Читать книгу DNA- and RNA-Based Computing Systems - Группа авторов - Страница 47

The biochemical applications of logic gates include biosensing and data processing [88]. The simplest logic gates are based on one input and correspond with YES and NOT gates. All fluorogenic (light‐up) RNA aptamers are macromolecules possessing YES logic functions since they turn ON only in the presence of input fluorophore molecules (Figure 5.2b). Examples of other common logic gates requiring two inputs implemented in binary algebra and their truth tables defining each function are summarized in Figure 5.3. The six most common binary logic gates include AND, OR, XOR, NAND, NOR, and XNOR. Among these gates, the OR logic gate yields 1 output when at least one of the inputs equals 1. The AND logic gate generates 1 output if and only if both inputs are 1. It is important to note that theoretically to form a complete set of all possible logic gates, only four basic YES, NOT, AND, and OR gates are required.

Figure 5.3 Common binary logic gate symbols and truth tables.

The RNA light‐up aptamers are the perfect system to develop devices operating in a gated manner. However, currently there only a handful of reports assessing fluorogenic aptamers from this perspective [46,89–92]. Recent work reported by Goldsworthy et al. [46] notably utilized MG‐binding RNA aptamers to design complex systems performing AND, OR, NAND, and NOR logic gate functions. The design principle is illustrated in Figure 5.4a. In AND and OR specific examples, the MG‐binding RNA aptamer sequence (shown in representative red color) was extended at each 5′‐ and 3′‐ends to have 26 nts programmed to interfere with the structure of the MG‐binding pocket. The resulting AND and OR gates with default setting 0‐0 (no inputs present) produced 0 output. Two short DNA oligonucleotides served as inputs (A and B) that hybridize with the interfering ends (shown in Figure 5.4 in green and blue colors). The AND logic achieved output 1 (ON conformation) only when two inputs were present at the same time. The OR logic function is converted into the ON state when at least one of the inputs was present. Since the RNA sequences for the AND and OR operations were the same, inputs of varying length were chosen to selectively manipulate the response from the MG‐binding RNA aptamer.

Figure 5.4 Logic gates design principle using MG‐binding RNA aptamer. (a) Representative AND and OR gates showing fluorescence response and 2D RNA structures with no inputs, presence of one or both inputs. (b) Examples of NAND and NOR gates with fluorescence readouts and 2D structures of the RNA construct in response to presence of either input and both inputs.

Source: (Panel b) Adapted from Goldsworthy et al. [46].

AND Boolean logic has been successfully applied by using split RNA aptamer systems [89–93]. The design does not utilize DNA as input to trigger conformational changes of the aptamer structure. Instead, the RNA aptamer itself is bisected into a two‐component system. This allows the split aptamer to perform the AND logic function because two halves are required to bind the dye.

The NAND and NOR gates were designed slightly differently. The extensions at the 5′‐ and 3′‐ends of the core MG‐binding aptamer region were shorter (18 nt at the 5′‐end and 17 nt at the 3′‐end). These are non‐interfering ends; thus they do not cause disruption of the MG‐binding pocket. Therefore, in the absence of both inputs, 0‐0, the output is 1 (Figure 5.4b). In NAND logic, the non‐interfering ends must be able to bind inputs A or B without changing the structure of the MG‐binding pocket. However, when both A and B are presented, the conformation of the aptamer needs to be sufficiently distorted to achieve OFF state. In NOR logic, the presence of either inputs significantly disrupts the conformation of the RNA molecule, rendering MG binding impossible. Hence, the output was “1” only in the absence of both potential DNA inputs.

Programs such as NUPAC and mfold are often used to design DNA–RNA aptamers with MG‐binding potential. Unfortunately, these programs cannot analyze stability of hybrid RNA–DNA interactions. Since the overall design relies primarily on the strand displacement reaction between RNA–RNA and RNA–DNA interactions, the folding predictions might be inaccurate. To overcome this, various concentrations of RNA gates and DNA inputs should be explored to gain better ON and OFF separation threshold.