Hardware

CRC Prognosis Kit: asymmetric RPA-based Dual-target Visualization Prognosis Kit

Working Mode Diagram of CRC Prognosis Kit

Note: please see Project Results for details of the experiment.

Overview
The CPU-CHINA research team, after an extensive literature review, brainstorming sessions, and proof-of-concept studies, successfully engineered a bacterium capable of dual-targeting for the treatment of Colorectal Cancer (CRC), particularly in cases presenting positive Fusobacterium nucleatum (Fn). However, we are unable to directly assess its therapeutic effect. Currently, the mainstream ways to determine the progression of CRC are endoscopy, tissue biopsy and imaging[1], which are generally invasive, complicated, expensive and dependent on large instruments, and the patient compliance is poor. To solve this problem, we developed the CRC Prognostic Kit as an adjunctive prognostic method for CRC. This user-friendly assay operates at ambient temperature and offers results in under 30 minutes. We adopt a dual-target nucleic acid detection strategy to improve the sensitivity and specificity of the test, which greatly reduces the problem of false positives of the kit. At the same time, the total cost of the kit is less than 10 dollars, and the test results can be directly associated with the WeChat applet for long-term detection of the condition. When the applet reminds the patient of the poor prognosis, the patient can go to the hospital for further examination and determine the treatment plan as soon as possible to prevent delays in disease.

Table 1.CRC Prognostic Kit parts list

Note: The RPA kit used by CPU-CHINA was purchased from TwistDx™.

2. Design
Based on the guidance of asymmetric RT-RPA model (see details of asymmetric RT-RPA model in Awards Model) and wet-lab exploration, we successfully constructed the CRC Prognostic Kit.The CRC Prognostic Kit consists of two modules: exponential amplification module and nucleic acid detection module. Exponential amplification module relies on nucleic acid isothermal amplification technology, and nucleic acid detection module relies on lateral flow dipstick (LFD) technology. The combination of RPA (Recombinase Polymerase Amplification, a nucleic acid isothermal amplification technology) and LFD can amplify the target gene to the 109 in just a few minutes of reaction at 37℃. Besides, it enables on-site visualization of nucleic acid amplification products, and the test results are obvious and can intuitively be recognized by the naked eye.It has the characteristics of high specificity, high sensitivity, simplicity, safety, and low cost.

Exponential Amplification Module: In the reaction tube accompanying the kit, with the help of reverse transcriptase added in the reaction system, microRNA-135b (miR-135b) is used as a template to synthesize cDNA (nusG1 itself is a cDNA). Then with isothermal amplification technology-RPA, with specially designed asymmetric amplification primers, a large amount of ssDNA is synthesized within 30min to achieve exponential amplification of targets.

MiR-135b is a 97-bp microRNA that targets the 3' untranslated region of Adenomatous Polyposis Coli (APC), a well-known tumor suppressor, repressing its expression and inducing downstream Wnt pathway activity. Furthermore, miR-135b was significantly upregulated in CRC accompanied by low levels of APC, suggesting that upregulation of miR-135 may be involved in CRC pathogenesis[2]. Clinical results showed that Stool miR-135b levels decreased significantly after resection of CRC or advanced adenomas[3]. Therefore, the team chose miR-135b as a molecular biomarker for assessing treatment efficacy.

Fusobacterium nucleatum (Fn) is a common oral commensal that plays an important role in promoting the activities of CRC, including carcinogenesis, progression and metastasis[4]. Meanwhile, Fn is significantly enriched in the stool of CRC patients[5]. Therefore, we considered Fn as a detection organism. Through literature research we found the specific gene nusG of Fusobacterium Nucleatum Subsp. Nucleatum and intercepted a conserved fragment of 300bp only present in Fusobacterium Nucleatum Subsp. Nucleatum as another important molecular biomarker for us to evaluate the therapeutic efficacy (Fig. 1), named as nusG1.

Figure 1. NusG1 sequence alignment results

RPA is an isothermal nucleic acid amplification technique, in which the reaction process is always maintained at a constant temperature. Different active enzymes and specific primers are added to achieve the purpose of rapid amplification of nucleic acids. These characteristics make it suitable for point-of-care testing (POCT). Compared with PCR, isothermal amplification has the advantages of being fast, efficient, specific and does not require specialized equipment, so it has been considered by many scholars to be an assay that is potentially comparable to PCR since its inception. RPA technology relies on three enzymes: recombinase that binds to single-stranded nucleic acids (oligonucleotide primers), single-stranded DNA-binding proteins (SSBs), and strand-substitution DNA polymerases. The mixture of these three enzymes is also active at room temperature, with the optimal reaction temperature being around 37℃. We did not stick to the conventional RPA amplification. Instead, the team innovatively proposed asymmetric RPA amplification, that is, changing the ratio of forward and reverse primer incorporation (ratio of 25:1) to obtain ssDNA amplification products for subsequent nucleic acid detection.

Figure 2. Asymmetric RT-RPA schematic diagram

Nucleic Acid Detection Module: The module mainly relies on LFD technology. After obtaining ssDNA, the reaction solution is added dropwise to the sampling well of the testing card. SsDNA, under the action of chromatography, binds gold nanoparticles (AuNPs) with tagged DNA (Tag@AuNPs) at the gold pad, forming ssDNA-Tag complex. When chromatography reaches the T line, the above complex will combine with the DNA tetrahedral structure attached to the T line to form a composite structure of (T line-DNA tetrahedron-ssDNA-Tag@AuNPs). AuNPs are enriched in the T line, forming a purple-red detection line that can be seen by the naked eye, while the C line is purple-red due to the adsorption of AuNPs by solution chromatography. The CRC Prognostic Kit designed by our team has one quality control line (C line) and two detection lines (T line), one line (TN line) corresponds to the detection of miR-135b, and another line (TS line) corresponds to the detection of nusG1, and the kit can satisfy the needs of simultaneous detection of CRC cells and Fn.

Figure 3. Pattern diagram of the involvement of DNA tetrahedron-Biotin in the formation of Tag-ssDNA-Probe complex

Figure 4. Pattern diagram of LFD-based nucleic acid detection

3. Software
The WeChat applet, “Bi-Bye” synergizes with the CRC Prognostic Kit, emphasizing longitudinal condition monitoring and therapy assessment. By clicking button “检测评估”, patients can take a photo to upload the test results of the kit, and then click button “AI入口”, Bi-Bye can visualize the test results uploaded by patients as the concentration of the two targets through semi-quantitative model (details of the model can be found in following PDF) to analyze the patient's status, and more accurately assess the therapeutic effect of the team's bacterial therapy. When the concentration of the target exceeds the warning line, the patient is reminded that the treatment result is poor and should consider the replacement of the treatment package in time.
“Bi-Bye” can also track the condition, through long-term monitoring, the patient can have an intuitive understanding of the effect of continuous treatment based on the curve of the historical test results, so that the patient will not be on edge due to the lack of understanding of the condition, and to stabilize their mood and state of mind, which will help the disease to recover. In the future, this applet can cooperate and associate with many hospitals, so that when patients need treatment, they can directly click the button from the “医生服务” of this applet and jump to the WeChat applet, public number and official webpage of the cooperating units for making appointments and operations related to online consultation, which provides convenience for CRC patients.

Figure 5. The working pages of “By-Bye” applet

4. User's Manual

Figure 6. User's Manual

5. defects & prospects
Due to the late start of the hardware lab and the heavy workload of the project, only the amplification and detection of the miR-135b target has been completed, while the combined use of dual targets is still being tested and is expected to achieve our goal. In addition, in order to expand the application scope of the kit, the team selected S and N genes, as targets for a series of analyses, including primer design, for different mutation sites and mutation status of the new coronavirus, indicating that the CRC Prognostic Kit, in addition to being able to detect biomarkers related to CRC and Fn, can theoretically be applied to the detection of a wide range of DNA/RNA viruses such as β-coronaviruses. In addition, our team anticipates that further work could combine the CRC Prognostic Kit with a Cortex-M4 microcontroller to build an electrical detection device. The conductivity on the T line was measured to construct a conductivity-miR-135b/nusG1 concentration relationship.

Figure 7. Mutations in Gene N.,Gene S. and orf1ab of mainstream mutant strains (alpha, beta, delta, gamma, lambda, omicron), endemic strains in our country from 2022.12-2023.4, and newly entered strains (BA.*, BQ.*, BF.*, XBB.*) a-b. Gene S; c-d, Gene N; e-f, orf 1ab

Figure 8. Free energy of active DNA strand hybridization complexes and secondary structure of DNA strand hybridization complexes
a, Free energy of DNA strand hybridization complexes designed for Gene S. b, Secondary structure of DNA strand hybridization complexes designed for Gene S. c, Free energy of DNA strand hybridization complexes designed for Gene N. d, Secondary structure of DNA strand hybridization complexes designed for Gene N.

Note: please see Project Results for details of the experiment.

A Microfluidic Tumor-on-a-Chip for In Vitro Reconstruction of Primary Colorectal Cancer

Overview
Constructing in vitro colorectal cancer (CRC) models that faithfully recapitulate the physiological and pathological microenvironment of human CRC holds tremendous potential for accelerating research into CRC mechanisms and the development of anti-CRC drugs. To this end, iGEMers from CPU-CHINA have presented a colorectal tumor-on-a chip (CRT-chip), which can be used to simulate the physiological effects of Fusobacterium nucleatum (Fn) on tumor progression and metastasis. Additionally, it also serves as a platform to explore the mechanisms and feasibility of simultaneous targeting of both Fn and colorectal cancer cells using rod protein. Furthermore, it facilitates a high-fidelity characterization of the practical effects of dual-target fusion antimicrobial peptides and the simulation of the interactions between engineered Bifidobacterium longum and the entire intestinal symbiotic microbiota.

The CRT-chip consists of four tightly bonded layers of polydimethylsiloxane (PDMS). From top to bottom, these layers include the inlet layer, simulation layer, culture layer, and infusion layer. The chip allows precise implantation of colorectal cancer cells into the central channel, mimicking cancer cell invasion mechanisms. It provides cells with continuous shear stress and simulates the tumor vascular perfusion mechanism. Under the action of gas, it replicates the mechanical peristalsis of the human colon. By regularly observing the biological characteristics of cells within the central channel, it more accurately models the development of human colorectal cancer.

The CRT-chip will serve a dual purpose - not only in its utilization for investigating the etiology and evolving dynamics of colorectal cancer but also in its deployment to authenticate the modes of action of Bifidobacterium longum. Furthermore, it will be employed to delve into the holistic dynamic progression of colon cancer-probiotic communities within the gastrointestinal tract subsequent to the introduction of genetically engineered bacteria. It offers advantages such as low cost, ease of fabrication, miniaturization, and ease of observation, demonstrating its significant potential for in vitro evaluation of anti-colorectal cancer drugs.

Background
In prior research on colorectal cancer (CRC), conventional in vitro methodologies predominantly relied on two-dimensional (2D) tumor cell lines and xenograft tumor models. However, these models possess intrinsic limitations. 2D models inadequately replicate the intricate spatial configuration and heterogeneity of in vivo tumors, as well as the interactions between tumor cells and the tumor microenvironment (TME). Xenograft tumor models, while closely resembling tumor structure and physiological characteristics, employ immunocompromised mice, rendering the exploration of tumor immunology and immunotherapeutic agents unfeasible. Additionally, xenograft models suffer from low success rates in modeling, protracted experimental timelines, and high costs, among other challenges.

In stark contrast to these conventional models, the advent of in vitro three-dimensional (3D) culture models has yielded a robust platform for emulating the spatial structure and growth milieu of in vivo tumors. These 3D models retain the pathological and genetic traits of tumor cells while faithfully capturing the intricate interplay between tumor cells and their microenvironment(Figure 1. A). Consequently, they have evolved into potent tools for dissecting tumor mechanisms, conducting drug screening, and tailoring cancer therapy to individual patients. Multiple techniques for 3D tumor modeling, including spheroids, organoids, and microfluidic devices, have reached an advanced stage of development. Furthermore, innovative technologies such as co-culturing, 3D bioprinting, and air-liquid interfaces have bolstered the verisimilitude of these models. Certain 3D models can even recapitulate elements of the tumor microenvironment, including endogenous immune components and vascular systems(Figure 1. B). With continual advancements in relevant technologies, 3D tumor models are poised to make substantial contributions to colorectal cancer research, drug development, therapeutic selection, and personalized treatment.

Figure 1
A.Schematic of a tumor system with a complex organization of cancer cells, fibroblasts, extracellular matrix, vasculature system, and multiple chemical factors. Tumor progression is affected by microenvironment factors including mechanical and chemical such as shear stress, hypoxia, chemotaxis, cell-cell interactions, etc. These factors could be incorporated into tumor-on-a-chip models.
B.To generate a tumor-on-a-chip model for mimicking metastasis, multiple critical metastatic factors could be recreated in vitro for different metastatic responses to be found.

Promising avenues for the advancement of 3D tumor models encompass spheroids, tumor organoids, and microfluidic devices, among other strategies. Microfluidic devices represent a technology that integrates fundamental units of biological sample preparation, reaction, separation, and detection on microscale chips, drawing upon precision engineering, biomaterials, tissue engineering, and associated methodologies. These devices permit precise manipulation of minuscule fluid volumes through microchannels, facilitating fluid transfer between different compartments and automating detection and analysis. Microfluidic devices provide the capability to culture a diverse array of cell types, organs, and tissues on a single platform, enabling precise control over the quantity, arrangement, and spatial relationships of each constituent. By tailoring the composition of the microfluid, and in conjunction with techniques such as spheroid and organoid culture, complex, accurate, and functionally robust in vitro 3D models can be established. Furthermore, these devices offer features such as high throughput, customization, minimal sample consumption, and high efficiency. Importantly, microfluidic devices not only excel in recapitulating intricate tumor microenvironments (TME) but also possess the unique advantage of simulating microvascular systems, effectively mitigating the shortcomings associated with prior 3D models.

Design
To emulate the characteristic of early-stage primary colorectal cancer typically manifesting at a singular site within the colorectum, as well as effectively encapsulate the intricate microenvironment of colorectal cancer(i.e., mechanical peristalsis of colorectum, shear stress from fluids, aberrant angiogenesis in tumors etc.), we have developed a colorectal tumor-on-a chip (CRT-chip) based on microfluidic technology. The CRT-chip is designed to encompass three pivotal features. Firstly, it permits site-specific implantation and growth of CRC cells at specific locations to replicate the usual occurrence of primary CRC at a single site within the colorectum. Secondly, it introduces continuous low-speed flow and peristaltic-like deformations on the colorectal epithelium, thereby respectively simulating fluid shear and peristalsis within the human colorectum. Lastly, it integrates a lower channel beneath the tumor implantation site to simulate the function of tumor vasculature, specifically facilitating the transport of nutrients to colorectal cancer cells. Through numerical simulations and experimental validation, the operational parameters of the CRT-chip can be systematically optimized, further ensuring a high degree of precision in the simulation.

The CRT-chip comprises four closely bonded layers of polydimethylsiloxane (PDMS). From top to bottom, these layers are referred to as the inlet layer 1, simulation layer 2, culture layer 3, and infusion layer 5(Figure 3). Among these, the CRT-chip comprises three core functional layers(Figure 2), namely, the colorectal tract layer (simulation layer 2), the tumor cell cultivation layer (culture layer 3), and the tumor vascular layer (infusion layer 5).

The colorectal tract layer contained three parallel channels(Figure 2). The central channel 21 was designed to accommodate the human normal colon epithelial cells (NCM-460) to mimic the human colorectum. The two gas chambers 22 were connected to a computer-controlled multifunctional fluid controller to apply cyclic peristalsis-like mechanical deformation to the central channel. In this manner, the colon epithelial cells in the central channel 21 were exposed to both flow and peristalsis stress, which mimicked the dynamic physiological environment of human colorectum. The tumor cell cultivation layer contained a circular microchamber, of which the lower side was attached to the porous membrane layer with an array of micropores. The microchamber allowed site-specific implantation and cultivation of the human colon cancer cells (HCT-116). The tumor vascular layer contained a single channel that was perpendicular to the central channel and right below the microchamber. It was designed to mimic the tumor vasculature, which provided nutrients to tumor cells by permeation through the porous membrane layer.

Figure 2. The analogy between the CRT-chips and the in vivo CRC

Further structural details are provided in the following text(Figure 3):

(1)The simulation layer 2 is equipped with a central channel 21 and gas chambers 22 on either side of it.
(2)The culture layer 3 contains culture microchambers 31 that are connected to the central channel 21.
(3)The infusion layer 5 includes a tumor infusion channel 51 that is in communication with the culture microchambers 31.
(4)The inlet layer 1 includes the inlet 11 and outlet 12 of the central channel, inlets 15 for the gas chambers, and the inlet 13 and outlet 14 for the tumor infusion channel. The inlet 11 and outlet 12 of the central channel are respectively connected to the first and second ends of the central channel 21. The gas chamber inlet 15 is connected to the first end of the gas chambers 22. The inlet 13 and outlet 14 for the tumor infusion channel are respectively connected to the first and second ends of the tumor infusion channel 51.
(5)The simulation layer 2 includes first through-holes 23 that connect the inlet 13 and outlet 14 of the tumor infusion channel with the first and second ends of the tumor infusion channel 51.
(6)Furthermore, there is a porous membrane layer 4 located between the culture layer 3 and the infusion layer 5. The porous membrane layer 4 includes third through-holes 42 that connect the inlet 13 and outlet 14 of the tumor infusion channel with the first and second ends of the tumor infusion channel 51. The porous membrane layer 4 also contains microhole membranes 41 that are in communication with the culture microchambers 31. The microholes in the microhole membranes 41 have a diameter of 20 μm, with a spacing of 10 μm between them.
(7)In addition, the incorporation of the micro-porous membrane 41 serves the purpose of providing attachment points for colorectal cancer cells, facilitating their deposition within the culture microchambers 31. Additionally, as colorectal cancer cells fill the culture microchambers 31, they effectively separate the central channel 21 from the tumor infusion channel 51. Given that different culture media are used for culturing human colorectal epithelial cells and colorectal cancer cells, this separation allows for the supply of nutrients to tumor cells through the tumor infusion channel 51 and the provision of nutrients to epithelial cells through the central channel 21.

Figure 3. A disassembly diagram of the chip

After assembling the CRT-chip(Figure 4,5), colorectal cancer cells are first pumped through the inlet of the central channel 21, ultimately depositing within the culture microchambers 31 that are interconnected with the central channel 21. Subsequently, residual cells within the central channel 21 are washed away. Following this, human normal colorectal epithelial cells are pumped through the inlet of the central channel 21 and adhere to the entire central channel 21. This process enables the precise implantation of colorectal cancer cells at a specific location within the colon-like central channel 21, thus mimicking cancer cell invasion mechanisms.Cells are continuously supplied with cell culture medium at a controlled rate through both the central channel 21 and the tumor infusion channel 51. This provides cells with sustained shear stress, effectively simulating the mechanism of tumor vascular perfusion. Under the influence of the gas chambers 22, the CRT-chip replicates the mechanical peristalsis of the human colon. As a result, the tumor cells within the culture microchambers 31 invade the normal colorectal epithelial cells, and by regularly observing the cellular characteristics within the central channel 21, a more accurate representation of the developmental process of human colorectal cancer is achieved.

The central channel 21 is separated from the gas chambers 22 by a distance of 320 μm. The middle section of the gas chambers 22 runs parallel to the central channel 21, with the first and second ends of the gas chambers 22 positioned on one side, away from the central channel 21. The first end of the gas chambers 22 is connected to the air pump, serving as both the inlet and outlet for the gas chambers 22. The middle section of the gas chambers 22 functions as the primary pressure transfer zone, influencing deformation within the central channel 21 to simulate the mechanical peristalsis of the intestine. The positioning of the first and second ends of the gas chambers 22, away from the central channel 21, serves to distribute the stress generated at both ends, more closely resembling the authentic mechanical forces experienced by organs within the digestive tract as they compress against each other.

The central channel 21 is oriented perpendicularly to the tumor infusion channel 51, and the culture microchambers 31 are located at their vertical intersection. The perpendicular arrangement of the two channels effectively separates them, preventing the permeation or contamination of one channel by cells or culture medium from the other. Additionally, the placement of the culture microchambers 31 at the vertical intersection of the two channels ensures even distribution of force, without impeding the flow of cells or culture medium.

The central channel 21 has a diameter of 600 μm, the tumor infusion channel 51 has a diameter of 1200 μm, and the apertures of the culture microchambers 31 are 400 μm in diameter. The central channel 21, tumor infusion channel 51, and culture microchambers 31 of the CRT-chip respectively simulate the intestinal tract, vasculature, and sites of cancer cell invasion, aligning with the mechanisms and progression of colorectal cancer cell invasion.

Figure 4. The three-dimensional structure diagram of the chip

Figure 5. Top view (top) and front view (bottom) of the chip

The fabrication method of the CRT-chip
(1) Prepare chip molds using photopolymerization-based 3D printing technology. Place the individual mold layers in a plastic vacuum desiccator, which is then positioned inside a fume hood. Insert a pipette tip containing 10 μl of methyltrichlorosilane into the desiccator, quickly seal the lid, and stop when the pressure gauge on the plastic vacuum desiccator reaches 0.8 kg/cm². Allow the setup to stand for 12 hours.

(2) Prepare a prepolymer mixture of curing agent and polydimethylsiloxane (PDMS) at a 10:1 ratio. Pour the mixture onto the molds, place them in the plastic vacuum desiccator, and use a vacuum pump to remove air bubbles for 20 minutes. Cure the PDMS by placing the molds horizontally in an oven at 60°C for 2 hours. Subsequently, peel the cured PDMS from the molds to obtain PDMS substrates.

(3) Use a 1.6 mm diameter puncher to create holes at the inlet and outlet of the channels corresponding to the chip. Clean the chip surface with lint-free paper, then apply a small amount of PDMS onto a glass slide, spread it evenly using a roller, and coat the chip surface with a thin layer of PDMS. Align the various chip layers and the glass slide under a microscope, gently press them together, and place the assembly on a hot plate to bond. Finally, cut the resulting PDMS into microfluidic chips measuring 1.5 cm × 2.0 cm along the designed boundaries.

The method for establishing a dynamic model of human colorectal cancer using a fabricated CRT-chip
(1) Cultivate human normal colorectal epithelial cells (NCM460) and colorectal cancer cells (HCT116) conventionally at 37.0°C and 5% CO2 in a CO2 incubator. For cells adherent to the culture flask, use 0.25% trypsin to perform digestion, followed by the addition of fresh cell culture medium. After thorough pipetting and mixing, dilute the cell suspension to an appropriate concentration with fresh culture medium.

(2) Place the microfluidic chip in an autoclave for 1 hour of high-temperature sterilization. Subsequently, insert it into a plasma cleaner and clean it at 80 watts of power for 120 seconds. After completing the plasma cleaning, rapidly add a few drops of distilled water to the four inlets of the microfluidic chip. Then, introduce a 1% (w/v) gelatin solution into the chip, allowing it to stand for 40 minutes, followed by 30 minutes of ultraviolet sterilization.

(3) Initially, introduce a cell suspension of a certain concentration of colorectal cancer cells through a liquid pump into the central channel of the microfluidic chip. Stop when a certain number of cancer cells have settled in the culture chamber and wash away other cells outside the culture chamber. Subsequently, introduce a cell suspension of a certain concentration of human normal colorectal epithelial cells into the central channel of the microfluidic chip. Allow it to sit for 1 hour, enabling the colorectal epithelial cells to adhere to the chip walls.

(4) Transfer the microfluidic chip to the live cell workstation's loading platform, allowing cells to incubate at 37.0°C and 5% CO2. Continuously supply fresh cell culture medium at a rate of 1.2 μL/min into both the central channel and tumor perfusion channel.

(5) Connect the two side cavities of the microfluidic chip to an air pump and use computer control to cyclically compress the central channel with a pressure of 300 mbar and a frequency of 0.1 Hz to simulate the mechanical peristalsis of the colorectal region. Regularly sample from the central channel to obtain a dynamic model of human colorectal cancer.

Conclusion
We have developed a multi-layered CRT-chip with three major functions. Firstly, it allows for the targeted implantation and growth of CRC cells, with continuous validation and optimization through numerical simulations and experimental procedures. Secondly, it introduces continuous flow and pressure-induced deformation on colonic epithelium to mimic the fluid shear and peristalsis within the human colon. To better emulate in vivo physiological conditions, numerical and experimental optimization can be carried out for fluid flow rates and applied pressures. Lastly, a dedicated bottom channel is designed to simulate the functionality of tumor vasculature, facilitating the delivery of nutrients to CRC cells. Based on this pathophysiological CRT-chip model, the therapeutic efficacy of ICG is validated and quantified.

Subsequently, our team will continually validate and optimize the CRT-chip, not only for its application in studying the occurrence and dynamic development mechanisms of colorectal cancer but also for further employing the CRT-chip in the validation of the mechanisms of action of Bifidobacterium longum, and to explore the overall dynamic development of colon cancer-probiotic populations within the intestine after the introduction of engineered bacteria.