A Role for Early-Phase Transmission in the Enzootic Maintenance of Plague

Abstract

Yersinia pestis, the bacterial agent of plague, is enzootic in many parts of the world within wild rodent populations and is transmitted by different flea vectors. The ecology of plague is complex, with rodent hosts exhibiting varying susceptibilities to overt disease and their fleas exhibiting varying levels of vector competence. A long-standing question in plague ecology concerns the conditions that lead to occasional epizootics among susceptible rodents. Many factors are involved, but a major one is the transmission efficiency of the flea vector. In this study, using Oropsylla montana (a ground squirrel flea that is a major plague vector in the western United States), we comparatively quantified the efficiency of the two basic modes of flea-borne transmission. Transmission efficiency by the early-phase mechanism was strongly affected by the host blood source. Subsequent biofilm-dependent transmission by blocked fleas was less influenced by host blood and was more efficient. Mathematical modeling predicted that early-phase transmission could drive an epizootic only among highly susceptible rodents with certain blood characteristics, but that transmission by blocked O. montana could do so in more resistant hosts irrespective of their blood characteristics. The models further suggested that for most wild rodents, exposure to sublethal doses of Y. pestis transmitted during the early phase may restrain rapid epizootic spread by increasing the number of immune, resistant individuals in the population.

Fig 1. Infection and blockage rates of O. montana fleas during a four-week period following a blood meal containing 5 to 8 x 10⁸ Y. pestis/ml in mouse blood (blue symbols) or rat blood (red symbols).

(A) The percentage of fleas still infected and (B) the bacterial load per infected flea at different times after infection. (C) The percentage of fleas that developed partial or complete proventricular blockage during the four-week period. The mean and range of three independent experiments using mouse blood and two experiments using rat blood (Table 1) are indicated.

Fig 2. Temporal distribution of the occurrence of partial and complete blockage in O. montana fleas following a single infectious blood meal containing 5 to 8 x 10⁸ Y.

pestis/ml in mouse or rat blood. The cumulative numbers from the three experiments using mouse blood (A) and two experiments using rat blood (B) are shown; see Table 1 for details.

Fig 3. The number of Y. pestis CFU transmitted by individual O. montana fleas 2 to 4 days after infection (early-phase) and after the development of partial or complete proventricular blockage.

Cumulative results from three experiments using mouse blood (blue symbols), two experiments using rat blood (red symbols), and two experiments using rat blood and the Y. pestis ΔhmsH mutant strain (open circles) are shown (see Table 1 for details); bars indicate the median number of Y. pestis transmitted per individual flea bite. The transmission probability (number of positive transmissions divided by the total number of trials) is indicated. All early-phase fleas were confirmed to have been infected when they fed for the transmission test. For both the partially blocked and blocked groups, differences in transmission probability and the number of CFU transmitted by fleas infected using mouse blood or rat blood were not statistically significant.

Fig 4

(A) Frequency distribution histogram of the numbers of Y. pestis CFU transmitted by individual O. montana fleas during the early phase (EPT) and by partially blocked (PB) and completely blocked (B) fleas. (B) Temporal distribution pattern of the number of CFU transmitted by individual partially blocked (PB) and completely blocked (B) fleas (positive transmission events only). Lines connecting data points indicate transmissions by the same flea on successive days. Cumulative results from three experiments using mouse blood (blue symbols) and two experiments using rat blood (red symbols) are shown; see Table 1 for details.

Fig 5. Flow chart of the flea vector-rodent host model.

The three transmission-competent stages of flea infection are highlighted, and the different possible outcomes following transmission to the rodent host are indicated. See text and Tables 2 and 3 for details. t_x = t_ep, t_pb, or t_b.; p_x = p_ep, p_pb, or p_b.

Fig 6. Model predictions of the incidence of infected-dead (% mortality) and infected-recovered (% recovered) hosts in populations with different levels of susceptibility [lethal dose (LD) of 1, 10, or 100 Y. pestis CFU].

Separate outcomes produced by fleas infected using mouse blood or rat blood in which both early-phase transmission and biofilm-dependent transmission by partially and completely blocked fleas are operative (All Tx); or in which only early-phase transmission (EPT only) or only biofilm-dependent transmission (B/PB only) are operative are indicated. All simulations were initiated with 9 susceptible hosts, 1 infected (highly bacteremic) host, and 50 uninfected fleas, monitored over a 100-day period. The results from two versions of the model using (A) unmodified parameters and (B) modified parameters for probability of transmission (p) and probability of transmission at or above a lethal dose (t) that account for cumulative transmission by simultaneous flea bites (S1 and S2 Figs). See text for details.

Fig 7

The number of O. montana fleas per host (m) required to realize different levels of host-to-host transmission (R₀) predicted by the SEIR model, with all transmission modes operative (A); or (B), with only early-phase transmission (EPT only) or only biofilm-dependent transmission (B/PB only) operative, shown relative to each other and to cumulative transmission by both modes (All). The results for fleas infected using mouse blood or rat blood and for three host populations with different susceptibilities to Y. pestis (lethal dose of 1, 10 or 100 CFU) are shown. Dashed lines indicate the intersects of the curves for R₀ = 1 (enzootic) and R₀ = 2 (epizootic) conditions.