Isolation of a nanobody specific to the PstS-1 protein and evaluation of its immunoreactivity with structural components of Mycobacterium tuberculosis granuloma

Abstract

Mycobacterium tuberculosis (Mtb) causes infectious granulomatous disease tuberculosis (TB), and existing in vitro TB-diagnosis is insensitive for extrapulmonary TB (EPTB) as well as paucibacillary TB due to low bacillary load; therefore, alternative non-invasive molecular imaging-based diagnostic tools are urgently required. Within TB granulomas, foci of Mtb secreted antigens anchored on the surface of either bacilli or host cells may serve as targetable biomarkers for antibody based molecular imaging of TB. Nanobody is better suited over conventional antibody or fragment derivatives for molecular imaging due to its quick localization in target tissue and rapid clearance from off-target organs. Here, we report the production of a high affinity nanobody against PstS-1 protein of Mtb which helps bacilli in phosphate uptake as well as host cell adhesion. C8 nanobody (C8Nb) was isolated from a phage displayed nanobody library which was constructed from a camel immunized with secreted proteins of Mtb. C8Nb was characterized in vitro and in vivo for immunoreactivity against PstS-1 protein. The ability of C8Nb to bind the PstS-1 protein, associated with the surface of Mtb bacilli or adhered on the macrophages, and its localization around BCG cells injected intramuscularly into mice, demonstrate its potential in the development of molecular imaging-based diagnostic tools for TB.

Keywords: PstS-1; T7 phage display; TB granuloma; molecular imaging; nanobody; tuberculosis.

Figure 1

Schematic map depicting construction of nanobody display library and production of nanobody.

Figure 2

Nanobody display library construction. (A) Generation of heavy chain antibodies in camel serum was confirmed by ELISA, where ELISA wells coated with rPstS-1 protein were incubated with serially diluted camel sera and bound antibodies were measured by incubating rabbit anti-camelid V_HH antibody-HRP conjugate. (B) In IgG PCR, genetic locus between FR1 region of variable domain and 5’ end of CH2 domain were amplified, which yielded 620bp, 690bp and 900bp size amplicons corresponding to IgG3, IgG2 and IgG1 subclasses, respectively. (C) In V_HH PCR, primers amplified the region between FR1 and FR4, which produced a broad band of ~400bp-500bp size amplicons. (D) Vector based PCR (Primers: T7SelectUp and T7SelectDown) on ligation reaction mixture produced broad band of ~450bp-550bp size amplicons, which confirmed ligation between nanobody gene inserts and T7 vector DNA arms. (E) Nanobody library size (~2.5×10⁵ clones) was measured by PFU assay. Image shows clear plaques of T7 phages on lawn of E. coli Rosetta-gami B5615.

Figure 3

Isolation of nanobody specific to PstS-1 protein of Mtb. (A) 38kDa size rPstS-1 protein of Mtb was overexpressed in E. coli BL21(DE3) and soluble intact protein was purified using Ni-NTA agarose column chromatography. Lane 1- Protein marker; Lane 2–8 represents eluates from 1 to 7, respectively. (B) Three cycle bio-panning was performed to enrich PstS-1 protein specific nanobodies. Stringent washing produced more than 100-fold phage enrichment in rPstS-1 protein coated wells over BSA coated wells. (C) Nanobody screening was carried out using two assay formats. Immunoreactivity of randomly selected nanobody clones was evaluated by reacting monoclonal phage lysates with coated rPstS-1 protein and bound phages were measured by incubating ¹²⁵I-anti-T7TF mAb. Out of 30 clones, 29 phages demonstrated strong positive binding. In both assays 1xPBS and non-specific (NS) phage were included as negative controls. (D) Reproducibility of immunoreactivity of nanobody clones which were selected in the first assay was confirmed by capturing phages in anti-T7TF mAb coated wells, followed by reacting wells with ¹²⁵I labelled rPstS-1 protein. Image inserts in the graphs depict immunocomplex formation in screening assays. (E) C8Nb was overexpressed in E. coli BL21(DE3) and soluble protein was purified using (I) Ni-NTA agarose column chromatography and eluates (E1-E5) containing nanobody were pooled together, concentrated and subjected to (II) DEAE cellulose column chromatography. Void fractions (V1-V5) containing unbound C8Nb were pooled together. (III) 14.33kDa size C8Nb was further purified using size exclusion chromatography on Superdex G75 Increase 10/300 GL column.

Figure 4

High affinity C8Nb specifically recognizes PstS-1 protein. (A) Bacterial protein mixtures or rPstS-1 protein were electrophoresed on SDS-PAGE and blotted on PVDF membrane, followed by sequential incubation with C8Nb and rabbit anti-camelid V_HH-HRP conjugate. Band development around 38kDa size marker was observed only in the lanes loaded with either rPstS-1 protein or protein mixtures of Mtb and BCG. (B) Saturation binding ELISA demonstrated K_D 0.198nM for C8Nb. (C) Binding of C8Nb to rPstS-1 was analyzed using a Biacore X100 SPR system. The rPstS-1 was immobilized on CM5 sensor chip and sensorgrams were obtained at increasing analyte (C8Nb) concentrations. Kinetic rate constants k_a (5.44x10⁵ M⁻¹ sec⁻¹), k_d(1.499x10⁴ sec⁻¹) and the equilibrium dissociation constant K_D (0.275nM) were determined using 1:1 binding model. (D) Immunoreactivity of ¹²⁵I-C8Nb tracer (T) was confirmed against rPstS-1 protein coated in ELISA wells by radioimmunoassay. Spiked tracer (ST), prepared by mixing the tracer with excess unlabeled C8Nb, was included to assess specificity of binding. All values are mean ± SD. Significance was calculated by one-way ANOVA followed by Dunnett’s multiple comparisons post hoc test; ***P <0.001. Data are representative of more than three independent experiments.

Figure 5

C8Nb recognizes PstS-1 protein adhered on the surface of either mycobacteria or macrophages. (A) When single cell suspensions of bacteria were probed with ¹²⁵I-C8Nb, only Mtb and BCG pellets demonstrated tracer binding. (B) Immunoreactivity of ¹²⁵I-C8Nb was confirmed with single cell suspensions of Mtb LAM and Mtb Beijing strains. (C) Protein mixtures prepared from bacterial lysates or rPstS-1 protein were adhered on formaldehyde fixed RAW 264.7 macrophages and then cells were incubated with ¹²⁵I-C8Nb. Tracer showed binding with the cells which were treated with either rPstS-1 protein or protein mixtures prepared from Mtb or BCG lysates. (D) In similar binding studies on THP-1 macrophages adhered with rPstS-1 protein, WT and TLR4 KO cells demonstrated comparable tracer binding, while tracer binding with TLR2 KO cells was low and comparable with WT cells treated with BSA. In the case of MR, spiking Mtb protein mixture with mannose (1mg/ml) did not affect tracer binding. Data suggested that the epitope on PstS-1 protein bound to only TLR2 is accessible to C8Nb for binding, while MR and TLR4 are not playing role in binding. In all assays (A-D), specific binding of ¹²⁵I-C8Nb tracer (T) was assessed by including spiked tracer (ST). (E) In fluorescence microscopy, THP-1 cells were sequentially incubated with rPstS-1 protein, C8Nb and rabbit anti-camelid V_HH cocktail-iFluor555. Analysis of acquired images suggested that TLR2 is involved in binding of C8Nb with PstS-1 protein treated THP-1 macrophages. Scale bars: 10μm. (A-D) all values are mean ± SD, and derived from more than three independent experiments. Significance was calculated by one-way ANOVA followed by Dunnett’s multiple comparisons or Bonferroni’s post hoc test; ***P <0.001.

Figure 6

In vivo localization of ¹²⁵I-C8Nb around intramuscularly injected mycobacterial cells. (A) To reduce uptake of free ¹²⁵I in thyroid gland, BALB/c mice were fed with KI solution. For in vivo localization study, animals (n=3 animals) were injected with single cell suspensions of BCG (right forelimb) and M. smegmatis (left forelimb) through intramuscular route and 25.91µCi of ¹²⁵I-C8Nb through catheterized tail vein. (B) Four hours post tracer administration, animals were sacrificed and limbs were dissected and accumulated activity was measured in gamma-photon counter. All mean ± SD. values are derived from more than three independent experiments. Significance was calculated by one tailed T-test; *P <0.05.