Unique nucleotide sequence-guided assembly of repetitive DNA parts for synthetic biology applications

Abstract

Recombination-based DNA construction methods, such as Gibson assembly, have made it possible to easily and simultaneously assemble multiple DNA parts, and they hold promise for the development and optimization of metabolic pathways and functional genetic circuits. Over time, however, these pathways and circuits have become more complex, and the increasing need for standardization and insulation of genetic parts has resulted in sequence redundancies--for example, repeated terminator and insulator sequences--that complicate recombination-based assembly. We and others have recently developed DNA assembly methods, which we refer to collectively as unique nucleotide sequence (UNS)-guided assembly, in which individual DNA parts are flanked with UNSs to facilitate the ordered, recombination-based assembly of repetitive sequences. Here we present a detailed protocol for UNS-guided assembly that enables researchers to convert multiple DNA parts into sequenced, correctly assembled constructs, or into high-quality combinatorial libraries in only 2-3 d. If the DNA parts must be generated from scratch, an additional 2-5 d are necessary. This protocol requires no specialized equipment and can easily be implemented by a student with experience in basic cloning techniques.

Figure 1. Overview of UNS-guided assembly

In UNS-guided assembly, DNA parts are first flanked with UNSs by cloning into standardized vectors, PCR or total synthesis. In the cloning approach, each part vector (“P.V.” in figure) contains a multiple cloning site (MCS), into which the sequences of interest are cloned, as well as a series of UNSs flanked by rare restriction sites (indicated by red carets). The vectors thus assembled can be digested to yield UNS-flanked parts for assembly. Once UNS-flanked dsDNA parts are generated, they are gel-purified, assembled via Gibson isothermal assembly and transformed into a recA mutant strains of E. coli. By including multiple versions of each UNS-flanked part in the assembly, combinatorial libraries can be generated.

Figure 2. Sequence considerations when using standard part vectors

When using currently available part vectors (Table 2), restriction sites lying just outside of the UNSs are digested to generate linear parts. Shown is an example in which pFLU1U2 (Table 2) containing “Part A” is digested with AscI and MauBI. This digestion leaves behind terminal nucleotides from the restriction sites (orange). The 5′ nucleotides at the termini are digested by T5 exonuclease during isothermal assembly, but those at the 3′ ends are left behind. When Part A anneals to another part, “Part B” (bottom), the terminal nucleotides in Part A anneal with nucleotides internal to the UNSs in Part B and vice versa. Caution must be taken to ensure these terminal nucleotides do not create mismatches that might interfere with assembly.

Figure 3. Restriction site locations in part and destination vectors

(A) Each part vector contains restriction sites at which it can be cleaved to produce U_N-U_N+1 or U_N-U_X-flanked parts. We label the enzymes used to cleave at near U_N as Enz1, near U_N+1 as Enz2 and near U_X as Enz3. Each restriction site contains both a primary, rare, 8-bp restriction site (site ‘A’) and a secondary, overlapping type IIS restriction site (‘B’) to be used as a backup. (B) Destination vectors contain restriction sites (recognized by Enz4 and Enz5) that must be digested to remove the DNA segment between U₁ and U_X. As in the part vectors, each of these DNA sequences has both a primary and secondary restriction site. In (A) and (B) restriction sites are indicated by red carets.

Figure 4. Constructing individual circuits and combinatorial libraries

(A) Combinatorial assembly of a 3-part biosynthetic pathway for deoxychromoviridans (an insoluble green alkaloid pigment) by UNS-guided assembly. Each part contained one of six promoters, a triple terminator, and one of the three genes required for deoxychromoviridans biosynthesis (vioB, vioA, vioE). (i) Analytical restriction digestion of a pool of 60 clones obtained by this method. The pool was digested to isolate the destination vector backbone (black arrow) from the inserts assembled into it. Indicated are the frequent, correctly-sized (green arrow) and rare, incorrectly-sized (red arrows) inserts. (ii) Transformation of empty pDestET into TOP10 competent cells yields a lawn of unpigmented TOP 10 E. coli. (iii) Transformation of the isothermal assembly reaction (containing the vioBAE library, indicated by “Vio Lib” in the figure) into TOP10 E. coli yields colonies with variable levels of green pigmentation. (B) Construction of individual, four-part circuits for and logical computation in mammalian cells. (i) Parts were assembled into the pDestRmceBAC destination vector, which is capable of single-copy integration into appropriate mammalian cell lines. The first three parts (A, B, C1) were mixed with either the D1 or D2 part. Each mixture was assembled on its own, transformed into E. coli, and five clones (indicated by the numbers 1–5 in Figure 4bii) selected from each assembly reaction for further analysis. (ii) Analytical restriction digestion of these ten clones revealed the correct pattern for 9 out of 10. This panel is reprinted from .