Murine model for Fusarium oxysporum invasive fusariosis reveals organ-specific structures for dissemination and long-term persistence

Abstract

The soil-borne plant pathogen Fusarium oxysporum causes life-threatening invasive fusariosis in immunocompromised individuals. The mechanism of infection in mammalian hosts is largely unknown. In the present study we show that the symptoms of disseminated fusariosis caused by F. oxysporum in immunosuppressed mice are remarkably similar to those reported in humans. Distinct fungal structures were observed inside the host, depending on the infected organ. Invasive hyphae developed in the heart and kidney, causing massive colonization of the organs. By contrast, chlamydospore-like survival structures were found in lung, spleen and liver. Systemically infected mice also developed skin and eye infections, as well as thrombosis and necrosis in the tail. We further show that F. oxysporum can disseminate and persist in the organs of immunocompetent animals, and that these latent infections can lead to lethal systemic fusariosis if the host is later subjected to immunosuppressive treatment.

Figure 1. Effect of inoculum size and timing of immunosuppression on the severity of F. oxysporum infection.

(A) BALB/c mice (n = 6) were inoculated intravenously with F. oxysporum microconidia. Immunosuppressive treatment (150 mg cyclophosphamide per kg body weight) was initiated three days prior to infection (day −3) and repeated every 3 days thereafter. Survival was recorded for 20 days. An inoculum dose of 2×10⁶ microconidia caused significantly lower mortality than 2×10⁷ microconidia (p = 0.0016); 10⁶ microconidia caused significant lower mortality than 2×10⁶ microconidia (p = 0.0024). (B) BALB/c mice (n = 6) were inoculated with 2×10⁷ microconidia, and immunosuppressive treatment was initiated either on day −3 or on day 0, and repeated every 3 days thereafter. Survival was recorded over 20 d. (C) Fungal burdens in mice immunosuppressed on day 0 and inoculated with 2×10⁷ microconidia. On each of the indicated days post-infection (dpi), one animal was sacrificed and organ homogenates quantified.

Figure 2. F. oxysporum displays distinct growth morphologies in different organs.

PAS-hematoxylin-staining of tissue sections from different organs. Mice were subjected to immunosuppressive treatment on day 0, and every 3 days thereafter. Organs were obtained on day 1 (A, D), 4 (B, E, G, H, I) or 14 (C, F) after infection with 2×10⁷ microconidia: heart (A–C), kidney (D–F), lung (G), spleen (H), liver (I). Fungal structures are indicated by arrows. G, microconidial germlings; M, mycelium; C, chlamydospores. Scale bar = 50 µm.

Figure 3. Invasive growth of F. oxysporum in the kidney.

(A) Fungal biomass on the surface of an aseptically removed kidney from a mouse subjected to immunosuppressive treatment on day 0, and every 3 days thereafter. The kidney was obtained 20 d after infection with 10⁶ microconidia. Visible fungal lesions are indicated by arrows. (B) Kidney cross section from a mouse subjected to immunosuppressive treatment on day 0, and every 3 days thereafter. The kidney was obtained 20 d post-infection with 10⁶ microconidia. Bright areas correspond to fungal biomass stained with the chitin-binding dye Calcofluor White (CFW). The white box indicates the area used for the tissue section shown in C. (C) PAS-hematoxylin staining of a tissue section. The area of invasive mycelial growth is surrounded by arrows. M, mycelium. Scale bars are indicated.

Figure 4. F. oxysporum causes infection in the tails and eyes of immunosuppressed mice.

Mice were subjected to immunosuppressive treatment on day 0, and every 3 days thereafter. (A) Tail of a non-infected immunosuppressed BALB/c mouse. (B–E) Tails of immunosuppressed BALB/c mice intravenously inoculated with 10⁶ conidia at 10 d post-infection. Macroscopic symptoms are indicated by arrows: swellings (B, D), open lesions (C), necrosis (D, E), and loss of the tip (E). (F, G) CFW staining of tissue samples taken from the lesion shown in (C). Fungal structures are indicated by arrows. H, hyphae; C, chlamydospores. (H–L) Infection of the eye tissue in immunosuppressed mice infected with F. oxysporum. (H) Uninoculated mouse; (I–K) mice inoculated with 10⁶ conidia at 10 dpi. (L) CFW staining of an eye removed from an immunosuppressed mouse infected with F. oxysporum. H, hyphae. Scale bar = 50 µm (F, G); 1 mm (L).

Figure 5. F. oxysporum disseminates and persists in immunocompetent mice.

(A) Fungal burdens were determined at day 4 post-infection for groups of 3 BALB/c mice, either immunocompetent or immunosuppressed on day 0 and every 3 days thereafter, and intravenously infected with 2×10⁷ microconidia. (B) Fungal burdens in infected immunocompetent mice sacrificed at 11 d post-infection. (C–F) PAS-hematoxylin-staining of tissue sections obtained from different organs of immunocompetent mice sacrificed at 4 d post-infection. (C) Heart, (D) lung, (E) kidney, (F) liver. (G) Spleen sample from an infected immunocompetent mouse sacrificed at 11 d post-infection was homogenized, KOH-treated and stained with CFW to visualize fungal structures (indicated by arrows). G, microconidial germlings; C, chlamydospores. Scale bar = 50 µm.

Figure 6. F. oxysporum persistence in immunocompetent mice can lead to systemic infection and death upon subsequent immunosuppression.

(A) BALB/c mice (n = 6) were intravenously inoculated with 2×10⁷ microconidia. Immunosuppressive treatment was initiated either on day 3 or 7 post-infection and repeated every 3 days thereafter. Survival was recorded over 20 d. Fungal burdens (B) and PAS-hematoxylin-staining (C–F) of organ sections from a mouse inoculated with 2×10⁷ F. oxysporum microconidia and subjected to immunosuppressive treatment starting at 7 dpi. The mouse that died was culled at 15 dpi. Organs: heart (C), kidney (D), spleen (E), liver (F). M, mycelium; C, chlamydospores. Scale bar = 50 µm.