Listeria monocytogenes traffics from maternal organs to the placenta and back

Abstract

Infection with Listeria monocytogenes is a significant health problem during pregnancy. This study evaluates the role of trafficking between maternal organs and placenta in a pregnant guinea pig model of listeriosis. After intravenous inoculation of guinea pigs, the initial ratio of bacteria in maternal organs to placenta was 10(3)-10(4):1. Rapid increase of bacteria in the placenta changed the ratio to 1:1 after 24 h. Utilizing two wild-type strains, differentially marked by their susceptibility to erythromycin, we found that only a single bacterium was necessary to cause placental infection, and that L. monocytogenes trafficked from maternal organs to the placenta in small numbers. Surprisingly, bacteria trafficked in large numbers from the placenta to maternal organs. Bacterial growth, clearance, and transport between organs were simulated with a mathematical model showing that the rate of bacterial clearance relative to the rate of bacterial replication in the placenta was sufficient to explain the difference in the course of listeriosis in pregnant versus nonpregnant animals. These results provide the basis for a new model where the placenta is relatively protected from infection. Once colonized, the placenta becomes a nidus of infection resulting in massive reseeding of maternal organs, where L. monocytogenes cannot be cleared until trafficking is interrupted by expulsion of the infected placental tissues.

Figure 1. Listeriosis in Pregnant Guinea Pigs and Mathematical Simulation of the Infection

CFUs per (A) individual placentas (on average 3–4 per animal), (B) fetal liver, maternal (C) liver, (D) spleen, and (E) blood were enumerated 6, 24, 48, and 72 h after i.v. inoculation of animals with 7.5 × 10⁵–10⁶ wild-type L. monocytogenes. 3/11 animals had no placental infection 72 h p.i. and were excluded from this figure. CFUs for maternal organs in pregnant animals without placental infection are shown in Figure S1. The black line represents the mathematical simulation of CFUs per compartment. ND = not detectable.

Figure 2. Listeriosis in Nonpregnant Guinea Pigs and Mathematical Simulation of the Infection

CFUs per (A) liver, (B) spleen, and (C) blood were enumerated 24, 48, and 72 h after i.v. inoculation of animals with 7.5 × 10⁵–10⁶ wild-type L. monocytogenes. The black line represents the mathematical simulation of CFUs per compartment. ND = not detectable.

Figure 3. Trafficking of L. monocytogenes in Guinea Pigs

Guinea pigs were inoculated i.v. with a 1:1 mixture of 10403S and DP-L3903 at a total dose of 7.5 × 10⁵–10⁶. Ratios between erm^R and erm^S CFUs from individual placentas (A), maternal liver (triangle), and spleen (circle) of pregnant (B) and nonpregnant (C), and maternal blood (D) of pregnant animals were calculated at 24, 48, and 72 h p.i.. In (A–D) “pure erm^R” or “pure erm^S” represents pure cultures of the erm^R or erm^S strain, respectively. (E) Ratio of placental ratio (pooled placentas from each animal) and ratio of maternal spleen (circle), liver (triangle), and blood (diamond). In (E) “erm^R” and “erm^S” represents pure cultures of the respective strains in the placenta and a ratio around 1 in maternal organs. Data points with pure cultures of the same strain in placenta and maternal organs were plotted as 1, data points with pure cultures in the placenta, but not in maternal organs, were excluded. All of the excluded data points (6/46) had a predominance of the same strain in all organs.

Figure 4. Mathematical Model for Growth, Killing, and Movement in an Infected Animal

Maternal liver (l), spleen (s), and placenta (p) are separate compartments connected by the maternal bloodstream (b), treated as a fourth compartment. The fetal liver (f), the fifth compartment, is connected to the placenta. Within each of the five compartments, the number of bacteria can change by four processes: replication (r), clearance (c), efflux to another compartment (e), and influx from another compartment (i). The connections among the compartments and the relevant first-order rate constants (k) are diagrammed.

Figure 5. Computational Simulation of Bacterial Load in Relation to Clearance Rate in the Placenta

Simulation of CFUs per (A) placenta and fetal liver and (B) maternal liver, spleen, and blood at 72 h as a function of the placental clearance rate (k_pc).

Figure 6. Listeriosis in Animals Treated with Gentamicin

(A) CFUs per spleen and liver in nonpregnant animals were enumerated at 72 hours p.i. with 7.5 × 10⁵–10⁶ L. monocytogenes. Animals treated with gentamicin (+G) are represented by hollow symbols, untreated control animals (−G) are represented by solid symbols. Black bars represent median values. P-values between treated and control animals are not statistically significant (0.18 and 0.94 for liver and spleen, respectively). (B) CFUs per placenta, maternal spleen, liver, and ml of blood from pregnant animals with placental infection were enumerated at 72 hours p.i. with 7.5 × 10⁵–10⁶ L. monocytogenes. P-values marked with * between treated and control animals were statistically significant (0.028 and 0.016 for liver and spleen, respectively). (C) CFUs per maternal liver, spleen, and blood were plotted against CFUs per placenta. The black lines represent the best-fit power law curve for liver (y = 4818^0.38; R² = 0.85), spleen (y = 278.7^0.44; R² = 0.78), and blood (y = 0.907^0.34; R² = 0.84).