Team:Tsinghua/Project
How aflatoxin hazards people
Class
I
cancerogen
Aflatoxin is known as the most toxic carcinogen.
It can covalently bind
adenine in DNA, thus causing mutations.
In nature
12
kinds of Aflatoxin
with in which B1 is the most toxic
and can be found in maize, peanuts, cotton seeds and many nuts.
Up to
155,000
liver cancer-incidents due to Aflatoxin
It can also cause stomach
cancer and esophageal cancer
how YeasyAFT can test
and how they help people?
Quick
detection
The detection of AFT can be accomplished in hours,
which is distinguished from those prevalent biochemical methods.
Low
cost
Utilization of genetic-engineered
yeast allows quite low cost.
High
sensitivity
We use single-chain Fv antibody to detect AFT
, and the signal is enlarged by yeast-two-hybrid system and hexose transporter.
Background
Aflatoxins (AFT)
Aflatoxins (AFT) are poisonous carcinogens that are produced by certain molds (Aspergillus flavus and Aspergillus parasiticus) which grow in soil, decaying vegetation, hay, and grains. They are regularly found in improperly stored staple commodities such as cassava, chili peppers, corn, cottonseed, millet, peanuts, rice, sesame seeds, sorghum, sunflower seeds, tree nuts, wheat, and a variety of spices. When contaminated food is processed, aflatoxins enter the general food supply where they have been found in both pet and human foods, as well as in feedstock for agricultural animals. Animals fed with contaminated food can pass aflatoxin transformation products into eggs, milk products, and meat[1]
History of AFT
AFT was coined around 1960 after its discovery as the source of "Turkey X disease"[2], which killed about 100,000 turkeys in England. In animals, after oxidized by cytochrome P450 enzyme in liver cells, AFT can covalently bind adenine in DNA, thus causing mutations. It is listed in Group I carcinogen by the International Agency for Research on Cancer (IARC)[3]. The carcinogenicity of AFT is 900 times more than that of dimethylamino azobenzene, one of the most famous carcinogen. In China, incidence and mortality of stomach cancer, esophageal cancer, and liver cancer are extremely high, second only to lung cancer.
So far, methods for mycotoxins detection mainly include thin layer chromatography, liquid chromatography, and immunological methods, such as enzyme-linked immunosorbent assay (ELISA), membrane-based immunoassay, fluorescent polarization, and so on[4]. However, these time-consuming methods cannot meet our demands for daily detection, taking into consideration that AFT can be nearly everywhere. What we need is to detect AFT in everyday foods such as cooking oil. To achieve such detection in our daily life, a simple and straightforward method is in great demand.
overview
With the rapid development of synthetic biology, we propose to construct a biosensor system for AFT detection. Saccharomyces cerevisiae is chosen as the model organism to design such a biosensor, for its rapid growth speed, minimal pathogenicity, ensembled selectable markers, well-defined genetic system, and most importantly, highly versatile DNA transformation system together with a highly efficient homologous recombination system[5]. Furthermore, we can easily make it into dry powder or test paper for daily use.
The system is composed of two components: sensor part and reporter part. In sensor part, we use single-chain Fv antibody (scFv) against AFT, fused with “trigger protein”, as a monitor. Once ScFv detects AFT, it can activate downstream gene expression in reporter part. For the convenience of readout, we choose hexose transporter (HXT) as the reporter gene.
Videos
Design
Sensor part
The sensor part of the AFT detection system was based on the yeast two-hybrid technique. The DNA-binding domain (BD) and the transcription-activation domain (AD) of GAL4, the activator, were incorporated with two different single-chain antibodies of AFT (ScFv1 and ScFv2), correspondingly. Without external AFT, GAL4 the activator was incomplete and unable to activate the expression of the downstream structure gene, hexose transporter (HXT). When AFT was added to the culture, the transcription-activation domain would be recruited to the promoter through the interactions between AFT and both single-chain antibodies, and HXT would be produced.
Reporter part
We chose HXT as the reporter gene because the change in the glucose concentration of the culture would be easy to detect by glucometer. To ensure that the autonomous glycometabolism would not interfere with our system, auxotroph that lacked glucose transporters was used, which utilized the maltose instead.
The auxotroph was cultivated in a culture with glucose as the only carbon source. Without the existence of AFT in the culture, the auxotroph would be unable to transport glucose, thus no significant change in the glucose concentration of the culture would be observed. When AFT was added, HXT would be produced in the auxotroph and glucose would be transported and utilized, thus a change in the glucose concentration of the culture would be detected by the glucometer. Furthermore, the concentration of AFT in the culture would be reflected by the consumption rate of the glucose.
Result
Sensor
The sensor part of our system includes two single-chain Fv antibodies (ScFv), fused with the activating domain (AD) and DNA binding domain (BD) from the yeast-two-hybrid system, respectively. Upon interacting with AFT, the two antibodies can pull AD to BD and the two domains, when functioning together, can activate the GAL promoter and its downstream reporter gene.
To successfully pull AD and BD together when and only when AFT is present, we need the two AFT antibodies to bind AFT jointly but not sequentially, which is different from most antibodies available in the ELISA kit. Consequently, we got our two ScFvs from disparately published source.[1-3] It has to be noted that, to get the ScFv2, we assembled the two chains of the published antibody sequences by ourselves and optimize the codons at the same time. The sequence of the two ScFvs was sent to synthesis in Wuxi Qinglan Biotech Co. Ltd. Upon receiving the DNA of the two ScFvs, we fused AD to ScFv1 and BD to ScFv2. It has to be noted that, because BD functions at its N-terminal, we fused AD and BD at the N-terminal of ScFvs.
Figure1 The design and validation of sensor system
A) The two ScFvs bind to AFT jointly but not sequentially
B) The two ScFvs are fused with either AD or BD and driven by constitutive promoter
C) His3 is used to test ScFv-AFT binding and activating capacity
D) The growth rate at 10hr, 15hr and 20hr, beginning with 0.15 OD or 0.015OD. In this experiment, yeast described in (C) is cultured in Sc-His, and 600nm OD is measured. The growth rate is calculated by dividing the OD value with the OD value in the first hour.
To see whether AD-ScFv1 and BD-ScFv2 function well with AFT present, we transfected the plasmids containing AD- ScFv1 and BD-ScFv2 into the JDY26 yeast strain, in which the intrinsic Gal4 and Gal80 are knocked out and a His selection marker is knocked in after the Gal1 promoter (named pGal1). It has to be noted that the yeast strain was generated previously by Junbiao Dai and his team. We expect when AFT is present, the ScFvs pull AD and BD together and activate the His3 gene at the downstream of pGal1. Then we can add yeast culture medium without His (Sc-His) and test the binding of ScFvs with AFT by monitoring the growth of yeast.
When we carried out the validation experiment, the growth of yeast is measured by 600nm OD, detected once per hour for 20 hours, and different concentrations of both yeast and AFT, with 3 replicas for each concentration, is tested. We calculated the growth rate, which is defined by the 600nm OD fold change compared to initial value, at 10, 15 and 20 hours. As it turns out, the growth rate at 1/10, 1/20, 1/40 are all larger than that without AFT added (defined as a negative control), with 1/20 at the peak. However, we’ve noticed that when we dilute AFT to 1/5 fold, the growth rate is similar to that of the negative control. We speculate that this may due to the toxicity of AFT to yeast. In conclusion, we’ve proved that the two ScFv can bind to AFT and their AD/BD can activate pGal1.
Reporter
The reporter part of our system includes a GAL promoter followed by hexose transporter (HXT). Upon activated by AD and BD, the GAL promoter can expression HXT, which will be distributed to the cell membrane and transfer glucose. When glucose is consumed, the decreased concentration can be measured by commercial glucometer.
We were firstly supposed to make sure that the HXT gene we used can be expressed normally and the transporter will be distributed onto the membrane afterward. In order to confirm our expectation, we transfected our target gene, either HXT2 or HXT5, into an HXT-related gene-knock-out yeast strain (EBY.VW4000), which cannot transport external glucose into the cell without the help of exogenous gene. We then carried out the serial dilution experiments on a solid medium which uses glucose as the only carbon source, using these two kinds of yeast as well as negative control. Theoretically, only the yeast with normal expression and distribution of HXT can grow on the medium. We were glad to see that proper colonies had grown on the medium, which demonstrates that our HXT genes have the proper function. We randomly selected HXT2 as the reporter gene.
B) Serial dilution result of HXT2/5 expressing yeast. The plate uses glucose as the only carbon source, so the yeast can only survive when HXT functions well. The first line is negative control, labeled “N.C.”
C) The accuracy of commercial glucometer is tested. The readouts form a relatively good linear relationship with glucose concentration. The spots are measured results and the dotted lines are regression curve. The solvent is noted in the graph.
D) The generation of the JDYY001 yeast strain. GAL4 and GAL80 in EBY.VW4000 is knocked out by homologous recombination.
E) Culturing yeast can obviously decrease glucose concentration. In this experiment, HXT is constitutively expressed. We get two replicas, each with a yeast concentration of about 0.6 OD. The yeast was diluted to 1/2, 1/4, 1/8 of the initial concentration, and the dilution coefficient is used as sample name. The glucose concentration is normalized to the initial concentration first, and then averaged between two replicas
It has to be noted that, to make EBY.VW4000 appropriate for the sensor part, we need to delete Gal4 and Gal80 first. So we transfer linear DNA, with homologous arms at both ends but no Gal4 or Gal80 inside, into the strain and pick the clone that has Gal4 and Gal80 depleted due to homologous recombination. The strain we get is named JDYY001. Following experiments are carried out in this yeast strain.
Till now, we still hold the concern that, upon yeast culture, the concentration change of glucose may not be suitable for commercial glucometer detection. So we transfect construct that can constitutively express HXT2 to JDYY001, and cultured them in the liquid medium, the glucose concentration of which had been tested by glucometer before. After incubated under 30℃, we can see that the concentration decrease of glucose is obvious upon glucometer. Moreover, it is evident that the higher concentration of the initial yeast in the medium and the longer we incubate, the more glucose concentration decrease, which is quite reasonable. Further modeling can be seen in the Model part.
Integration
Now that our sensor and reporter systems are all proved to work well, we need a well-qualified GAL promoter to link them together. Previously Junbiao Dai and his team have developed an artificial Gal promoter, named pGALS. Since the system is well developed, we decided to use pGALS as our Gal promoter. We cloned it on the upstream of HXT2 and adjusted the marker to differentiate it from that of the two ScFvs’ plasmids. To make sure that pGALS really works, we transfected pGALS-HXT2 and Gal4 into JDYY001 and cultured it on the plate with glucose as its only carbon source. The yeast does grow.
Then we want to validate that pGALS can also a response to ScFv-fused AD/BD and see the relationship between pGALS gene expression efficiency and AFT concentration. So we add GFP at the downstream of pGALS, and transfer it into the JDYY001, along with the two ScFv’s plasmids. We are still waiting for the confocal data and this will be shown in the final presentation.
Then we transfer the two ScFv’s plasmids and the pGALS-HXT plasmid into JDYY001 and monitor the glucose concentration change, with different AFT and yeast concentration applied. We are still waiting for the data and this will be shown in the final presentation.
reference
[1] Li, X., Li, P. and Zhang, Q., Molecular analysis of monoclonal antibodies against aflatoxins: a prediction of functional amino acid which influences antibody affinity. Chinese Academy of Agricultural Sciences, 2011.[2] Pei, S.a.S., D., Construction and identification of mouse Anti-AFB1 phage single-chain antibody library. Food and Biological Engineering College, 2010.
[3] Li, X., Li, P. and Zhang, Q., Molecular characterization of monoclonal antibodies against aflatoxins: a possible explanation for the highest sensitivity. Chinese Academy of Agricultural Sciences, 2011.
[4] Fratamico P M, Bhunia A K, Smith J L, et al. Foodborne pathogens: microbiology and molecular biology.[J]. Foodborne Pathogens Microbiology & Molecular Biology, 2005, 6:334.
[5] Wannop C C. The histopathology of Turkey 'X' disease in Great Britain.[J]. Avian Diseases, 1961, 5(4):371-381.
[6] IARC Monographs on the evaluation of carcinogenic risks to human. Some traditional herbal medicine, some mycotoxins, naphthalene, and styrene. IARC, France.
[7] Placinta CM, D’Mello JPF, Macdonald AMC (1999) A review of worldwide contamination of cereal grains and animal feed with Fusarium mycotoxins. Anim Feed Sci Technol 78:21–37
[8] Sherman, F. (1991). Getting started with yeast. Methods in Enzymology, 194, 3-21.
Start following YeasyAFT on wechat now
