https://orcid.org/0000-0002-6240-7936
Figures
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.
Author summary
The ecology of bubonic plague is complex but depends largely on flea-borne transmission. Certain susceptible rodents experience periodic epizootics that can decimate local populations, but the conditions that lead to these episodes are not fully understood. Fleas can transmit Yersinia pestis, the bacterial agent of plague, during two different phases: an early phase within the first few days after their infectious blood meal and again sometime later after the development of a Y. pestis biofilm in the flea foregut. The relative contribution of these two transmission modes to plague ecology has not been systematically examined. Our results indicate that in most ecological contexts early-phase transmission is too inefficient to drive an epizootic, but instead acts to reduce the number of susceptible individuals in a population, thereby favoring a more stable enzootic state.
Citation: Mitchell CL, Schwarzer AR, Miarinjara A, Jarrett CO, Luis AD, Hinnebusch BJ (2022) A Role for Early-Phase Transmission in the Enzootic Maintenance of Plague. PLoS Pathog 18(12): e1010996. https://doi.org/10.1371/journal.ppat.1010996
Editor: Igor E. Brodsky, University of Pennsylvania, UNITED STATES
Received: August 17, 2022; Accepted: November 10, 2022; Published: December 15, 2022
This is an open access article, free of all copyright, and may be freely reproduced, distributed, transmitted, modified, built upon, or otherwise used by anyone for any lawful purpose. The work is made available under the Creative Commons CC0 public domain dedication.
Data Availability: All relevant data are within the manuscript and its Supporting Information files.
Funding: This research was funded by the Intramural Research Program of the NIH (to BJH), NIAID (ZIA AI000796-25) and by the National Science Foundation (2109828 to ADL). The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.
Competing interests: The authors have declared that no competing interests exist.
Introduction
Plague is a vector-borne zoonotic disease that primarily afflicts rodents, notably members of the Muridae (rats, mice, gerbils) and Sciuridae (squirrels, marmots, prairie dogs), and persists in extensive natural foci throughout the world. Yersinia pestis, the plague bacillus, circulates within these host populations via rodent-flea-rodent transmission cycles that involve several flea vector species. Plague exhibits an enigmatic pattern in which dramatic epizootics in highly susceptible rodent populations periodically flare up from the normal relatively quiescent enzootic background state of a plague focus or from reintroduction of Y. pestis into a population. The ecological factors and biological mechanisms that initiate and drive these periodic epizootics versus those that tend toward enzootic maintenance have been subjects of long-standing interest [1–5]. Metapopulation structure, fluctuations in population densities of rodents and fleas, and the immune status and degree of susceptibility to overt disease within and among reservoir host species have been theorized to be important [1, 6–12].
One crucial component of plague cycle dynamics is the transmission rate from infected to uninfected hosts. Although it can sometimes be transmitted by direct contact, ingestion, or aerosol, Y. pestis has evolutionarily adapted to the flea-borne transmission route on which it now depends. Fleas can transmit Y. pestis in three stages following an infectious blood meal. Transmission can occur the very next time they feed during the first week after infection, a phenomenon referred to as early-phase transmission. The second and third stages of transmission are effectuated after Y. pestis forms a bacterial biofilm in the proventriculus, a valve between the esophagus and midgut. As the biofilm grows it gradually restricts the passage of blood through the proventriculus and impedes valvular function. In this partially blocked state, fleas are able to ingest some blood but because the proventricular valve is unable to close completely, blood contaminated with bacteria from the flea digestive tract can backflow into the bite site [13, 14]. Eventually, in the final stage, the biofilm can fill the entire proventriculus and completely block the flow of incoming blood into the midgut. When such completely blocked fleas attempt to feed, blood flowing into the esophagus is stopped in front of the blocked proventriculus, the esophagus initially expands, and then the blood, mixed with some of the bacteria washed from the proventricular biofilm, is recoiled back into the bite site [14, 15]. Early-phase transmission was historically referred to as mass transmission, because it is rarely observed unless groups of 5 to 10 or more infected fleas feed simultaneously on a naïve animal. Although long assumed to be a form of mechanical transmission, early-phase transmission also appears to occur via regurgitation of bacteria from a heavily infected proventriculus even though it does not depend on formation of a mature biofilm [14, 16–18]. Thus, bacterial obstruction of the lumen of the proventricular valve and its normal function, to a greater or lesser extent, is the underlying mechanism common to all three modes of transmission [14].
The several rodent flea species implicated in plague transmission cycles vary in their vector efficiency [19]. However, the relative efficiency of the different phases of transmission by individual fleas has not been empirically evaluated in a systematic way. Early-phase transmission and transmission by blocked fleas have been studied in separate experiments and comparisons made between them, but significant differences in experimental design among the studies make the conclusions problematic. Different infectious doses and blood meal sources, both of which greatly affect flea infection and transmission outcomes, have been employed. Transmission by partially blocked fleas has for the most part been neglected; often the blockage rate has been used as the surrogate indicator for transmission beyond the early phase. Mathematical modeling based on parameterization of these available data has supported a role for early-phase transmission in driving epizootics in some ecological contexts [20, 21]. However, there is a recognized need to reexamine the relative vector competency of different flea species and efficiencies of the different transmission modes using newer standardized, more exacting methods [22].
In this study, we examined transmission of Y. pestis by the North American ground squirrel flea, Oropsylla montana. Several aspects of transmission dynamics were recorded during a one-month period following a single infectious mouse or rat blood meal, including infection and mortality rates and the incidence of proventricular blockage. The number of Y. pestis transmitted by individual fleas during the early phase (2 to 4 days after infection) and after the development of partial or complete blockage was also determined. The data were used to parameterize a simple deterministic susceptible-exposed-infected-recovered (SEIR) model of plague transmission dynamics that we developed. Simulations using the model provided an assessment of the relative contributions of early-phase transmission and biofilm-dependent transmission to disease incidence ensuing from a single cohort of fleas that feed on a highly bacteremic host. The influence of infectious blood source (mouse vs. rat) on plague transmission dynamics was also evaluated.
Results
Quantitative evaluation of flea infection and transmission parameters
Cohorts of O. montana fleas were infected by feeding on highly bacteremic mouse or rat blood (containing 5 to 8 x 108 Y. pestis/ml) and thus began with a 100% infection rate at an average of 3 to 6 x 104 Y. pestis per flea on day 0. Fleas were maintained at 21°C and provided twice-weekly mouse or rat sterile blood meals for 4 weeks and monitored for mortality, infection, and proventricular blockage status (Fig 1). Host blood source had a large effect on infection rate. Consistent with a previous study [23], over half of the fleas infected using mouse blood had completely cleared the infection by day 3. In contrast, around 90% of fleas infected using rat blood remained infected (Fig 1A). The bacterial load of the chronically infected fleas was similar, however (Fig 1B). In correlation with their higher infection rate, a higher percentage of fleas infected using rat blood developed proventricular blockage during the experiments than fleas infected using mouse blood (21% and 11%, respectively; Table 1, Fig 1C). Fleas that appeared to be partially or even completely blocked were seen as early as 2 to 3 days after infection, although the highest incidence occurred between 1 and 3 weeks after infection (Fig 2).
(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.
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.
Transmission efficiency of fleas during the early phase (2 to 4 days after the infectious blood meal) and of partially and completely blocked fleas was determined by allowing individual fleas to feed on a small reservoir of sterile mouse or rat blood for 1 h, and then recovering and plating the blood for CFU (colony-forming unit) count. Fleas were examined for evidence of feeding and for proventricular blockage status immediately after the 1 h access period. The rodent source of the infectious blood meal had a large effect on early-phase transmission efficiency. Only 1 of 38 fleas (3%) infected using mouse blood transmitted during their first blood meal 2 to 4 days later, and only 1 CFU was recovered from the blood that this flea fed upon (Fig 3). In contrast, 8 of 33 (24%) of fleas infected using rat blood transmitted 3 to 2,000 CFU during early-phase transmission trials. Notably, however, 5 of these 8 fleas were observed to be completely blocked and 2 appeared to be partially blocked after this initial post-infection feeding on day 3 or 4. For this reason, we repeated early-phase transmission trials with fleas infected with a Y. pestis hms mutant strain, which is unable to block fleas but is fully transmissible in the early phase [17, 24]. To clearly distinguish the two transmission modes, we concentrated on results with this hms mutant [4 of 42 fleas (10%) transmitted 3 to 72 CFU (median 17 CFU); Fig 3] to estimate the early-phase transmission efficiency.
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.
Our estimate of early-phase transmission efficiency for fleas infected using rat blood was in line with those reported previously for O. montana infected similarly that were used to challenge mice [17, 20, 25–27]. However, our estimate for fleas infected using mouse blood (3%) was much lower than we found with the mouse challenge model (18%; [27]). This may in part be due to a lower sensitivity of the in vitro transmission model. The number of CFUs transmitted in the early phase by fleas infected using mouse blood was clearly very low, but some transmissions of one or a few bacteria may have been missed because the efficiency of plating and recovering very low CFU numbers from a blood reservoir is likely not 100%. For the early-phase probability of transmission parameters (pep) described below, therefore, we chose to use the higher estimates reported for the mouse challenge model (18% and 14% for O. montana infected using mouse or rat blood, respectively) [27]. These parameter values were used in conjunction with our empirical data on the number of wild-type and hms mutant Y. pestis transmitted by individual early-phase fleas infected using mouse or rat blood, respectively (Fig 3).
Transmission by partially blocked fleas was also rather inefficient–only about 10% of feeding events resulted in transmission, regardless of the infectious blood source (Fig 3). Completely blocked fleas had the highest transmission efficiency, and it was similar for fleas infected using mouse or rat blood (50% and 67%, respectively; p = 0.1). Transmission tests were repeated periodically for blocked fleas as long as they remained alive, and several of these fleas transmitted on more than one day (Fig 4). In one case, a single blocked flea transmitted four different times over a period of 12 days.
(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.
Another aspect of transmission efficiency is the number of bacteria transmitted per bite, which we were able to assess on an individual flea basis. The number of Y. pestis transmitted by blocked fleas can be highly variable [28, 29]. Our results confirm that, with a range from 1 to >10,000 CFU transmitted by individual fleas, and also indicate that the number of Y. pestis transmitted by early-phase and partially blocked fleas can vary widely (Figs 3 and 4) An exception may be early-phase transmission by fleas infected using mouse blood, which was of such low efficiency that only small numbers are likely to be transmitted (Fig 3). A summary of the results of the flea infection and transmission experiments is given in Table 1.
Model simulations of flea-borne transmission dynamics
We developed a deterministic SEIR model to compare the relative contribution of three transmission stages on plague dynamics in the context of two host blood sources that are known to influence infection of fleas and early-phase transmission [16, 30]. Conceptual design of the model is illustrated in Fig 5. Our experimental results summarized above were used to estimate the parameter values of the flea vector submodel (Table 2). Parameter values for the rodent host submodel (Table 3) are from published sources. We examined different model structures based on fleas infected using bacteremic mouse or rat blood, host susceptibility to fatal plague [lethal dose (LD100) of 1, 10, or 100 Y. pestis CFUs], and transmission mode (early-phase transmission or biofilm-dependent transmission by blocked or partially blocked fleas). The starting condition was 9 susceptible hosts, 1 infected (highly bacteremic) host, and 50 uninfected fleas; model outputs were recorded over a simulated 100-day period (S1 Fig). Fig 6A shows the predicted mortality for the various scenarios. With all transmission modes operative, plague epizootics (arbitrarily defined as ≥ 50% cumulative mortality during the 100-day simulation) ensued in the most susceptible host population, but at the second tier of susceptibility (LD = 10 CFU) the output was host-blood dependent. Fleas infected using rat blood were predicted to cause an epizootic, but fleas infected using mouse blood were not. In the more resistant population (lethal dose of 100 CFU), enzootic scenarios (<50% mortality) ensued with both sets of fleas. Most of the mortality was attributable to transmission by blocked fleas–when early-phase transmission parameters were set to zero, mortality was equivalent or even slightly higher than when both transmission modes were operative. Conversely, when blockage-dependent transmission was removed, predicted mortality was 2.7- to 3.6-fold lower in all cases. The strong influence of the infectious blood source on early-phase transmission (Fig 3) was reflected in the model output (Figs 6A and S1). Early-phase-dependent mortality effected by fleas infected using mouse blood was 36% in the highly susceptible host population (LD = 1 CFU) but nil in the other two populations–the only fatality was the single infected host used to initiate the model simulations. In contrast, early-phase transmission generated higher mortality (93% and 24%) in the two susceptible populations in the rat blood context, although it was still negligible in the more resistant population (LD = 100 CFU).
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. tx = tep, tpb, or tb.; px = pep, ppb, or pb.
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.
Recent studies of early-phase transmission efficiency have used a model in which groups of ~10 fleas infected using highly bacteremic rat blood fed simultaneously on highly susceptible laboratory mice (LD50 <10 CFU) [20, 40, 41]. To explicitly examine this scenario, we adjusted our early-phase transmission parameters, which are based on an individual flea bite, to reflect the cumulative probability of 10 simultaneous flea bites transmitting a lethal dose. In addition, since a blocked flea makes repeated feeding attempts, we likewise adjusted the blocked flea transmission parameters to reflect the cumulative probability of transmitting a lethal dose in two consecutive feeding attempts. The adjusted parameters are listed in Table 2. In this scenario, early-phase transmission by fleas infected using mouse blood was sufficient to produce an epizootic only in the most susceptible host population, and in populations with the first two levels of susceptibility by fleas infected using rat blood (Figs 6B and S2). High mortality ensued in the more resistant host population only when transmission by blocked and partially blocked fleas was included in the model. Notably, with both modes of transmission active, mortality in this more resistant population was less than half that (27%) of the mortality predicted when blockage-dependent-transmission operated alone (early-phase transmission parameters set to zero; 66% mortality). Correspondingly, infected but recovered hosts were greater in this population when both modes of transmission were operative than when early-phase transmission was eliminated (59% and 25% recovery rate, respectively; Figs 6 and S2).
Estimation of flea densities required for enzootic and epizootic transmission dynamics
The basic reproduction number (R0) is defined as the number of secondary cases that ensue from a single case in a naïve population. We used the next-generation matrix method [42] on our SEIR model to estimate R0 values for the various scenarios and to estimate the flea density per host (m) that would be required to sustain an enzootic or epizootic state (defined here as R0 = 1 and R0 = 2, respectively). With all transmission modes in effect, an m value of five or fewer fleas per host was sufficient to achieve an enzootic state (R0 = 1) for all the scenarios (Fig 7A). To reach an epizootic state (R0 = 2), m values of 8, 16, and 20 were estimated from the model, respectively, for fleas infected using mouse blood with the three host populations (LD of 1, 10, or 100 Y. pestis CFU). For fleas infected using rat blood, estimated m values required for R0 = 2 were 4, 5, and 11 (Fig 7A).
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.
Flea densities required for enzootic and epizootic conditions were also estimated separately for early-phase transmission only and for transmission by blocked fleas only. The results from the SEIR model are shown in Fig 7B. In all cases, the flea burdens required for a given R0 level were lower for blocked-flea transmission than for early-phase transmission. For example, in a host population with rat blood characteristics for which the LD = 10 Y. pestis CFUs, 6 fleas per host were sufficient to drive an epizootic (R0 = 2) by the proventricular blockage mechanism, whereas early-phase transmission would require ~25 fleas per host. Early-phase transmission alone was not sufficient for even enzootic maintenance–blocked fleas were required (Fig 7B).
Discussion
In this study we systematically examined flea-borne transmission dynamics during a one-month period following an infectious blood meal. Cohorts of O. montana fleas were infected uniformly, using two different host blood sources, and then transmission by individual fleas in three progressive stages of transmission competence was evaluated. Two aspects of transmission efficiency were quantified: the probability of transmission and the number of Y. pestis transmitted during a single flea bite. The standardized experimental design enabled a more stringent head-to-head comparison of the early-phase and proventricular-blockage dependent transmission modes. Previous comparisons have depended on separate studies that were not standardized as to infectious dose, blood source, or the suitability of the experimental conditions for the general fitness of a particular flea species, all of which can affect infection, blockage rates, and transmission dynamics [16, 23, 29, 30, 33, 40, 43]. Early-phase transmission dynamics following rat blood meals containing ~109 Y. pestis/ml have been compared with blocked-flea transmission dynamics following infection by feeding on guinea pigs with a much lower bacteremia level, such that only a minority of fleas (~20 to 32%) that fed on them became infected [44–46]. This low infection rate is consistent with a bacteremia of only about 107 Y. pestis/ml [29, 33], below the threshold for efficient early-phase transmission [43]. Another problematic area has been the essentially preliminary, and sometimes discordant, available data on blockage rates of different flea vectors, often based on single trials with small sample sizes performed under a variety of conditions and lacking important controls [22, 40, 44]. For example, we recently showed that the prairie dog flea, Oropsylla hirsuta, can become blocked at a rate much higher than previously reported [47]. Other studies incorrectly reported that O. montana rarely becomes blocked and transmits beyond the early phase [20, 28, 33, 48]. As shown here and in previous studies however, O. montana readily becomes blocked and transmits efficiently by the proventricular blockage mechanism [23, 49–51]. Blocked O. montana are efficient vectors and can transmit large doses of Y. pestis (Fig 3).
The focus of this study was the transmission rate from flea to rodent, which we maximized by infecting the fleas with blood with the high bacteremia level that has been used for early-phase transmission efficiency experiments [20, 52, 53]. Because host blood source can influence infection and early-phase transmission rates [16, 30], we compared fleas infected using mouse blood or rat blood. Fleas in three transmission-competent states were examined individually. Early-phase transmission by fleas infected using mouse blood was rare and inefficient but was much better if the fleas were infected using rat blood. These results are consistent with a previous study that evaluated early-phase transmission by groups of fleas [16]. Rat blood promotes early-phase transmission because it induces a phenomenon termed post-infection esophageal reflux (PIER), in which a mixture of partially digested blood, hemoglobin crystals, and Y. pestis is refluxed into the foregut soon after an infectious blood meal [16]. Mouse blood, characterized by a more soluble hemoglobin molecule and a faster digestion rate in the flea gut, does not engender PIER. As in the previous study [16], some fleas appeared to be completely blocked as early as three days after infection using rat blood, suggesting that early-phase can overlap temporally with biofilm-dependent transmission (Fig 2). For this reason, we based the early-phase transmission estimate on fleas infected with a Y. pestis hms mutant, which is incapable of producing proventricular blockage but fully capable of early-phase transmission [17, 24].
A lower percentage of fleas infected using mouse blood developed partial or complete blockage (Fig 1); but this can be attributed to their lower infection rate compared to fleas infected using rat blood, consistent with results reported previously [23]. Transmission by partially blocked fleas was surprisingly inefficient, comparable to early-phase transmission, which may be because the transmission mechanisms are more or less the same (transient or partial impedance of blood feeding), although partial blockage, unlike early-phase, develops later and is biofilm-dependent [14, 16, 18]. However, no effect of blood source on transmission by partially blocked fleas was evident. Transmission by completely blocked fleas was the most efficient of the three stages, both in terms of transmission rate and the number of CFUs transmitted per flea (Fig 3). At this stage also, no difference was apparent between blocked fleas infected using mouse or rat blood. After their 1-h feeding attempts, 50 to 67% of individual blocked fleas had transmitted, and ~10% of these transmitted >1,000 CFUs, with >10,000 CFU recovered in three instances. On average, the number of CFU transmitted by a blocked O. montana flea is greater than the number transmitted by a blocked Xenopsylla cheopis flea [23, 29].
A second aspect of this study was to use the experimentally derived transmission efficiency data to parameterize an SEIR model of plague dynamics that we developed. Here the main goal was to compare the relative importance and contribution of the early-phase and proventricular blockage-dependent transmission modes in determining epizootic outbreaks. The model outputs indicated that blockage-dependent transmission was most important in producing epizootics, regardless of host blood source. Our modeling indicated that early-phase transmission alone resulted in significantly less mortality, and a large effect of host blood source was evident. Early-phase fleas infected using rat blood could drive an epizootic, but only in the most susceptible host population (LD = 1 CFU). Early-phase transmission by fleas infected using mouse blood was insufficient by itself to produce epizootic conditions in any host population.
Model parameter values were estimated on an individual flea bite basis. However, a feature of flea-borne transmission in nature is that several infected fleas feeding simultaneously on a single host in the early-phase window, or a blocked flea making repeated feeding attempts in succession can produce an augmented, cumulative transmission efficiency. For example, a blocked flea will bite continuously and persistently in the few days before it dies from starvation; and blocked flea transmission efficiency estimates based on single, short term exposure trials are acknowledged to be underestimates [8, 28]. The probability of a blocked flea cumulatively transmitting significant numbers of CFUs in repeated bites likely approaches 100%. In contrast, early-phase transmission is primarily limited to the first feeding after infection, with reduced transmission during later feeds unless reinfected [52]. Therefore, in a second version of the model we incorporated a specific scenario for effective efficiency of cumulative early-phase transmission (simultaneous bites by 10 fleas modeled) and also of blocked-flea transmission (conservatively, 2 consecutive bites by a single blocked flea modeled). In this version of the model, early-phase transmission by fleas infected using mouse blood was sufficient to drive an epizootic, but only in the most susceptible population (LD = 1 CFU). Early-phase transmission by fleas infected using rat blood was sufficient to drive an epizootic in hosts in which the lethal dose was 10 CFU or less. In either version of the model, however, early-phase transmission was incapable of driving an epizootic in a more resistant population (LD = 100 CFU), which is more representative of most wild rodents. Blockage-dependent transmission was essential for epizootic scenarios in those populations. Notably, predicted mortality increased when early-phase transmission was set to zero, whereas the number of recovered hosts was higher when both transmission modes were operative. This suggests that early-phase transmission would act to dampen outbreaks in more resistant populations because the sublethal doses usually delivered lead to recovery and immunity rather than disease. In a previous experiment in which laboratory mice were fed on by 3 to 12 infected O. montana early-phase fleas, the majority of transmissions were detected by seroconversion only [27]. Thus, blocked flea transmission is likely the driving force behind epizootics in moderately resistant populations. Based on some prior reports that O. montana rarely becomes blocked, this mode of transmission was discounted a priori in an earlier model of plague dynamics as vectored by this flea [54]. Our results show that this needed to be reevaluated, because in this and a previous study [23] we show that blockage-dependent transmission by O. montana is more efficient than early-phase transmission. In accordance, another study reported that O. montana transmits at higher efficiency 7 to 21 days after infection than in the early phase [25].
The higher vector competence of blocked O. montana fleas was also reflected in the estimated flea burden required to drive an epizootic, which was lower than for early-phase transmission, particularly in a more resistant host population (Fig 7B). The value we estimated for early-phase transmission by O. montana infected using rat blood and a susceptible population was similar to a previous estimate for X. cheopis that used the same infection scenario [40]. That study came to an opposite conclusion than we did for blocked flea transmission, maintaining that the flea burden would have to be higher than for early-phase transmission. However, the parameter values for blocked flea transmission were based on a single separate study in which the fleas were infected by feeding on mice with varying levels of bacteremia, such that their blockage rate was only 7% [33]. When X. cheopis feed on highly bacteremic blood, as used for the early-phase experiments, however, several studies have shown that the blockage rate is 35 to 40% or higher [18, 23, 24, 28, 29, 55–58].
Early-phase transmission, although the first mode of transmission to be described, received relatively little attention after the blockage-dependent transmission mode was discovered. Some, however, proposed that early-phase transmission could add momentum to an epizootic when the plague incidence was high enough to reduce a population size, leading to increased flea burdens on remaining hosts [28, 59]. According to this scenario, the large number of questing fleas that had last fed on a host that died from plague septicemia could transmit when they next fed a new host in a phenomenon called mass transmission (the original name for early-phase transmission, reflecting the fact that many fleas feeding simultaneously are required for a high probability of transmission). Our results also indicate that a high flea burden is required for productive early-phase transmission. Transmission by blocked and partially blocked fleas was still considered to be essential for stable ecological maintenance of flea-borne plague [28, 60]. This premise is supported by the fact that all Y. pestis strains maintain the genes required for flea biofilm formation, even though they are not required for virulence in mammals [24, 61, 62]. A previous theoretical modeling study also concluded that gaining the ability to block fleas was evolutionarily adaptive for Y. pestis because transmissibility is superior to that of unblocked (e.g. early-phase) fleas [63].
More recently, a series of studies concluded that early-phase transmission might be the principal driving force of epizootics, particularly in rodent-flea cycles in which the primary flea vector reportedly develops proventricular blockage at low incidence (reviewed in [21]). These conclusions were based on transmission results with highly susceptible laboratory mice challenged by groups of ~10 fleas that had been infected 1 to 4 days earlier by feeding on highly bacteremic rat blood. Our simulations also indicate that early-phase transmission could be sufficient to drive an epizootic in those conditions. However, early-phase transmission varies with host blood–it is less efficient following infection with bacteremic mouse blood, (Fig 3; [16]), and may be important in driving epizootics only with hosts whose blood, like that of brown rats, induce PIER in fleas. Furthermore, most wild rodents are more resistant to Y. pestis than laboratory mice. For example, the LD50 of Y. pestis for the ground squirrel hosts of O. montana is reportedly at least 250 to >1,000 CFU [64–66], although this value is likely to vary locally as well as within a given population. It is not known whether transmission efficiency in conjunction with ground squirrel blood would be more like mouse or rat blood, and it will be important to examine that in future studies. In either case, our results suggest a different ecological role for early-phase transmission within most rodent populations. By transmitting a sublethal dose, early transmission may serve to increase the number of resistant individuals in a population in which plague is introduced, in a sense vaccinating them against an otherwise lethal challenge by the later, more efficient transmission by blocked fleas.
Long-term focal persistence of plague requires stable host-pathogen dynamics to maintain an enzootic state. A pathogen as virulent as Y. pestis, which depends on producing a lethal, high-density bacteremia in the host to infect its flea vectors would seemingly be at risk of burning through host populations too quickly for stable ecological maintenance. This epizootic scenario can occur in the most susceptible hosts. However, our models suggest that exposure of most wild rodents to sublethal, immunizing doses of Y. pestis transmitted during the early phase may ameliorate rapid epizootic spread by reducing the number of susceptible individuals in the population. The coexistence of resistant and susceptible hosts in a population has long been thought to be a factor in the enzootic persistence of plague despite its high virulence [11]. Thus, in many situations early-phase transmission may be more important in maintaining the enzootic state than in driving an epizootic. The effect of early-phase transmission in generating partially immune populations would also increase the time span of enzootic prevalence, during which the movement of individuals could spread the disease to new populations in new areas [67].
Summary
Transmission efficiency of individual O. montana fleas was measured over a four-week period after they had fed on mouse or rat blood containing 5 to 8 x 108 Y. pestis/ml. Both early-phase and proventricular blockage modes of transmission were monitored for cohorts of fleas following a single infectious blood meal. The results indicate that: 1) blockage-dependent transmission is much more efficient than early-phase transmission, both in terms of the probability of transmission and the number of CFUs transmitted. The recurrent biting behavior of blocked fleas is a significant force multiplier of transmission [68], as a single blocked flea can cumulatively transmit thousands of CFUs before it dies of starvation. 2) Host blood source strongly affects early-phase transmission efficiency, but not blockage-dependent transmission efficiency. However, host blood source can influence the infection rate and therefore the subsequent blockage rate. 3) Early-phase transmission could drive an epizootic only in naïve, very susceptible host populations and when the flea burden is high. 4) The low CFU numbers typically transmitted in the early phase may “immunize” many individuals of more resistant host species, acting to limit epizootic spread and promote an enzootic state.
Materials and methods
Ethics statement
Fleas were from colonies maintained according to a protocol approved by the Institutional Animal Care and Use Committee of the Rocky Mountain Laboratories (RML) [23]. Animal work adhered to the institution’s guidelines for animal use, the guidelines and principles in the United States Public Health Service Policy on Humane Care and Use of Laboratory Animals, and the Guide for the Care and Use of Laboratory Animals and was conducted by certified staff in an Association for Assessment and Accreditation of Laboratory Animal Care (AAALAC) International accredited facility.
Flea infections
O. montana fleas were from a laboratory colony originally established at the CDC, Fort Collins [69] and maintained at RML since 2011. Groups of ~300 O. montana that had not fed for 5 days were infected by allowing them to feed on 5 ml of defibrinated mouse or rat blood containing 5 to 8 × 108 CFU/ml of Y. pestis KIM6+ with pAGFP1 (Clontech/Takara Bio) a plasmid that encodes the green fluorescent protein and carbenicillin-resistance or with KIM6+ ΔhmsH (pAcGFP1) through a Parafilm M membrane stretched across an artificial feeding device, following a previously established standard protocol [23, 70, 71]. The pAcGFP1 plasmid is stably maintained by Y. pestis in infected fleas for at least 31 days in the absence of antibiotic pressure. After a 1-h feeding period, fleas were individually examined microscopically, and those that took an infectious blood meal (denoted by the presence of fresh red blood in the midgut) were collected; 20 of them were placed at -80°C for later determination of the initial infectious dose and the rest were placed in capsules containing a layer of sawdust and maintained at 21° C, 75% relative humidity [23].
Evaluation of flea infection, mortality, blockage, and transmission rates after infection
To assess early-phase transmission, 12–48 individual fleas (equal numbers of males and females) were allowed to feed for 1 h on ~300 μl of sterile defibrinated mouse or rat blood (corresponding to the infectious blood meal source), using miniaturized versions of the artificial feeding device, on days 2 to 4 after infection [71]. Each flea was then examined for evidence of feeding and for partial or complete proventricular blockage [24]. Partial blockage is diagnosed by the presence of fresh blood in the esophagus and midgut and complete blockage by the presence of fresh blood only in the esophagus, usually pooled just anterior to the proventriculus [71]. The blood in the individual reservoirs was removed and spread onto blood agar plates containing 100 μg/ml carbenicillin. CFUs were counted after 48 h at 28° C to determine the number of bacteria transmitted. Fleas that fed were individually frozen at -80°C for subsequent bacterial load determinations.
All other fleas that took an infectious blood meal were provided maintenance feeds twice weekly (Monday and Thursday) beginning 2 to 4 days after infection for one month on a neonatal mouse or on sterile defibrinated rat blood in the same feeding system used to infect them. After the 1-h feeding period fleas were individually examined and the feeding rate and the incidence of partial or complete proventricular blockage was recorded. Flea mortality was also recorded throughout the experiment, and samples of 20 fleas were removed at 7, 14, and 30 days after infection and frozen for bacterial load determination. Fleas showing signs of partial or full blockage were removed from the group and stored separately. These partially and completely blocked fleas were permitted to feed individually on a miniaturized artificial feeder every 1 to 3 days, and evidence of feeding, proventricular blockage status, and mortality were determined each time. Blood was collected from the reservoirs corresponding to those fleas that showed evidence of feeding or attempted feeding and plated to determine the number of CFUs transmitted. Transmissions by the same partially or completely blocked flea on multiple days were each considered as separate transmission events. Between-group differences in transmission probability and number of Y. pestis CFU transmitted were analyzed by chi-square and Mann-Whitney test, respectively (GraphPad Prism 9 software).
To determine infection rates and the average bacterial load per infected flea, samples of 20 fleas that had been collected and stored at -80°C at different times after the infectious blood meal were thawed, surface-sterilized, individually triturated in PBS, and dilutions plated in BHI soft agar overlays for CFU counts (lower limit of detection = 40 CFU/flea) as previously described [70, 71].
Deterministic host-vector model
We developed a continuous time, deterministic model formulated to reflect the dynamics of infection in the vector population as they relate to progression of disease in reservoir hosts (Fig 5). The model is based on Bailey’s single host-single vector and Ross-Macdonald models [72, 73] and uses ordinary differential equations to describe the flow of individual vertebrate hosts and vectors through different infection categories (S1 Text). Our model is focused on understanding the relative contribution of the three transmission-competent states of infection in fleas to the dynamics of infection in a host population. Thus, we did not include a birth rate for either flea or rodent compartments.
The flea vector compartment follows a susceptible-infected (SI) pattern with the infected state divided into three successive transmission-competent stages: early-phase (Iep; here defined as the first four days after the infectious blood meal), partially blocked (Ipb), and completely blocked (Ib) [14, 23]. Additionally, we account for the subset of fleas that clear themselves of infection but that may become reinfected. However, there is no final recovery state because the endstage of biofilm development in the flea (complete blockage) is invariably fatal.
Hosts follow a susceptible-exposed-infected-recovered (SEIR) pattern of disease; however, the exposed-recovered progression is disjointed from the development of overt disease. When the number of CFUs transmitted is less than a lethal dose (LD), we expect that a susceptible host will resolve the infection and recover without ever developing the terminal, high-density bacteremia (>107 Y. pestis/ml of peripheral blood) that is required to reliably infect fleas that feed on it [29, 33, 43]. Thus, transmission by fleas in any of the three transmission-competent stages (Iep, Ipb, Ib) to a susceptible host can lead to either exposed but never infectious stages (E, R), or to an infectious stage (I) characterized by a fatal bacteremia. We also included a latent stage (L) prior to the infectious stage to account for the time to develop high bacteremia after transmission.
Transmission parameter values and conditions for model simulations
Model simulations were performed using the programming language R [74] and were based on parameters generated from experimental results reported here or derived from literature sources. To simulate the cumulative transmission by 10 infected fleas feeding simultaneously in an early-phase challenge, the probability of transmission was calculated as 1 - (1- pep)10 and the probability of transmitting a lethal dose was calculated as 1 - (1- tep)10, with pep and tep being the probabilities for a single flea bite (Table 2). It was further assumed that an individual blocked flea would (conservatively) make at least two successive feeding attempts, and the pb and tb values were similarly adjusted to 1 - (1- pb)2 and 1 - (1- tb)2.
Simulation conditions were set for 100 days at 0.3-day time steps to capture acute infection dynamics. The flea population is represented by O. montana, and host populations with different susceptibilities (LD100 = 1, 10 or 100 Y. pestis CFUs) were considered. Model populations consisted of 10 hosts (1 bacteremic and 9 uninfected) and 50 uninfected fleas. When infected and dead hosts exceeded 50%, we considered the infection scenario an epizootic. Conversely, when more hosts survived than died from infection, we classified the infection as enzootic. See S1 Text for details of the model and R code.
To understand the role of the different modes of flea-borne transmission of Y. pestis in maintaining enzootic levels of the pathogen versus stimulating epizootic bursts, we evaluated all flea transmission states together and then systematically compared the individual capacity of early-phase transmission to transmission by blocked/partially blocked fleas by artificially setting one or the other transmission probabilities to zero.
Estimation of flea density required for enzootic and epizootic transmission
We formulated an expression for R0 from our model using the next-generation matrix method [42]. See S1 Text for details and R code. Based on our parameter estimates (Tables 2 and 3, and Fig 3), the number of fleas per host (m) that would be required for an enzootic (R0 = 1) or epizootic (R0 ≥ 2) was calculated.
Supporting information
S1 Text. The deterministic host-vector model; calculating R0; and model code.
https://doi.org/10.1371/journal.ppat.1010996.s001
(PDF)
S1 Fig. Model output of the dynamics of plague in host populations with different levels of susceptibility.
https://doi.org/10.1371/journal.ppat.1010996.s002
(PDF)
S2 Fig. Model output of the dynamics of plague in host populations with different levels of susceptibility using the modified parameters to account for cumulative transmission by simultaneous flea bites.
https://doi.org/10.1371/journal.ppat.1010996.s003
(PDF)
S3 Fig. Model output of the dynamics of flea infection, blockage, and mortality in scenarios with different levels of host susceptibility.
https://doi.org/10.1371/journal.ppat.1010996.s004
(PDF)
S4 Fig. Model output of the dynamics of flea infection, blockage, and mortality in scenarios with different levels of host susceptibility using the modified parameters to account for cumulative transmission by simultaneous flea bites.
https://doi.org/10.1371/journal.ppat.1010996.s005
(PDF)
Acknowledgments
The authors acknowledge the 2015 Clinic on Dynamical Approaches to Infectious Disease Data (DAIDD) instructors and participants, which is part of the International Clinics on Infectious Disease Dynamics and Data (ICI3D) Program, for their instructional support in developing the mathematical model. DAIDD 2015 was supported by the NIH National Institute for General Medical Sciences (R25GM102149) and by the South African DST-NRF Centre of Excellence in Epidemiological Modelling and Analysis (SACEMA).
References
- 1.
Gage KL, Kosoy M. Recent trends in plague ecology. In: Roelle J, Miller B, Godbey J, Biggins D, editors. Recovery of the black-footed ferret, progress and continuing challenges. Fort. Collins, CO: U. S. Geological Survey; 2006. p. 213–31.
- 2. Eisen RJ, Gage KL. Adaptive strategies of Yersinia pestis to persist during inter-epizootic and epizootic periods. Vet Res. 2009;40(2):1.
- 3. Wimsatt J, Biggins DE. A review of plague persistence with special emphasis on fleas. J Vector Borne Dis. 2009;46:85–99. pmid:19502688
- 4. Dubyanskiy VM, Yeszhanov AB. Ecology of Yersinia pestis and the epidemiology of plague. Adv Exp Med Biol. 2016;918:101–70.
- 5. Zeppelini CG, de Almeida AM, Cordeiro-Estrela P. Zoonoses as ecological entities: a case review of plague. PLoS Negl Trop Dis. 2016;10(10):e0004949. pmid:27711205
- 6. Keeling MJ, Gilligan CA. Metapopulation dynamics of bubonic plague. Nature. 2000;407:903–6. pmid:11057668
- 7. Davis S, Trapman P, Leirs H, Begon M, Heesterbeek JA. The abundance threshold for plague as a critical percolation phenomenon. Nature. 2008;454:634–7. pmid:18668107
- 8. Durham DP, Casman EA. Threshold conditions for the persistence of plague transmission in urban rats. Risk Anal. 2009;29:1655–63. pmid:19878483
- 9. Reijniers J, Davis S, Begon M, Heesterbeek JA, Ageyev VS, Leirs H. A curve of thresholds governs plague epizootics in Central Asia. Ecol Lett. 2012;15:554–60. pmid:22449078
- 10. Salkeld DJ, Salathe M, Stapp P, Jones JH. Plague outbreaks in prairie dog populations explained by percolation thresholds of alternate host abundance. Proc Natl Acad Sci USA. 2010;107:14247–50. pmid:20660742
- 11. Gascuel F, Choisy M, Duplantier JM, Debarre F, Brouat C. Host resistance, population structure and the long-term persistence of bubonic plague: contributions of a modelling approach in the Malagasy focus. PLoS Comput Biol. 2013;9(5):e1003039. pmid:23675291
- 12. Andrianaivoarimanana V, Kreppel K, Elissa N, Duplantier JM, Carniel E, Rajerison M, et al. Understanding the persistence of plague foci in Madagascar. PLoS Negl Trop Dis. 2013;7(11):e2382. pmid:24244760
- 13. Bacot AW. Further notes on the mechanism of the transmission of plague by fleas. J Hygiene Plague Suppl 4. 1915;14:774–6.
- 14. Hinnebusch BJ, Jarrett CO, Bland DM. "Fleaing" the plague: Adaptations of Yersinia pestis to its insect vector that lead to transmission. Annu Rev Microbiol. 2017;71:215–32. pmid:28886687
- 15. Bacot AW, Martin CJ. Observations on the mechanism of the transmission of plague by fleas. J Hygiene Plague Suppl 3. 1914;13:423–39.
- 16. Bland DM, Jarrett CO, Bosio CF, Hinnebusch BJ. Infectious blood source alters early foregut infection and regurgitative transmission of Yersinia pestis by rodent fleas. PLoS Pathog. 2018;14(1):e1006859.
- 17. Vetter SM, Eisen RJ, Schotthoefer AM, Montenieri JA, Holmes JL, Bobrov AG, et al. Biofilm formation is not required for early-phase transmission of Yersinia pestis. Microbiology. 2010;156:2216–25.
- 18. Dewitte A, Bouvenot T, Pierre F, Ricard I, Pradel E, Barois N, et al. A refined model of how Yersinia pestis produces a transmissible infection in its flea vector. PLoS Pathog. 2020;16(4):e1008440. pmid:32294143
- 19. Eisen RJ, Gage KL. Transmission of flea-borne zoonotic agents. Ann Rev Entomol. 2012;57:61–82. pmid:21888520
- 20. Eisen RJ, Bearden SW, Wilder AP, Montenieri JA, Antolin MF, Gage KL. Early-phase transmission of Yersinia pestis by unblocked fleas as a mechanism explaining rapidly spreading plague epizootics. Proc Natl Acad Sci USA. 2006;103:15380–5.
- 21. Eisen RJ, Dennis DT, Gage KL. The role of early-phase transmission in the spread of Yersinia pestis. J Med Entomol. 2015;52:1183–92.
- 22. Eisen RJ, Eisen L, Gage KL. Studies of vector competency and efficiency of North American fleas for Yersinia pestis: state of the field and future research needs. J Med Entomol. 2009;46:737–44.
- 23. Hinnebusch BJ, Bland DM, Bosio CF, Jarrett CO. Comparative ability of Oropsylla montana and Xenopsylla cheopis fleas to transmit Yersinia pestis by two different mechanisms. PLoS Negl Trop Dis. 2017;11(1):e0005276. pmid:28081130
- 24. Hinnebusch BJ, Perry RD, Schwan TG. Role of the Yersinia pestis hemin storage (hms) locus in the transmission of plague by fleas. Science. 1996;273:367–70. pmid:8662526
- 25. Williams SK, Schotthoefer AM, Montenieri JA, Holmes JL, Vetter SM, Gage KL, et al. Effects of low-temperature flea maintenance on the transmission of Yersinia pestis by Oropsylla montana. Vector Borne Zoonotic Dis. 2013;13:468–78. pmid:23590319
- 26. Johnson TL, Hinnebusch BJ, Boegler KA, Graham CB, MacMillan K, Montenieri JA, et al. Yersinia murine toxin is not required for early-phase transmission of Yersinia pestis by Oropsylla montana (Siphonaptera: Ceratophyllidae) or Xenopsylla cheopis (Siphonaptera: Pulicidae). Microbiology. 2014;160:2517–25. pmid:25187626
- 27. Bosio CF, Jarrett CO, Scott DP, Fintzi J, Hinnebusch BJ. Comparison of the transmission efficiency and plague progression dynamics associated with two mechanisms by which fleas transmit Yersinia pestis. PLoS Pathog. 2020;16(12):e1009092. pmid:33284863
- 28. Burroughs AL. Sylvatic plague studies. The vector efficiency of nine species of fleas compared with Xenopsylla cheopis. J Hygiene. 1947;45:371–96.
- 29. Lorange EA, Race BL, Sebbane F, Hinnebusch BJ. Poor vector competence of fleas and the evolution of hypervirulence in Yersinia pestis. J Inf Dis. 2005;191:1907–12.
- 30. Eisen RJ, Vetter SM, Holmes JL, Bearden SW, Montenieri JA, Gage KL. Source of host blood affects prevalence of infection and bacterial loads of Yersinia pestis in fleas. J Med Entomol. 2008;45:933–8.
- 31. Burroughs AL. Sylvatic plague studies X. Survival of rodent fleas in the laboratory. Parasitology. 1953;43:35–48.
- 32.
Murphy K, Travers P, Walport M, Janeway C. Janeway’s Immunobiology. 8th ed. New York: Garland Science; 2012.
- 33. Engelthaler DM, Hinnebusch BJ, Rittner CM, Gage KL. Quantitative competitive PCR as a technique for exploring flea-Yersina pestis dynamics. Am J Trop Med Hyg. 2000;62:552–60.
- 34. Jarrett CO, Sebbane F, Adamovicz JJ, Andrews GP, Hinnebusch BJ. Flea-borne transmission model to evaluate vaccine efficacy against naturally acquired bubonic plague. Infect Immun. 2004;72:2052–6. pmid:15039326
- 35. Sebbane F, Jarrett CO, Gardner D, Long D, Hinnebusch BJ. Role of the Yersinia pestis plasminogen activator in the incidence of distinct septicemic and bubonic forms of flea-borne plague. Proc Natl Acad Sci USA. 2006;103:5526–30.
- 36. Sebbane F, Jarrett C, Gardner D, Long D, Hinnebusch BJ. The Yersinia pestis caf1M1A1 fimbrial capsule operon promotes transmission by flea bite in a mouse model of bubonic plague. Infect Immun. 2009;77:1222–9.
- 37. Sebbane F, Jarrett C, Gardner D, Long D, Hinnebusch BJ. Role of the Yersinia pestis yersiniabactin iron acquisition system in the incidence of flea-borne plague. PloS One. 2010;5(12):e14379. pmid:21179420
- 38. Bosio CF, Viall AK, Jarrett CO, Gardner D, Rood MP, Hinnebusch BJ. Evaluation of the murine immune response to Xenopsylla cheopis flea saliva and its effect on transmission of Yersinia pestis. PLoS Negl Trop Dis. 2014;8(9):e3196. pmid:25255317
- 39. Dobay A, Pilo P, Lindholm AK, Origgi F, Bagheri HC, Konig B. Dynamics of a tularemia outbreak in a closely monitored free-roaming population of wild house mice. PloS One. 2015;10(11):e0141103. pmid:26536232
- 40. Eisen RJ, Wilder AP, Bearden SW, Montenieri JA, Gage KL. Early-phase transmission of Yersinia pestis by unblocked Xenopsylla cheopis (Siphonaptera: Pulicidae) is as efficient as transmission by blocked fleas. J Med Entomol. 2007;44:678–82. pmid:17695025
- 41. Wilder AP, Eisen RJ, Bearden SW, Montenieri JA, Tripp DW, Brinkerhoff RJ, et al. Transmission efficiency of two flea species (Oropsylla tuberculata cynomuris and Oropsylla hirsuta) involved in plague epizootics among prairie dogs. EcoHealth. 2008;5:205–12. pmid:18787922
- 42. Diekmann O, Heesterbeek JA, Roberts MG. The construction of next-generation matrices for compartmental epidemic models. J R Soc Interface. 2010;7(47):873–85. Epub 2009/11/07. pmid:19892718; PubMed Central PMCID: PMC2871801.
- 43. Boegler KA, Graham CB, Johnson TL, Montenieri JA, Eisen RJ. Infection prevalence, bacterial loads, and transmission efficiency in Oropsylla montana (Siphonaptera: Ceratophyllidae) one day after exposure to varying concentrations of Yersinia pestis in blood. J Med Entomol. 2016;53:674–80. pmid:26843450
- 44.
Eskey CR, Haas VH. Plague in the western part of the United States. Washington, D.C.: U.S. Public Health Service; 1940.
- 45. Webb CT, Brooks CP, Gage KL, Antolin MF. Classic flea-borne transmission does not drive plague epizootics in prairie dogs. Proc Natl Acad Sci USA. 2006;103:6236–41. pmid:16603630
- 46. Wilder AP, Eisen RJ, Bearden SW, Montenieri JA, Gage KL, Antolin MF. Oropsylla hirsuta (Siphonaptera: Ceratophyllidae) can support plague epizootics in black-tailed prairie dogs (Cynomys ludovicianus) by early-phase transmission of Yersinia pestis. Vector Borne Zoonotic Dis. 2008;8:359–67. pmid:18454591
- 47. Miarinjara A, Eads DA, Bland DM, Matchett MR, Biggins DE, Hinnebusch BJ. Reevaluation of the role of blocked Oropsylla hirsuta prairie dog fleas (Siphonaptera: Ceratophyllidae) in Yersinia pestis transmission. J Med Entomol. 2022;59:1053–9.
- 48. Kartman L, Prince FM. Studies on Pasteurella pestis in fleas. V. The experimental plague-vector efficiency of wild rodent fleas compared with Xenopsylla cheopis, together with observations on the influence of temperature. Am J Trop Med Hyg. 1956;5:1058–70.
- 49. Douglas JR, Wheeler CM. Sylvatic plague studies. II. The fate of Pasteurella pestis in the flea. J Inf Dis. 1943;72:18–30.
- 50. Wheeler CM, Douglas JR. Sylvatic plague studies V. The determination of vector efficiency. J Inf Dis. 1945;77:1–12.
- 51. Lemon A, Cherzan N, Vadyvaloo V. Influence of temperature on development of Yersinia pestis foregut blockage in Xenopsylla cheopis (Siphonaptera: Pulicidae) and Oropsylla montana (Siphonaptera: Ceratophyllidae). J Med Entomol. 2020;57:1997–2007. pmid:32533162
- 52. Eisen RJ, Lowell JL, Montenieri JA, Bearden SW, Gage KL. Temporal dynamics of early-phase transmission of Yersinia pestis by unblocked fleas: secondary infectious feeds prolong efficient transmission by Oropsylla montana (Siphonaptera: Ceratophyllidae). J Med Entomol. 2007;44:672–7. pmid:17695024
- 53. Eisen RJ, Holmes JL, Schotthoefer AM, Vetter SM, Montenieri JA, Gage KL. Demonstration of early-phase transmission of Yersinia pestis by the mouse flea, Aetheca wagneri (Siphonaptera: Ceratophylidae), and implications for the role of deer mice as enzootic reservoirs. J Med Entomol. 2008;45:1160–4. pmid:19058643
- 54. Buhnerkempe MG, Eisen RJ, Goodell B, Gage KL, Antolin MF, Webb CT. Transmission shifts underlie variability in population responses to Yersinia pestis infection. PloS One. 2011;6(7):e22498. pmid:21799873
- 55. Kartman L, Quan SF, McManus AG. Studies on Pasteurella pestis in fleas. IV. Experimental blocking of Xenopsylla vexabilis hawaiiensis and Xenopsylla cheopis with an avirulent strain. Exp Parasitol. 1956;5:435–40.
- 56. Kartman L, Prince FM, Quan SF. Studies on Pasteurella pestis in fleas VII. The plague-vector efficiency of Hystrichopsylla linsdalei compared with Xenopsylla cheopis under experimental conditions. Am J Trop Med Hyg. 1958;7:317–22.
- 57. Hinnebusch BJ, Rudolph AE, Cherepanov P, Dixon JE, Schwan TG, Forsberg Å. Role of Yersinia murine toxin in survival of Yersinia pestis in the midgut of the flea vector. Science. 2002;296:733–5.
- 58. Lemon A, Sagawa J, Gravelle K, Vadyvaloo V. Biovar-related differences apparent in the flea foregut colonization phenotype of distinct Yersinia pestis strains do not impact transmission efficiency. Parasit Vectors. 2020;13(1):335. pmid:32611387
- 59. Kartman L, Prince FM, Quan SF, Stark HE. New knowledge on the ecology of sylvatic plague. Ann N Y Acad Sci. 1958;70:668–711. pmid:13559927
- 60.
Pollitzer R. Plague. Geneva: World Health Organization; 1954.
- 61. Lillard JW, Bearden SW, Fetherston JD, Perry RD. The haemin storage (Hms+) phenotype of Yersinia pestis is not essential for the pathogenesis of bubonic plague in mammals. Microbiology. 1999;145(:197–209. pmid:10206699
- 62. Sun YC, Jarrett CO, Bosio CF, Hinnebusch BJ. Retracing the evolutionary path that led to flea-borne transmission of Yersinia pestis. Cell Host Microbe. 2014;15:578–86.
- 63. Gandon S, Heitzmann L, Sebbane F. To block or not to block: The adaptive manipulation of plague transmission. Evol Lett. 2019;3:152–61. pmid:31161047
- 64. McCoy GW. Studies upon plague in ground squirrels. Publ Hlth Bull. 1911;43:3–51.
- 65. Williams JE, Moussa MA, Cavanaugh DC. Experimental plague in the California ground squirrel. J Infect Dis. 1979;140:618–21. pmid:512421
- 66. Quan TJ, Barnes AM, Carter LG, Tsuchiya KR. Experimental plague in rock squirrels, Spermophilus variegatus (Erxleben). J Wildl Dis. 1985;21:205–10. pmid:4032620
- 67. Pulliam JRC, Dushoff JG, Levin SA, Dobson AP. Epidemic enhancement in partially immune populations. PloS One. 2007;2(1) e165. pmid:17225866
- 68. Tedrow RE, Zimmerman PA, Abbott KC. Multiple blood feeding: a force multiplier for transmission. Trends Parasitol. 2019;35:949–52. pmid:31585840
- 69. Engelthaler DM, Gage KL, Montenieri JA, Chu M, Carter LG. PCR detection of Yersinia pestis in fleas: comparison with mouse inoculation. J Clin Microbiol. 1999;37:1980–4.
- 70. Bland DM, Hinnebusch BJ. Feeding behavior modulates biofilm-mediated transmission of Yersinia pestis by the cat flea, Ctenocephalides felis. PLoS Negl Trop Dis. 2016;10(2):e0004413. pmid:26829486
- 71. Bland DM, Brown LD, Jarrett CO, Hinnebusch BJ, Macaluso KR. Methods in Flea Research [Internet]. BEI Resources. 2017. Available from: https://www.beiresources.org/Portals/2/Methods%20in%20Flea%20Research%20v2.pdf.
- 72.
Bailey NTJ. The Biomathematics of Malaria. London: Charles Griffin & Co. Ltd.; 1982.
- 73. Smith DL, Battle KE, Hay SI, Barker CM, Scott TW, McKenzie FE. Ross, Macdonald, and a theory for the dynamics and control of mosquito-transmitted pathogens. PLoS Pathog. 2012;8(4):e1002588. pmid:22496640
- 74.
R Core Team. R: A language and environment for statistical computing. Vienna, Austria: R Foundation for Statistical Computing; 2019.