Development of novel chimeric toxins for targeted therapy by genetic engineering

 
Principal Investigator :  Janendra K Batra

Ph D Students
Divya Seth
Paroma Ghosh
Punyatirtha Dey
Shveta

Collaborators
Raj K Puri, CBER, FDA, USA

Design and development of recombinant protein toxins for targeted therapy is the theme of research. Members of fungal ribotoxin, plant ribosome-inactivating proteins, and ribonuclease-A families are being analysed for structure-function relationships to understand their molecular mechanism of action with an aim to design knowledge-based chimeric toxins.

A.    Studies on the molecular mechanism of action of ribonucleolytic and ribosome-inactivating protein toxins

The objective of the project is to investigate structure-function relationship of Aspergillus ribotoxin, restrictocin; ribosome inactivating protein, saporin; human pancreatic ribonuclease, and eosinophil-derived neurotoxin to understand their mechanism of catalysis and cytotoxicity.

Saporin has been shown to bind to a2-macroglobulin receptor, also called as low-density-lipoprotein-receptor-related protein (LRP), expressed particularly in fibroblasts, monocytes and hepatocytes. LRP is responsible for the uptake and clearance of macromolecular complexes between proteinases and a2-macroglobulin. We showed earlier that the cytotoxic activity of saporin-6 results due to the combined manifestation of its N-glycosidase and internucleosomal DNA-fragmentation activities. The mechanism of saporin cytotoxicity was further investigated using recombinant saporin-5 and saporin-6 isoforms, expressed in E. coli and purified from the inclusion bodies. Saporin-6 was found to be about ten-fold more active than saporin-5 in inhibiting in vitro protein synthesis. Both the isoforms specifically depurinated the target rRNA, however, saporin-5 had a lower activity compared to saporin-6. In U937 cells compared to saporin-6, saporin-5 appeared to be significantly less active in causing genomic DNA fragmentation. The cytotoxic activity of saporin isoforms was studied on LRP-positive cell lines, J774A.1 and U937, and LRP-negative cell lines HUT 102 and HeLa. Saporin-6 was more active than saporin-5 on all cell lines tested; the difference of activity between the two isoforms varied between 5-20-fold. However, we found LRP-negative cell lines to have sensitivities similar to LRP-positive cell lines towards saporin toxicity. To find out if the difference in activities of saporin isoforms on various cell lines was due to a difference in their internalization efficiencies, the cytotoxicity of isoforms was tested on HeLa cells permeabilised by adenoviral infection. Saporin-6 was 8-fold more active than saporin-5 in adenovirus infected cells indicating the difference in the cytotoxic activity to be a reflection of the difference in the catalytic activity of the two isoforms. To further investigate the involvement of LRP in the cytotoxic activity of saporin, the toxicity of saporin-6 was assayed on LRP-positive chinese hamster ovary cell line, CHO-K1, and its mutant CHO-13-5-1, which has been shown to have no detectable LRP mRNA or protein. Both CHO-K1 and CHO-13-5-1 cells were found to have similar sensitivities towards saporin-6 toxicity suggesting that binding and internalization of saporin are not mediated through LRP. In addition, the similar sensitivities of LRP-positive and LRP-negative cells towards saporin toxicity indicate the entry of toxin into LRP-negative cells through receptors or mechanisms other than LRP-mediated endocytosis.

To investigate if the intracellular translocation of saporin-6 requires acidic endosomal pH and the involvement of Golgi, cytotoxicity of saporin-6 was assayed in the presence of ammonium chloride or brefeldin-A on J774A.1 cells. Brefeldin-A did not affect the toxicity of saporin-6, whereas ammonium chloride protected the cells from saporin-6 toxicity indicating that saporin-6 requires acidic endosomal pH for the manifestation of its cytotoxicity, however, intact Golgi is not required. To investigate if saporin translocates to nucleus to degrade DNA, J774A.1 cells were treated with iodinated saporin-6 for different time periods and presence of radio-labelled protein was checked in the nuclear, cytosolic and membrane fractions. The concentration of saporin-6 increased in cytosol up to 6 hours, decreased gradually and became negligible by 16 hours. The membrane fraction did not show any significant amount of protein at any time point. The concentration of the protein in the nuclear fraction was found to be comparable to that in the cytosolic fraction up to 9 hours, however, by 16 hours concomitant with the decrease in cytosol, most of the labeled saporin-6 localized in the nucleus. The study shows that after internalization, initially the protein stays in the cytosol and subsequently migrates to the nucleus.

Eosinophil-derived neurotoxin (EDN) and eosinophil cationic protein (ECP) perform various biological functions that are dependent on their RNase activity, however, they are much weaker enzymes compared to RNase A, which does not have any special bioaction. The loop regions of EDN and ECP, which differ from RNase A loops in structure and sequence, have been proposed to play a role in the physiological function of EDN and ECP. One of these loops, loop l7, contains a nine residue insertion, AspGlnArgArgAspProProGlnTyr in EDN and AspProArgAspSerProArgTyr in ECP as compared to the amino acid sequence of RNase A. Residues Arg117, Pro120 and Gln122 in loop l7 of EDN were mutated to alanine by site-directed mutagenesis. The mutants were expressed in E.coli and purified from the inclusion bodies. In a Western blot the mutants reacted with anti–EDN antibody as well as EDN did. The CD-spectral analysis revealed that the overall structure of the mutants was moderately altered as compared to EDN. The mutants showed ribonucleolytic activity comparable to that of EDN on yeast tRNA, Poly (U) and Poly (C). All the three mutants inhibited protein synthesis in a manner similar to that of EDN in a cell-free assay indicating that these residues per se are not critical for the catalytic activity of EDN. However, further analysis is required to delineate the overall contribution of this unique insertion in EDN function.

B.    Construction and evaluation of ribonuclease and ribosome-inactivating protein-based chimeric toxins

The objective of the project is to develop chimeric toxins with ribosome-inactivating and ribonucleolytic toxins, and to characterize their in vitro and in vivo cytotoxic activity. The project also aims to rationally design and engineer the chimeric toxins based on the knowledge from the structure-function analysis to improve their biological activity.

We have shown earlier that after internalization the intracellular translocation of restrictocin takes place through the retrograde pathway. It has been shown that the cytotoxicity of many protein toxins, translocated via the retrograde pathway, increases remarkably if a terapeptide tail, LysAspGluLeu (KDEL), is attached at their carboxy terminus. To enhance the cytotoxic activity of IL13 containing chimeric toxins we have made two constructs, IL13-restrictocin-KDEL and IL13-spacer-restrictocin-KDEL, in which the C-terminal end of restrictocin is extended by incorporating the KDEL tetrapeptide. The characterization of these chimeric toxins is ongoing.

We have also developed two saporin-based immunotoxins namely IL13-saporin, and saporin-IL13, targeted at the human IL13 receptor. The fusion proteins contain IL13 fused either at the amino terminus or the carboxy terminus of saporin. The fusion proteins are expressed in E. coli and localized in the inclusion bodies. The purification and characterization of saporin constructs is underway.

Publications

Original peer-reviewed articles

1.     Bagga S, Seth D and Batra JK (2003) The cytotoxic activity of ribosome-inactivating protein saporin-6 is attributed to its rRNA N-glycosidase and internucleosomal DNA fragmentation activities. J Biol Chem 278:4813-4820.

2.     Gaur D, Seth D and Batra JK (2002) Glycine 38 is crucial for the ribonucleolytic activity of human pancreatic ribonuclease on double stranded RNA. Biochem Biophys Res Commun 297:390-395.

3.     Goyal A, Seth D and Batra JK (2002) Role of cis prolines 112 and 126 in the functional activity of ribonucleolytic toxin restrictocin. Biochem Biophys Res Commun 295:812-817.