A bioengineered probiotic for the oral delivery of a peptide Kv1.3 channel blocker to treat rheumatoid arthritis

Abstract

Engineered microbes for the delivery of biologics are a promising avenue for the treatment of various conditions such as chronic inflammatory disorders and metabolic disease. In this study, we developed a genetically engineered probiotic delivery system that delivers a peptide to the intestinal tract with high efficacy. We constructed an inducible system in the probiotic Lactobacillus reuteri to secrete the Kv1.3 potassium blocker ShK-235 (LrS235). We show that LrS235 culture supernatants block Kv1.3 currents and preferentially inhibit human T effector memory (T_EM) lymphocyte proliferation in vitro. A single oral gavage of healthy rats with LrS235 resulted in sufficient functional ShK-235 in the circulation to reduce inflammation in a delayed-type hypersensitivity model of atopic dermatitis mediated by T_EM cells. Furthermore, the daily oral gavage of LrS235 dramatically reduced clinical signs of disease and joint inflammation in rats with a model of rheumatoid arthritis without eliciting immunogenicity against ShK-235. This work demonstrates the efficacy of using the probiotic L. reuteri as a novel oral delivery platform for the peptide ShK-235 and provides an efficacious strategy to deliver other biologics with great translational potential.

Keywords: Kv1.3 channel; drug delivery; synthetic biology.

Fig. 1.

Supernatants from LrS235, but not from LrGusA, block Kv1.3 currents and inhibit the proliferation of human CCR7⁻ T_EM cells. (A) Representativewhole-cell recordings of L929 cells stably expressing mKv1.3 before (control) and after the addition of supernatants diluted 1/10 of LrGusA or LrS235. (B) Percentage of remaining mKv1.3 currents after addition of LrGusA () or LrS235 () supernatants diluted 1/10. Mean ± SEM, each data point represents a different measurement. (C) Representative flow cytometry plots of CellTrace Violet dye dilution and CCR7 expression of CD3⁺ cells from human peripheral blood MNC stimulated for 7 d without any bacterial supernatants (Left) or in the presence of supernatants from LrGusA (Middle) or LrS235 (Right). (D) Percent of divided human CCR7⁻ T_EM and CCR7⁺ naïve/T_CM cells in the absence of stimulation () and after anti-CD3 induced stimulation in the presence of Lr medium (■) or supernatants of LrGusA () or LrS235 () diluted 1/100. Mean ± SEM, N = 4 different buffy coat donors. *P < 0.05, **P < 0.01.

Fig. 2.

LrS235 secretes sufficient ShK-235 in the intestines for detection in the circulation of healthy rats. Healthy rats received an oral bolus of 1 × 10⁹ cfu of LrGusA (▲) or LrS235 (▼), or an enteric-coated capsule filled with ShK-235 (●, 2 mg/kg body weight) or gelatin (o). Blood was drawn at different time points, and a single-cell patch-clamp was used to assess the ability of the serum to block Kv1.3 currents. A, Representative traces before (control) and after addition of serum diluted 1/100 from the 6-h time point. B, Current block of serum samples collected from rats at the 6-h time point. Each data point represents an individual rat. N = 3 to 4 measurements per rat. Serum dilution: 1/100. C, Current block of serum samples collected at the indicated time points. Each data point represents a rat. N = 3 to 4 measurements per rat. Serum dilution: 1/10. D, An active DTH reaction was induced against ovalbumin and rats received a single bolus of the following immediately before ear challenge: 1 × 10⁹ cfu of LrGusA (▲) or LrS235 (▼) orally, an enteric-coated capsule filled with ShK-235 (●, 2 mg/kg body weight) or gelatin (o) orally, 1 mL of LrS235 culture supernatant orally (▽), or subcutaneous injection of 0.1 mg/kg synthetic ShK-235 (■) or vehicle (2). N = 6 rats per group (three males, three females). **P < 0.01, ***P < 0.001.

Fig. 3.

LrS235 stops disease progression and reduces bone and joint damage and inflammation in rats with CIA. A. Clinical scores of paw inflammation from rats with CIA treated with vehicle (2) or with 100 μg/kg ShK-235 () every other day starting disease onset, 1 × 10⁹ cfu LrGusA () or LrS235 gavage daily (). B. Hematoxylin and eosin (Left) and safranin O/fast green (Right) staining and histology scoring (C) of joints from paws from CIA rats received different treatments. Original magnification, 10×, scale bars, 100 μm. Refer to “Histology and micro-CT” in the Materials and Methods for more details of the scoring system. D. Representative micro-CT of paws from CIA rats treated with vehicle, synthetic ShK-235 every other day, or oral gavage with LrGusA, LrS235 daily. Data presented as mean ± SEM. N = 7 to 10 rats per group. Asterisks indicate areas of cartilage erosions. *P < 0.05; **P < 0.01, ***P < 0.001, and ****P < 0.0001.

Fig. 4.

ShK-235, delivered via injection or LrS235, is not immunogenic A. ELISA plates were coated with 10 μg/mL of either porcine collagen II, ShK-235, or HsTX1[R14A]. Sera from rats with CIA treated with vehicle, ShK-235, LrGusA, or LrS235 were tested at tenfold dilution steps starting from a 1:100 dilution. The serum from non-immunized rats was used as a negative control. Each column represents the mean OD of four individual rat sera ± SEM. B. Data from different PBMC donors were normalized to the unstimulated cells with 100% CellTrace Violet. The higher dilution of CellTrace Violet indicates a higher division rate. Data are present as mean ± SEM.