SPR Biacore molecular interaction

In System 1, we initially conducted Surface Plasmon Resonance (SPR) Biacore molecular interaction assays to screen which small molecule could better bind to the LilrB3 receptor. Based on the simulated molecular docking results provided by the modeling group (for details, see model), we selected candidate substances that are more suitable for human use and have fewer side effects: tetrahydrofolate and beta-carotene.

We procured LilrB3 protein, tetrahydrofolate, and beta-carotene small molecules, as well as GM-5 chips, and performed the experiments on the SPR Biacore molecular interaction instrument. Initially, we immobilized the LilrB3 protein onto the GM-5 chips using the instrument. Subsequently, we prepared solutions of tetrahydrofolate and beta-carotene at corresponding concentration gradients for experimental analysis.

In the experiment with tetrahydrofolate and LilrB3 protein, the results indicated that the interaction between the LilrB3 protein and tetrahydrofolate is consistent with the modeling predictions, with a dissociation constant (KD) value of 3.183 × 10^-8 M. (Figure 1-2)

Figure 1. Result of interaction between lilrb3 protein and Tetrahydrofolate-1

Figure 2. Result of interaction between lilrb3 protein and Tetrahydrofolate-2

This indicates that tetrahydrofolate can effectively bind to the LilrB3 protein.

However, the results for beta-carotene were not as expected. As the concentration increases, the binding rate decreases, indicating nonspecific binding. This suggests that in this system, there may be no significant interaction between beta-carotene and the LilrB3 protein. This could be due to the lipophilic nature of beta-carotene, which may not be conducive to its binding with the LilrB3 protein under the conditions used in this system. (Figure 3)

Figure 3. Result of interaction between lilrb3 protein and beta-carotene

Based on the aforementioned results, we have decided to select tetrahydrofolate as the binding molecule for the LilrB3 protein.

Our objective is to enhance the expression of tetrahydrofolate (THF) to target the LilrB3 receptor. As a critical target in Alzheimer's disease (AD), the receptor LilrB3 strongly binds to APOE4. By disrupting the APOE4-LilrB3 binding, this system may relieve the formation and deposit of amyloid protein to ameliorate AD pathology.

Figure 4. Tetrahydrofolate Module Mechanism Diagram [1]

We choose Lactobacillus plantarum L168 as the chassis and pSIP403 as the vector. By searching its synthetic pathway, we found the key enzyme folKE and its coding genes folK and folE. Then, we inserted the gene folKE into our designed plasmids to complete the construction and synthesis of plasmids. In view of biosafety, we designed a kill switch which is activated under hypoxic conditions, include a Hypoxia-inducible promoter, a RBS, a coding of SrpR repressor that inhibits transcription of the SrpR-dependent promoter and a terminator.

DNA level

Lactobacillus plantarum L168, a strain we obtained from our PI, Xingyin Liu's lab, has been proven to be able to alleviate social behavior deficits and reduce gut inflammation in animal models.[2,3]

We constructed the pSIP403-P9-folKE in DH5α, transformed it into Lactobacillus plantarum L168, and extracted the plasmid to prove the success of our transformation.

Initially, we designed primers to linearize the plasmid and amplify the target gene from the genome of L168 via PCR. The vector utilized in this process was pSIP403. For homologous recombination, we strategically designed primers and conducted PCR using the homology arm located on the plasmid backbone.

Primer	Sequence
HWY86F	tctagactcgaggaattcggtacc
HWY86R	atctaaaatctccttgtaatagtattttatagaatacatatatgctggc
HWY69F	atactattacaaggagattttagatATGGCAAGTAGGGAAGAACGG
HWY97R	cgaattcctcgagtctagaCTACTTTGCGATTCGCTGTAAGAATTCT

Then we transformed it into E.coli DH5α. To confirm the efficacy of our transformation, we devised primers, conducted plasmid PCR and colony PCR, and subsequently analyzed the results via agarose gel electrophoresis.

Function Verification

We achieved successful plasmid transformation into L168, and the validation through PCR and agarose gel electrophoresis confirmed the success of this process. Subsequently, we cultured L168 and performed RNA extraction. At the same time, the culture supernatant was employed for metabolite measurement.

Subsequently, reverse transcription and quantitative polymerase chain reaction (qPCR) experiments were conducted to quantify the gene expression levels of L168 that contained the pSIP403-P9-folKE plasmid or not.

As shown in the figure 5, these are our qPCR results.

Figure 5. Relative Expression Analysis of genes folK and folE by qPCR.

The left group represents two different transcription levels of the gene folK. In this group, the right column shows the baseline L168 in its normal state, serving as the control group, which harbors the empty vector pSIP403. We can see the successful expression of the gene folK in L168 from the left column with the significantly increased level of the gene folK. The right group represents two different transcription levels of the gene folE, indicating the preliminary effect of the gene folE in L168.

The above experiment analyzed the relative expression levels of two genes using qPCR. The figure illustrates that the relative expression of two genes (folK and folE) was significantly upregulated in pSIP403-P9-folKE plasmid in L168 compared to the control group (p ＜ 0.0001).

We also used Liquid Chromatograph Mass Spectrometer (LC-MS) to examine the production of THF in the supernatant of L168.

As shown in the figure 6, these are our LC-MS results.

Figure 6. Relative levels of THF between two groups by LC-MS.

It's easy to see the level of THF increased dramatically in L168 containing the pSIP403-P9-folKE plasmid compared to the control group, demonstrating that as expected, the folKE gene was successfully expressed in L168 and produced THF.

SPR Biacore molecular interaction

In System 2, we initially conducted Surface Plasmon Resonance (SPR) Biacore molecular interaction assays to screen which small molecule could better bind to the LilrB3 receptor. Based on the simulated molecular docking results provided by the modeling group (for details, see model 【放model链接】), we selected candidate substances that are more suitable for human use and have fewer side effects: inosine, uridine 5'-monophosphate and flavin adenine dinucleotide.

We procured nNOS protein, inosine, uridine 5'-monophosphate and flavin adenine dinucleotide small molecules, as well as GM-5 chips, and performed the experiments on the SPR Biacore molecular interaction instrument. Initially, we immobilized the nNOS protein onto the GM-5 chips using the instrument. Subsequently, we prepared solutions of inosine, uridine 5'-monophosphate and flavin adenine dinucleotide at corresponding concentration gradients for experimental analysis.

In the experiment with inosine and nNOS protein, the results indicated that the interaction between the nNOS protein and inosine is consistent with the modeling predictions, with a dissociation constant (KD) value of 2.524×10-6, with chi2 within a credible range.(Figure 1-2)

Figure 1. Result of interaction between nNOS protein and inosine-1

Figure 2. Result of interaction between nNOS protein and inosine-2

This indicates that inosine can effectively bind to the nNOS protein.

However, the results for uridine 5'-monophosphate were not as expected. As the concentration increases, the binding rate decreases, indicating nonspecific binding. This suggests that in this system, there may be no significant interaction between uridine 5'-monophosphate and the nNOS protein.

Figure 3. Result of interaction between nNOS protein and uridine 5'-monophosphate-1

Repeating the experiment with increased uridine concentration, the results are still unsatisfactory.

Figure 4. Result of interaction between nNOS protein and uridine 5'-monophosphate-2

Figure 5. Result of interaction between nNOS protein and uridine 5'-monophosphate-3

After our discussion with our teacher, we thought that the high refractive index of uridine itself might be the reason it's not suitable as a component of the mobile phase. Consequently, the SPR molecular interaction assay may not effectively capture the binding affinity between uridine and the protein.

Ultimately, we proceeded to screen the last candidate small molecule, flavin adenine dinucleotide (FAD), for its molecular interaction with neuronal nitric oxide synthase (nNOS) protein. However, we encountered some challenges: flavin adenine dinucleotide is lipid-soluble and was dissolved in DMSO. Consequently, we utilized a PBS buffer system containing 0.02% DMSO for our experiments.

Figure 6. Result of interaction between nNOS protein and flavin adenine dinucleotide-1

Figure 7. Result of interaction between nNOS protein and flavin adenine dinucleotide-2

The results indicated that the outcomes were not satisfactory. The lack of significant interaction observed in this buffer suggests that the experimental results might have been affected by DMSO buffer.

Based on the aforementioned results, we have decided to select inosine as the binding molecule for the nNOS protein.

Our objective is to enhance the expression of inosine to target the neuronal nitric oxide synthase (nNOS) receptor. Increased SERT-nNOS coupling in the dorsal raphe nucleus is a significant trigger for depression, so we overexpress gsk to produce inosine, which will bind to the nNOS receptor to uncouple SERT from nNOS.

We identified the sequence of the key gene gsk in the synthesis of inosine, and designed both upstream and downstream primers for gsk. The selected vector for this purpose was pSIP403. Subsequently, we designed and constructed the pSIP403-P9-gsk plasmid in the DH5α prior to introducing it into the sensory Lactobacillus plantarum L168 to produce inosine. Moreover, we also add the kill switch mentioned above into the vector for biosafety.

DNA level

Lactobacillus plantarum L168, a strain we obtained from our PI, Xingyin Liu's lab, has been proven to be able to alleviate social behavior deficits and reduce gut inflammation in animal models.[1,2]

We generated the pSIP403-P9-gsk plasmid in DH5α, subsequently performing its transformation into Lactobacillus plantarum L168 and conducting plasmid extraction to confirm the successful transformation.

Initially, we designed primers to linearize the plasmid and amplify the target gene from the genome of L168 via PCR. The vector utilized in this process was pSIP403. For homologous recombination, we strategically designed primers and conducted PCR using the homology arm located on the plasmid backbone.

Primer	Sequence
HWY86F	tctagactcgaggaattcggtacc
HWY86R	atctaaaatctccttgtaatagtattttatagaatacatatatgctggc
HYY96F	cgaattcctcgagtctagaTTAACGATCCCAATAACTTTCTTCCAAACTATCT
HWY87R	atactattacaaggagattttagatATGAAGTTTCCAGGTAAACGGAAGAGT

Then we transformed it into E.coli DH5α. To confirm the efficacy of our transformation, we devised primers, conducted plasmid PCR and colony PCR, and subsequently analyzed the results via agarose gel electrophoresis.

Function Verification

We achieved successful plasmid transformation into L168, and the validation through PCR and agarose gel electrophoresis confirmed the success of this process. Subsequently, we cultured L168 and performed RNA extraction. At the same time, the culture supernatant was employed for metabolite measurement.

Subsequently, reverse transcription and quantitative polymerase chain reaction (qPCR) experiments were conducted to quantify the gene expression levels of L168 that contained the pSIP403-P9-gsk plasmid or not.

As shown in the figure 8, these are our qPCR results.

The right column (L168) represents the baseline L168 in its normal state, serving as the control group. The qPCR results elucidate that our designed plasmid effectively and significantly upregulated the gene expression of inosine in L168.

Figure 8. Relative Expression Analysis of gene gsk by qPCR.

We also used Liquid Chromatograph Mass Spectrometer (LC-MS) to examine the production of inosine in the supernatant of L168.

As shown in the figure 9, these are our LC-MS results.

Figure 9. Relative levels of inosine between two groups by LC-MS.

It’s easy to see the level of inosine increased dramatically in L168 containing the pSIP403-P9-gsk plasmid compared to the control group, demonstrating that as expected, the gsk gene was successfully expressed in L168 and produced inosine.

There is sufficient research evidence indicating that L-theanine and niacin have alleviating effects on depression and Alzheimer's disease, both of which also exist in the supernatant of L168 according to our previous experimental data. The key gene of theanine synthesis glnA and the key gene of niacin pncA were independently integrated into two different plasmids, namely pSIP403-P9-glnA and pSIP403-P9-pncA. We identified the sequences of the key genes glnA and pncA, and designed both upstream and downstream primers for them. The selected vector for this purpose was pSIP403. Subsequently, we designed and constructed the two plasmids in the DH5α prior to introducing it into the sensory Lactobacillus plantarum L168 to produce theanine and niacin.

DNA level

Lactobacillus plantarum L168, a strain we obtained from our PI, Xingyin Liu's lab, has been proven to be able to alleviate social behavior deficits and reduce gut inflammation in animal models.[1,2]

We generated and verified the pSIP403-P9-glnA plasmid and pSIP403-P9-pncA plasmid in DH5α, subsequently performing their transformation into two groups of Lactobacillus plantarum L168 and conducting plasmid extraction to confirm the successful transformation.

Initially, we designed primers to linearize the plasmid and amplify the target genes from the genome of L.plantarum L168 via PCR. The vector utilized in this process was pSIP403. For homologous recombination, we strategically designed primers and conducted PCR using the homology arm located on the plasmid backbone.

Then we transformed it into E.coli DH5α. To confirm the efficacy of our transformation, we devised primers, conducted plasmid PCR and colony PCR, and subsequently analyzed the results via agarose gel electrophoresis.

Function Verification

We achieved successful plasmid transformation into L168, and the validation through PCR and agarose gel electrophoresis confirmed the success of this process. Subsequently, we cultured L168 and performed RNA extraction. At the same time, the culture supernatant was employed for metabolite measurement.

Subsequently, reverse transcription and quantitative polymerase chain reaction (qPCR) experiments were conducted to quantify the gene expression levels of L168 that contained the recombinant plasmid or not.

As shown in the figure 1, these are our qPCR results. The left group represents the transcription level of glnA, and the right one represents pncA. The qPCR results indicate that our designed plasmid effectively and significantly upregulated the key gene expression of both theanine and niacin in L168 compared to the control group respectively.

Figure 1. Relative Expression Analysis of genes glnA and pncA by qPCR.