Preconceptual Priming Overrides Susceptibility to Escherichia coli Systemic Infection during Pregnancy

Abstract

Maternal sepsis is a leading cause of morbidity and mortality during pregnancy. Escherichia coli is a primary cause of bacteremia in women and occurs more frequently during pregnancy. Several key outstanding questions remain regarding how to identify women at highest infection risk and how to boost immunity against E. coli infection during pregnancy. Here, we show that pregnancy-induced susceptibility to E. coli systemic infection extends to rodents as a model of human infection. Mice infected during pregnancy contain >100-fold-more recoverable bacteria in target tissues than nonpregnant controls. Infection leads to near complete fetal wastage that parallels placental plus congenital fetal invasion. Susceptibility in maternal tissues positively correlates with the number of concepti, suggesting important contributions by expanded placental-fetal target tissue. Remarkably, these pregnancy-induced susceptibility phenotypes are also efficiently overturned in mice with resolved sublethal infection prior to pregnancy. Preconceptual infection primes the accumulation of E. coli-specific IgG and IgM antibodies, and adoptive transfer of serum containing these antibodies to naive recipient mice protects against fetal wastage. Together, these results suggest that the lack of E. coli immunity may help discriminate individuals at risk during pregnancy, and that overriding susceptibility to E. coli prenatal infection by preconceptual priming is a potential strategy for boosting immunity in this physiological window of vulnerability.IMPORTANCE Pregnancy makes women especially vulnerable to infection. The most common cause of bloodstream infection during pregnancy is by a bacterium called Escherichia coli This bacterium is a very common cause of bloodstream infection, not just during pregnancy but in all individuals, from newborn babies to the elderly, probably because it is always present in our intestine and can intermittently invade through this mucosal barrier. We first show that pregnancy in animals also makes them more susceptible to E. coli bloodstream infection. This is important because many of the dominant factors likely to control differences in human infection susceptibility can be property controlled for only in animals. Despite this vulnerability induced by pregnancy, we also show that animals with resolved E. coli infection are protected against reinfection during pregnancy, including having resistance to most infection-induced pregnancy complications. Protection against reinfection is mediated by antibodies that can be measured in the blood. This information may help to explain why most women do not develop E. coli infection during pregnancy, enabling new approaches for identifying those at especially high risk of infection and strategies for preventing infection during pregnancy.

Keywords: Escherichia coli; preconceptual; pregnancy; prenatal infection; vaccination.

FIG 1

Pregnancy confers increased susceptibility to systemic E. coli infection in mice. (A) Schematic outlining the susceptibility to E. coli intravenous infection in virgin mice compared with mice midgestation (E10–12) during allogeneic pregnancy; (B) recoverable E. coli CFU in the spleen or liver 48 h after infection for the mice described in panel A; (C) E. coli CFU in the blood at each postinfection time point for the mice described in panel A; (D) percent fetal wastage among individual litters of mice 48 h after maternal E. coli infection at midgestation (E10–12) compared with that of no-infection control pregnant mice; (E) recoverable E. coli CFU in the placenta and concepti for each litter 48 h after maternal E. coli infection at midgestation; (F) number of live pups born at term among individual litters of mice 48 h after maternal E. coli infection at midgestation (E10–12) compared with that of no-infection control pregnant mice. Each point represents the data from an individual mouse, combined from at least two independent experiments, both with similar results. i.v., intravenous.

FIG 2

Susceptibilities to systemic E. coli infection are comparable during allogeneic and syngeneic pregnancies. (A) Schematic outlining the use of BALB/c (H-2^d) or C57BL/6 (H-2^b) males to establish allogeneic and syngeneic pregnancies, respectively, in C57BL/6 (H-2^b) female mice; (B) recoverable E. coli CFU in the spleen or liver 48 h after infection at midgestation (E10–12) for the mice described in panel A; (C) percent fetal wastage among individual litters of mice 48 h after maternal E. coli infection at midgestation for the mice described in panel A; (D) percent concepti with recoverable E. coli CFU 48 h after maternal E. coli infection at midgestation for the mice described in panel A; (E) average number of recoverable E. coli CFU among concepti in each litter 48 h after maternal E. coli infection at midgestation for the mice described in panel A. Each point represents the data from an individual mouse, combined from at least two independent experiments with similar results.

FIG 3

Maternal E. coli susceptibility during pregnancy directly correlates with the number of concepti in each litter. (A) Regression analysis comparing the number of concepti in each litter with recoverable E. coli in the maternal spleen and liver; (B) regression analysis comparing fetal wastage with the average number of recoverable E. coli CFU in the concepti of each litter. Each point represents the data from an individual mouse, combined from at least two independent experiments with similar results.

FIG 4

E. coli infection primes protective immunity against reinfection in mice. (A) Schematic outlining the susceptibility to E. coli high-dose (4 × 10⁷ CFU) challenge of specific-pathogen-free naive mice compared with that of mice infected 20 days prior with a sublethal E. coli inoculum; (B) survival for each group of mice described in panel A after high-dose E. coli challenge; (C) recoverable E. coli CFU in the spleen or liver 24 h after high-dose E. coli challenge for the mice described in panel A. Each point represents the data from an individual mouse, combined from at least two independent experiments with similar results.

FIG 5

Preconceptual infection overrides pregnancy-induced E. coli infection susceptibility in mice. (A) Schematic outlining the susceptibility to E. coli prenatal challenge in specific-pathogen-free naive mice compared with mice infected 20 days prior to mating to establish allogeneic pregnancy; (B) recoverable E. coli CFU in the spleen or liver 48 h after infection for the mice described in panel A; (C) E. coli CFU in the blood at each postinfection time point for the mice described in panel A; (D) percent fetal wastage 48 h after maternal E. coli infection at midgestation (E10–12) among E. coli-naive (no preconceptual infection) and primed female mice with resolved E. coli infection prior to pregnancy; (E) recoverable E. coli CFU in the placenta and concepti for each litter 48 h after maternal E. coli infection at midgestation. Each point represents the data from an individual mouse, combined from at least two independent experiments with similar results.

FIG 6

Serum cytokine levels after E. coli prenatal infection. Levels of each cytokine in the serum prior to or 8, 24, or 48 h after intravenous injection of E. coli (2 × 10⁵ CFU of uropathogenic strain UTI89) for virgin control mice (gray squares), pregnant midgestation (E10–12) mice without prior E. coli exposure (black circles), and pregnant midgestation (E10–12) mice with resolved E. coli infection prior to pregnancy (red circles). These data are representative of 4 to 5 mice per group at each time point, combined from two independent experiments with similar results.

FIG 7

Serum containing E. coli-specific antibodies from mice with resolved infection transfers protection to naive recipient mice. (A) Schematic outlining when immune cells (splenocyte and lymph node cells) and serum (200 μl after heat inactivation) are harvested from E. coli-primed donor mice and transferred (5 × 10⁷ splenocytes plus lymph node cells and/or 200 μl heat inactivation serum) to each group of naive recipient mice; (B) survival for each group of mice described in panel A after high-dose (4 × 10⁷ CFU) E. coli challenge; (C) recoverable E. coli CFU in the spleen or liver 24 h after high-dose E. coli challenge for mice administered serum 1 day prior to infection compared with that in naive control mice given no serum; (D) E. coli-specific IgG antibody titers in the sera of E. coli-primed mice (20 days after infection) compared with those of naive control mice; (E) optical density of each antibody type with E. coli specificity after administration of a 1:2,000 dilution of the serum from E. coli-primed mice (20 days after infection) compared with that of naive control mice. Each point represents the data from an individual mouse, combined from at least two independent experiments with similar results.

FIG 8

Serum containing E. coli-specific antibodies transfers protection to naive pregnant mice. (A) Schematic outlining when serum is harvested from E. coli-primed nonpregnant donor mice and transferred into pregnant mice midgestation (E10–12) during allogeneic pregnancy; (B) recoverable E. coli CFU in the spleen or liver 48 h after prenatal E. coli challenge for mice administered serum (200 μl after heat inactivation) 1 day prior to infection compared with that of naive control pregnant mice given no serum; (C) percent fetal wastage for the mice described in panels A and B; (D) average number of recoverable E. coli CFU among concepti in each litter 48 h after maternal E. coli prenatal challenge for the mice described in panels A and B. Each point represents the data from an individual mouse, combined from at least two independent experiments with similar results.