ORIGINAL ARTICLES: ANDROLOGY

Ability of Escherichia coli to produce

hemolysis leads to a greater
pathogenic effect on human sperm
Rodrigo Boguen, M.Sc.,a Favian Treulen, M.Sc.,a Pamela Uribe, M.Sc.,a and Juana V. Villegas, Ph.D.a,b
a
Centre of Reproductive Biotechnology (CEBIOR), Scientific and Technological Bioresources Nucleus (BIOREN), Temuco; and
b
Department of Internal Medicine, Faculty of Medicine, Universidad de La Frontera, Temuco, Chile

Objective: To determine the effect on human sperm of Escherichia coli strains separated on the basis of their ability to produce
hemolysis.
Design: Experimental study.
Setting: University-based laboratory.
Patient(s): Semen samples from healthy donors.
Intervention(s): Five million sperm, selected via the swim-up method, were incubated with 3 E. coli concentrations to obtain ratios of
sperm to E. coli of 1:2, 1:16, and 1:128. The E. coli strains were: a hemolytic isolated strain (H), a nonhemolytic American Type Culture
Collection strain (NH-ATCC), and a nonhemolytic isolated strain (NH-I).
Main Outcome Measure(s): Aliquots of human sperm were used to measure progressive motility using computer-aided sperm analysis,
mitochondrial membrane potential (DJm) with a JC-1 (5,50 ,6,60 tetrachloro-1,10 ,3,30 -tetraethylbenzamidazolocarbocyanin iodide) and
propidium iodide stain, and intracellular reactive oxygen species (iROS) with a dihydroethidium (DHE) stain. Sperm DJm and iROS
were measured by flow cytometry. Sperm vitality was considered the mean of propidium iodide–negative and DHE-negative cells.
Result(s): Sperm incubated with the H strain in a 1:2 sperm to bacteria ratio demonstrated a significant decrease in motility and DJm,
and an increase of iROS. The NH-ATCC strain decreased sperm motility and DJm, but in a ratio of sperm to bacteria of 1:128; it
increased iROS at a ratio of 1:16. The NH-I strain did not affect the analyzed sperm functions, even at a 1:128 sperm to bacteria ratio.
Conclusion(s): Results show a greater pathogenic effect on human sperm of E. coli strains with,
versus without, hemolytic capacity. (Fertil SterilÒ 2015;103:1155–61. Ó2015 by American So-
ciety for Reproductive Medicine.) Use your smartphone
Key Words: Human spermatozoa, uropathogenic Escherichia coli, mitochondrial membrane to scan this QR code
potential, motility, reactive oxygen species and connect to the
discussion forum for
this article now.*
Discuss: You can discuss this article with its authors and with other ASRM members at http://
fertstertforum.com/boguenr-ecoli-hemolysis-pathogenicity-sperm/ * Download a free QR code scanner by searching for “QR
scanner” in your smartphone’s app store or app marketplace.

in this group; Escherichia coli represents Escherichia coli is a broad species

T
he presence of bacteria in semen,
in contrast to the semen in 60%–85% of isolates from prostatic that colonizes environmental and ani-
men without infection, has been secretion (3). Other authors have mal niches (6). There are many intestinal
associated with male infertility (1). reported that E. coli represents 69% of and extraintestinal pathogenic E. coli
Additionally, males with urinary tract the microorganisms isolated from strains, of which the most important is
infection have impaired sperm motility semen in patients with bacterial the uropathogenic (UPEC; reviewed in
and vitality (2). Gram-negative bacteria prostatitis (4), and it has been associated [7]). The O serogroups that are frequent
have been associated with male with non–sexually transmitted epidi- in UPEC are: O1, O2, O4, O6, O7, O8,
accessory gland infection and included dymitis (5). O16, O18, O25, and O75 (8). These are
the same E. coli O serogroups that
Received November 20, 2014; revised January 26, 2015; accepted January 30, 2015; published online have been reported to cause prostatitis
March 4, 2015. (9). The most-frequent O serogroups of
R.B. has nothing to disclose. F.T. has nothing to disclose. P.U. has nothing to disclose. J.V.V. has
nothing to disclose.
E. coli found in the semen of infertile
Supported by grant DI12-0102 from the Universidad de La Frontera (to J.V.V.) and National Doctoral patients are O1, O2, O4, and O6 (10).
Scholarship 21110764 (to R.B.). The UPEC strains have virulence factors
Reprint requests: Juana V. Villegas, Ph.D., Universidad de La Frontera, Francisco Salazar 01145,
Temuco, Chile (E-mail: juana.villegas@ufrontera.cl). such as an adherence system, sidero-
phores, and toxins, highlighting the
Fertility and Sterility® Vol. 103, No. 5, May 2015 0015-0282/$36.00
Copyright ©2015 American Society for Reproductive Medicine, Published by Elsevier Inc.
alpha hemolysin (11), which produces
http://dx.doi.org/10.1016/j.fertnstert.2015.01.044 erythrocyte lysis (12).

VOL. 103 NO. 5 / MAY 2015 1155

ORIGINAL ARTICLE: ANDROLOGY
Uropathogenic E. coli impairs sperm quality because it
induces sperm motility loss (13), which has been linked to
an adherence effect of E. coli on sperm (14), accompanied
by sperm agglutination, a mannose-dependent effect (15).
Another hypothesis is that E. coli affects the sperm
membrane architecture (16). In addition, sperm motility
has been shown to be impaired by reactive oxygen species
(ROS), which are in turn induced by leukocytes that respond
to E. coli (17). In addition to sperm motility, UPEC has
been reported to impair sperm mitochondrial membrane
potential (DJm) (18). Uropathogenic E. coli has been
reported to decrease sperm DJm and induce lipid
membrane scrambling (19). E. coli toxins, such as lipo-
polysaccharide and porins, have been found to induce cell
death in sperm (20). In addition, lipopolysaccharide induces
DNA (deoxyribonucleic acid) fragmentation in human
sperm (21).
We have reported recently that serogroups are not
relevant to predicting the pathogenicity of E. coli on human
sperm in vitro (22). To date, the identity of the toxins that
determine higher pathogenicity in human sperm is unknown;
therefore, the objective of this work was to determine the
in vitro effect on human sperm of E. coli strains that have
been separated on the basis of their ability to produce
hemolysis.
MATERIALS AND METHODS
Sperm Collection and Selection
Human semen was obtained from apparently healthy men,
via masturbation, according to recommendations from the
World Health Organization (23). Semen donors were
interviewed and signed the informed consent form, which
was reviewed and approved (file 45/012) by the Ethics
Committee of the Faculty of Medicine at the Universidad de
La Frontera, Temuco, Chile.
Sperm selection was conducted by the direct swim-up
method. Briefly, 150 mL of semen were deposited in the
bottom of a sterile conical polystyrene tube containing
500 mL of human tubal fluid (HTF) (24), without antibiotics,
and incubated for 1 hour at 37 C at a 45 angle.
After swim-up incubation, 400 mL of the top phase was
collected and transferred to another sterile tube. Next, a cell
count of the collected sperm was performed in a Neubauer
chamber.
Escherichia coli Collection, Identification,
Typification, and Growth
Strains of UPEC were isolated from the urine of patients who
had a urinary tract infection. The E. coli were identified
according to Bergey's Manual (25), and the hemolytic
condition of E. coli was established according to its growth
in blood agar. Typification of E. coli was performed with O
antigen–specific antisera (SSI Diagnostica), according to
manufacturer instructions.
The experiments were conducted with 3 UPEC strains: an
isolated hemolytic E. coli strain (H), an isolated nonhemolytic
E. coli strain (NH-I), and nonhemolytic American Type
1156
Culture Collection (NH-ATCC) strain 25922. Escherichia coli
O-antigen typification was: O25 to the H strain, O75 to the
NH-I strain, and O6 to the NH-ATCC strain. The UPEC and
sperm were incubated in HTF without antibiotics, because
streptomycin could inactivate the metabolism of E. coli
strains and would interfere with the pathogenic activity of
the UPEC strains. The 3 UPEC strains were incubated
with HTF, and with HTF without antibiotics, to test growth
capacity. The growth of the UPEC strains at 1 hour of
incubation with both media was observed to be
negligible (data not shown); therefore the initial number of
bacteria added to each was kept constant during the
incubation time.
Experimental Procedure
Five million sperm selected via the swim-up method were
incubated with each of the 3 E. coli strains, separately: H,
NH-I, and NH-ATCC. Each UPEC strain was used at 10 x
106 colony-forming units (CFU), 80 x 106 CFU, and 640 x
106 CFU, to obtain a ratio of sperm to E. coli of 1:2, 1:16,
and 1:128. As a baseline control, sperm were incubated
without E. coli (a sperm to E. coli ratio of 1:0).
The 3 experimental groups, and the baseline control
without bacteria, were adjusted to 1 mL with HTF without
antibiotics and were incubated for 1 hour at 37 C. Later,
aliquots of sperm suspension were removed from each
experimental group and the baseline control to assess
progressive motility, vitality, DJm, and intracellular ROS
(iROS). A total of 4 independent replicates were performed
on different days for each E. coli strain.
Sperm motility. Ten microliters of each sperm suspension
were used to measure sperm progressive motility by
conducting computer-aided sperm analysis with Integrated
Sperm Analysis System (ISAS) software, version 1 (Proiser).
Each measurement was taken twice.
Sperm mitochondrial membrane potential. Changes in
sperm DJm were measured with 5,50 ,6,60 tetrachloro-
1,10 ,3,30 -tetraethylbenzamidazolocarbocyanin iodide (JC-1;
Mît-E-J mitochondrial permeability detection kit; Enzo
Life Sciences Inc). The JC-1 stain enters all the cell
compartments in a monomeric form, fluorescing green.
When JC-1 enters mitochondria, it fluoresces orange, because
it aggregates and cannot exit the mitochondrial matrix,
owing to differences in charges between the matrix and the
mitochondrial intermembrane space (26). Thus, changes in
sperm DJm can be measured as changes in the fluorescence
of JC-1 aggregates.
The JC-1 stain was combined with propidium iodide
(Sigma Aldrich) as the vital stain, which enters cells that
have membrane disruptions, binds DNA, and fluoresces red.
Briefly, 1 x 106 sperm suspended in 1 mL of HTF were stained
with 1 mL of JC-1 100X. After 15 minutes, 1 mL of 1 mmol/L
propidium iodide was added to each test and incubated for
2 minutes at 37 C. After that, sperm were centrifuged at
500 g for 5 minutes. The supernatant was discarded and the
cells were suspended with 300 mL of phosphate buffered saline
1X, to be measured by flow cytometry (see later section,
VOL. 103 NO. 5 / MAY 2015
 Fertility and Sterility®

FIGURE 1

Flow cytometry graph to determine sperm DJm, iROS and vitality. (A[i] and B[i]) First, the sperm population was gated in a size vs. internal
complexity dot-plot. Sperm were gated as a low debris fraction from swim-up–selected sperm. (A[ii] and B[ii]) Live sperm were gated from
sperm-negative population to their respective vital stain. (A[iii]) The MFI of JC-1 aggregates in live sperm was considered sperm DJm. (B[iii])
The MFI of DHE in live sperm was considered the sperm iROS. (C) Sperm vitality was calculated as the mean of live sperm from the iROS and
DJm determinations, i.e., PI-negative and SYTOX green–negative sperm. DHE ¼ dihydroethidium; iROS ¼ intracellular reactive oxygen species;
JC-1 ¼ 5,50 ,6,60 tetrachloro-1,10 ,3,30 -tetraethylbenzamidazolocarbocyanin iodide; MFI ¼ mean fluorescent intensity; PI ¼ propidium iodide;
DJm ¼ mitochondrial membrane potential.
Boguen. Hemolytic E. coli and sperm damage. Fertil Steril 2015.

VOL. 103 NO. 5 / MAY 2015 1157

ORIGINAL ARTICLE: ANDROLOGY

Cytofluorimetric analysis). Sperm DJm was measured as

FIGURE 2
mean fluorescence intensity (MFI, geometric mean of histo-
gram) of JC-1 orange aggregates, through a 585/42-nm filter,
discarding previously propidium iodide–positive sperm
through an LP670-nm filter (Fig. 1A). Each test was analyzed
twice.
Sperm intracellular reactive oxygen species. The iROS
measurement was performed with dihydroethidium (DHE,
Molecular Probes, Life Technologies), which enters cells and
is oxidized by ROS, principally superoxide anion. Oxidized
DHE binds DNA and fluoresces orange; therefore, changes
in DHE fluorescence are proportional to iROS production
(27). SYTOX green (Molecular Probes) was used as the vital
stain; it enters cells that have membrane disruptions, binds
DNA, and fluoresces green. Briefly, 1 x 106 sperm were
stained with 1 mL of DHE at 2 mmol/L, and 1 mL of SYTOX
green at 0.5 mmol/L in 1 mL of HTF as the final volume.
After 15 minutes at 37 C, sperm were centrifuged at 500 g
for 5 minutes; the supernatant was discarded and the
sperm pellet was suspended in 300 mL of phosphate buffered
saline 1X to be measured by flow cytometry (see later section,
Cytofluorimetric analysis). The MFI (geometric mean of
histogram) of DHE was measured through a 585/42-nm
filter, discarding previously SYTOX green–positive sperm
through a 530/30-nm filter. This was considered a
measurement of iROS (Fig. 1B). Each measurement was taken
twice.
Effect of Escherichia coli strains on sperm vitality (A) and motility (B).
Sperm vitality. Sperm vitality was considered to be the mean Three E. coli strains were tested: an isolated hemolytic (H) strain, an
percentage of sperm negative to the 2 fluorescent vital stains: isolated nonhemolytic (NH-I) strain, and a nonhemolytic American
Type Culture Collection (NH-ATCC) strain. Sperm were incubated
propidium iodide in sperm DJm, and SYTOX green in sperm for 1 hour with 3 ratios of sperm to E. coli: 1:2, 1:16, and 1:128.
iROS (Fig. 1C). The figure summarizes 4 independent experiments, with
measurements taken twice, and SD bars. *, P<.05 compared with
Cytofluorimetric analysis. Flow cytometry measurements its control without E. coli (sperm to E. coli ratio of 1:0).
were developed in a FACSCanto II flow cytometer (Becton, Boguen. Hemolytic E. coli and sperm damage. Fertil Steril 2015.
Dickinson and Company). A total of 10,000 sperm events
were acquired from each measurement. Acquisition was
performed with a sample aspiration speed of 60 mL/minute. RESULTS
Samples were acquired and analyzed with the software Effect of E. coli on Sperm Vitality
FACSDiva, version 6.1.3 (Becton, Dickinson and Company).
Sperm vitality decreased with the H strain, from 90.6% 
The green FITC (fluorescein isothiocyanate) channel
4.8% to 86.2%  3.3%, compared with the control without
(530/30-nm filter) was used to read SYTOX green–positive
bacteria (P< .05), but only when used in a sperm to E. coli ra-
stain. The PE (phycoerythrin) channel (585/42-nm filter)
tio of 1:128 (Fig. 2A). Conversely, the percentage of viable
was used to read the JC-1 orange aggregates and DHE.
sperm did not decrease with the 3 ratios used for the NH-I
The far red PerCP (peridinin chlorophyll protein) channel
and NH-ATCC strains.
(LP670 nm) was used to read propidium iodide fluores-
cence. All the fluorophores were excited with a 488-nm
argon laser. Effect of E. coli on Sperm Motility
All concentrations of the H strain decreased sperm motility
significantly, compared with the control without E. coli
Statistical Analysis
(Fig. 2B). However, the NH-ATCC strain decreased sperm
Analyses and graphs were completed with the statistical motility only when it was incubated in a sperm to E. coli ratio
software Prism 6 (GraphPad). Gaussian distribution was of 1:128, and none of the 3 ratios of the NH-I strain decreased
analyzed with D'Agostino's test. Numeric results were the percentage of sperm with progressive motility (Fig. 2B).
transformed to a logarithmic scale. Transformed data were
analyzed with a 2-way ANOVA, to compare both strains
and concentrations of E. coli. The Bonferroni test of multiple Effect of Escherichia coli on Sperm Mitochondrial
comparisons was performed to provide a contrast with the Membrane Potential
baseline control without bacteria. A P value < .05 (P< .05) To evaluate the effect of E. coli on changes in sperm DJm,
was considered statistically significant. JC-1 stain was used. The MFI of JC-1 aggregate represents

1158 VOL. 103 NO. 5 / MAY 2015


FIGURE 3
Effect of Escherichia coli (E. coli) strains on mitochondrial membrane
potential (DJm) (A) and intracellular reactive oxygen species (iROS)
(B). Three E. coli strains were tested: an isolated hemolytic (H)
strain, an isolated nonhemolytic (NH-I) strain, and a nonhemolytic
American Type Culture Collection (NH-ATCC) strain. Sperm were
incubated for 1 hour with 3 ratios of sperm to E. coli: 1:2, 1:16,
and 1:128. The figure summarizes 4 independent experiments,
with measurements taken twice, and SD bars. AU ¼ arbitrary units;
DHE ¼ dihydroethidium; JC-1 ¼ 5,50 ,6,60 tetrachloro-1,10 ,3,30 -
tetraethylbenzamidazolocarbocyanin iodide; MFI ¼ mean
fluorescent intensity. *, P<.05 compared with its control without
E. coli (sperm to E. coli ratio of 1:0).
Boguen. Hemolytic E. coli and sperm damage. Fertil Steril 2015.
sperm DJm variations in live cells. Figure 3A shows that the
H strain decreased the MFI of JC-1 aggregates significantly in
sperm, at the 3 E. coli concentrations used; the NH-ATCC
strain decreased MFI from 185.3  7.2 to 163  9.4 in a sperm
to E. coli ratio of 1:128, whereas none of the 3 concentrations
of the NH-I E.coli strain used here had any effect on the MFI
of the JC-1 aggregates.
Effect of Escherichia coli on Sperm Intracellular
Reactive Oxygen Species
To evaluate the effect of various strains and concentrations of
E. coli on iROS production in sperm, DHE stain was used. Var-
iations in the MFI of DHE in cells represent various levels of
iROS production. The hemolytic strain induced a significant
increase in iROS sperm production, with all the E. coli concen-
trations tested in this study (Fig. 3B). The NH-ATCC strain
increased sperm iROS only in the sperm to E. coli ratios of
1:16 and 1:128, but the NH-I strain did not significantly
VOL. 103 NO. 5 / MAY 2015
Fertility and Sterility®
change the iROS production of sperm with any of the E. coli
concentrations tested.
DISCUSSION
We have reported previously that the in vitro effect of various
E.coli isolates on human sperm is highly heterogeneous and
that this heterogeneity is not O-antigen dependent (22).
Consequently, we looked for other conditions that could pre-
dict the pathogenic in vitro effect of E. coli. One possibility is
the toxin alpha hemolysin, which renders E. coli strains
hemolytic (28).
In this study, we evaluated 2 isolated E. coli strains, 1
hemolytic (H), and the other nonhemolytic (NH-I). In addition,
we included the commercial strain ATCC 25922, which is
nonhemolytic as well. Our results show that the hemolytic
strain altered all the sperm variables studied here, demon-
strating a greater pathogenic effect than the nonhemolytic
strains. The E. coli strains used in this work have the antigens
O25 (H), O75 (NH-I), and O6 (NH-ATCC). All of these O antigens
have been described as being prevalent in UPEC strains (8).
Bacteria such as E. coli have been confirmed to cause
infertility by reducing sperm vitality (2, 29). However, here
we observed distinct effects of the various E. coli isolates. In
addition, we observed that strains described as pathogenic
to sperm function did not show any important effects under
our experimental conditions. We found a clear difference
between the hemolytic and nonhemolytic strains in their
effects on various sperm parameters.
In the experimental conditions used, sperm vitality
decreased only with the H strain. The nonhemolytic strains
did not reduce vitality, even in a ratio of 128 bacteria per sper-
matozoa. The effect of E. coli on sperm vitality has been
analyzed by Diemer and colleagues (16). With the use of elec-
tronic microscopy, these authors observed sperm plasma
membrane damage after sperm were incubated with an O6
E. coli strain. One interesting detail is that the O6 E. coli strain
was hemolytic (13).
In view of our current results, we now suggest that the
damaging effect of E. coli on the sperm observed by Diemer
and colleagues (16), rather than being determined by the O6
antigen as these authors affirmed, can be attributed to its he-
molytic condition. This possibility is corroborated in the cur-
rent study, in which we work with an O6 strain as well—the
ATCC 25922 strain. As this strain is nonhemolytic, it did
not affect sperm vitality, even in a ratio of 128 bacteria per
spermatozoa. Although we did not use electronic microscopy,
we used vital stains that are based on accurately detecting the
integrity of plasma membrane. Our results are consistent with
those of Diemer and colleagues (16), but we suggest that their
reported direct effect of E. coli on sperm vitality can be attrib-
uted to alpha hemolysin production.
Analysis of sperm motility showed that, as with sperm vi-
tality, it was differentially affected by the 3 E. coli strains
studied. Sperm motility decreased in all the sperm to bacteria
ratios with the H strain, whereas the NH-ATCC strain
decreased sperm motility only in a sperm to E. coli ratio of
1:128. The NH-I strain did not alter sperm motility in any
sperm to E. coli ratio used. Our results show a
1159
ORIGINAL ARTICLE: ANDROLOGY
concentration-dependent effect of the NH-ATCC E. coli strain
on human sperm, as has been described by other authors (13,
30). We have reported that the O antigen of E. coli is not
important in pathogenicity affecting human sperm motility
(22). We now suggest that the hemolytic condition of E. coli
strains may be a pathogenic characteristic relevant to sperm
motility.
In addition to hemolytic condition, other factors affecting
the virulence of E. coli toward human sperm have been
described. Agglutination of sperm by type 1 fimbriae UPEC
has been reported as a cause of reduced sperm motility (15).
Uropathogenic E. coli can agglutinate sperm by type 1 and
type p fimbriae, which bind mannose and a-D-galp-1-4-b-
D-galp receptors, respectively. Both receptors have been
found in sperm (31). However, we have not observed sperm
agglutination with any of the 3 UPEC strains.
A decrease in motility has been linked to DJm loss in hu-
man sperm (32, 33). Our results show an association of
motility and sperm DJm with all the E. coli strains
analyzed in this study. The NH-I strain did not decrease sperm
motility or DJm; the NH-ATCC strain altered motility and
DJm with a sperm to E. coli ratio of 1:128; and the H strain
reduced sperm motility and DJm, even in a ratio of 2 bacteria
per spermatozoa. Two reports in the literature that did not
define the hemolytic or nonhemolytic condition of the E.
coli, used in vitro, demonstrated a negative effect of E. coli
on human sperm DJm (18, 19). We can assume they used
H strains.
The results with the NH-I E. coli strain are consistent with
our previous report, according to which none of the E. coli
strains used decreased sperm DJm (22). Our results now sug-
gest a strain-dependent effect of E. coli on sperm motility and
DJm, confirming that not all E. coli alter sperm functions in
the same way. Sperm iROS production increased in the sperm
to E. coli ratios of 1:16 and 1:128 with the NH-ATCC strain,
and with all the concentrations of the H strain. The NH-I
strain did not increase sperm iROS at any concentration
tested. These results again show a pathogenic predominant
effect of the H strain on human sperm.
Previously, the pathogenic effects of E. coli were attrib-
uted to the increase in extracellular ROS produced by
infiltrating seminal leukocytes (17), an increase that addition-
ally induces phosphatidylserine translocation in human
sperm (34). The increase in iROS seems to be associated
with the loss of sperm DJm, as sperm DJm has been
observed to decrease in infertility cases in which ROS are
increased (32, 35). Therefore, a decrease in sperm DJm
could be due to increased iROS induced by the NH-ATCC
and H E. coli strains. The NH-I strain seems to confirm here
the association between iROS and sperm DJm, because
neither of the variables was impaired with this strain.
Conversely, in somatic cells, bacteria such as E. coli are
recognized by Toll-like receptors (TLR)-1, TLR-2, and TLR-4,
which induce tumor necrosis factor receptor–associated fac-
tor 6 translocation to mitochondria, where evolutionarily
conserved signaling intermediates in Toll pathways promote
mitochondrial ROS generation (reviewed in [36]). The
presence of TLR-2 and TLR-4 have been described in
human sperm, and they respond to lipopolysaccharide and
1160
peptidoglycan with motility loss and DNA fragmentation
(37). Therefore, an iROS increase could be the first conse-
quence of E. coli infection, and iROS would subsequently
induce sperm DJm dissipation.
However, we have shown that not all E. coli strains in-
crease sperm iROS, so we think that the effect of bacteria
may be increased through alpha hemolysin activity inactivat-
ing the PI3k/Akt (phosphatidylinositol 3'-kinase–protein
kinase B) pathway (38), as would occur with the hemolytic
E. coli strain. By contrast, the nonhemolytic ATCC strain
increased sperm iROS as well; therefore, other mechanisms
must help increase iROS, and this is a line of inquiry to be
pursued.
Our results showed differences between the 2 nonhemo-
lytic strains of E. coli. The NH-I strain does not seem to affect
human sperm negatively, at least in the concentrations and
experimental conditions tested here. The NH-ATCC strain im-
pairs sperm parameters such as motility and DJm, and in-
creases ROS production, indicating that other mechanisms,
aside from alpha hemolysin secretion, may be damaging the
sperm. For example, another mechanism that impairs sperm
function is agglutination by fimbriae (15), as previously
mentioned, but in our work, sperm agglutination was not
observed. Thus, we propose that an interesting approach
would be to analyze phenotypic differences, such as the pro-
duction of other toxins in the 2 nonhemolytic strains, to
determine which mechanism may be further involved in
sperm-function impairment.
Although our results were obtained with an in vitro sys-
tem, and normally, sperm are considered to have a very short
contact time with seminal plasma, the effects of E. coli
observed on human sperm after 1 hour of incubation may
be similar to what might occur in epididymitis, in which
sperm and bacteria have a prolonged interaction. In fact, an
experimental infection with E. coli in the epididymis of rats
impaired even spermatogenesis (39), with cell-death path-
ways and necrosis involved (40). Therefore, a bacterial infec-
tion would be expected to have an effect on mature sperm
stored in the epididymis.
In conclusion, hemolytic, versus nonhemolytic, E. coli
was associated with a major pathogenic effect on human
sperm. We observed that the hemolytic E. coli strain altered
sperm motility, DJm, and iROS levels to a greater extent,
compared with the nonhemolytic strains, and even decreased
sperm vitality, although only at the highest concentration of
bacteria used. Differences were found between the 2 nonhe-
molytic strains, indicating that other mechanisms are
involved in sperm damage, which should be investigated.
Acknowledgments: The authors thank Dr. Helen Lowry, a
native English speaker, who reviewed and corrected this
article.
REFERENCES
1. Merino G, Carranza-Lira S, Murrieta S, Rodriguez L, Cuevas E, Moran C. Bac-
terial infection and semen characteristic in infertile men. Arch Androl 1995;
35:43–7.
2. Sanocka-Maciejewska D, Ciupinska M, Kurpisz M. Bacterial infection and
semen quality. J Reprod Immunol 2005;67:51–6.
VOL. 103 NO. 5 / MAY 2015

3. Weidner W, Jantos C, Schiefer HG, Haidl G, Friedrich HJ. Semen parameters
in men with and without proven chronic prostatitis. Arch Androl 1991;26:
173–83.
4. Weidner W, Schiefer HG, Krauss H, Jantos C, Friedrich HJ,
Altmannsberger M. Chronic prostatitis: a thorough search for etiologically
involved microorganisms in 1,461 patients. Infection 1991;19 Suppl 3:
119–25.
5. Weidner W, Krause W, Ludwig M. Relevance of male accessory gland infec-
tion for subsequent fertility with special focus on prostatitis. Hum Reprod
Update 1999;5:421–32.
6. Wiles TJ, Kulesus RR, Mulvey MA. Origins and virulence mechanisms of ur-
opathogenic Escherichia coli. Exp Mol Pathol 2008;85:11–9.
7. Croxen MA, Finlay BB. Molecular mechanisms of Escherichia coli pathoge-
nicity. Nat Rev Microbiol 2010;8:26–38.
8. Lloyd AL, Rasko DA, Mobley HL. Defining genomic islands and
uropathogen-specific genes in uropathogenic Escherichia coli. J Bacteriol
2007;189:3532–46.
9. Terai A, Yamamoto S, Mitsumori K, Okada Y, Kurazono H, Takeda Y, et al.
Escherichia coli virulence factors and serotypes in acute bacterial prostatitis.
Int J Urol 1997;4:289–94.
10. Bartoov B, Ozbonfil D, Maayan MC, Ohad E, Nitzan Y. Virulence character-
istics of male genital tract Escherichia coli isolated from semen of suspected
infertile men. Andrologia 1991;23:387–94.
11. Brzuszkiewicz E, Bruggemann H, Liesegang H, Emmerth M, Olschlager T,
Nagy G, et al. How to become a uropathogen: comparative genomic anal-
ysis of extraintestinal pathogenic Escherichia coli strains. Proc Natl Acad Sci U
S A 2006;103:12879–84.
12. Bhakdi S, Mackman N, Nicaud JM, Holland IB. Escherichia coli hemolysin
may damage target cell membranes by generating transmembrane pores.
Infect Immun 1986;52:63–9.
13. Diemer T, Weidner W, Michelmann HW, Schiefer HG, Rovan E, Mayer F. In-
fluence of Escherichia coli on motility parameters of human spermatozoa
in vitro. Int J Androl 1996;19:271–7.
14. Rosenthal L. Agglutinating properties of Escherichia coli: agglutination of
erythrocytes, leucocytes, thrombocytes, spermatozoa, spores of molds,
and pollen by strains of E. Coli. J Bacteriol 1943;45:545–50.
15. Wolff H, Panhans A, Stolz W, Meurer M. Adherence of Escherichia coli to
sperm: a mannose mediated phenomenon leading to agglutination of
sperm and E. coli. Fertil Steril 1993;60:154–8.
16. Diemer T, Huwe P, Michelmann HW, Mayer F, Schiefer HG, Weidner W. Es-
cherichia coli-induced alterations of human spermatozoa. An electron mi-
croscopy analysis. Int J Androl 2000;23:178–86.
17. Diemer T, Huwe P, Ludwig M, Schroeder-Printzen I, Michelmann H,
Schiefer HG. Influence of autogenous leucocytes and Escherichia coli on
sperm motility parameters in vitro. Andrologia 2003;23:100–5.
18. Schulz M, Sanchez R, Soto L, Risopatron J, Villegas J. Effect of Escherichia coli
and its soluble factors on mitochondrial membrane potential, phosphatidyl-
serine translocation, viability, and motility of human spermatozoa. Fertil
Steril 2010;94:619–23.
19. Fraczek M, Piasecka M, Gaczarzewicz D, Szumala-Kakol A, Kazienko A,
Lenart S, et al. Membrane stability and mitochondrial activity of
human-ejaculated spermatozoa during in vitro experimental infection
with Escherichia coli, Staphylococcus haemolyticus and Bacteroides ureo-
lyticus. Andrologia 2012;44:315–29.
20. Galdiero F, Gorga F, Bentivoglio C, Mancuso R, Galdiero E, Tufano MA. The
action of LPS porins and peptidoglycan fragments on human spermatozoa.
Infection 1988;16:349–53.
VOL. 103 NO. 5 / MAY 2015
Fertility and Sterility®
21. Gorga F, Galdiero M, Buommino E, Galdiero E. Porins and lipopolysaccha-
ride induce apoptosis in human spermatozoa. Clin Diagn Lab Immunol
2001;8:206–8.
22. Boguen R, Uribe P, Treulen F, Villegas JV. Distinct isolates of uropathogenic
Escherichia coli differentially affect human sperm parameters in vitro. An-
drologia 2014;46:943–7.
23. World Health Organization (WHO). WHO laboratory manual for the exami-
nation and processing of human semen. 5th ed. Geneva: WHO Press; 2010.
24. Quinn P, Kerin J, Warnes G. Improved pregnancy rate in human in vitro fertil-
ization with the use of a medium based on the composition of human tubal
fluid. Fertil Steril 1985;44:493–8.
25. Holt JG. Bergey's manual of determinative bacteriology. 9th ed. Baltimore,
MD: Williams Wilkins; 1994.
26. Reers M, Smith TW, Chen LB. J-aggregate formation of a carbocyanine as a
quantitative fluorescent indicator of membrane potential. Biochemistry
1991;30:4480–6.
27. Carter WO, Narayanan PK, Robinson JP. Intracellular hydrogen peroxide
and superoxide anion detection in endothelial cells. J Leukoc Biol 1994;
55:253–8.
28. Springer W, Goebel W. Synthesis and secretion of hemolysin by Escherichia
coli. J Bacteriol 1980;144:53–9.
29. Gopalkrishnan K, Joseph R, Sheth AR. Alteration of semen characteristics
and regulatory factors in human semen with bacterial infection. Arch Androl
1994;32:213–8.
30. Huwe P, Diemer T, Ludwig M, Liu J, Schiefer HG, Weidner W. Influence of
different uropathogenic microorganisms on human sperm motility parame-
ters in an in vitro experiment. Andrologia 1998;30 Suppl 1:55–9.
31. Monga M, Roberts J. Spermagglutination by bacteria: receptor-specific in-
teractions. J Androl 1994;15:151–6.
32. Marchetti C, Obert G, Deffosez A, Formstecher P, Marchetti P. Study of mito-
chondrial membrane potential, reactive oxygen species, DNA fragmentation
and cell viability by flow cytometry in human sperm. Hum Reprod 2002;17:
1257–65.
33. Espinoza JA, Schulz MA, Sanchez R, Villegas JV. Integrity of mitochondrial
membrane potential reflects human sperm quality. Andrologia 2009;41:
51–4.
34. Villegas J, Schulz M, Soto L, Iglesias T, Miska W, Sanchez R. Influence of reac-
tive oxygen species produced by activated leukocytes at the level of
apoptosis in mature human spermatozoa. Fertil Steril 2005;83:808–10.
35. Wang X, Sharma RK, Gupta A, George V, Thomas AJ, Falcone T, et al. Alter-
ations in mitochondria membrane potential and oxidative stress in infertile
men: a prospective observational study. Fertil Steril 2003;80(Suppl 2):844–50.
36. West AP, Shadel GS, Ghosh S. Mitochondria in innate immune responses.
Nat Rev Immunol 2011;11:389–402.
37. Fujita Y, Mihara T, Okazaki T, Shitanaka M, Kushino R, Ikeda C, et al. Toll-like
receptors (TLR) 2 and 4 on human sperm recognize bacterial endotoxins and
mediate apoptosis. Hum Reprod 2011;26:2799–806.
38. Wiles TJ, Dhakal BK, Eto DS, Mulvey MA. Inactivation of host Akt/protein ki-
nase B signaling by bacterial pore-forming toxins. Mol Biol Cell 2008;19:
1427–38.
39. Pilatz A, Ceylan I, Schuppe HC, Ludwig M, Fijak M, Chakraborty T, et al.
Experimental Escherichia coli epididymitis in rats: assessment of testicular
involvement in a long-term follow-up. Andrologia 2015;47:160–7.
40. Lu Y, Bhushan S, Tchatalbachev S, Marconi M, Bergmann M, Weidner W,
et al. Necrosis is the dominant cell death pathway in uropathogenic Escher-
ichia coli-elicited epididymo-orchitis and is responsible for damage of rat
testis. PLoS One 2013;8:e52919.
1161