Proof of concept in HEK293T Verification of RADAR system At this stage, our objective is to validate the effectiveness of the RADAR system reported by K. Eerik Kaseniit and explore optimal conditions and experimental setups for RADAR, thereby laying the foundation for the development of our RADAR system for breast cancer surveillance. We selected two RADAR sensors, Sensor 1 and Sensor 2, which have been documented by K. Eerik Kaseniit to perform well, for experimental validation. Subsequently, we used the AND gate sensor—— Sensor 1+2, formed by connecting Sensor 1 and Sensor 2 together, to verify its function of simultaneously detecting two distinct Triggers. At the same time, a sequence with the stop codon (UAG) replaced by a tryptophan codon (UGG) was used as a positive control to evaluate the switching efficiency of RADAR system. The positive control sensor is labeled by adding the suffix "-UGG", such as Sensor 1-UGG[1]. We transfected HEK293T cells with plasmids carrying the Sensors, the target RNAs (Trigger RNA) of the Sensors, and ADAR. After 48 hours, we assessed transfection efficiency and effectiveness of RADAR system by detecting mCherry and EGFP expression. Amplification and verification of plasmids Plasmids carrying Sensors (including Sensor 1, Sensor 2, Sensor 1+2 and their positive control), triggers (including Trigger 1 and Trigger 2) and hADAR1p150 were synthesized according to the sequences provided (Table 1)[1-2].
Detailed information of plasmids
Short Name Full Name Length (bp)
ADAR pcDNA3.1-hADAR1p150 9028
Sensor 1 pcDNA3.1/Hygro(+)-PSFFV-mCherry-Sensor 1-EGFP 6591
Sensor 1-UGG pcDNA3.1/Hygro(+)-PSFFV-mCherry-Sensor 1(PC)-EGFP 6591
Sensor 2 pcDNA3.1/Hygro(+)-PSFFV-mCherry-Sensor 2-EGFP 6591
Sensor 2-UGG pcDNA3.1/Hygro(+)-PSFFV-mCherry-Sensor 2(PC)-EGFP 6591
Sensor 1+2 pcDNA3.1/Hygro(+)-PSFFV-mCherry-AND Sensor-EGFP 6687
Sensor 1+2-UGG pcDNA3.1/Hygro(+)-PSFFV-mCherry-AND Sensor(PC)-EGFP 6687
Trigger 1 pcDNA3.1/Hygro(+)-mTagBFP2-Trigger 1 5981
Trigger 2 pcDNA3.1/Hygro(+)-mTagBFP2-Trigger 2 5981
These plasmids were transformed into DH5α for amplification. Monoclonal colonies were cultured in LB medium, followed by plasmid Maxiprep from the bacterial cultures. The extracted plasmid was then verified by restriction digestion, and the digested plasmids were analyzed by agarose gel electrophoresis to confirm the correct size of the plasmids (Figure 1).
A-I Maps of plasmids;
J-R Plasmids digested with single/double restriction endonucleases, 1% agarose gel electrophoresis.
Single input RADAR system transfection and detection of fluorescent protein To verify that the RADAR sensor can specifically respond to its corresponding Trigger RNA and induce the expression of the output gene (EGFP), we transfected HEK293T cells with plasmids using Lipofectamine™ 3000 according to the ratios specified in Table 2.
Transfection system of single input RADAR
System of sensor 1 System of sensor 2
Group Plasmid Amount (ng) Group Plasmid Amount (ng)
Sensor 1 pcDNA3.1/Hygro(+)-PSFFV-mCherry-Sensor 1-EGFP 200 Sensor 2 pcDNA3.1/Hygro(+)-PSFFV-mCherry-Sensor 2-EGFP 200
Sensor 1+Trigger 1 pcDNA3.1/Hygro(+)-PSFFV-mCherry-Sensor 1-EGFP 200 Sensor 2+Trigger 1 pcDNA3.1/Hygro(+)-PSFFV-mCherry-Sensor 2-EGFP 200
pcDNA3.1/Hygro(+)-mTagBFP2-Trigger 1 250 pcDNA3.1/Hygro(+)-mTagBFP2-Trigger 1 250
Sensor 1+Trigger 2 pcDNA3.1/Hygro(+)-PSFFV-mCherry-Sensor 1-EGFP 200 Sensor 2+Trigger 2 pcDNA3.1/Hygro(+)-PSFFV-mCherry-Sensor 2-EGFP 200
pcDNA3.1/Hygro(+)-mTagBFP2-Trigger 2 250 pcDNA3.1/Hygro(+)-mTagBFP2-Trigger 2 250
Sensor 1+ADAR pcDNA3.1/Hygro(+)-PSFFV-mCherry-Sensor 1-EGFP 200 Sensor 2+ADAR pcDNA3.1/Hygro(+)-PSFFV-mCherry-Sensor 2-EGFP 200
pcDNA3.1-hADAR1p150 50 pcDNA3.1-hADAR1p150 50
Sensor 1+Trigger 1+ADAR pcDNA3.1/Hygro(+)-PSFFV-mCherry-Sensor 1-EGFP 200 Sensor 2+Trigger 1+ADAR pcDNA3.1/Hygro(+)-PSFFV-mCherry-Sensor 2-EGFP 200
pcDNA3.1/Hygro(+)-mTagBFP2-Trigger 1 250 pcDNA3.1/Hygro(+)-mTagBFP2-Trigger 1 250
pcDNA3.1-hADAR1p150 50 pcDNA3.1-hADAR1p150 50
Sensor 1+Trigger 2+ADAR pcDNA3.1/Hygro(+)-PSFFV-mCherry-Sensor 1-EGFP 200 Sensor 2+Trigger 2+ADAR pcDNA3.1/Hygro(+)-PSFFV-mCherry-Sensor 2-EGFP 200
pcDNA3.1/Hygro(+)-mTagBFP2-Trigger 2 250 pcDNA3.1/Hygro(+)-mTagBFP2-Trigger 2 250
pcDNA3.1-hADAR1p150 50 pcDNA3.1-hADAR1p150 50
Sensor 1-UGG pcDNA3.1/Hygro(+)-PSFFV-mCherry-Sensor 1(PC)-EGFP 200 Sensor 2-UGG pcDNA3.1/Hygro(+)-PSFFV-mCherry-Sensor 2(PC)-EGFP 200
48 hours post transfection, fluorescence images were performed using excitation wavelengths of 488 nm (for EGFP) and 580 nm (for mCherry).
Images of HEK293T 48h after transfection.
Fluorescence images reveal that cells transfected with either Sensor 1 or Sensor 2 exhibit strong red fluorescence, confirming high transfection efficiency. Most important, EGFP expression from sensors were only induced in the presence of their matching trigger. Notably, when ADAR was transfected, Trigger-induced EGFP expression was obviously increased, indicating that exogenous ADAR enhanced the switching effeciency of “UAG” to “UGG” of the RADAR system (Figure 2). These results demonstrate that the RADAR system can be specifically activated by matching triggers to induce the expression of output genes. Double input RADAR system transfection and detection of fluorescent protein Next, to further investigate whether an AND gate function can be achieved by linking the sequences of Sensor 1 and Sensor 2 together, we transfected HEK293T cells with plasmids carrying the Sensor 1+2, along with the corresponding Triggers. Since we confirmed that additional ADAR expression can enhance the efficiency of RADAR system, an additional 50 ng of the ADAR-expressing plasmid (pcDNA3.1-hADAR1p150) was included in each transfection group except Sensor 1+2-UGG, as specified in Table 3.
Transfection system of double input RADAR system verification
Group Plasmid Amount (ng)
Sensor 1+2 +ADAR pcDNA3.1/Hygro(+)-PSFFV-mCherry-AND Sensor-EGFP 200
pcDNA3.1-hADAR1p150 50
Sensor 1+2+Trigger 1+ADAR pcDNA3.1/Hygro(+)-PSFFV-mCherry-AND Sensor-EGFP 200
Sensor 1+2+Trigger 1+ADAR 125
pcDNA3.1-hADAR1p150 50
Sensor 1+2+Trigger 2+ADAR pcDNA3.1/Hygro(+)-PSFFV-mCherry-AND Sensor-EGFP 200
pcDNA3.1/Hygro(+)-mTagBFP2-Trigger 2 125
pcDNA3.1-hADAR1p150 50
Sensor 1+2+Trigger 1+Trigger 2+ADAR pcDNA3.1/Hygro(+)-PSFFV-mCherry-AND Sensor-EGFP 200
pcDNA3.1/Hygro(+)-mTagBFP2-Trigger 1 125
pcDNA3.1/Hygro(+)-mTagBFP2-Trigger 2 125
pcDNA3.1-hADAR1p150 50
Sensor 1+2-UGG pcDNA3.1/Hygro(+)-PSFFV-mCherry-AND Sensor(PC)-EGFP 200
48 hours post transfection, fluorescence imagings were performed using excitation wavelengths of 488 nm (for EGFP) and 580 nm (for mCherry).
Images of HEK293T 48h after transfection of Sensor 1+2 and other plasmids.
Fluorescence images reveal that EGFP expression from the Sensor 1+2 was triggered only upon co-transfection with both Trigger 1 and Trigger 2, which demonstrates that an AND gate constructed by linking two sensors together can be activated only when both triggers were present (Figure 3). Therefore, we decided to design our own RADAR sensor with AND gate function by linking two sensors together. Verification of Gaussia Luciferase (Gluc) Humanized Gaussia Luciferase (Gluc), a small secretory protein (19.9 kD), catalyzes the reaction of coelenterazine to coelenteramide to emit luminescence, which makes it an ideal reporter for our ABCS project[3]. At this stage, our objective is to explore and optimize the experimental setup for Gluc expression, thereby laying the groundwork for the development and testing of Gluc as a output reporter gene with RADAR system. In our experiments, we transfected HEK293T cells with plasmids carrying the humanized Gluc gene, and collect the conditioned medium at 24 h, 48 h and 72 h post transfection. Luminescence intensity was measured using a multifunctional microplate reader, and luminescence images were obtained using a fully automated chemiluminescence imaging analysis system. Amplification and verification of plasmids We used Gluc expressing plasmid (pCMV-Gluc-1), which carries the humanized Gluc gene (Figure 4 A). This plasmid was transformed into DH5α for amplification. Monoclonal colonies were then expanded in LB medium, followed by extraction of plasmids from the bacterial culture. The extracted plasmids were digested with restriction endonucleases and analyzed by agarose gel electrophoresis to confirm the correct plasmid size (Figure 4 B).
A Map of pCMV-Gluc-1;
B pCMV-Gluc-1 digested with single/double restriction endonucleases, 1% agarose gel electrophoresis.
Transfection and luminescence detection To evaluate the expression and kinetics of Gaussia Luciferase (Gluc) in HEK293T cells, we transfected HEK293T cells in 24-well plates with 500 ng of Gluc plasmid per well, including a non-transfected control group (NC). The conditioned medium were collected from the transfected HEK293T cells at 24, 48, 72 hours post transfection. Supernatants of the conditioned medium were obtained by centrifuging at 1000 rpm for 5 minutes, and subsequently analyzed using a multifunctional microplate reader after adding coelenterazine as substrate.
Chemiluminescence analysis of culture supernatant from HEK293T cells collected at 24h, 48h and 72h after transfection. *p < 0.05, **p < 0.01, ***p < 0.001.
Results showed that Gluc was successfully expressed and secreted from the transfected HEK293T cells, which remained at high levels at 24, 48 and 72h, with the peak time at 48h (Figure 5). To visualize the luminescence produced by the Gluc-catalyzed reaction, we also performed luminescence images using both a fully automated chemiluminescence imaging analysis system and a cell phone. Similar as results shown in Figure 5, while supernatant from non-transfected control exhibited no luminescence, supernatant from cells transfected with Gluc expression plasmid displayed bright luminescence (Figure 6A). Importantly, such luminescence could also be captured by a cellphone, indicating that, in the future, the monitoring of occurence of breast cancer by our ABCS system could be easily done with a cell phone at home (Figure 6B).
A Luminescence images captured by fully automated chemiluminescence imaging analysis system;
B Luminescence images captured by a cellphone.
Furthermore, to investigate the flash kinetic characteristics of Gluc, we analyzed these supernatants using the kinetic mode of a multifunctional microplate reader. A key finding was that all samples exhibited similar kinetic profiles—reaching maximum luminescence intensity (peak RLU) within milliseconds after substrate addition, clearly demonstrating its ultra-rapid reaction capability (Figure 7).
Analysis of sustained luminescence over 25 minutes between kinetics of luminescence production from Gluc.
Based on these results, we decided to collect supernatant at 48h post transfection and measure luminescence immediately after the adding of coelenterazine in subsequent experiments using Gluc as output reporter. Validation of RADAR system with Gluc So far, we have validated the function of the RADAR system in HEK293T cells and confirmed that Gluc can be expressed and secreted by HEK293T cells as an efficient reporter. To investigate whether Gluc can be used as an efficient output reporter for RADAR system, we constructed RADAR sensor plasmids with Gluc as output reporter and performed functional validation in HEK293T cells. Amplification and verification of plasmids Based on the plasmid used in the previous experiments, we constructed plasmids carrying Sensor 1-Gluc and Sensor 1-Gluc-UGG. Additionally, to validate the efficiency of the AND gate sensor using Gluc as the reporter gene in simultaneously detecting two target triggers, we constructed plasmids carrying Sensor 1+2-Gluc and Sensor 1+2-Gluc-UGG (Table 4).
Detailed information of plasmids carrying RADAR sensor with Gluc
Short Name Full Name Length (bp)
Sensor 1-Gluc pcDNA3.1/Hyg1ro(+)-PSFFV-mCherry-Sensor 1-Gluc 6429
Sensor 1-Gluc-UGG pcDNA3.1/Hygro(+)-PSFFV-mCherry-Sensor 1(PC)-Gluc 6429
Sensor 1+2-Gluc pcDNA3.1/Hygro(+)-PSFFV-mCherry-AND Sensor-Gluc 6525
Sensor 1+2-Gluc-UGG pcDNA3.1/Hygro(+)-PSFFV-mCherry-AND Sensor(PC)-Gluc 6525
These plasmids were transformed into DH5α for amplification. Monoclonal colonies were cultured in LB medium, followed by extraction of plasmid DNA from the bacterial cultures. The extracted plasmids were digested with restriction enzyme and analyzed by agarose gel electrophoresis to confirm the correct plasmid size (Figure 8).
A-D Maps of plasmids;
E-H Plasmids digested with single/double restriction endonuclease, 1% agarose gel electrophoresis.
Next, we transfected HEK293T cells according to the system outlined in Table 4.
Transfection system of single input RADAR system with Gluc as output gene
Group Plasmid Amount (ng)
Sensor 1-Gluc+ADAR pcDNA3.1/Hygro(+)-PSFFV-mCherry-Sensor 1-Gluc 200
pcDNA3.1-hADAR1p150 50
Sensor 1-Gluc+Trigger 1+ADAR pcDNA3.1/Hygro(+)-PSFFV-mCherry-Sensor 1-Gluc 200
pcDNA3.1/Hygro(+)-mTagBFP2-Trigger 1 250
pcDNA3.1-hADAR1p150 50
Sensor 1-Gluc+Trigger 2+ADAR pcDNA3.1/Hygro(+)-PSFFV-mCherry-Sensor 1-Gluc 200
pcDNA3.1/Hygro(+)-mTagBFP2-Trigger 2 250
pcDNA3.1-hADAR1p150 50
Sensor 1-Gluc-UGG pcDNA3.1/Hygro(+)-PSFFV-mCherry-Sensor 1(PC)-Gluc 200
48 hours post transfection, we collected conditioned medium from transfected cells and measured chemiluminescence following the addition of coelenterazine.
Luminescence of HEK293T cells transfected with Sensor 1-Gluc and other plasmids.
We observed that, luminescence intensity from cells transfected with Sensor 1-Gluc and Trigger 1 was significantly higher than that from cells transfected with Sensor 1-Gluc but no Trigger 1. In addition, luminescence intensity from cells transfected with Sensor 1-Gluc and Trigger 1 was significantly higher than that from cells transfected with Sensor 1-Gluc and Trigger 2, demonstrating that Sensor 1-Gluc specifically detects Trigger 1 and drives Gluc expression efficiently (Figure 9). These results validated that Gluc could serve as a effitient output reporter of RADAR system. To further evaluate the sensitivity and specificity of an AND gate Sensor with Gluc as output reporter, we transfected HEK293T cells with plasmids as listed in Table 5, and assessed luminescence intensity for validation.
Transfection system of double input RADAR sensor with Gluc as output gene
Group Plasmid Amount (ng)
Sensor 1+2-Gluc+ADAR pcDNA3.1/Hygro(+)-PSFFV-mCherry-Sensor 1+2-Gluc 200
pcDNA3.1-hADAR1p150 50
Sensor 1+2-Gluc+Trigger 1+ADAR pcDNA3.1/Hygro(+)-PSFFV-mCherry-Sensor 1+2-Gluc 200
pcDNA3.1/Hygro(+)-mTagBFP2-Trigger 1 125
pcDNA3.1-hADAR1p150 50
Sensor 1+2-Gluc+Trigger 2+ADAR pcDNA3.1/Hygro(+)-PSFFV-mCherry-Sensor 1+2-Gluc 200
pcDNA3.1/Hygro(+)-mTagBFP2-Trigger 2 125
pcDNA3.1-hADAR1p150 50
Sensor 1+2-Gluc+Trigger 1+Trigger 2+ADAR pcDNA3.1/Hygro(+)-PSFFV-mCherry-Sensor 1+2-Gluc 200
pcDNA3.1/Hygro(+)-mTagBFP2-Trigger 1 125
pcDNA3.1/Hygro(+)-mTagBFP2-Trigger 2 125
pcDNA3.1-hADAR1p150 50
Sensor 1+2-Gluc-UGG pcDNA3.1/Hygro(+)-PSFFV-mCherry-Sensor 1+2(PC)-Gluc 200
48 hours post transfection, conditioned medium from transfected cells were collected and subjected to chemiluminescence detection after adding coelenterazine (Figure 10).
Chemiluminescence of conditioned medium from HEK293T cells transfected with Sensor 1+2-Gluc and other plasmids.
We observed that the luminescence intensity of the conditioned medium from cells co-transfected with both Trigger 1, Trigger 2 and the Sensor 1+2 was significantly higher than that from cells transfected with only one Trigger or no Trigger, which demonstrates that the tandem AND-gate sensor with Gluc as output reporter functions effectively and specifically in detecting the presence of both Triggers (Figure 10).
Luminescence of conditioned medium from transfected HEK293T cells that were collected at indicated times.
Furthermore, to investigate the kinetic characteristics of the Gluc expression from AND gate RADAR sensor, we collected conditioned medium at 12, 24, 36, 48, 60 and 72 hours post transfection and measured luminescence. The results showed a continuous generation of luminescence over the 72-hour period following transfection, with a peak time at 60 hours post transfection (Figure 11). Validation of chassis-adipocytes Induced differentiation of preadipocytes To obtain mature adipocytes for subsequent experiments, we chose to induce differentiation of preadipocytes. For the induction of preadipocytes, a commercial adipogenic differentiation kit was used, specific steps are described in the Protocol. To confirm successful differentiation, we performed Oil Red O staining and examined lipid droplet formation under microscopy. Following 14 days of induction, successful differentiation was observed via Oil Red O staining, thereby establishing a reliable way to generate differentiated adipocytes for subsequent experiments (Figure 12).
Induced differentiation of preadipocytes.
Validation of rAAV9 delivery We utilized recombinant adeno-associated virus serotype 9 (rAAV9) to deliver the Sensor 1-Gluc, Trigger 1 and ADAR to investigate whether RADAR system can be successfully transduced into adipocytes and remain functional within this cell type. Infection and detection of Gluc expression Following adipocyte differentiation, cells were infected with rAAV carrying Sensor 1, Trigger 1, and ADAR to assess both rAAV delivery efficiency and RADAR function in adipocytes. Conditioned medium were collected two days post infection and luminescence was meatured by adding coelenterazine, thereby evaluating the expression of Gluc in differentiated adipocytes (Figure 13).
Luminescence of the conditioned medium from adipocytes infected by rAAVs carrying Sensor 1 and Trigger 1.
The result demonstrates that RADAR system can be effectively delivered and function appropriately in adipocytes. Test the viability of infected adipocytes The viability of adipocytes after rAAV infection is another critical aspect we must consider. We examined the viability of infected adipocytes to ensure that rAAV infection did not cause severe detrimental effects results on the viability of adipocytes. Detection of Engineered Adipocyte Viability by CCK-8 Assay To ensure that rAAV infection did not induce significant toxicity or necrosis in engineered adipocytes, cell viability was measured using the CCK-8 assay. In this assay, the water-soluble tetrazolium salt WST-8 is reduced by mitochondrial dehydrogenases in viable cells to an orange formazan product. The reduction rate is proportional to the number of living cells, and absorbance at 450 nm is used to quantify metabolic activity. OD450 was measured 2 hours after adding CCK-8 reagent to assess cell viability.
The viability of infected and uninfected adipocytes revealed by CCK-8 assay.
The result shows no significant difference in viability between rAAV infected and non infected adipocytes, indicating that rAAV infection might be a safe way to engineer adipocytes (Figure 14). Test of our RADAR in adipocytes Following the proof-of-concept experiments of RADAR system in HEK293T cells and validation of adipocytes as the chassis, we proceeded to develop our own RADAR system to detect the emerging of cancer associated adipocytes. Since literature review and database analysis by our model group showed that PLOD2 and LIF are optimal biomarkers of cancer associated adipocytes, we decided to use them as target genes for our RADAR-based detection. Verification of target gene expression As literatures reported that PLOD2 and LIF expression in adipocytes could be induced by PAI-1 and CXCLs in breast tumor microenvironment[4-5], we validated these observation with experiments. To investigate the efficiency of induced expression of PLOD2 and LIF, we treated differentiated adipocytes with recombinant PAI-1 and CXCL8 proteins. Cytokine treatment and qPCR Differentiated adipocytes were treated with PAI-1 and CXCL8 individually or in combination. The expression levels of PLOD2 and LIF were measured via qPCR with GAPDH as internal control.
Expression of PLOD2 and LIF in adipocytes treated with PAI-1 and CXCL8.
qPCR results revealed that PAI-1 treatment significantly increased PLOD2 expression, while CXCL8 treatment significantly increased LIF expression (Figure 15). Test of single input RADAR systems Through analysis of RNA secondary structure with iPKnot++, we obtained the sensor sequences capable of targeting PLOD2 and LIF (details provided in the Model). These sensors, with Gluc as the output gene, were constructed. Using rAAV, we delivered these sensors into adipocytes and treated the infected adipocytes with PAI-1 and CXCL8. The luminescence generated by Gluc were then measured to validate the function of these two RADAR sensors. The function validation of PLOD2 sensor We infected the differentiational adipocytes with rAAV9 carrying PLOD2 sensor and ADAR. Two days after infection, we treated the cells with PAI-1, and then collected conditioned medium to measure the luminescence by adding coelenterazine.
Luminescence of conditioned medium supernatant from adipocytes infected by rAAV carrying PLOD2 sensor, with or without PAI-1 treatment.
We found that the luminescence intensity of the PAI-1-treated group was significantly higher than that of the untreated group (Figure 16). Together with the result shown in Figure 15, these results suggested that our designed PLOD2 sensor could detect PAI-1 induced expression of PLOD2 and activated the expression of output reporter gene Gluc. The function validation of LIF sensor We infect adipocytes with rAAV carrying LIF sensor and ADAR. We treated cells with CXCL8 and then collected conditioned medium to measure the luminescence by adding coelenterazine.
Luminescence of conditioned medium supernatant from adipocytes infected by rAAV carrying LIF sensor, with or without CXCL8 treatment.
We found that the CXCL8-treated group exhibited significantly higher luminescence intensity compared to the untreated group (Figure 17). Together with the result shown in Figure 15, these results suggested that our designed LIF sensor could detect CXCL8-induced expression of LIF and activate the expression of output reporter gene Gluc. The successful validation of the effectiveness of PLOD2 sensor and LIF sensor establishes a foundation for linking these two sensor sequences together to enable simultaneous detection of PLOD2 and LIF. Test of double input RADAR system Following the validation of PLOD2 sensor and LIF sensor, we proceeded to verify the function of our designed AND-gate sensor, i.e. PLOD2-LIF sensor. After delivering the PLOD2-LIF sensor into adipocytes using rAAV, we treated the cells with PAI-1 and CXCL8 to assess its detection capability. Subsequently, these adipocytes were treated with the conditioned medium from breast cancer cell MDA-MB-231 to further evaluate the ability of sensor to detect cytokines secreted from breast cancer cells. The function validation of PLOD2-LIF sensor Differentiated adipocytes were infected with rAAV9 carrying the PLOD2-LIF sensor. On day 2 post infection, cells were treated as follows: no cytokines, PAI-1 alone, CXCL8 alone, or both cytokines combined. Conditioned medium were collected for luminescence detection. Results demonstrated that the luminescence intensity was significantly higher in the conditioned medium from cells treated with both PAI-1 and CXCL8 than those from cells treated with PAI-1 only or CXCL8 only, indicating that Gluc expression from PLOD2-LIF sensor was induced only when both PLOD2 and LIF were simultaneously up-regulated. This result demonstrates that the PLOD2-LIF sensor responds to the presence of both PAI-1 and CXCL8 and activates the expression of output reporter gene Gluc.
Luminescence of conditioned medium supernatant from adipocytes infected by rAAV carrying PLOD2-LIF sensor, with or without cytokines treatment.
To better simulate the interaction between breast cancer cells and adipocytes, conditioned medium from breast cancer cell MDA-MB-231 was used to assess the performance of the engineered adipocytes. Human breast cancer cells MDA-MB-231 were cultured in medium with 0.2% FBS for 24 hours. The conditioned medium was then collected to treat the adipocytes infected with rAAV carrying PLOD2-LIF sensor. After 48 hours of treatment, the supernatant was collected for luminescence detection.
Luminescence of the conditioned medium from adipocytes infected by rAAV carrying PLOD2-LIF sensor, with or without treatment of conditioned medium from MDA-MB-231 cells.
Treatment with conditioned medium from MDA-MB-231 cells significantly increased the luminescence generated from conditioned medium of adipocytes with PLOD2-LIF sensor, indicating that these engineered adipocytes could respond to the cytokines secreted from breast cancer cells (Figure 19). This result suggests that these engineered adipocytes have the potential to detect breast cancer cells. Conclusion We have constructed engineered adipocytes for detection of cytokines secreted from breast cancer cell. Initially, we verified in HEK293T cells the function of established RADAR systems, and the effectiveness of the RADAR sensor using Gluc as reporter. Subsequently, the RADAR system was efficiently delivered via rAAV9 into differentiated human adipocytes—confirmed by Oil Red O staining. Finally, by deploying RADAR sensors targeting PLOD2 and LIF in adipocytes, we confirmed that our designed sensors specifically responds to cytokines secreted from breast cancer cells and activates the expression of Gluc as output reporter. Discussion So far, we have completed the full workflow from sensor design, adipocyte engineering to functional validation of RADAR system. Our engineered adipocytes perform effectively without significant impairment to viability, demonstrating that the ABCS has the potential of long-term monitoring of breast cancer, thereby offering a novel approach to breast cancer surveillance. Although our current experiments are confined to the laboratory stage, we constructed a model that simulates the Glomerular Filtration Barrier (GFB) to investigate the renal metabolic characteristics of Gaussia luciferase (Gluc) in the human body (Click Model to learn more). While the model simulation provides valuable insights for in vivo applications, issues such as the long-term safety and stability of engineered cells in real living environments still require further investigation. In addition, the sensitivity and specificity of the sensors can be further optimized—for instance, by screening more efficient ADAR isoforms or refining sensor sequence structures to reduce background noise and improve the signal-to-noise ratio. Looking forward, we will focus on developing improved adipocytes engineering strategies to enhance the survival and functional persistence of engineered adipocytes. Concurrently, we aim to refine sensor performance through iterative design and testing, enabling high-accuracy and high-specificity detection in complex in vivo microenvironments. The ultimate goal is to advance this technology toward preclinical translation, providing a feasible solution for monitoring of breast cancer occurence. Reference
Reference
1 Kaseniit KE, Katz N, Kolber NS, Call CC, Wengier DL, Cody WB, Sattely ES, Gao XJ. Modular, programmable RNA sensing using ADAR editing in living cells. Nat Biotechnol. 2023 Apr;41(4):482-487.
2 Gayet RV, Ilia K, Razavi S, Tippens ND, Lalwani MA, Zhang K, Chen JX, Chen JC, Vargas-Asencio J, Collins JJ. Autocatalytic base editing for RNA-responsive translational control. Nat Commun. 2023 Mar 11;14(1):1339.
3 Wang T, Chen Y, Goodale D, Allan AL, Ronald JA. A survivin-driven, tumor-activatable minicircle system for prostate cancer theranostics. Mol Ther Oncolytics. 2021 Jan 20;20:209-219.
4 Wei X, Li S, He J, Du H, Liu Y, Yu W, Hu H, Han L, Wang C, Li H, Shi X, Zhan M, Lu L, Yuan S, Sun L. Tumor-secreted PAI-1 promotes breast cancer metastasis via the induction of adipocyte-derived collagen remodeling. Cell Commun Signal. 2019 Jun 6;17(1):58.
5 Zhou C, He X, Tong C, Li H, Xie C, Wu Y, Wang L, Yan X, Luo D, Tang Y, Cheng Z, Xiong X. Cancer-associated adipocytes promote the invasion and metastasis in breast cancer through LIF/CXCLs positive feedback loop. Int J Biol Sci. 2022 Jan 16;18(4):1363-1380.