Luciferase-Specific Coelenterazine Analogues for Optical Contamination-Free Bioassays

Spectral overlaps among the multiple optical readouts commonly cause optical contamination in fluorescence and bioluminescence. To tackle this issue, we created five-different lineages of coelenterazine (CTZ) analogues designed to selectively illuminate a specific luciferase with unique luciferase selectivity. In the attempt, we found that CTZ analogues with ethynyl or styryl groups display dramatically biased bioluminescence to specific luciferases and pHs by modifying the functional groups at the C-2 and C-6 positions of the imidazopyradinone backbone of CTZ. The optical contamination-free feature was exemplified with the luciferase-specific CTZ analogues, which illuminated anti-estrogenic and rapamycin activities in a mixture of optical probes. This unique bioluminescence platform has great potential for specific and high throughput imaging of multiple optical readouts in bioassays without optical contamination.

The C-2 hydroxyphenyl forms hydrophobic interactions with Leu165 and Phe180 (cyan). Removal of the C-2 para-hydroxy group in the G3 analogue 6pi-OH-2H selectively activated RLuc8.  perhaps by enabling an additional hydrophobic interaction with Met174 (yellow circle). However, the C-2 group is likely to possess more conformational freedom in the product than the substrate, and this structure may not represent a biologically relevant conformation. ALucs have very low sequence identity to RLuc (ca. 17%), and therefore most likely contain variant residues that interact with the C-2 group making ALucs sensitive to C-2 group substitutions. The C-8 phenylmethyl group is buried in a hydrophobic pocket consisting of Trp156 (green), Asp162 C (not shown), Ile163, Ile166, Phe181, Val185 (salmon), Met174 and Phe180 (cyan). Chemical modifications to this group are likely to cause charge repulsion and steric hindrance. We suspect this hydrophobic pocket is highly conserved due to the abrogation of catalytic activity in all tested luciferases with the G5 analogue containing a C-8 parahydroxy group (8phenol-CTZ). Increases in steric bulk to the C-6 hydroxyphenyl that produce a kink (G2 and G3 analogues) appear to be accommodated via orientation into free space deep in the active site pocket (magenta rectangle). The rigid increase in steric bulk associated with an ethynyl moiety (G1 analogues) might cause a steric clash with the putative catalytic triad (Asp120, Glu144 and His285) (blue ellipse). It should be noted that the putative catalytic triad is not in the immediate proximity of the imadazopyrazine backbone, so it is not clear if catalysis could occur in this conformation. It is possible that the substrate inserts deeper into the active site pocket, and that the 2PSJ crystal structure captured the coelenteramide product during its exit from the pocket.

Synthesis of 3-benzyl-5-(phenylethynyl)pyrazin-2-amine (23)
3-Benzyl-5-bromopyrazine-2-amine (2) (160.00 mg, 0.60 mmol, 1 eq.) was dissolved in DMF (12.8 ml) and TEA (1.0 ml) and stirred at room temperature. Phenylacetylene (0.20 ml, 1.82 mmol, 3 eq.) was added into the reaction mixture. After vacuum deaeration, a catalytic amount of tetrakis(triphenylphosphine)palladium(0) and CuI was added into the solution and the mixture was deaerated again and stirred for 18 hours at 120 . After cooling to room temperature, the solution was filtered through a Celite pad to remove the catalysts. The solution was extracted with ethyl acetate, and the organic phase was washed with water and brine, dried over Na 2 SO 4 and evaporated. The

Synthesis of 3-benzyl-5-((trimethylsilyl)ethynyl)pyrazin-2-amine (28)
3-Benzyl-5-bromopyrazine-2-amine (2) (267.00 mg, 1.01 mmol, 1 eq.) was dissolved in DMF (12.0 ml) and TEA (1.5 ml) and stirred at room temperature. Ethynyltrimethylsilane (0.57 ml, 4.03 mmol, 4 eq.) was added into the reaction mixture. After vacuum deaeration, a catalytic amount of tetrakis(triphenylphosphine)palladium(0) and CuI was added into the solution and the mixture was deaerated again and stirred for 20.5 hours at 120 . After cooling to room temperature, the solution was filtered through a Celite pad to remove the catalysts. The solution was extracted with ethyl acetate, and the organic phase was washed with water and brine, dried over Na 2 SO 4 and evaporated. The We first determined the optical intensities of luciferases in the full pH range to see the optical profile according to pH (Suppl. Figure 6) and then focused on a narrow range from pH 7 to 10 (Suppl. Figure 7).
For the experiment of Suppl.  Figure 6).
The column-purified ALuc16 in Suppl. Figure 7 was obtained from our previous study 19 . Briefly The CTZ analogues exert distinctive luciferase selectivity according to pH ranges.
The optical intensities of ALuc23 and ALuc34 in lysates were heavily suppressed to the background intensity in an acidic pH span lower than pH 5 (Suppl. Figure 6), and in the highly basic pH range above pH 11 (data not shown). The maximal optical intensities of nCTZ and 6piOH-CTZ with ALucs were found at pH 8 and 9, respectively.
For determining the precise pH-driven feature of ALucs, ALuc16 was expressed in E.
coli and was column-purified (Suppl. Figure 7), and the pH-driven optical feature was highlighted in the pH 7-10 range. The optical intensities of ALuc16 with 6etOH-CTZ and 6H-2OH-CTZ were found to be greatly influenced by varying pH, whereas the intensities of nCTZ and CTZh were relatively stable and unaffected by varying pH. Interestingly, the optical intensity of ALuc16 with 6etOH-CTZ was suppressed almost to the background at pH 10 (ca. 3% of the maximal intensity at pH 8): i.e., at pH 10, ALuc16 selectively luminesces with nCTZ, 6H-2OH-CTZ, CTZh, but not with 6etOH-CTZ. The result indicates that the pH can act as a key ingredient for postulating a luciferase-specific assay scheme with minimized optical contamination by combining the pH environment with a specific luciferase-CTZ analogue pair.

Discussion on the pH-driven optical intensities of the CTZ analogures. It is intriguing
to interpret the dramatic pH-driven feature in the luciferase-CTZ activities (Suppl. Figure   6). It is noted that CTZ analogues bearing multiple OH groups should be more strongly influenced by pH variations than those bearing single or no OH groups. Furthermore, it should be noted that the OH group on the C-2 of CTZ analogues plays a key role for interacting with ALucs, in contrast to the OH group at the C-6 position (Suppl. Figures   7(A)). Hence, as a rule for the ALuc-CTZ activity, it was determined that CTZ analogues 27 bearing an OH group at the C-2 position (e.g. nCTZ, 6etOH-CTZ, 6H-2OH-CTZ), were more influenced by pH than other analogues that did not contain an OH group at the C-2 position, for example CTZh as shown in (Suppl. Figures 7(A)).
The optical intensity profile in Suppl. Figure 6 reveals that ALuc-CTZ activity is completely suppressed in the acidic pH range, sharply enhanced in the neutral range (pH 7), and reaches a plateau at a weak basic pH range from pH 8 to 9. In the neutral and basic pH ranges, the OH group of CTZ analogues is considered to be deprotonated, and interacting with the residues in the active site of marine luciferases, as the pKa value of phenols in nCTZ is around 10 17 . This view is supported by a previous study, where the deprotonated OH group forms a triad bond with the active site residues of luciferase, and acts as an activity center for marine luciferase 13 . Thus, the dramatic pH-driven features of ALuc-CTZ activity may be explained by the pH susceptibility of the OH groups of CTZs deprotonated at each pKa value.