(A) Direct ELISA of trastuzumab and 13 against HER2 extracellular domain. active mimic binds to cells with a 10-fold lower avidity than the least active variant suggesting that structure plays a large role in their efficacy. This work suggests that the megamolecule approach can be used to prepare antibody mimics having a broad structural diversity. Monoclonal antibodies (mAbs) are an important class of therapeutics and have emerged as the decades new blockbuster drugs. Yet, nearly all approved therapeutic mAbs are based on structures that are genetically encoded, and the majority are based on the natural Y-shaped immunoglobulin (IgG) scaffold. Protein engineering approaches have increased the number of mAb variants but are still quite limited in accessing a broader structural space.1,2 For example, the IgG scaffold enforces a constant distance and orientation of the two Fab domains that are organized for recognition of a pair of identical epitopes on the cell surface, and efforts to vary this distance, expand valency, or create multispecific variants of the core structure are still challenging.3 In this paper, we demonstrate the use of megamolecule assembly to prepare a family of antibody mimics that present two Fabs for the HER2 receptor and show that these analogs vary in their activity, and that the most active analog halts tumor growth activity of trastuzumab mimic 13. (A) Direct ELISA of trastuzumab and 13 against HER2 extracellular domain. (B) In vitro AP24534 (Ponatinib) cytotoxicity data for 13 and trastuzumab in three HER2 positive and one HER2 negative cell lines. (C) A plot of mean tumor volume vs time for the mouse BT474 xenograft model. We then used a cytotoxicity assay to compare the activity of 13 with trastuzumab in the HER2 positive cell lines BT474, SKBR3, and MDA-MB-135-VII. We treated cells cultured in 96-well plates with trastuzumab and 13 in concentrations ranging from 400 nM to 4 pM for 96 h and then measured their viability after this period using the Alamar Blue reagent. The results of these experiments are shown in Figure 2B. The determined EC50 values were similar in all cases: 2.8, 1.4, and 2.9 nM for 13 and 1.6, 0.7, and 0.8 nM for trastuzumab in the three cell lines, respectively. Both molecules showed no cytotoxicity in a HER2 negative carcinoma cell line A431, indicating insignificant off-target effects associated with mimic 13 in this assay. Rabbit Polyclonal to CSPG5 Control experiments using just the Fab and the VH-cutinase and VH-SnapTag fusions gave EC50 values that were similar to each other and were approximately 10-fold higher than those for the bivalent molecule 13 (Supporting Information, Figure S1). This result shows that the fusion partners did not interfere in the engagement of HER2 or enhance the blocking effect of the Fab domains. As an additional comparison, we monitored the stability of trastuzumab and 13 by SDS-PAGE in serum-containing cell culture media at 37 C and in PBS at 55, 65, and 75 C and found that the two had comparable stability under the conditions of the cytotoxicity experiments, yet different stability at 65 C with 13 decomposing within 1 AP24534 (Ponatinib) h and trastuzumab slowly decomposing over 10 h (Figures S4 and S5). Having established that our mimic displays comparable activity to trastuzumab in assays, we next characterized the efficacy of 13 in a mouse xenograft BT474 tumor model. After establishing the tumor, SCID-beige mice (= 5 per treatment group) were treated once daily for 28 days with 13 (2.5 mg/kg) or with doxorubicin (2.5 mg/kg). As megamolecule 13 lacks the Fc domain and thus the ability to bind the neonatal Fc receptor (FcRn),14,15 its half-life is short (= 0.034) (Figure 2C). Additionally, no statistically significant difference in total body or organ weight was observed between the treatment group and control suggesting that the molecule was well-tolerated at this dose (Figure S3). We selected antibody 13 from a panel of 16 trastuzumab mimics we prepared. Each of these molecules presents two copies of the Fab on a scaffold formed by joining a cutinase and SnapTag domain with the linker, AP24534 (Ponatinib) as explained earlier. The structures differ in the spacing and relative orientations of the Fab domains because each fusion can be assembled by tethering the enzyme to either the heavy or light chain of the Fab, and either the N- or C-terminus of each chain. Hence the antibody mimics were prepared by joining one of the four Fab-cutinase fusions with one of the four AP24534 (Ponatinib) Fab-SnapTag fusions to give.
Categories:Hexokinase