Intracellular trafficking of furin enhances cellular intoxication by recombinant immunotoxins based on Pseudomonas exotoxin A

Abstract

Furin is a mammalian serine protease with important roles in cellular homeostasis and disease. It cleaves and activates numerous endogenous and exogenous substrates, including the SARS-CoV-2 viral spike protein and protein toxins such as diphtheria toxin and Pseudomonas exotoxin A (PE). Recombinant immunotoxins (RITs) are toxin conjugates used as cancer therapeutics that connect tumor-directed antibodies with toxins for targeted cell killing. RITs based on PE have shown success in treating a variety of cancers, but often suffer from safety and efficacy concerns when used clinically. We have explored furin as a potential limiting factor in the intoxication pathway of PE-based RITs. Although the furin has widely recognized importance in RIT intoxication, its role is incompletely understood. Circumstantial evidence suggests that furin may act as a transporter for RITs in addition to its role of activation by cleavage. Here, we describe the creation of a CRISPR-engineered furin-deficient HEK293 cell line, ΔFur293. Using ΔFur293 and derivatives that express mutant forms of furin, we confirm the importance of furin in the PE RIT intoxication pathway and show that furin trafficking has a significant impact on RIT efficacy. Our data support the hypothesis that furin acts as a transporter during RIT intoxication and suggest furin as a target for modification to improve the effectiveness of RITs.

Keywords: Pseudomonas exotoxin A; Cancer therapy; Furin; Gene knockout; Intracellular trafficking; Recombinant immunotoxin.

Fig. 1.

HEK293 FRT furin knockout. HEK293 FRT cells were transfected with plasmids containing genes coding for Cas9 and three furin sgRNAs. Surviving cells were clonally selected by serial dilution in a 96-well plate. Lysates from HEK293 FRT (WT) and four mutant clones were evaluated for furin expression by western blot. Clones 1, 2, and 4 showed no discernable furin expression, and clone 1 (boxed) was selected for further study as ΔFur293.

Fig. 2.

Cytotoxicity assays. HEK293 FRT and ΔFur293 cells were treated with the anti-transferrin receptor/PE24 RIT HB21-LR in the presence and absence of 1 μM PPCI furin inhibitor. EC₅₀ (pM) values from four separate paired assays for each condition are plotted. Dotted lines denote the average value for each condition and error bars indicate the standard deviation. Significant P-values (P<0.05) are reported from a one-way ANOVA performed as described. A representative cytotoxicity assay is shown in Fig. S3.

Fig. 3.

Furin complementation. HEK293 FRT and ΔFur293 cells were stably transfected with the gene for Fur or CAT. All cell lines were then treated with the anti-transferrin receptor/PE24 RIT HB21-LR and evaluated for cytotoxicity. EC₅₀ (pM) values from five separate paired assays for each line are plotted. Dotted lines denote the average value for each condition and error bars indicate the standard error. Significant P-values (P<0.05) are reported from a one-way ANOVA performed as described.

Fig. 4.

Cleavage assays. HEK293 FRT cells, ΔFur293 cells, and ΔFur293 cells stably expressing transgenic wild-type furin (ΔFur293/Fur) were incubated for various time intervals from 0.5 to 8 h in culture with the anti-transferrin receptor/PE24 RIT HB21-LR. Whole cell lysates were evaluated for full length and cleaved HB21-LR by western blot (panel A) and densitometry (panel B) as described. Also shown are untreated (U) cell lysates for each cell line, HB21-LR with (+) and without (−) furin treatment in vitro, and the β-actin loading control. The ratio between the furin-cleaved band intensity and the total intensity of all RIT bands at each time point is plotted in panel B. The individual densitometric analysis values (points) and mean (bar) for at least two separate assays of each cell line are shown.

Fig. 5.

Complementation with mutant furin. ΔFur293 cells were stably transfected with genes for furin that contained mutations designed to impair its intracellular trafficking or catalytic function. Mutations S773A/S775A (ADA) and S773D/S775D (DDD) alter furin trafficking, while the N295A mutant (Ala-295) inhibits catalytic activity. Two separate clonal lineages stably transfected with mutant or wild-type furin were treated with the anti-transferrin receptor/PE RIT HB21-LR to assess cytotoxicity. The EC50 (pM) values from at least four separate assays for each line were normalized for furin expression levels and plotted. Dashed lines denote the average value for each clone and error bars indicate the standard error. The largest significant P-values (P<0.05) between sets of clones from a one-way ANOVA performed as described are indicated. All P-values are reported in Table S3.

Fig. 6.

Cleavage by mutant furin. ΔFur293 cells stably expressing transgenic furin mutants (FurADA, FurDDD, and FurAla-295) were incubated for various time intervals from 0.5 to 8 h in culture with the anti-transferrin receptor/PE24 RIT HB21-LR. Whole cell lysates were evaluated for full length and cleaved HB21-LR by western blot (panel A) and densitometry as described. Also shown are untreated (U) cell lysates for each cell line, HB21-LR in vitro with (+) and without (−) furin treatment, and the β-actin loading control. The ratio between the furin-cleaved band intensity and the total intensity of all RIT bands at each time point is plotted in panel B. The individual densitometric analysis values (points) and mean (bar) for at least two internalization and cleavage assays in each cell line are shown.