo synthesis but otherwise ligating correctly. This demonstrates the ability to directionally control ligations and select against self-ligation products.

[0042] Example 2: Preventing carry over of unwanted products during Architect- directed DNA Synthesis.

[0043] Referring now to Figure 3, different Architect sequences and 5’ modifications of the Architect oligos can be used to block carry over of unwanted products from cycle to cycle in ADS. In this embodiment, we demonstrate that Architect switching prevents the amplification of uncleaved products in subsequent cycles while 5’ hairpins on the oligo sequences prevent unwanted ligations. To demonstrate this, we performed 3 first cycle reactions, 2 of which used Architect 1 and 2 and the other using Architect 3 and 4, each with different “donated” sequences. We cleaved one of the Architect 1 and 2 products with the Type IIS enzyme Bsal to remove Architect 2 and cleaved the other two products with SapI to remove Architect 1 and 3, respectively. Each overhang was filled in during an Extend reaction. The Bsal cleaved product was then ligated with each of the SapI cleaved products, as illustrated in Figure 3a and 3b. Each of these reactions was also performed with and without 5’ hairpins on the Architect oligos, as illustrated in Figure 3c and 3d, respectively. As shown in the gel image in Figure 3e, Architect switching was essential to achieving the correct product, while Architects containing 5’ hairpins blocked unwanted ligations as demonstrated by the clear band in the 4th lane of the gel.

[0044] Example 3: Demonstrating 3 cycles of Architect-directed DNA Synthesis.

[0045] Referring now to Figure 4, 3 cycles of Architect-directed DNA synthesis are demonstrated. For this demonstration, Casl2a was used during cleavage steps due to higher efficiency. A well-documented issue with Casl2a is “wobble” cleavage in which it does not cleave the same location relative to the target PAM site each time but instead cleaves in several locations (Lei et al. 2017). In this example, we demonstrate the principles of ADS cycling but do not solve the “wobble” cleavage and therefore don’t expect to see “sequence correct” product but rather“structurally correct” product in which each of the expected sequences are present in the correct order (but with extra sequences in between as an effect of “wobble” cleavage). As shown in Figure 4a, Donor oligos encoding 30bp of “donated” DNA (labeled 1 through 8) in combination with different Architect sequences (labeled 1 through 6) are used as starting material. Over 3 cycles of ADS, these 8 starting Donor oligos were joined together to create first 60bp “donated” DNA, then 120bp, and finally 240bp. As shown in the gel in Figure 4b, each reaction within the three cycles produced the expected sized DNA band. We performed long-read sequencing on the Cycle 3 product and confirmed that over half of the product was structurally correct (Figure 4c), demonstrating that we can perform 3 cycles of synthesis.

[0046] Example 4: Using a minimal set of Donor and Architect oligos to synthesize any sequence.

[0047] Referring now to Figure 5, we illustrate the minimal set of Donor and Architect oligos that could be used to synthesize any sequence. We illustrate this with six unique Architect sequences and single nucleotides of “donated” DNA, resulting in a set of 30 oligos that could be used to make any sequence. We show six unique Architect sequences but different numbers could be used. Similarly, we show single nucleotide “donor” sequences (A, T, C, and G) but 2 or more nucleotides could be used to reduce the overall number of reactions required to make a particular sequence.

[0048] Example 5: Different architectures of starting DNA can be used.

[0049] Referring now to Figure 6, different architectures of starting Donor and Architect DNA can be used. In one embodiment, as shown in Figure 6a and 6b, Architect and Donor oligos are single stranded DNA with no modifications and the first cycle begins with an Extend step in which the Architect and Donor oligos anneal to each other and the overhang is filled in with a polymerase, as previously described. In another embodiment (Figure 6c and 6d), everything stays the same except the Architect oligo contains a 5’ hairpin. Another embodiment (Figure 6e and 6f), the Donor and Architect oligo are combined into a single double-stranded oligo, encoding both strands of the Architect and Donor sequences and joined by a hairpin on the side of the Architect sequence. In this example, the Extend step would be skipped in the first cycle. In another embodiment, Architect and Donor sequences can be cloned into plasmids to create clonal populations that can then be used as a template for PCR, as described in Figure 6g. In this example, the cloned DNA is shown with different Architect sequences on either side and from a single PCR,one of two starting Donor / Architect combinations can be achieved based on how the PCR product is cleaved (Figure 6h). While not shown, another embodiment of this could rely on Phi29 DNA polymerase and a protelomerase to generate “doggybone DNA” that is covalently closed, linear DNA that could encode both the Architect and Donor DNA, similar to that illustrated in Figure 6g but covalently sealed.

[0050] Example 6: Different methods can be used to ligate DNA.

[0051] As described so far, ADS relies on ligation of blunt ends. However, other embodiments of this method could rely on sticky overhangs generated by Type IIS enzymes or Casl2a cleavage for the ligations. This would skip the Extend step. Additionally, DNA that links the Architect sequences might be used to increase ligation efficiency, as described previously PCT / US2023 / 60131)

[0052] Example 7: Extend, Ligate and Cleavage steps can be multiplexed

[0053] In one embodiment of this technology, the Extend and Ligate steps could be multiplexed and PCR could be used to amplify the specific desired products using a single pair of primers targeted to the Architect sequences. Additionally, the Cleave step could be multiplexed by pooling PCR products in which the same Architect or Cut Site was being cleaved using Cast 2a and Type IIS enzymes, respectively. The effect of pooling on reducing the number of Extend, Ligate, and Cleavage reactions can be seen in Table 2.

[0054] Table 2: Comparing how Architect-directed ligations reduce the number of reactions required.

[0055] Example 8: Controlling Casl2a cleavage precision to prevent “wobble” cleavage.

[0056] In one embodiment, Cast 2a is used with sequence specific crRNA for cleavage during ADS cycles. While Type IIS enzymes can be used, Casl2a has higher sequence specificity and higher cleavage efficiency than Type IIS enzymes. However, Casl2a is well-known to induce “wobble” cleavage, in which the specific cut site within the target sequence is not a single position (eg, the 19th bp) but rather a set of cut sites in which certain positions might dominate (eg. most cleavage at the 19th bp but some at the 17th or 14th or anywhere in between). Different crRNA appear to have different “wobble” specificities and different lengths of crRNA impact “wobble” as well. This suggests it will be possible to find crRNA that are highly specific or to make crRNA more specific by incorporating modifications such as locked nucleic acids (LNAs). (Cromwell et al. 2018) Additionally, different Casl2a mutants or species-specific variants may have higher specificity. (Chen et al. 2020) However, previous work has shown that Casl2a cleavage can be mostly inhibited by incorporation of phosphorothioate bonds (PS bonds; in which a sulfur atom is substituted for a non-bridging oxygen in the backbone between two nucleotides) into the long stretches of the target and non-target strands. (Swarts and Jinek 2019) As shown in Figure 7, these PS bonds can be used to block cleavage in certain positions, while allowing cleavage in other positions. Therefore, an embodiment of ADS may include the use of strategic PS bonds in theArchitect primers used for the Amplification of dsDNA, enabling a specific cut site outside of the crRNA targeting region. Alternatively, other methods of blocking nucleolytic activity in specific locations along the Architect primers may be used.

[0057] Example 9: Incorporating error correction to enable sustained fidelity across many cycles.

[0058] In one embodiment of ADS, an error correction step can be incorporated into ADS. Standard methods of error correction work on double-stranded DNA and since ADS relies on double-stranded DNA throughout every step of synthesis, error correction can be incorporated every cycle or every n cycles, where n can be 2 or more as needed. A major advantage of the repeated application of error correction to ADS cycles is illustrated in Figure 8. Repeated application of error correction can enable sustained fidelity above a certain threshold (Figure 8c), such that DNA fidelity does not drop off with sequence length.

[0059] Example 10: Different polymerases

[0060] An embodiment wherein the polymerase used in the Extend step does not leave a terminal adenine, such as Q5 Polymerase, Q5U Polymerase, Phusion Polymerase, and Vent Polymerase.

[0061] Example 11: Different methods of cleaving DNA

[0062] An embodiment wherein Cut Site 1 and 2 are cleaved by Type IIS restriction enzymes, as shown in Example 2 where Cut Site 1 is cleaved by SapI and Cut Site 2 is cleaved by Bsal.

[0063] An embodiment wherein Cut Site 1 and 2, Architect sequences (with Cut Sites removed), or a combination of Architect and Cut Site are cleaved by a CRISPR endonuclease or mutant thereof wherein cleavage is outside of the crRNA target sequence or otherwise enables scarless ADS cycling. In this example, different crRNA would be used to cleave different Cut Site or Architect sequences. Methods discussed in Example 8 would be used to control “wobble” cleavage.

[0064] Example 12: Replacing Amplification step with solid support-based purifications.

[0065] In one embodiment, the Amplify step could be replaced with a method that selects for directionally correct ligations. For example, in an Extend and Ligate reaction, one Architect / Donor pair could be covalently linked to a biotin via the 5’ end or near to the 5’ end (eg.,in a hairpin loop) of the Architect oligo while the other Architect / Donor pair was free in solution. The biotin could then be bound by streptavidin that was covalently linked to solid supports or magnetic beads. Other forms of covalent attachment could be used as well. The ligation could then proceed in such a way that the Architect / Donor pair in solution can either ligate to each other (unwanted) or to the Architect / Donor pair to form the desired product. The unwanted products could then be washed away. The remaining desired product could then be cleaved to remove one or the other Architect sequences and either moved to mix with a new Architect / Donor pair attached to solid supports or remain in place and have another Architect / Donor pair added to the solution. This could be used in combination with, or alternating with, reactions in which Amplification steps are used.

[0066] Example 13: Final DNA can be provided ready-to-use

[0067] An embodiment wherein the final synthesized DNA sequence is amplified after the final Architect-directed ligation by using primers that bind to the Architect sequences and amplify the synthesized DNA. Subsequent cleavage of Cut Site 1 and Cut Site 2 by Type IIS enzymes (or by Cast 2a) releases the synthesized DNA with 5’ overhangs that can be filled in using a polymerase. The synthetic DNA could then be purified if needed using multiple methods known to one skilled in the arts, such as column-based purifications, bead-based approaches, or HPLC.

[0068] An embodiment wherein the final synthesized DNA sequence is amplified after the final Architect-directed ligation by using primers specific to the synthesized sequence.

[0069] An embodiment wherein the final synthesized DNA sequence includes a plasmid origin of replication and antibiotic resistance gene and is self-ligated following amplification such that it can be transformed into E. coli or other organism for further propagation or use.

[0070] Example 14: Linear and exponential synthesis

[0071] An embodiment wherein Architect-directed ligations are used to linearly generate the desired sequence of DNA by repeatedly adding Donor oligonucleotides to a growing strand of DNA.

[0072] An embodiment wherein Architect-directed ligations are used to exponentially generate the desired sequence of DNA by repeatedly combining products of Architect-directed ligations such that the length of the product DNA doubles after each round of Architect-directed ligations.

[0073] Example 15: Incorporating linear pre-made DNA as a donor

[0074] An embodiment wherein a PCR step is used to generate Donor DNA out of any existing template by incorporating a Cut Site and Architect sequence into the 5’ end of one of the primers used to amplify the target DNA. By selecting which primer to incorporate the Architect sequence with, the orientation of the subsequent Architect-directed ligation can be controlled.

[0075] Example 16: starting material could be a mixture (Mutant library / sequence)

[0076] An embodiment wherein a mixture of Donor oligonucleotides could be used to generate a targeted mutant library as the synthesized DNA product.

[0077] The complete disclosure of all patents, patent applications, and publications, and electronically available material cited herein are incorporated by reference. The foregoing detailed description and examples have been given for clarity of understanding only. No unnecessary limitations are to be understood therefrom. The invention is not limited to the exact details shown and described, for variations obvious to one skilled in the art will be included within the invention defined by the claims.

Claims

CLAIMS1. A method for synthesizing DNA consisting of at least one cycle of synthesis, wherein each cycle of synthesis comprises the following steps: a. a “ligation” reaction is used to join two sequences that each consist of: i. A unique Architect sequence that is used to direct this and subsequent ligations and amplifications and is different for each of the two sequences being joined, ii. A Donor sequence consisting of 1 or more nucleotides that becomes part of the final synthesized DNA, iii. A sequence that is either part of the Architect sequence or separate from the Architect sequence and enables cleavage between the Architect and Donor sequences; b. an “amplification” reaction that uses PCR to amplify the desired ligation product with primers that bind to the Architect sequences; c. a “cleavage” reaction that sequence specifically cleaves one of the two Architect sequences into two sequences: one sequence consists of just an Architect sequence and the other sequence consists of both of the Donor sequences and the other Architect sequence; d. an “extend” reaction that fills in any overhangs created by the “cleavage” step creating a blunt end.

2. The method of claim 1, wherein the initial “ligation” reaction joins sequences that contain 1 or more nucleotides of Donor sequence DNA.

3. The method of claim 1, wherein the cycles are repeated until the Donor sequence DNA reaches the desired length and sequence of nucleotides.

4. A method of claim 3 wherein, after a number of cycles, a final sequence is provided as linear DNA by cleaving off both Architect sequences and filling in the overhangs generated, or alternatively, the final sequence is provided as circular plasmid DNA by cleaving off both Architect sequences and ligating the ends together, wherein the donor DNA encodes an origin of replication and antibiotic marker for propagation in bacteria.

5. The method of claim 4, wherein the final step further comprises a “ligation” step between the sequence encoding the desired synthesized DNA and a single Architect sequence and a sequence encoding a plasmid backbone that encodes an origin of replication and an antibiotic marker and an Architect sequence, wherein the “ligation” step joins the Donor DNA to the plasmid DNA to result in a sequence can be ligated and transformed into bacteria.

6. The method of claim 1, wherein the first “ligation” reaction of step a is preceded by an “extend” reaction composed of two overlapping oligos that have a 5’ overhang as follows: a’ a first oligonucleotide sequence comprised of an Architect sequence at the 3’ end, a “donor” sequence at the 5’ end, a 5’ phosphate group, and a sequence that is either part of the Architect sequence or in between the Architect and Donor sequence that allows the DNA to be cleaved between the Donor sequence and the rest of the sequence b’ a second oligonucleotide sequence comprised of an Architect sequence that is the reverse complement and binds to the Architect sequence in the prior sequence, resulting in a 5’ overhang that consists of the Donor sequence and, optionally, a sequence that allows DNA to be cleaved between the Donor sequence and the rest of the sequence.

7. The method of claim 7, wherein the second oligonucleotide encodes a 5’ blocking group that prevents unwanted ligations to the single-stranded end or to the double-stranded end of the Architect (distal to the “donor” sequence) formed by the first and second oligonucleotide binding to each other.

8. The method of claim 1, wherein the primers used in the “amplification” step encode a 5’ blocking group that prevent ligations at that end of the PCR product.

9. The method of claim 8, wherein: a” the 5’ blocking group is a 5’ hairpin sequence; and b” the encoded hairpin has a sufficiently high annealing temperature that the hairpin structure is formed during extend steps in a given PCR cycle or reaction; and c” the polymerase used in the “amplification” step is not strand displacing and does not have 5’ to 3’ nuclease activity.

10. The method of claim 1, wherein the Donor and Architect sequences of step a are combined into a single double-stranded oligo, encoding both strands of the Architect and Donor sequences and joined by a hairpin in the Architect sequence.

11. The method of claim 1, wherein the Architect and Donor sequences are cloned into plasmids to create clonal populations that substitute for the sequences of step a of the method.

12. The method of claim 1, wherein the “cleavage” reaction is performed by restriction enzymes wherein each of the two sequences in the “ligation” reaction have a unique restriction enzyme cut site between the Architect and Donor sequences.

13. The method of claim 1, wherein the “cleavage” reaction is performed by Casl2a and a guide RNA that targets the Architect sequence.

14. The method of claim 1, wherein the primers in the “amplification” step include phosphorothioate bonds at the 3’ end such that cleavage by Casl2a does not happen inside the primer sequence.

15. The method of claim 1, wherein any of the “ligation”, “cleavage”, and “extend” steps are multiplexed with 2 or more sequences and unique pairs of Architect sequences such that each intended product can be amplified by specific pairs of primers that bind to the Architect sequences during the “amplification” step.

16. The method of claim 1, wherein the Architect sequences are alternated between cycles such that no pairing of Architects is used in the cycle immediately after it or in the cycle after that.

17. The method of claim 1, wherein a purification tag is added to the synthesized DNA such that the cleaved product intended for the subsequent ligation can be separated from the unwanted cleavage product and uncleaved products and the purification tag is a biotin molecule such that:a’” one of the primers in the “amplification” step is linked at the 5’ end to a biotin; and b”” after the “cleavage” step, this biotin is captured by streptavidin-coated magnetic beads and steps are taken to separate the biotin-tagged DNA from the rest of the DNA.

18. A method of claim 1, wherein enzymatic error correction is applied every n cycles, where n is 1 or more, such that a specific threshold of sequence correctness is maintained.

19. A method of claim 18, wherein the enzymatic error correction involves the use of a mismatch cleaving enzyme, such as T7 endonuclease I, to cleave mismatched sequences followed by a PCR reaction using a polymerase with 3’ to 5’ exonuclease activity to amplify correct sequences.

20. A method of claim 19, wherein the enzymatic error correction involves the use of a mismatch binding enzyme, such as mutS or similar, to capture and discard mismatched sequences.

21. The method of claim 1, wherein the polymerase used during the “extend” reaction does not leave an adenine at the 3’ end.

22. The method of claim 1, wherein the “ligation” reaction is performed using any DNA ligase capable of ligating double-stranded DNA.

23. The method of claim 22, wherein the ligase is T4 DNA ligase or a mutant therein.

Images