The RIOT Responder circuit is used in conjunction with the RIOT Transponder molecule. These two components work in series to enable the generation of a bacterial gene expression output in response to the presence of extracellular target cells in the immediate vicinity. The antibody adhesion module of RIOT Transponder recognises and binds to specific biomarker receptors on the target cell while the HasA module acts as a signal activator of the RIOT Responder circuit, thus serving as a bridging transducer between the target cells and the RIOT system bacteria. The RIOT Responder circuit comprises four main parts:

1. The membrane surface receptor HasR which binds to the HasA module of RIOT Transponder and activates downstream gene expression
2. The repressor molecule HasS, which constitutively inactivates the transcription cofactor HasI and inhibits gene expression at the HAS promoter unless HasR binds to HasA
3. The sigma transcription cofactor HasI which initiates transcription at the HAS promoter after it is released from HasS repression when HasR binds to HasA
4. The HAS promoter under control of the HAS system

Overview of working mechanism of RIOT Transponder and Responder

CD44 is a cell surface receptor that is involved in cell to cell signalling and adhesion. The variant isoform CD44v6 is upregulated in many forms of cancer and contributes the metastatic effect of cancer cells. We have chosen CD44v6 as the cancer cell target biomarker of the RIOT transponder and RIOT responder circuit. The antibody module of RIOT transponder would thus be an anti-CD44v6 antibody. The anti-CD44v6 was used to probe for CD44v6 on cancer cell surfaces as seen in the immunostaining results in Fig. 1, as well as in Western Blots of various cancer cell lysates in Fig. 2.

Fig 1 provides confirmation of CD44v6 being found on cancer cells in MDA-468, thus making it a suitable biomarker target for the RIOT system. The cells were fixed with 4% paraformaldehyde (PFA) and non-specific sites blocked with 0.5% Fetal Bovine Serum (FBS). The cells were then incubated with anti-CD44v6 antibodies and subsequently with anti-mouse secondary antibodies with emission at 488 nm (green). The coverslips were then mounted onto microscope slides with mounting media and observed under the fluorescence microscope.

Total cell lysates of the respective cell lines were prepared using 0.5% Triton X100. SDS-PAGE of the samples was run on a polyacrylamide gel (4% stacking, 10% resolving) at 70V for 30 minutes and followed by 100V for 1 hour. Subsequently, semi-dry transfer was conducted at 15V for 1 hour to transfer the protein bands to a PVDF membrane. After blocking and incubation with the relevant primary and secondary antibodies, the blot was visualised with chemiluminescence using the Amersham Imager 600.

As seen in Fig 2, all cell line lysates produced a definite band at approximately 80kDa when immunoblotting was carried out with anti-CD44v6 as the primary antibody. This implies the presence of CD44v6 protein in all the assayed cell lines as the expected size is 81.5 kDa (http://www.uniprot.org/uniprot/P16070).

MDA-MB-231 lysate produced the thickest band for CD44v6 and suggests that it may have comparatively the highest proportion of CD44v6. All cell line lysates exhibited the band for GAPDH to approximately equal intensities with each other.

The signalling component of the RIOT Transponder is the hemophore HasA. When bound to a heme group, the HasA is able to act as an extracellular signal for the activation of the RIOT Responder HAS operon and subsequently initiate downstream expression. We constructed an expression plasmid (BBa_K1897001) for the production of HasA (BBa_K1897000) under control of a T7 promoter and RBS (BBa_K525998) and a double terminator (BBa_B0015). The HasA expression plasmid was transformed into E.coli BL21(DE3) cells for production of HasA.

The hasA gene was joined to the Promoter and RBS sequence BBa_K525998 and double terminator BBa_B0015 by RE digest and ligation. The plasmid was then transformed into E. coli DH5α and cultured overnight. To confirm the presence of the insert, restriction digest of the plasmid was carried out using XbaI and PstI. An expected insert size of approximately 800 bp and the linearised pSB1C3 backbone band of around 2 kbp was observed (Fig 4: band a and b respectively). Further verification was carried out using PCR with BioBrick primers VF2 and VR. The expected band size of approximately 1 kbp was obtained (band e). Comparison with RE digest and PCR of HasA coding sequence part BBa_K1897000 (bands c, d, and f) suggested that the the Promoter and RBS sequence BBa_K525998 and double terminator BBa_B0015 had been successfully inserted as well. The part sequences were confirmed by Sanger sequencing.

Recombinant E. coli BL21(DE3) expressing HasA was cultured in 200mL of LB liquid media at 37°C. When OD 0.6 was reached, IPTG was added to final concentration of 1mM to induce HasA production. After 4 hours, the bacterial cells were harvested and lysed by addition of B-PER. The lysate was passed through a nickel column and the flowthrough, first wash, second wash, and three elutions (1mL each) were collected. SDS-PAGE of the samples was run on a polyacrylamide gel (4% stacking, 10% resolving) at 70V for 30 minutes and followed by 100V for 1 hour. Subsequently, semi-dry transfer was conducted at 15V for 1 hour to transfer the protein bands to a PVDF membrane. After blocking and incubation with the relevant primary and secondary antibodies, the blot was visualised with chemiluminescence using the Amersham Imager 600.

As seen in Fig 5, intense bands were observed in the total lysate and elute fractions, with the thickest band found in elute 2. The approximate size of the purified product was 19 to 25 kDa, suggesting that the his-tagged HasA was successfully produced and purified.

The RIOT Responder circuit (BBa_K1897007) is designed to receive an extracellular signal from the HasA hemophore signal module of the RIOT Transponder molecule. The circuit primarily borrows its concept from the Heme Acquisition System (HAS) operon pathway of Serratia marcescens. Ultimately, when the HasA signal is received at the outer membrane, the circuit activates transcription of the luxR gene. The LuxR protein is the first activator of the RIOT Invader module of the complete RIOT system.

The HAS operon comprises four main parts: The membrane surface transceptor HasR, the sigma factor silencer HasS, the sigma-like factor HasI, and the HAS promoter pHas. As seen in the schematic Fig 6, HasR HasS, and HasI are constitutively expressed. Upon receiving the HasA signal from RIOT Transponder, HasR initiates the release of HasI from HasS, allowing HasI to enhance transcription initiation at pHas.

The full RIOT Responder circuit was custom synthesised in three smaller fragments (Seq1, Seq2, and Seq3). The fragments were sequentially inserted into the pSCB13 BioBrick backbone by RE digest and ligation. Intermediates were transformed into E. coli DH5α and cultured overnight to produce more plasmids for the next insertion step.

Fig 7 shows the verification RE digest bands at each insertion step. Band c represents the full RIOT Responder part BBa_K1897007 of approximately 7.5kbp. In all cases, the linearised pSB1C3 BioBrick backbone was present as a 2kbp band.

The complete RIOT Responder circuit part was verified through Sanger sequencing.

The production of the main protein components of the HAS operon (HasR, HasS, and HasI) in RIOT Responder bacteria was confirmed by Western Blot. The components were all tagged with hexa-histidine sequences for purification.

RIOT Responder E. coli was cultured in 400mL of LB liquid media at 37°C. The bacterial cells were harvested and lysed by addition of B-PER. The lysate was passed through a nickel column and the flowthrough, first wash, second wash, and elute were collected. SDS-PAGE of the samples was run on a polyacrylamide gel (4% stacking, 10% resolving) at 70V for 30 minutes and followed by 120V for 1 hour. Subsequently, semi-dry transfer was conducted at 15V for 1 hour to transfer the protein bands to a PVDF membrane. After blocking and incubation with the relevant primary and secondary antibodies, the blot was visualised with chemiluminescence using the Amersham Imager 600.

As seen in Fig 8, bands corresponding to the approximate sizes of each protein component (HasR: 98kDa, HasS: 34kDa, HasI: 20kDa) were observed. The result implies that the HAS operon pathway was present and expressed in the RIOT Responder bacteria.

To test the ability of the RIOT Responder circuit to receive a signal from HasA and activate transcription downstream of pHas, a reporter rfp gene for red fluorescence protein was included. RIOT Responder E. coli bacteria was added with three concentrations of HasA (0M, 10^-4M, 10^-5M) and observed under fluorescence microscopy.

The results suggested a positive relationship between HasA concentration and intensity of fluorescence, implying that the RIOT Responder circuit had indeed been activated by HasA. The RIOT Responder circuit is thus a possible working component of the complete RIOT system.

Fig 9 shows the fluorescent microscopy images at the 2 hour timepoint. There was more RIOT Responder bacteria cells exhibiting fluorescence when more concentrated HasA was present. Fig 10 shows the images captured after overnight induction. Although there was some background fluorescence in the negative control, fluorescence was more intense in the sample with 10^-5M HasA.

The fluorescence of the RIOT Responder cells induced at the different HasA concentrations were compared as shown in Fig. 11. The results suggest that higher concentration of HasA corresponded to higher fluorescent intensities and implies that the RIOT Responder circuit can be activated by addition of HasA.