Journal of Medical Entomology, XX(X), 2018, 1–7

doi: 10.1093/jme/tjy055
Vector/Pathogen/Host Interaction, Transmission Research

Effects of Bacterial Dose and Fly Sex on Persistence and

Excretion of Salmonella enterica serovar Typhimurium
From Adult House Flies (Musca domestica L.; Diptera:
Muscidae)
Dana Nayduch,1,4 Klara Zurek,2 Jessica L. Thomson,2 and Kathleen M. Yeater3

1
USDA-ARS, Arthropod-Borne Animal Diseases Research Unit, Center for Grain and Animal Health Research, 1515 College Ave,
Manhattan, KS 66502, 2Department of Entomology, Kansas State University 123 W. Waters Hall, Manhattan, KS 66506, 3USDA-ARS-
PA-NRRC, Office of the Director, 2150 Centre Avenue, Building D, Suite 300, Fort Collins, CO 80526, and 4Corresponding author,
e-mail: [email protected]

Subject Editor: Lane Foil

Received 30 January 2018; Editorial decision 20 March 2018

Abstract
Salmonella Typhimurium (Le Minor and Popoff 1987; Enterobacteriales: Enterobacteriaceae) is a pathogen
that causes gastroenteritis in humans and can be harbored by house flies. Factors influencing excretion of S.
Typhimurium from infected flies have not been elucidated but are essential for assessing transmission potential. We
determined the persistence and excretion of a green fluorescent protein (GFP) expressing strain of S. Typhimurium
from house flies. Individual male and female flies were fed either sterile Luria-Bertani (LB) broth (controls) or
cultures of “high” (~105 colony forming units [CFU]) or “low” (~104 CFU) doses of bacteria (treatments). Bacterial
persistence was determined over 16 h by culturing whole-fly homogenate. Both sex and dose affected persistence
between 6 and 12 h postingestion. In a separate experiment, fly excretion events were monitored during this
time interval and excreta droplets were individually cultured for bacteria. Female flies had more excretion events
than males across treatments. We observed interactions of fly sex and bacterial abundance (dose), both on the
proportion of Salmonella-positive droplets and the CFU shed per droplet (CFU/droplet). In the low-dose treatment,
males excreted a greater proportion of positive droplets than females. In the high-dose treatment, males excreted
more CFU/droplet than females. High-dose male flies excreted more CFU/droplet than low-dose males, but low-
dose females excreted more CFU/droplet than high-dose females. Irrespective of sex, low-dose flies excreted a
greater dose-adjusted CFU (CFU droplet/CFU fed) than high-dose flies. This study demonstrates that both bacterial
abundance and fly sex may influence excretion of bacteria from flies, and should be considered when assessing the
risk of house fly transmission of pathogens.

Key words: Musca domestica, bacteria, pathogen, vector, excreta

Due to their reproductive and trophic link to microbe-rich substrates biological vectors by shedding ingested bacteria in their excreta. This
such as manure and garbage, house flies have long been implicated type of vectoring requires persistence and proliferation of pathogens
as reservoirs and disseminators of pathogenic bacteria and other within the vector followed by transmission, which, in the case of
microbes (reviewed in Nayduch and Burrus 2017). Adult flies inti- flies, would occur via regurgitation or defecation of excreta droplets.
mately associate with and feed upon the septic substrates they visit. Ingested bacteria that survive in the fly gut and exit in excreta have
As a result, numerous species of microbes, including bacterial patho- the potential to be disseminated over time and space as the fly moves
gens, have been isolated from wild-caught flies in surveys (Nazni within the environment.
et al. 2005, Förster et al. 2007, Gupta et al. 2012, Ommi et al. 2017). Bacteria ingested by flies and harbored in the alimentary canal
House flies are gregarious, synanthropic and zoophilic, serving as face methods of elimination, primarily by destructive forces such as
a bridge between filth and human or animal habitation. Transmission digestive enzymes and the immune response, followed by peristaltic
of bacteria is often by mechanical methods, such as contaminated expulsion (Terra et al. 1988, Nayduch et al. 2013, Fleming et al. 2014,
tarsi and mouthparts (Barro et al. 2006), but flies can also serve as Gill et al. 2017). The immune response likely underlies the previously

Published by Oxford University Press on behalf of Entomological Society of America 2018. 1

This work is written by (a) US Government employee(s) and is in the public domain in the US.

Downloaded from https://academic.oup.com/jme/advance-article-abstract/doi/10.1093/jme/tjy055/4967820

by guest
on 17 May 2018
 2 Journal of Medical Entomology, 2018, Vol. XX, No. XX

observed phenomenon that survival and persistence of bacteria in study carries a pGFPuv plasmid (Clontech Laboratories, Mountain
the gut was correlated with the number of cells that were ingested View, CA) with both kanamycin and ampicillin resistance genes
(Kumar and Nayduch 2016). The homeostatic mechanism of the local (McGaughey and Nayduch 2009) and previously has been used
gut immune response has been well described in fruit flies (Zaidman- in fly-feeding assays (Chifanzwa and Nayduch 2017). Fly-feeding
Rémy et al. 2006, Buchon et al. 2013) and involves mechanisms for assays were performed using Luria-Bertani (LB) broth containing
scavenging peptidoglycan (PGN) shed from ingested bacteria such that antibiotics for GFP plasmid maintenance (50 µg/ml kanamycin and
unnecessary immune responses are not mounted to low-abundance 50 µg/ml ampicillin, Fisher Scientific; Amp/Kan LB). On the day
populations in the gut. However, if bacterial abundance is high, either prior to each assay, bacteria were cultured overnight in 50 ml Amp/
due to a large amount being ingested or subsequent to proliferation of Kan LB broth at 37°C with 60 rpm rotation. On the day of the
ingested cells, the PGN scavenging capability is overwhelmed, and the experiment, 100 µl of overnight culture was inoculated into 15 ml
immune response is activated by free PGN components. The result is Amp/Kan LB. These cultures were incubated as mentioned previ-
that the flies mount an immune response proportionate to the number ously and monitored via spectrophotometry (OD600) for desired den-
of cells (as sensed by the amount of PGN) in the gut, which allows for sity. Growth curves for GFP S. Typhimurium were performed prior
control of bacterial population overgrowth while preventing needless, to fly-feeding assays (data not shown) and were used to determine
persistent immune induction. Whether a similar mechanism exists in time and OD600 dynamics relative to the desired bacterial abundance
other flies remains to be determined, but the house fly gut epithelium (as CFU, colony forming units) to be used in the assays (i.e., high and
mounts a local immune response to ingested bacteria (Joyner et al. low doses, described in the following paragraphs). Bacterial abun-
2013, Nayduch et al. 2013, Fleming et al. 2014), which indicates that dance was confirmed by culturing the suspension after all flies had
bacteria presence is detected. Nonetheless, some species of bacteria fed on bacterial droplets.
evidently withstand or avoid these defensive assaults and can prolifer-
ate in the fly gut, sometimes persisting for several days (Greenberg Fly Droplet Feeding for Assays
et al. 1970, Nayduch et al. 2002, Joyner et al. 2013, El-Bassiony et al. For each replicate in each assay, individually housed flies (n = 3
2016, Chifanzwa and Nayduch 2017). males, n = 3 females) were presented a 2 µl droplet which contained
Salmonella enterica serovar Typhimurium (S. Typhimurium) is a either sterile culture media (Amp/Kan LB broth) and/or Amp/Kan
zoonotic enteropathogen carried in the digestive tracts of several live- LB broth containing high or low doses of bacteria (details for each
stock species, including cattle, pigs, and chickens (Tauxe 1991, Van assay follow in the subsequent paragraphs). Each assay was repli-
Duijkeren et al. 2002). Salmonella spp. were isolated from house flies in cated twice.
surveys from agricultural and domestic habitats (Olsen and Hammack
2000, Ugbogu et al. 2006, Choo et al. 2011). Salmonella spp. survive
Enumeration of GFP S. Typhimurium From Whole
on both the surfaces and within the digestive tracts of experimen-
Homogenized House Flies
tally exposed house flies (Pava-Ripoll 2012, Thomson et al. 2017).
Individually housed male and female flies were fed a 2 µl droplet of
Greenberg (1964) also demonstrated that house flies exposed to feces
Amp/Kan LB which contained either low (4.27 ± 2.00 × 104 CFU) or
inoculated with S. Typhimurium can transmit the pathogen to humans
high (2.31 ± 0.52 × 105 CFU) doses of GFP S. Typhimurium. In each
via contaminated food. The effects of ingested bacterial dose on the
treatment and replicate, bacteria-fed flies (n = 3 of each sex) were
persistence, proliferation, and temporal survival of a GFP-expressing
chill-immobilized at 3, 6, 12, and 16 h postingestion (PI). Flies were
strain of S. Typhimurium in house flies have been investigated previ-
surface sanitized by washing in 10% household bleach (Great Value,
ously (Chifanzwa and Nayduch 2017). However, fly sex was not con-
Walmart, Bentonville, AR) for 5 min followed by 1 min in 70%
sidered in study designs or analyses and effects of dose on excretion
ethanol, then individually homogenized with a motorized pestle in
of bacteria was not determined. Thus, we investigated the role that
500 µl sterile phosphate-buffered saline (PBS; per 1 l: 8 g NaCl, 0.2 g
both fly sex and bacterial abundance (dose) play in the persistence and
KCl, 1.44 g Na2HPO4, 0.24 g KH2PO4, pH 7.4; Fisher Scientific).
excretion of GFP S. Typhimurium from male and female house flies.
Homogenate was serially diluted and cultured in duplicate on Amp/
Kan LB agar. Cultures were incubated at 37°C for 24 h for GFP S.
Typhimurium enumeration.
Materials and Methods Statistical analyses of the bacterial abundance (CFU) were per-
House Fly Rearing and Preparation formed using the GLIMMIX procedure in SAS/STAT 14.1 (SAS
Late stage pupae of M. domestica were collected from colonies estab- Institute Inc., Cary, NC). The LSMEANS and LSMESTIMATE
lished at Kansas State University in 2011. Pupae were sanitized by options were invoked to compare means (1) within sex and dose
sequential washes in 10% sodium hypochlorite, 70% ethanol, and ster- treatment across PI time points and (2) within each PI time point
ile deionized water, 2 min each. Pupae were kept individually in sterile (a) across sex, within dose treatment and (b) across dose treatment
15 ml conical vials at 28°C for imago emergence. Newly emerged flies within sex. In order to include zero CFU values, the measured CFU
were sorted by sex, transferred with sterile forceps to individual sterile variable was transformed, Log10 (CFU + 0.01). The sex, dose treat-
60 × 15 mm plastic petri dishes (Fisher Scientific, Atlanta, GA) and ment, and PI time points were all fixed effects and the replicate was
fasted 24 h at 21°C under a 14:10 light:dark photoperiod. Following random with each fly identified as the experimental unit.
this, flies were fed a 5µl droplet of 10% sterile sucrose on a piece of
Parafilm (Fisher Scientific). Flies were then fasted an additional 24 h Observation of Excretion Events and Enumeration
at 21°C as mentioned previously, after which they were fed bacteria of GFP S. Typhimurium in Excreta Droplets
cultures or broth as described in the following sections. Individually housed male and female flies were fed a 2 µl droplet
of Amp/Kan LB (broth, controls) or Amp/Kan LB which contained
Bacterial Culture either low (3.26 ± 0.01 × 104 CFU) or high (1.39 ± 0.01 × 105 CFU)
The strain of Salmonella enterica serovar Typhimurium doses of bacteria (treatments). Flies were then left at room tempera-
(S. Typhimurium) SR-11 (Schneider and Zinder 1956) used in this ture (~21°C) and observed continuously from 6 to 12 h PI. Beginning

Downloaded from https://academic.oup.com/jme/advance-article-abstract/doi/10.1093/jme/tjy055/4967820

by guest
on 17 May 2018
 Journal of Medical Entomology, 2018, Vol. XX, No. XX 3

at 6 h PI, flies were transferred to fresh, sterile 60 × 15 mm Petri in SAS/STAT 14.1. The fixed effects were sex and dose treatments
dishes with sterile forceps. For the next 6 h, individual flies were and the replicate was random. The time to the excreta event was
monitored and excretion events (droplets that were either regurgi- modeled as a repeated measure with individual fly identified as the
tated or defecated) were documented and circled on each Petri dish subject unit. The covariance of the excretion time from observation
at the time of excretion. At 30-min time intervals, each fly with an to observation within each fly is a compound symmetry relationship.
excretion event was briefly chill-immobilized by covering the dish Post hoc comparisons of the fixed effects and their interaction were
with crushed ice and transferred to a new Petri dish at the start of the determined with the LSMEANS and ESTIMATE options. (4) Does
next interval. After the fly was transferred, individual excreta drop- the dose-adjusted mean CFU per droplet differ between low- and
lets were washed from the Petri dish with sterile PBS via repeated high-dose treatments, and is there an effect of sex within treatment
pipette aspiration, and each was cultured on Amp/Kan LB agar. and/or an effect of treatment within sex? For this analysis, raw CFU
Culture plates were incubated at 37°C for 24 h for enumeration of per positive droplet were divided by the amount fed in that treat-
GFP S. Typhimurium. ment (i.e., this dose-adjustment gives values that represent the frac-
tion of the amount ingested that was present in excreta). These data
Statistical Analysis of Excretion Events and Bacteria were analyzed using the GLIMMIX procedure in SAS/STAT 14.1.
Enumeration The fixed effects were sex and dose treatments and the replicate was
random. Post hoc comparisons of the fixed effects and their interac-
There were four questions regarding the excretion events and enu-
tion were determined with the LSMEANS and ESTIMATE options
meration of bacteria from excreta across fly sex and treatment (low,
in SAS.
high doses of bacteria). (1) Do the number of excretion events differ
across treatments [control (broth), low-dose bacteria, high-dose bac-
teria] within each fly sex? This analysis was performed with a two- Results
way ANOVA in the Fit Model application of JMP 12 (SAS Institute
Inc.). The dose treatment and sex were fixed effects and the replicate
Enumeration of GFP S. Typhimurium From
was random. (2) Does the proportion of “positive” droplets (i.e., Homogenized House Flies
droplets positive for GFP S. Typhimurium out of the total number Bacteria were enumerated from whole, surface-sanitized male and
of droplets) differ between low- and high-dose treatments, and is female flies that were fed high and low doses of bacteria (Fig. 1).
there an effect of sex within treatment and/or an effect of treatment Overall, there were significant effects of sex (F = 4.16, df = 1,75,
within sex? This analysis was performed using the GLIMMIX pro- P = 0.045), treatment (F = 15.57, df = 1,75, P = 0.0002), and time
cedure in SAS/STAT 14.1. The fixed effects of the sex and dose treat- (F = 6.15, df = 1,75, P = 0.0009) on the mean CFU recovered from
ments were modeled on the proportion of infective excreta droplets flies. Post hoc comparisons were made across time points within sex
of the total events using the binomial distribution. The replicate and treatment. Additionally, CFU was compared within each time
was a random effect in the model. The post hoc comparisons of point both within treatment across sex and within sex across treat-
sex, dose treatments, and their interaction were calculated with the ment as follows.
LSMEANS option. (3) Does the mean abundance of bacteria (mean
CFU) per positive droplet differ between low- and high-dose treat- CFU comparisons across time points
ments, and is there an effect of sex within treatment and/or an effect Interestingly, bacterial persistence patterns were similar across male
of treatment within sex? For this analysis, CFU from bacteria-posi- and female flies fed the high dose of bacteria (Fig. 1). In both sexes,
tive excreta droplets were natural log-transformed prior to statisti- there was a significant difference in the mean CFU recovered from
cal analysis. Analysis was performed using the GLIMMIX procedure flies between 3 and 16 h PI (males: t = 2.21, df = 75, P = 0.0298;

Low dose male (104 CFU) Low dose female (104 CFU)
6 5
High dose male (10 CFU) High dose female (105 CFU)
Log10 (CFU+1) bacteria per fly

0
0 4 8 12 16
time post ingestion (h)

Fig. 1. Persistence of GFP S. Typhimurium in male and female house flies. Individual flies were fed Luria-Bertani (LB) broth containing low (4.27 ± 2.00 × 104
CFU) or high (2.31 ± 0.52 × 105 CFU) doses of GFP S. Typhimurium. At 3, 6, 12, and 16 h postingestion, whole flies were collected and homogenized for bacteria
culture and enumeration as described in the methods. Statistical analyses of bacteria recoveries are discussed in the text. Mean ± SEM Log10 (CFU + 1) bacteria
shown for n = 6 flies per time point.

Downloaded from https://academic.oup.com/jme/advance-article-abstract/doi/10.1093/jme/tjy055/4967820

by guest
on 17 May 2018
 4 Journal of Medical Entomology, 2018, Vol. XX, No. XX

females: t = 2.03, df = 75, P = 0.0455). In contrast, mean CFU recov- treatment (Fig. 2A) as well as overall (mean number of events:
ered from male and female flies fed the low dose of bacteria had females, 8.11; males, 4.94).
somewhat dissimilar patterns of persistence. In male flies fed the low The remaining analyses, in the following paragraphs, were con-
dose of bacteria, there was a significant difference in the mean CFU ducted on flies in the bacteria treatment groups only. CFU of GFP
recovered between 3 and 12 h (t = 3.54, df = 75, P = 0.0007) and S. Typhimurium enumerated from these flies and times of excretion
3 and 16 h PI (t = 3.84, df = 75, P = 0.0003), and only 67% of events shown in Fig. 3 were used for these analyses. We noted that
flies had recoverable GFP S. Typhimurium at both 12 and 16 h PI. female flies in either treatment shed a larger number of excreta drop-
However, the mean CFU cultured from female flies that were fed the lets without GFP S. Typhimurium than males (Fig. 3). For example,
low dose of bacteria was not significantly different between any of in the low dose, only two male flies shed excreta without bacteria
the time points. (five total droplets), whereas four of six females shed excreta without
bacteria (34 total droplets), and three of those females shed no bac-
CFU comparisons within time points teria in any of their excreta. In the high dose, three of six males shed
The only significant differences in mean CFU recovered from flies excreta without bacteria (12 total droplets), with one male shed-
were seen in comparisons to the low-dose male flies, who differed ding no bacteria; however, five of six females shed excreta without
from high-dose males at 12 h (t = −3.55, df = 75, P = 0.0007) and bacteria (24 total droplets), and two of those females shed no bac-
16 h PI (t = −2.32, df = 75, P = 0.0231) and from low-dose females teria in any of their excreta. Although these data were not analyzed
also at these same time points (12 h: t = −2.14, df = 75, P = 0.0358; separately, they are incorporated within the proportion analysis that
16 h: t = −2.66, df = 75, P = 0.0095). follows.

Excretion Events From Male and Female Flies Fed Proportion of Salmonella-positive excreta droplets
Control Broth and Low and High Doses of GFP There was no overall effect of treatment on the proportion of posi-
S. Typhimurium tive excreta droplets (those containing GFP S. Typhimurium) shed
Excretion events by flies (F = 0.01, df = 1,19, P = 0.9335). The mean ± SEM propor-
There was no difference in the number of excretion events (Fig. 2A) tions of positive droplets in each group (irrespective of sex) were:
between flies in control, low-dose and high-dose treatments high dose, 0.63 ± 0.29; low dose, 0.64 ± 0.29. However, there was
(F = 0.1815, df = 2, 2, P = 0.8350), and there was no interaction a sex and treatment interaction (F = 5.26, df = 1,19, P = 0.0333),
between treatment and sex (F = 0.1913, df = 2, 29, P = 0.8269). which was apparent in the low-dose treatment (t = 3.83, df = 19,
However, there was a sex effect (F = 0.1913, df = 1, 29, P = 0.0004), P = 0.0011). Notably, the proportion of the positive droplets was
and female flies had more excretion events than male flies in each greatest in low-dose males (0.85 ± 0.17) (Fig. 2B).

A. B.
15
Control (Broth) 1.2
***
Proportion positive droplets

Low dose (104 CFU) 1.0

Excretion events

High dose (105 CFU)

10 0.8

0.6

5 0.4

0.2

0 0.0
Male Female Male Female

C. D.
***
1200 2.0
* **
*
1000
Dose-adjusted CFU
Mean CFU/droplet

1.5 **
800

600 1.0

400
0.5
200

0 0.0
Male Female Male Female

Fig. 2. House fly sex and bacterial-dose effects on excretion of GFP S. Typhimurium. Flies (n = 6 per sex/treatment) were fed LB broth or broth containing low
(3.26 ± 0.01 × 104 CFU) or high (1.39 ± 0.01 × 105 CFU) doses of GFP S. Typhimurium, and individual excreta were collected at 6–12 h postfeeding for culture. (A)
Excretion events from flies fed control broth, low and high doses of bacteria. (B) Proportion of Salmonella-positive droplets from bacteria-fed flies (droplets
containing GFP S. Typhimurium/total droplets). (C) CFU of GFP S. Typhimurium per bacteria-positive droplet. (D) Dose-adjusted CFU (CFU in droplet/CFU fed).
Differences within sex across treatment and within treatment across sex were examined via two-way ANOVA (panel A) or GLIMMIX procedure with LSMEANS
option (panels B, C, and D). Overall, female flies had more excretion events than male flies (panel A; P = 0.0004). For pairwise comparisons in panels B, C, and
D: * P ≤ 0.05, **P ≤ 0.01, ***P ≤ 0.001. Mean ± SEM are shown.

Downloaded from https://academic.oup.com/jme/advance-article-abstract/doi/10.1093/jme/tjy055/4967820

by guest
on 17 May 2018
 Journal of Medical Entomology, 2018, Vol. XX, No. XX 5

Fig. 3. Excretion events from house flies fed GFP S. Typhimurium. Male (A, C) and female (B, D) house flies (n = 6 per sex/treatment) were fed broth containing
low (3.26 ± 0.01 × 104 CFU; A, B) or high (1.39 ± 0.01 × 105 CFU; C, D) doses of GFP S. Typhimurium and excretion events were continuously monitored during
the 6- to 12-h postingestion interval. Time and number of events were recorded, and excreta were collected for bacteria culture and enumeration, as described
in the methods. Log10 (CFU + 1) shown. Control flies fed LB broth also were monitored for time and number of excretion events but contained no CFU of GFP
S. Typhimurium and are not represented here.

Abundance of GFP S. Typhimurium shed in fly excreta in a treatment effect on the CFU excreted in both females and male
There was no overall effect of treatment on the abundance of bac- flies. In females, high-dose flies excreted 152.67 ± 88.01 CFU and
teria (CFU) shed by flies in excreta droplets (F = 0.07, df = 1,75, low-dose flies excreted 374.94 ± 67.72 CFU (t = −2.71, df = 75,
P = 0.7909). The mean ± SEM CFU GFP S. Typhimurium in excreta P = 0.0084). In males, high-dose flies excreted 792.67 ± 254.22 CFU
droplets in each treatment (irrespective of sex) were: high dose, and low-dose flies excreted 280.70 ± 111.28 CFU (t = 2.46, df = 75,
448.05 ± 134.97; low dose, 322.07 ± 68.78. However, there was a sex P = 0.0162). Within treatment (dose), CFU excreted in droplets
and treatment interaction (F = 13.4, df = 1,75, P = 0.0005) evidenced from male and female flies differed only in the high-dose treatment

Downloaded from https://academic.oup.com/jme/advance-article-abstract/doi/10.1093/jme/tjy055/4967820

by guest
on 17 May 2018
 6 Journal of Medical Entomology, 2018, Vol. XX, No. XX

(t = 3.66, df = 75, P = 0.0005). Of note, high-dose males excreted the (e.g., positive droplets/total droplets by mean CFU/positive drop-
greatest number of bacteria (Fig. 2C). let), among all treatment groups, the greatest vectoring potential
was seen in the high-dose male flies while the lowest was seen in
Dose-adjusted GFP S. Typhimurium shed in fly excreta high-dose females.
There was an overall effect of treatment on the dose-adjusted CFU Our results revealed both sex and bacterial-dose effects on bac-
shed by flies (CFU in the droplet/CFU ingested), with low-dose flies terial persistence and excretion. Physiological differences between
excreting a greater abundance of dose-adjusted CFU than high-dose male and female flies that may underlie sex effects remain to be
flies (F = 7.19, df = 1,75, P = 0.009). The mean ± SEM dose-adjusted determined, but sex-specific nutritional needs such as female protein
CFU in excreta droplets in each treatment (irrespective of sex) were: requirements for vitellogenesis (Kobayashi 1934) could putatively
high dose, 0.28 ± 0.34; low dose, 0.88 ± 0.34. There was not an over- impact digestive processes and peristalsis rates. Our future studies
all sex and treatment interaction (F = 5.92, df = 1,75, P = 0.1206). are aimed at determining if there is a sex effect when bacteria are fed
However, within females, dose-adjusted CFU differed between high- in other types of media and substrates, especially those that would
and low-dose treatments (t = −2.98, df = 75, P = 0.0039), and within better represent bacterial acquisition in natural conditions.
the high-dose treatment, dose-adjusted CFU differed between male Dose effects on bacteria persistence and excretion in both male
and female flies (t = 2.33, df = 75, P = 0.0226) (Fig. 2D). and female flies may be mediated by the local epithelial immune
response in the gut. Insight into this effect is evident in the compari-
sons of dose-adjusted CFU across treatment groups. Irrespective of
Discussion sex, flies that ingested the low dose of GFP S. Typhimurium shed
Culture of whole, surface-sanitized flies revealed similar patterns of a larger fraction of what they ingested as compared to flies who
persistence of GFP S. Typhimurium in both male and female flies fed ingested the high dose. This phenomenon may occur if the low dose
the high dose of bacteria (~105 CFU per fly), but sex effects were seen of bacteria were below the threshold of local immune detection
in flies from the low-dose treatment groups (~104 CFU per fly). Apart and/or immune stimulation in the gut. In addition to digestion, pH
from our statistical analyses, we noticed distinct patterns of bacte- extremes, and entrapment by the peritrophic matrix (Nayduch and
rial persistence between high- and low-dose treatments within 6 and Burrus 2017), bacteria ingested by flies also face destruction by the
12 h PI (Fig. 1). During this time interval, bacterial abundance was house fly immune response, which has been shown to be induced
static (or slightly proliferating) in flies in the high-dose treatments, on both the mRNA and protein level in the alimentary canal of
whereas bacterial abundance declined in flies that ingested the low flies that have ingested bacteria (Joyner et al. 2013, Nayduch et al.
dose. Thus, this time interval was selected to determine if excretion 2013, Fleming et al. 2014). Previous studies have proposed that this
of bacteria also differed across fly sex and ingested bacterial dose. local immune response mediates dose-dependent effects on bacte-
Our results showed that female flies that ingested either control rial survival in house flies (Kumar and Nayduch 2016, Chifanzwa
broth, or low or high doses of bacteria had more excretion events and Nayduch 2017), where low doses (i.e., abundances below a
than male flies between 6 and 12 h PI (Fig. 2A). Although we did “threshold” amount that remains to be determined) do not induce
not note any obvious differences in the size of excreta droplets from the immune response and therefore either have enhanced persis-
male and female flies, the volume of each droplet should be meas- tence (surviving through the gut to excretion) or may even replicate
ured in future studies in addition to the number of events in order to if nutritional resources are available. However, should the bacteria
determine whether the total volume excreted (e.g., the sum of all the populations rise above this tolerated threshold, the immune response
excreta droplets) is different across treatments. would be induced to subsequently reduce the populations below the
Although female flies had more excretion events than males in threshold level. The feedback mechanism between the local immune
the 6-h interval, our subsequent analyses revealed that male flies response in the gut and microbiome homeostasis has been described
posed a greater risk of being potential vectors. In the high-dose in fruit flies (Zaidman-Rémy et al. 2006, Buchon et al. 2013) and
treatments, there was no significant difference in the CFU cultured involves interactions between sensor and scavenger peptidoglycan
from male and female flies over time (Fig. 1). Likewise, there was recognition proteins (PGRPs) that effectively detect levels of bacteria
no difference between the proportion of positive (containing GFP S. (via PGN) and modulate the appropriate response. PGRP orthologs
Typhimurium) droplets that were shed (Fig. 2B). However, male flies have been identified in the house fly genome (Scott et al. 2014), as
fed the high dose of bacteria excreted the greatest mean abundance have the rest of the components of the immune deficiency and Toll
of GFP S. Typhimurium per droplet compared to all of the other innate immune pathways, making it plausible that a similar mecha-
treatment groups (Fig. 2C). Thus, male and female flies who ingested nism operates in house flies. To further address this hypothesis, we
the high dose had the same dissemination potential (i.e., proportions are currently investigating dose and sex effects on innate immune
of positive excreta droplets), but high-dose males posed the greater gene expression in the alimentary canal of house flies that have
risk of pathogen transmission because they shed more bacteria in ingested bacteria.
their excreta droplets. Although our results show that male flies posed a greater risk of
The greater vectoring potential of male flies also was evident in vectoring and disseminating GFP S. Typhimurium than female flies,
our low-dose bacteria treatments. Male flies who ingested the low further studies are needed to bolster these conclusions. Excretion
dose of GFP S. Typhimurium lost significant numbers of bacteria events were only observed during 6–12 h PI and therefore excretion
over time (Fig. 1), but excreted more than twice as many positive that occurred during the remainder of the experimental period was not
droplets as low-dose females (85% vs 36%, respectively; Fig. 2B). captured. Future studies should be aimed at observing excretion events
However, the mean CFU per droplet was the same in both males during the entire time course after ingesting bacteria. Additionally, it
and females that ingested the low dose (Fig. 2C). Thus, our results would be interesting to determine if the ingestion of a second meal
show that male flies that ingested the low dose of bacteria showed affects peristalsis rates and excretion as well as bacteria population
the greatest dissemination potential, but that both sexes posed the dynamics, as the meal may provide additional nutrition to both the
same risk of pathogen transmission per droplet. If dissemination microbes and the flies. We have previously shown that male and female
potential and pathogen transmission risk are considered together flies differentially acquire and harbor GFP S. Typhimurium and GFP

Downloaded from https://academic.oup.com/jme/advance-article-abstract/doi/10.1093/jme/tjy055/4967820

by guest
on 17 May 2018
 Journal of Medical Entomology, 2018, Vol. XX, No. XX 7

E. coli from inoculated manure (Thomson et al. 2017), but did not Gupta, A. K., D. Nayduch, P. Verma, B. Shah, H. V. Ghate, M. S. Patole,
investigate excretion from the flies in those experiments. In the future, and Y. S. Shouche. 2012. Phylogenetic characterization of bacteria in
we aim to determine if sex (and possibly dose) affects not only the the gut of house flies (Musca domestica L.). FEMS Microbiol. Ecol. 79:
581–593.
excretion of bacteria from flies but also the acquisition from natural
Joyner, C., M. K. Mills, and D. Nayduch. 2013. Pseudomonas aeruginosa in
sources and substrates, rather than culture media, through to excretion
Musca domestica L.: temporospatial examination of bacteria population
and dissemination. Finally, sex effects on synanthropic and zoophilic
dynamics and house fly antimicrobial responses. Plos One. 8: e79224.
behaviors have not been identified to our knowledge but also are inte- Kobayashi, H. 1934. The influence of foods on the fecundity of Musca domes-
gral in fully assessing vector potential and vectorial capacity of flies. tica. Keio J. Med. 5: 35–67.
Kumar, N. H. V., and D. Nayduch. 2016. Dose-dependent fate of GFP-
expressing Escherichia coli in the alimentary canal of adult house flies.
Acknowledgments
Med. Vet. Entomol. 30: 218–228.
We thank Dr. John J. Maurer of University of Georgia (Athens, GA, U.S.A.) for McGaughey, J., and D. Nayduch. 2009. Temporal and spatial fate of GFP-
supplying S. Typhimurium SR-11 and Dr. Brian Weiss, Yale University (New expressing motile and nonmotile Aeromonas hydrophila in the house fly
Haven, CT, U.S.A.) for the pGFPuv-kanamycin plasmid construct. This work digestive tract. J. Med. Entomol. 46: 123–130.
was supported by USDA-ARS National Program 104, project number 3020- Nayduch, D., and R. G. Burrus. 2017. Flourishing in filth: house fly–microbe
32000-007 to DN. Mention of trade names or commercial products in this interactions across life history. Ann. Entomol. Soc. Am. 110: 6–18.
publication is solely for the purpose of providing specific information and Nayduch, D., G. P. Noblet, and F. J. Stutzenberger. 2002. Vector potential of
does not imply recommendation or endorsement by the U.S. Department houseflies for the bacterium Aeromonas caviae. Med. Vet. Entomol. 16:
of Agriculture. Any opinions, findings, conclusion, or recommendations 193–198.
expressed in this publication are those of the author(s) and do not necessarily Nayduch, D., H. Cho, and C. Joyner. 2013. Staphylococcus aureus in the
reflect the view of the U.S. Department of Agriculture. house fly: temporospatial fate of bacteria and expression of the antimicro-
bial peptide defensin. J. Med. Entomol. 50: 171–178.
Nazni, W. A., B. Seleena, H. L. Lee, J. Jeffery, T. A. T Rogayah, and
References cited M. A. Sofian. 2005. Bacteria fauna from the house fly, Musca domestica
Barro, N., S. Aly, O. C. Tidiane, and T. A. Sababénédjo. 2006. Carriage of (L.). Trop. Biomed. 22: 225–231.
bacteria by proboscises, legs, and feces of two species of flies in street food Olsen, A. R., and T. S. Hammack. 2000. Isolation of Salmonella spp. from
vending sites in Ouagadougou, Burkina Faso. J. Food Prot. 69: 2007–2010. the housefly, Musca domestica L., and the dump fly, Hydrotaea aenescens
Buchon, N., N. A. Broderick, and B. Lemaitre. 2013. Gut homeostasis in (Wiedemann) (Diptera: Muscidae), at caged-layer houses. J. Food Prot.
a microbial world: insights from Drosophila melanogaster. Nat. Rev. 63: 958–960.
Microbiol. 11: 615–626. Ommi, D., B. Hemmatinezhad, T. T. Hafshejani, and F. Khamesipour. 2017.
Chifanzwa, R., and D. Nayduch. 2017. Dose-dependent effects on replication Incidence and antimicrobial resistance of Campylobacter and Salmonella
and persistence of Salmonella enterica serovar Typhimurium in house flies from houseflies (Musca domestica) in kitchens, farms, hospitals and
(Diptera: Muscidae). J. Med. Entomol. 55: 225–229. slaughter houses. Proc. Natl. Acad. Sci. India B. 87: 1285–1291.
Choo, L. C., A. A. Saleha, S. S. Wai, and N. Fauziah. 2011. Isolation of Pava-Ripoll, M., R. E. Pearson, A. K. Miller, and G. C. Ziobro. 2012. Prevalence
Campylobacter and Salmonella from houseflies (Musca domestica) in and relative risk of Cronobacter spp., Salmonella spp., and Listeria mono-
a university campus and a poultry farm in Selangor, Malaysia. Trop. cytogenes associated with the body surfaces and guts of individual filth
Biomed. 28: 16–20. flies. Appl. Environ. Microbiol. 78: 7891–7902.
van Duijkeren, E., W. J. Wannet, D. J. Houwers, and W. van Pelt. 2002. Schneider, H. A., and N. D. Zinder. 1956. Nutrition of the host and natural
Serotype and phage type distribution of salmonella strains isolated from resistance to infection: V. an improved assay employing genetic markers in
humans, cattle, pigs, and chickens in the Netherlands from 1984 to 2001. the double strain inoculation test. J. Exp. Med. 103: 207–223.
J. Clin. Microbiol. 40: 3980–3985. Scott, J. G., W. C. Warren, L. W. Beukeboom, D. Bopp, A. G. Clark, S. D. Giers,
El-Bassiony, G. M., V. Luizzi, D. Nguyen, J. G. Stoffolano, Jr, and A. E. Purdy. M. Hediger, A. K. Jones, S. Kasai, C. A. Leichter, et al. 2014. Genome of
2016. Vibrio cholerae laboratory infection of the adult house fly Musca the house fly, Musca domestica L., a global vector of diseases with adapta-
domestica. Med. Vet. Entomol. 30: 392–402. tions to a septic environment. Genome Biol. 15: 466.
Fleming, A., H. V. Kumar, C. Joyner, A. Reynolds, and D. Nayduch. 2014. Tauxe, R. V. 1991. Salmonella: a postmodern pathogen. J. Food Prot. 54:
Temporospatial fate of bacteria and immune effector expression in house 563–568.
flies fed GFP-Escherichia coli O157:H7. Med. Vet. Entomol. 28: 364–371. Terra, W. R., F. P. Espinoza Fuentes, and C. Ferreira. 1988. Midgut amylase,
Förster, M., S. Klimpel, H. Mehlhorn, K. Sievert, S. Messler, and K. Pfeffer. lysozyme, aminopeptidase, and trehalase from larvae and adults of Musca
2007. Pilot study on synanthropic flies (e.g. Musca, Sarcophaga, domestica. Arch. Insect. Biochem. Physiol. 9: 283–297.
Calliphora, Fannia, Lucilia, Stomoxys) as vectors of pathogenic microor- Thomson, J. L., K. M. Yeater, L. Zurek, and D. Nayduch. 2017. Abundance
ganisms. Parasitol. Res. 101: 243–246. and accumulation of Escherichia coli and Salmonella Typhimurium pro-
Gill, C., S. Bahrndorff, and C. Lowenberger. 2017. Campylobacter jejuni in cured by male and female house flies (Diptera: Muscidae) exposed to cattle
Musca domestica: an examination of survival and transmission potential manure. Ann. Entomol. Soc. Am. 110: 37–44.
in light of the innate immune responses of the house flies. Insect Sci. 24: Ugbogu, O. C., N. C. Nwachukwu, and U. N. Ogbuagu. 2006. Isolation of
584–598. Salmonella and Shigella species from house flies (Musca domestica L.) in
Greenberg, B. 1964. Experimental transmission of Salmonella typhimurium Uturu, Nigeria. Afr. J. Biotechnol. 5: 1090–1091.
by houseflies to man. Am. J. Hyg. 80: 149–156. Zaidman-Rémy, A., M. Hervé, M. Poidevin, S. Pili-Floury, M. S. Kim,
Greenberg, B., J. A. Kowalski, and M. J. Klowden. 1970. Factors affecting the D. Blanot, B. H. Oh, R. Ueda, D. Mengin-Lecreulx, and B. Lemaitre. 2006.
transmission of salmonella by flies: natural resistance to colonization and The Drosophila amidase PGRP-LB modulates the immune response to
bacterial interference. Infect. Immun. 2: 800–809. bacterial infection. Immunity. 24: 463–473.

Downloaded from https://academic.oup.com/jme/advance-article-abstract/doi/10.1093/jme/tjy055/4967820

by guest
on 17 May 2018