Fig 1.
Cat fleas feed more often than rat fleas when given unrestricted access to blood.
50 male and 50 female C. felis or X. cheopis fleas were starved for 5 days, placed into a clean capsule, and allowed to feed for either 4 or 24 hours on defibrinated sheep blood in an artificial membrane system. Images show the amount of fecal material that soiled the bottom of the capsule at the end of the feeding period.
Fig 2.
Feeding frequency influences colonization of the cat flea digestive tract by Y. pestis.
Fleas that took a single infectious blood meal that contained ~ 2x108 CFU/ml of Y. pestis were subsequently provided uninfected maintenance feeds at low frequency (twice weekly for 4h), moderate frequency (five times weekly for 4h), or high frequency (daily for 18-22h) according to the schedule shown in (A). Black and grey bars indicate the infectious and maintenance blood meals, respectively; and the width of the bar corresponds to the 4h or 18-22h feeding period. (B) The bacterial load in individual fleas from the low-, moderate-, and high-frequency groups was determined in samples of 10–20 fleas collected immediately after the 4h infectious blood meal (day 0) or at 7–28 days after infection. The low- and high-frequency groups fed on the same infectious blood meal prior to segregating them. Horizontal bars represent the mean CFU/flea; dotted lines indicate the limit of detection (40 CFU). The cumulative results of three independent experiments are shown (Day 0, n = 60; Day 7, n = 60; Day 14, n = 54, 60, 60); Day 28 n = 50, 60, 40). *p <0.05 by Student’s t-test (Day 0) or by Kruskal-Wallis test with Dunn’s multiple comparison (Day 7–28). (C) Infection rate (percentage of fleas from which CFUs were recovered) at different times after infection. The mean and range of 3 experiments are indicated; *p <0.0166 by chi-square test with Bonferonni’s post-test. (D) Cumulative mortality of infected fleas and uninfected control fleas at the end of the 28-day experiments. The mean and standard deviation of 2–3 experiments with each of the three feeding frequency groups are indicated (low, n = 312, 310; moderate, n = 292, 276; high, n = 184, 214; cumulative number of infected fleas and uninfected control fleas, respectively). *p <0.0055 by chi-square test with Bonferonni post-test; NS = not significant.
Fig 3.
Cat flea feeding frequency influences severity of proventricular infection.
38–60 mixed-sex cat fleas were dissected 7 days after a blood meal containing ~2x108 Y. pestis KIM6+ (pAcGFP1) CFU/ml and the digestive tracts were visualized with fluorescence microscopy for the presence and localization of a Y. pestis biofilm. Representative images of an uninfected flea (A; no visible bacteria in the digestive tract; the spines of the proventriculus [PV] autofluoresce yellow); a flea with an infection only in the midgut (MG) (B; no green-fluorescent bacteria observable in the PV); and fleas with a light-moderate PV infection (C; <75% bacterial coverage of the PV) or a heavy PV infection (D; ≥75% bacterial coverage of the PV). Areas that appear white are due to fluorescence overexposure when acquiring the image. Scale bars = 100 μm (A, B) or 50 μm (C, D). (E-H) The percentage of infected fleas (E), and of those, the percentages with midgut-only infections (F), and with midgut plus light-moderate (G) or heavy (H) PV infections in three independent infection experiments for the three feeding-frequency groups are shown; bars indicate the mean. Each symbol represents an individual experiment. One experiment with the high-frequency group had no infected fleas at 7 days (thus only two symbols shown for this group in F-H). Numbers in parentheses indicate the number of positives/total number of fleas examined. * p < 0.0166 by chi-square test with Bonferonni post-test; NS = not significant.
Fig 4.
Y. pestis is capable of producing proventricular blockage in C. felis.
(A) Female cat flea blocked with Y. pestis expressing GFP, examined immediately after a feeding attempt. Fresh red blood is present in the esophagus only (arrow), with none in the midgut, which contains dark, digested blood from previous blood meals. (B) Bright field and (C) fluorescent microscopy images of the dissected digestive tract of the same blocked flea, showing the dense GFP+ bacterial mass that fills and blocks the proventriculus (PV) and also extends into the esophagus (E) and midgut (MG). Fresh red blood that was blocked from entering the midgut is still observable in the esophagus. (D-F) The same image series for an uninfected cat flea, which has fresh red blood in the midgut and an absence of a dark bacterial mass in the proventriculus. Scale bar = 100 μm.
Table 1.
Summary of C. felis infection experiments.
Fig 5.
Cat fleas transmit Y. pestis after the early phase at an efficiency that correlates with feeding frequency and proventricular blockage rate.
The number of Y. pestis CFUs recovered from the blood meal reservoir after maintenance feeds by ~100 cat fleas fed for 28 days at either A) low frequency (twice a week for 4 hours), B) moderate frequency (five times a week for 4 hours), or C) high frequency (daily for 18–22 hours) from 3 independent experiments (top to bottom, experiments 1 through 3). Blood was plated after a 4-hour feeding period for the low- and moderate-frequency groups (grey bars) or after an 18–22 hour feeding period for the high-frequency group (hatched bars). All infected flea groups initiated their maintenance feeding schedules 24 hours after the infectious blood meal. The numbers above each bar indicate the number of non-blocked and blocked fleas that fed on that day (# non-blocked (dash) # blocked). Blocked fleas were removed from the feeding pool upon identification. N, not fed on the indicated day; arrows indicate expected peak occurrence of early-phase transmission (EPT); X indicates that feeding began on day 1 post-infection but extended into day 2 (see Fig 2A). The results of three independent experiments with each group are shown.
Table 2.
Transmission efficiency of infected cat fleas.
Fig 6.
Transmission by individual blocked cat fleas.
Frequency distribution of CFUs recovered from blood meal reservoirs fed on by individual blocked cat fleas (n = 25) after a 4-hour feeding period. DNF = did not attempt to feed.
Fig 7.
The cat flea has a narrower esophagus and thicker proventricular musculature than the rat flea, X. cheopis.
Digestive tracts dissected from recently fed, uninfected female C. felis (A) or X. cheopis (B) fleas in 20 μl of PBS were overlaid with a cover slip and the base of the esophagus (E) and the widest part of the proventriculus (P), and the proventriculus plus surrounding musculature (P+M) were measured. Dotted red lines indicate the location where measurements were taken. (C-E) Width of the esophagus, proventriculus, and proventricular muscle layer (P+M–P)/2) of individual fleas. (F) Ratio of proventricular width to esophageal width (P/E). Horizontal bars indicate the mean and error bars the standard deviation (n = 20). *p > 0.0001 by Student’s t-test. Scale bar = 50 μm.
Fig 8.
Proventricular musculature surrounds and constricts the base of the esophagus of C. felis, but not of X. cheopis.
Digestive tracts dissected from uninfected female C. felis (A) or X. cheopis (B) fleas in 20 μl of PBS were photographed without the application of a cover slip. Arrows indicate a layer of muscle surrounding the base of the esophagus that is present in C. felis but absent in X. cheopis. (C-F) Actin stains of fixed, permeablized C. felis (C, E) and X. cheopis (D, F) digestive tracts were visualized by bright field and fluorescence microscopy (red fluorescence = actin). Representative images of 14–18 stained digestive tracts from 2 independent experiments are shown. Scale bar = 50 μm.
Fig 9.
The cat flea esophagus is less prone to blockage-induced distension than the rat flea esophagus.
Digestive tracts dissected from blocked female C. felis (A) or X. cheopis (B) fleas in 20 μl of PBS with (A, B) or without (C, D) the application of a cover slip. Representative images of 12–18 (with coverslip) or 15 (no coverslip) blocked fleas are shown. The width of the esophagus (E) and proventriculus (F) of individual blocked fleas was measured, and the proventriculus::esophagus width ratio (P/E) was calculated (G). Horizontal bars indicate the mean and error bars the standard deviation (n = 12–18). *p > 0.0001 by Student’s t-test. Scale bar = 50 μm.