Outline of the monoclonal stoichiometric (MOSIC) ODN production method. a A schematic representation of an example pseudogene for MOSIC ODN production. Three different sequences (green, magenta and blue) are hard-coded in a 3:3:1 stoichiometric ratio directly on the pseudogene. The exact order of the sequences is not important and can be shuffled like in this example to facilitate gene synthesis. b,c The single stranded form of the pseudogene can be amplified in two different ways: b by in vitro rolling circle amplification (RCA) or like in c, by helper phage rescue. In the in vitro variant (b), the pseudogene is cut out with restriction enzymes, re-ligated into smaller circles that are subsequently nicked to form a template for RCA using a Phi29 polymerase. In the phagemid protocol (c) the entire vector, including pseudogene, is packaged in phage particles in single-stranded form and extracted. In both cases, the final digestion product contains the monoclonal ODN in the desired stoichiometry. d Details of the hairpins used for MOSIC ODN production. A BtsCI restriction site is flanked on one side by a GAA loop and a GCGC stem, and on the cutting side by two hybridized base pairs. These base pairs are formed by encoding two complementary bases to the last bases at the 3’-end of ODN 1 into the hairpin sequence. Note that these two bases are encoded in the hairpin by-product, and consequently that the sequence of the resulting ODNs do not depend on the neighboring ODNs.

Digestion of the amplified crystal pseudogene releases the correct length ODNs. Denaturing polyacrylamide gel images stained with Sybr Gold. a In vitro amplification (uncropped gel in Supplementary Fig. S5): 1 10bp ladder, nominal sizes in nucleotides indicated to the left. 2 RCA product after in vitro amplification. 3 The same sample as in lane 2 after digestion with BseGI. b: 4 BseGI digested ssDNA amplified in E. coli in a phagemid vector. The pseudogene has not been separated from the vector prior to digestion and high molecular weight bands attributed to vector DNA can be seen. 5 10bp ladder, nominal sizes in nucleotides indicated to the right.

Comparison of synthetic, commercial DNA with MOSIC oligonucleotides for crystallization. The commercial sample was a mixture of the 21-mers and 14-mers (no hairpins included) in the proper relative stoichiometry. a HPLC analysis of synthetic, HPLC-purified DNA as-provided by a commercial vendor (dotted, black line) and MOSIC ODNs (red line). Two peaks corresponding to the 14- (left peak) and 21- mers (right peak) are visible, however the hairpin sequences, being double-stranded, show very weak retention in the column and are not visible here. b Denaturing PAGE of HPLC-purified, synthetic oligonucleotides as-provided by a large commercial vendor (lane 2) and MOSIC ODNs (lane 3). Lane 1: ladder (Gene Ruler Ultra Low, Fermentas). Nominal sizes in number of nucleotides are shown to the left.

A DNA crystal produced with MOSIC ODNs. a The entire, unpurified, pool of digestion products from an in vitro MOSIC preparation (same as lane 3) was folded by cooling over the course of one week to produce macroscopic crystals of about 70 to 150 µm in diameter. b Final X-ray model of MOSIC produced oligonucleotides where the 3:3:1 stoichiometry is highlighted by individual coloring of the 14:21:21-mer in magenta:green:cyan respectively. c Same structure as in b with the 4.79 Å Sigma-A weighted 2fo-fc map overlay, contoured at 1.5 sigma. d Comparison of the asymmetric unit of the MOSIC-produced pdb-id:4B8D (green carbons) model to the template model pdb-id: 3GBI (magenta carbons) used for molecular replacement.

DNA origami using MOSIC ODNs. a Denaturing polyacrylamide gel (20%) of digestion of 10 helix bundle staple sets with BseGI after amplification by helper phage rescue. 1 20bp ladder, nominal sizes in nucleotides indicated to the left. 2 to 5 BseGI digestions of phagemid pseudogene DNA encoding staple sets 1 to 4. b DNA after gel purification 6 20bp ladder, nominal sizes in nucleotides indicated to the left. 7 Gel purified staple set. 8 Synthetic staple set 1. c TEM image of 10 helix bundle folded with all the synthetic staple strands. d TEM image of 10 helix bundle lacking 72 staple strands. e TEM images of 10 helix bundle after addition of the lacking 72 staple strands produced by the MOSIC method. All scale bars 100 nm.

Applying the MOSIC method for production of very long DNA oligonucleotides. a Denaturing 5% polyacrylamide gel showing the digestion product (lane 2) of a pseudogene that releases a 378 nt MOSIC ODN, red arrow. Lane 1 is a 100 bp ladder with the nucleotide count indicated to the left. b Ultra-small origami folding: The 378 nt strands were used as scaffold and folded into 17 helices forming a 23 bp long, ultra-small, DNA origami brick that polymerizes into an extended 17 helix bundle. c TEM micrographs of the polymerized ultra-small DNA origamis, and a CanDo structure prediction (to scale) of 25 monomers. One monomer is highlighted in blue. The rightmost micrograph in cshows850 nm of tube corresponding to over 100 polymerized monomers. d–e Similar designs where the 378 nt MOSIC ODN was folded into 6 helix bundles or 9 helix bundles and polymerized. All scale bars are 20 nm.