Genetic Passive Immunization with Adenoviral Vector Expressing Chimeric Nanobody-Fc Molecules as Therapy for Genital Infection Caused by Mycoplasma hominis

Abstract

Developing pathogen-specific recombinant antibody fragments (especially nanobodies) is a very promising strategy for the treatment of infectious disease. Nanobodies have great potential for gene therapy application due to their single-gene nature. Historically, Mycoplasma hominis has not been considered pathogenic bacteria due to the lack of acute infection and partially due to multiple studies demonstrating high frequency of isolation of M. hominis samples from asymptomatic patients. However, recent studies on the role of latent M. hominis infection in oncologic transformation, especially prostate cancer, and reports that M. hominis infects Trichomonas and confers antibiotic resistance to Trichomonas, have generated new interest in this field. In the present study we have generated specific nanobody against M. hominis (aMh), for which the identified target is the ABC-transporter substrate-binding protein. aMh exhibits specific antibacterial action against M. hominis. In an attempt to improve the therapeutic properties, we have developed the adenoviral vector-based gene therapy approach for passive immunization with nanobodies against M. hominis. For better penetration into the mucous layer of the genital tract, we fused aMh with the Fc-fragment of IgG. Application of this comprehensive approach with a single systemic administration of recombinant adenovirus expressing aMh-Fc demonstrated both prophylactic and therapeutic effects in a mouse model of genital M. hominis infection.

Fig 1

A—Western blot analysis of binding selected nanobodies to LAMPs from M. hominis. LAMPs from M. hominis H-34 was boiled with Laemmli sample buffer, loaded on 12% Laemmli SDS-PAGE; and Western blot was performed for binding with nanobodies (nb 1–6) at working concentration 1 mkg/ml. B—Cell-based ELISA analysis of binding selected nanobody nb 6 (named aMh) to mycoplasma infected cells. M. hominis positive (experimentally infected) and control M. hominis negative A549 cells were fixed with paraformaldehyde and ELISA with selected aMh (nb6) was performed. aMh—incubation with aMh, isotype control—incubation with control nanobody, cells background—fixed cells background. C—Amino acid sequence of the selected nanobody aMh. Hypervariable regions CDR1, CDR2 and CDR3 are underlined and highlighted in bold. Positions of characteristic amino acids that are specific for single domain antibodies are highlighted in red (these amino acids differ in variable domains of classical type antibodies).

Fig 2. Dimer formations of aMh-FcG2a.

Protein samples aMh-FcG2a were prepared before electrophoresis in sample buffer with or without DTT (as reducing agent), canonical mouse IgG2a was prepared with DTT and used as control. Protein bands ~42 and ~84 kDa corresponding monomer and dimer forms of aMh-Fc; protein bands ~54 kDa and 25 kDa corresponding heavy and light chains of canonical mouse IgG2a.

Fig 3

A—aMh-FcG2a specificity against different mycoplasma species. LAMPs from different mycoplasma species were coated on the immunoplate at a concentration of 100 ng/well and ELISA was performed with aMh-Fc at working concentration 10nM, BSA was coated at a concentration of 100 ng/well as negative control. B—aMh-FcG2a binding with different clinical isolates of M. hominis. Pellet of M. hominis cells from 0.1 ml broth culture (titer ~10⁶−10⁷ CCU/ml) was boiled with Laemmli sample buffer, loaded on 12% Laemmli SDS-PAGE. Western blot was performed for binding with aMh-Fc at working concentration 10nM. 1 –reference strain H-34, 2–6 –clinical isolates.

Fig 4. Determination of affinity constants.

Binding of antigen to aMh-FcG2a was determined by surface plasmon resonance using Biacore 3000 (GE Healthcare). Antigen concentration series 5.95 nM; 11.9 nM; 59.5 nM and 595 nM (colour) and 1:1 fitting (black) interaction antigen to aMh-Fc. The fitted constant are k_a = 1.78⁴ M^-1s^-1 and k_d = 1.06⁻⁴ s^-1 which results K_D = 5.94*10⁻⁹ M (R_max = 428 RU; chi² = 11.8). Evaluation included double reference subtraction.

Fig 5. Inhibition of M. hominis growth with the direct addition of aMh-FcG2a to STE2 medium.

aMh-FcG2a was added in different concentrations to M. hominis inoculated in STE2 medium. M. hominis titer was determined with CCU counting.

Fig 6. rAd5-CMV-PLAP-aMh-FcG2a and control rAd5-CMV-PLAP-aMh-ILZ-HA constructions and purification.

(A) Scheme of expression cassettes for rAd5-CMV constructions; (B) sizing of purified Ad5-CMV-PLAP-aMh-FcG2a corresponding to virions suspensions with a particle size of ~100 nM.

Fig 7. Titer of aMh-FcG2a and aMh-ILZ-HA in serum (A) and in vaginal washes (B) from mouse system inoculated with rAd5-CMV-PLAP-aMh-FcG2a or rAd5-CMV-PLAP-aMh-ILZ-HA.

The DBA/2 mice were injected into the ophthalmic venous sinus with 10⁷ PFU of rAd5-CMV-PLAP-aMh-FcG2a or rAd5-CMV-PLAP-aMh-ILZ-HA previously dialyzed against PBS. As a control, we used mice injected with PBS.

Fig 8. Titers of M. hominis in vaginal washes with “prophylactic” scheme of rAds inoculation.

Vaginal washes were collected 5 days after M. hominis inoculation (6 days after rAds inoculation). Amount of M. hominis was evaluated with real-time PCR. The rAd5-CMV-PLAP-aMh-FcG2a group exhibited a significantly lower M. hominis titer (Student's t-test = 3.5; p<0.01). Ad-null n = 5, PBS n = 15, rAd5-CMV-PLAP-aMh-FcG2a n = 10, rAd5-CMV-PLAP-aMh-ILZ-HA n = 11. In the “prophylactic” scheme of rAds inoculation, the M. hominis titer was significantly decreased in the rAd5-CMV-PLAP-aMh-FcG2a group. The prophylactic scheme, however, is not practical for treating mycoplasma infection. In the “therapeutic” scheme, the number of animals diagnosed as positive at 7 days after inoculation with rAd5-CMV-PLAP-aMh-FcG2a (12 days after M. hominis inoculation) was significantly lower than that of the other groups. Nevertheless the mycoplasma titer was not significantly different among the infected animals in any of the groups and not informative due to the large variability in counts and the low number of infected animals in the rAd5-CMV-PLAP-aMh-FcG2a group (n = 2 at 3 days and n = 1 at 7 days).

Fig 9

(A) Titers of M. hominis in vaginal washes with “therapeutic” scheme of rAds inoculation. Vaginal washes were collected 12 days after M. hominis inoculation (7 days after rAds inoculation). Amount of M. hominis was evaluated with both CCU and real-time PCR. (B) The same data for 7 days after rAds inoculation through qualitative way (as M. hominis–positive or M. hominis–negative sample). The decrease in the number of M. hominis—positive samples was statistically significant at 7 day after rAd5-CMV-PLAP-aMh-FcG2a inoculation with Fisher's exact test ϕ* = 3.959 (ϕ_0.05 = 1.64; ϕ_0.01 = 2.31). The mycoplasma titer was not significantly different among infected animals in the groups due to the large variability in counts and the low number of infected animals in the rAd5-CMV-PLAP-aMh-FcG2a group (n = 2 at 3 days and n = 1 at 7 days).

Fig 10. Hypothetical mechanism of blocking nutrient acquisition through ABC transporter with aMh-FcG2a.