Ureaplasma parvum serovar 3 multiple banded antigen size variation after chronic intra-amniotic infection/colonization

Abstract

Ureaplasma species are the microorganisms most frequently associated with adverse pregnancy outcomes. The multiple banded antigen (MBA), a surface-exposed lipoprotein, is a key virulence factor of ureaplasmas. The MBA demonstrates size variation, which we have shown previously to be correlated with the severity of chorioamnion inflammation. We aimed to investigate U. parvum serovar 3 pathogenesis in vivo, using a sheep model, by investigating: MBA variation after long term (chronic) and short term (acute) durations of in utero ureaplasma infections, and the severity of chorioamnionitis and inflammation in other fetal tissues. Inocula of 2 × 10(7) colony-forming-units (CFU) of U. parvum serovar 3 (Up) or media controls (C) were injected intra-amniotically into pregnant ewes at one of three time points: day 55 (69d Up, n = 8; C69, n = 4); day 117 (7d Up, n = 8; C7, n = 2); and day 121 (3d Up, n = 8; C3, n = 2) of gestation (term = 145-150d). At day 124, preterm fetuses were delivered surgically. Samples of chorioamnion, fetal lung, and umbilical cord were: (i) snap frozen for subsequent ureaplasma culture, and (ii) fixed, embedded, sectioned and stained by haematoxylin and eosin stain for histological analysis. Selected fetal lung clinical ureaplasma isolates were cloned and filtered to obtain cultures from a single CFU. Passage 1 and clone 2 ureaplasma cultures were tested by western blot to demonstrate MBA variation. In acute durations of ureaplasma infection no MBA variants (3d Up) or very few MBA variants (7d Up) were present when compared to the original inoculum. However, numerous MBA size variants were generated in vivo (alike within contiguous tissues, amniotic fluid and fetal lung, but different variants were present within chorioamnion), during chronic, 69d exposure to ureaplasma infection. For the first time we have shown that the degree of ureaplasma MBA variation in vivo increased with the duration of gestation.

Figure 1. Amniotic fluid and fetal lung fluid pH analysis.

AF from 69 day chronic ureaplasma exposed animals showed a higher pH when compared to control animals (*p = 0.03) and 3 day acute animals (*p = 0.02). FL fluid from 69 day chronic ureaplasma exposed animals also showed a statistically higher pH (*p = 0.0058) when compared to the pH of FL from7d and 3d acute animals and control animals.

Figure 2. Lung Pressure-Volume (PV) curve, performed on fetal lungs at time of preterm delivery.

Animals exposed to chronic (69d) ureaplasma infection (* p = 0.0365) showed greater lung compliance in comparison to the media control group; the 3d acute (p = 0.165) and the 7d acute (p = 0.49) treatment groups.

Figure 3. H & E staining of tissues selected from three animals exposed to chronic, 69d, intra-amniotic ureaplasma infection.

Variable levels of inflammation were observed in chorioamnion, umbilical cord, and FL tissues. These animal tissue images were selected based on the severity of inflammation within the chorioamnion: A: severe chorioamnionitis; B: mild chorioamnionitis; C: scarring of chorioamnion; and D: Uninfected (control), minimal inflammation. The severity of inflammation observed in the FL and CORD tissues corresponded consistently with the severity of histological chorioamnionitis.

Figure 4. P1 Western blot demonstrating MBA antigenic variation.

Western blots comparing antigenic variation of P1 ureaplasma isolates from animals colonized/infected with U. parvum serovar 3 for 69 days (chronic infection), 7 days and 3 days (acute infection). The number of antigenic variants (single bands) within A. FL (L samples), B. AF (A samples), and C. chorioamnion (C samples) P1 cultures are compared. 442S = serovar 3 initial inoculum control; 10B = 10B media negative control; M = Precision Plus Dual Colour Protein Standard (BioRad, Gladesville, NSW).

Figure 5. Western blots of FL clone 2 ureaplasma isolates.

A: In specimen L29, 3 ureaplasma size variants (150 kDa, 140 kDa, 50 KDa) were detected in this severely inflamed lung tissue; B: Specimen L30: Mildly inflamed tissue demonstrating 2 size variants (105 kDa and 75 kDa); C: L60: Scarred tissue demonstrating only 1 size variant (50 kDa). 442S = serovar 3 initial inoculum control; Marker = Precision Plus Dual Colour Protein Standard (BioRad, Gladesville, NSW). The numbers identify the lung tissue (L29, L30 and L60) and the C2 isolate number.