Polymyxins as novel and safe mucosal adjuvants to induce humoral immune responses in mice

Abstract

There is currently an urgent need to develop safe and effective adjuvants for enhancing vaccine-induced antigen-specific immune responses. We demonstrate here that intranasal immunization with clinically used polypeptide antibiotics, polymyxin B (PMB) and colistin (CL), along with ovalbumin (OVA), increases OVA-specific humoral immune responses in a dose-dependently manner at both mucosal and systemic compartments. Enhanced immunity by boosting was found to persist during 8 months of observation. Moreover, mice intranasally immunized with OVA plus various doses of PMB or CL showed neither inflammatory responses in the nasal cavity and olfactory bulbs nor renal damages, compared to those given OVA alone. These data suggest that polymyxins may serve as novel and safe mucosal adjuvants to induce humoral immune responses. The polymyxin adjuvanticity was found to be independent of endotoxins liberated by its bactericidal activity, as indicated by similar enhancing effects of PMB in lipopolysaccharide (LPS)-hyporesponsive and LPS-susceptible mice. However, despite the presence of preexisting anti-PMB antibodies, we observed no reduction in the adjuvant function of polymyxins when they were given intranasally. Furthermore, the titers of OVA-specific Abs in mice intranasally immunized with OVA plus PMB or CL were significantly higher than those in mice administered with polymyxin analogues, such as polymyxin B nonapeptide and colistin methanesulfonate. The levels of released β-hexosaminidase and histamine in mast cell culture supernatants stimulated by PMB or CL were also significantly higher than those stimulated by their analogues. These results suggest that both the hydrophobic carbon chain and hydrophilic cationic cyclic peptide contribute to the mucosal adjuvanticity of PMB and CL.

Figure 1. Structures of polymyxins and their analogues.

PMB, polymyxin B sulfate; CL, colistin sulfate; CLMS, colistin sodium methanesulfonate; PMBN, polymyxin B nonapeptide hydrochloride.

Figure 2. OVA-specific Ab responses in mucosal secretions and plasma.

Mice were intranasally immunized with ovalbumin (OVA) plus 0, 2.5, 5, 10, 25, 50, 100, 250, or 500 µg of PMB or CL 3 times at weekly intervals. Mucosal secretions and plasma were collected at 1 week after the last immunization, and the OVA-specific IgA and IgG Ab levels were determined by endpoint ELISA. Each dot represents an individual mouse. Horizontal and vertical bars indicate mean ± standard deviation (SD) and median ± interquartile range (IQR), respectively, from 2 independent experiments (with 8 mice in each experimental group). Significant differences between the OVA alone and OVA plus PMB or CL groups are indicated with asterisks (*p<0.05 and **p<0.005).

Figure 3. Analysis of OVA-specific AFCs in the mucosal and systemic tissues.

Mice were intranasally immunized with OVA either alone (open column) or with 500 µg of PMB (closed) or CL (gray). One week after the last immunization, mononuclear cells isolated from various mucosal and systemic tissues were analyzed using an OVA-specific ELISPOT assay. n-LP, nasal lamina propria; NALT, nasopharynx-associated lymphoid tissue; i-LP, small intestinal lamina propria; MLNs, mesenteric lymph nodes; SMGs, submandibular glands; SMLNs, submandibular lymph nodes; and ALNs, axially lymph nodes. Data represent mean ± SD values of the number of AFCs/10⁶ cells from 3 independent experiments (with 12 mice in each experimental group). Significant differences between the OVA-alone and OVA plus PMB or CL groups are indicated with asterisks (* p<0.05 and ** p<0.005). ND, not detected.

Figure 4. PMB-specific Ab responses in plasma.

A. The PMB-specific Ab titers were determined by ELISA. To establish the ELISA system, 1,000 ng/well of anti-PMB mAb and twofold serial dilutions were tested. Standard anti-PMB was titrated on PMB-coated plates with a sensitivity of 20 ng/well. B. Plasma samples from mice intranasally immunized with OVA plus 0, 5, 50, or 500 µg of PMB 3 times at weekly intervals were assessed by PMB-specific endpoint ELISA. Each dot represents an individual mouse. Horizontal and vertical bars indicate median and IQR, respectively, from 2 independent experiments (with 8 mice in each experimental group).

Figure 5. Long-lasting humoral immune responses by intranasal immunization of OVA plus PMB.

Mice were intranasally immunized with OVA plus 500 µg of PMB 3 times at weekly intervals. One group was intranasally boosted with OVA plus 500 µg of PMB at 13 weeks after the last immunization (closed circle), whereas the other group was intranasally administered with normal saline (open circle). A. The levels of OVA-specific IgA Abs in fecal extracts, as well as IgG and IgA Abs in plasma, were determined by ELISA. Data represent mean ± SD values from 8 mice in each experimental group. B. After 31 weeks, mononuclear cells isolated from the n-LP, i-LP, SMGs, and spleen were determined using an OVA-specific ELISPOT assay. Data represent mean ± SD values of the number of AFCs/10⁶ cells from 8 mice per experimental group. Significant differences between non-boosted (open column) and boosted (closed column) groups are indicated with asterisks (* p<0.05 and ** p<0.005). ND, not detected.

Figure 6. Comparison of immune responses between C3H/HeN and C3H/HeJ mice immunized with OVA and PMB.

C3H/HeN and C3H/Hen mice were intranasally immunized with OVA plus 500 µg of PMB 3 times at weekly intervals. Mucosal secretions (fecal extracts, nasal washes, and saliva) and plasma were collected at 1 week after the last immunization. The OVA-specific IgA and IgG Ab levels were determined by OVA-specific endpoint ELISA. Each dot represents an individual mouse. Horizontal and vertical bars indicate mean and SD, respectively, from 2 independent experiments (with 8 mice in each experimental group).

Figure 7. Comparison of immune responses and mast cell degranulation induced by polymyxins and their analogues.

A. Mice were intranasally immunized with OVA plus 500 µg of PMB, PMBN, CL, or CLMS 3 times at weekly intervals. Mucosal secretions and plasma were collected at 1 week after the last immunization, and the OVA-specific IgA and IgG Ab levels were determined by OVA-specific endpoint ELISA. Each dot represents an individual mouse. Horizontal and vertical bars indicate either mean and SD or median and IQR, respectively, from 2 independent experiments (with 8 mice in each experimental group). Significant differences between the PMB and PMBN groups or between the CL and CLMS groups are indicated with asterisks (*p<0.05 and **p<0.005). B. Degranulation of MC/9 cells was assessed by a β-hexosaminidase release assay. Left panel, the percentages of β-hexosaminidase release elicited by compound 48/80 (a positive control); right panel, the percentages of β-hexosaminidase release stimulated by PMB (closed square), CL (open square), PMBN (closed triangle), or CLMS (open triangle). Data represent mean ± SD values from 3 different experiments. C. Histamine release of MC/9 cells was assessed by EIA. MC/9 cells were cultured with PMB, CL, PMBN, or CLMS. Data represent mean ± SD values from 3 different experiments. Significant differences between the PMB and PMBN groups or between the CL and CLMS groups are indicated with an asterisk (*p<0.05).