TNF expression studied with fluorescent reporter system in vitro and in vivo
Tumor necrosis factor (TNF) is a pro-inflammatory and immunoregulatory cytokine that plays an important role in protection against pathogens. However dysregulation and excessive TNF production may cause autoimmune and inflammatory pathologies. Study of TNF biology in disease may play an important role in understanding disease mechanisms. Fluorescent imaging provides an opportunity to visualize components of the immune system in vivo. One of the most common approaches for the in vivo imaging is utilization of fluorescent proteins that are bound or fused to the target molecule, or are co-expressed with it. Fused fluorescent proteins composed of the single-domain antibody against TNF and the red fluorescent protein Katushka might be used for evaluation of expression levels and localization of TNF production . In this study we used two genetic constructs encoding fluorescent proteins VHH41-kat and VHHAnti-hTNF-kat. VHH41-kat protein is able to bind TNF, whereas the protein VHHAnti-hTNF-kat not only binds but also neutralizes the biological activity of TNF.
The first aim of this study was to identify the optimal strain of E. coli to produce antibodies VHH41-kat and VHHAnti-hTNF-kat. The expression levels of antibodies in the following E. coli strains: Rosetta 2 (DE3) pLysS, BL21 (DE3), BL21 Codon Plus (DE3), Lemo21 (DE3) and B834 (DE3) were compared. The BL21 (DE3) was chosen as the most appropriate E. coli strain for production of the studied proteins.
To measure the blocking activity of fused fluorescent proteins in vitro the cytotoxic MTT-test with mouse fibrosarcoma cell line WEHI 164 was used. Recombinant human TNF was added to the cell culture at the concentration 200U/ml. Serial dilutions of tested proteins were added to the cell culture in concentration range from 2pM to 1μM per well. Commercial TNF blocker Remicade was used as a control. According to these measurements fluorescent blocker VHHAnti-hTNF-kat possessed a strong anti-TNF activity (LD50 = 3917 pM), while VHH41-kat did not.
To test the ability of VHH41-kat and VHHAnti-hTNF-kat to bind macrophage TNF, the cytofluorometric analysis was performed using bone marrow derived macrophages (BMDM) from humanized TNF mice. Macrophages were activated by LPS (100 ng/ml for 4 hours) in the presence of Brefeldin A and stained for TNF using intracellular cytokine staining protocol. The cytofluorometric analysis showed that both fluorescent antibodies were able to bind TNF produced by macrophages. The fluorescence intensity of VHHAnti-hTNF-kat was higher than the fluorescence intensity of VHH41-kat. Thus, VHH41-kat antibody might be used as a marker of TNF production, and the antibody VHHAnti-hTNF-kat may be considered as a theranostics agent due to its high fluorescence intensity and the ability to block TNF.
To evaluate the blocking activity of the fluorescent proteins in vivo experimental model of acute hepatotoxicity in the humanized mice was used.
Mice that were injected intraperitoneally (i/p) with 150pM/g and 300 pM/g of VHHAnti-hTNF-kat 30 minutes after i/p injection of LPS and D-galactosamine (400 ng/g and 800 μg/g, respectively) remained viable until the end of the experiment (24 hours). In the group of animals receiving the lowest dose (75 mg/g) a partial mortality was observed.
Mice that were injected with 150pM/g and 600 pM/g of VHH41-kat or PBS (as a control) died 6 hours after LPS and D-galactosamine administration. Partial lethality was observed in the group which received 300pM/g of VHH41-kat. Thus, VHHAnti-hTNF-kat at 150 pM/g provides 100% protection of mice.
In a parallel in vivo approach we used transgenic "reporter" animals which are able to endogenously co-express TNF and the reporter protein. Specifically, in the reporter TNF-2A-Kat mice immune cells produced red fluorescent protein Katushka under the control of regulatory sequences of the TNF gene. This model allows to follow TNF expression by fluorescent in vivo imaging and can be used for detailed study of TNF physiological functions in normal conditions and in TNF-dependent pathologies . Embryos were analyzed on the 17th day of gestation. Analysis of the fluorescence signal of newborns was performed for three days after birth.
Fluorescent analysis of embryos, adult healthy animal organs and newborns of TNF-2A-Kat reporter mice and wild type mice did not show any difference in the intensity of tissue fluorescence. The lack of significant differences was presumably due to the absence of microbiota.
On the contrary, the analysis of TNF production in the adult healthy TNF-2A-Kat reporter mice clearly showed tissue fluorescence presumably due to skin microbiota. Antibacterial drug Neosporin® was applied to the open skin area on the left shoulder of adult TNF-2A-Kat reporter mice and wild type mice for 9 days. A statistically significant difference in fluorescence intensity between normal skin area and the site of Neosporin® application was detected. We concluded that skin microbiota contributes to constitutive levels of TNF expression in adult healthy TNF-2A-Kat reporter mice.
 G. Efimov, Z. Khlopchatnikova, A. Sazikin, M. Drutskaya, A. Kruglov, E. Shilov,A. Kuchmiy, S. Nedospasov, and S. Tillib, “Isolation and characteristics of a new recombinant single domain antibody that specifically binds to human TNF,” Russ J Immunol, 2012, vol. 6(5), №4, pp. 337–345.
 Y. V Shebzukhov, A. A. Kuchmiy, A. A. Kruglov, F. Zipp, V. Siffrin, and S. A. Nedospasov, “Experimental Applications of TNF-Reporter Mice with Far-Red Fluorescent Label,” in The TNF Superfamily: Methods and Protocols, J. Bayry, Ed. New York, NY: Springer New York, 2014, pp. 151–162.