1623
Journal of General Microbiology (1988), 134, 1623-1 628. Printed in Great Britain
The Effect of Hormones on Trichomonas vagindis
By B A R R E T T S U G A R M A N ' * , 2 A N D N A N C Y M U M M A W '
Departments of Medicine' and Microbiology and Public Health2, Michigan State University,
East Lansing, Michigan 48824, USA
(Received 18 August 1987; revised 23 November 1987)
~~
~
~
The hormonal milieu can alter susceptibility to infection. The effect of hormones on
Trichomonas vaginalis was studied utilizing axenically cultured clinical isolates. Oestrogens, in
physiological concentrations, decreased the growth of the organisms and their attachment to
mammalian cells in vitro, and acted as a chemorepellent. The specificity of these effects was
verified by their being blocked with anti-oestrogens, by the dose- and time-dependency of the
responses, and by the lack of effect with other hormones. These results suggest that oestrogens
may decrease the virulence of T. vaginalis; however, interactions between oestrogens and
mammalian cells may promote the development of infection. Thus complicated interactions
between hormones, micro-organisms and mammalian cells must determine whether exposure to
oestrogens predisposes to or prevents the development of infection.
INTRODUCTION
The hormonal milieu may alter host susceptibility to numerous bacterial infections (Botta,
1979; Forslin et al., 1979; Sugarman & Agbor, 1986). Oestrogens and corticosteroids have been
studied most intensively in this regard. Previously identified mechanisms by which hormones
can alter the susceptibility to infection include hormone-mediated alterations in the immune
response of the host, alterations in other host defence factors such as the production of cervical
mucus in the genitourinary tract, and alterations of host cell morphology (Kita et al., 1985; Pung
et al., 1984, 1985). Recent studies have suggested that the binding of micro-organisms to
mammalian receptor cells may also be modified by hormones, especially oestrogens (Botta,
1979; Forslin et al., 1979; Sugarman & Epps, 1982). Still other, undefined, mechanisms may be
necessary to explain altered susceptibility to infection, or outcome after infection, related to the
presence of oestrogens (Sugarman & Agbor, 1986).The presence of hormone-binding proteins in
numerous micro-organisms lends support to the hypothesis that hormones may directly alter the
virulence of micro-organisms (Loose & Feldman, 1982; Powell et al., 1983; Schar et al., 1986;
Stover et al., 1986; Weiss et al., 1983), including protozoa (Arnold et al., 1969; Csaba et al., 1978,
1985;Hamana & Iwai, 1971). Studies with Trichomonasvaginalis, a frequent genitourinary tract
pathogen of women, suggest that infections with this flagellate protozoon are more common
after exposure of female mammals to oestrogens (Azuma, 1968; Cappuccinelli et al., 1974; ElBoulaqi et al., 1984). Here we report a systematic study on the effect of oestrogens on the growth,
adherence and chemotaxis of T . vaginalis. We used axenic T. vaginalis isolates and cultured
mammalian cells not known to respond to oestrogens in order to avoid any oestrogenic
(secondary) effects on the host.
METHODS
Trichomonas vaginalis isolates. These were cultured from women who were positive for T . vaginalis when saline
wet mounts were examined by light microscopy.Sterile cotton swabs were used to inoculate vaginal secretionsinto
8 x 100 mm borosilicate' glass screw-capped tubes containing 6 ml Diamond's DL8 defined medium
supplemented with 8% (v/v) human serum, 40 pg gentamicin ml-l and 2-5 pg Amphotericin B ml-l (Diamond,
1983; Linstead, 1981). Cultures were maintained in DL8 with 4%(v/v) heat-inactivatedequine serum at 36 "Cand
were subcultured every 3-4 d.
0001-4376 0 1988 SGM
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Thu, 15 Jun 2017 15:45:22
1624
B. SUGARMAN AND N. MUMMAW
Hormones. Hormones (from Sigma) and the anti-oestrogen tamoxifen citrate (graciously provided by Stuart
Pharmaceuticals, Wilmington, DE, USA) were dissolved in pH 7-3 phosphate-buffered saline (PBS: 0 1 5 Msodium chloride, 015 M-sodium phosphate). Oestrogens, tamoxifen, progesterone and testosterone were initially
dissolved in 100% ethanol, then PBS, with final ethanol dilutions less than 01 % (v/v). Hormones were frozen at
— 70 °C until utilized. Baseline T. vaginalis media contained less than 0· 1 % of the hormone concentrations as listed
in Table 1. Concentrations of hormones used in Table 1 are approximately ten times the maximum normal human
non-pregnant serum concentrations.
Growth assay. Stock cultures of T. vaginalis were centrifuged at 250 # for 5 min and resuspended in 1-0 ml
warmed DL8 medium. After vortexing, 01 ml was added to each tube containing 6 ml of DL8 with the specified
concentrations of hormones. The tubes were incubated at 36 °C for up to 3 d. They were then centrifuged, the
supernatant was removed, and the trichomonads were resuspended in 10 ml warmed PBS before counting with a
haemocytometer.
Radiolabelling. Exponential-phase T. vaginalis isolates were radiolabelled with [3H]thymidine [5 μ θ (185 kBq)
m l - 1 ; specific activity 78-3 Ci mmol - 1 (2-9 TBq mmol" 1 ), New England Nuclear] for 18 h at 36 °C (Alderete &
Garza, 1985). Organisms were washed three times by centrifugation in PBS before use in further assays.
Adherence assay. Mammalian cells were cultured in 24-well (16 mm diameter) Costar culture plates in an
atmosphere of air/C0 2 (95 :5, v/v) at 37 °C. McCoy cells were grown in pH 7-2 Eagle's minimum essential medium
with Earle's salts (Gibco) plus 10% (v/v) foetal bovine serum (Hyclone), glutamine and antibiotics. HeLa cells
(Flow Laboratories, No. 03-117) were grown in pH 7-2 McCoy modified 5A medium (Gibco) with 10% (v/v) foetal
bovine serum (Hyclone), glutamine and antibiotics. After removal of medium from confluent cell monolayers
(containing 2-5 x 105 cells per well), 2 x 106 radiolabelled T. vaginalis (washed three times by centrifugation in
PBS and exposed to hormones for various times) in 1Ό ml medium consisting of two parts cell culture medium and
one part T. vaginalis medium were added to cell monolayers and maintained at 37 °C with 5% C 0 2 for 45 min.
After incubation, non-adherent T. vaginalis were removed by rinsing each well three times with warmed PBS
(Martinotti et al., 1986). The contents of each well were then digested with 0-5 ml 01 M-NaOH for 30 min on a
rotator. The digest was neutralized with HC1 and transferred to vials; scintillation fluid was added and
radioactivity measured in a liquid scintillation counter (Sugarman & Epps, 1982).
Chemotaxis assay. Radiolabelled T. vaginalis were washed in PBS, resuspended in 0-5 ml medium to a
concentration of 1 x 105 cells ml - 1 , and placed in the bottom of a 24-well Transwell plate (Costar). A 0-45 μπι,
porosity membrane filter insert was placed on top of the T. vaginalis suspension and 0-5 ml medium was placed on
top of the membranes. Chemoattractant test substances were added to the bottom medium, the top medium, or not
at all (controls). After incubation at 36 °C for 1 h, during which migrating T. vaginalis were trapped in the filter,
the membrane insert was removed, inverted and any excess fluid was removed by blotting with filter paper. The
membranes were removed from the inserts and solubilized with 0-5 ml 0-5 M-Protosol (DuPont, NEN Research
Products); scintillation fluid was added and radioactivity was measured in a liquid scintillation counter.
Data conversion and statistical analysis. Duplicate experiments were always done with known quantities of T.
vaginalis counted in a liquid scintillation counter to allow conversion of c.p.m. values to number of trichomonads.
All assays were done at least six times and data analysed with (non-parametric) Wilcoxon signed rank tests.
RESULTS AND DISCUSSION
Growth of T. vaginalis
Numerous isolates of T. vaginalis were incubated with hormones and the growth of the
organisms was measured. Only the oestrogens had a significant, inhibitory, effect on growth.
Results for three isolates are shown in Table 1. Growth inhibition was maximal after 2-3 d
incubation with oestrogens at 1-2 ng ml -1 (3·7-7·4 χ 10~9 M). The specificity of this effect was
verified by its concentration dependency: no effect on growth was noted with concentrations of
oestradiol less than 2-8 x 10~10M, but above this value growth inhibition increased with
oestradiol concentration up to 3-7xl0~ 8 M; higher concentrations produced no further
diminution of growth. Further, the anti-oestrogen tamoxifen had no effect on the growth of the
T. vaginalis isolates at concentrations of up to 100 ng ml - 1 ; however, at 1 ng ml"1 it completely
blocked the growth inhibition that occurred in the presence of oestrogens (Table 2). Earlier
studies with oestrone (Christow, 1971) and with 17a- and 17/?-oestradiol (Martinotti & Savoia,
1985) reported no effect on growth (oestrone) and increased growth (about 50%) (oestradiol)
after hormone exposure. Because these studies used single isolates, and different incubation
media and conditions, the results are not strictly comparable. The reported increased growth
after oestradiol exposure (Martinotti & Savoia, 1985) occurred at 20 h incubation and was not
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Thu, 15 Jun 2017 15:45:22
1625
Hormones and Trichomonas vaginalis
Table 1. Eflect of hormones on the growth of T. vaginalis
Cell numbers were determined by direct counts after 3 d growth in the standard assay (see Methods).
The results are means f SD (n = 6).
x No. of cells ml-I
I
Hormone
None (PBS control)
17f?-Oestradiol (1 ng ml-I)
Oestriol (1 ng ml-I)
Hydrocortisone (0.4 pg ml-I)
Prednisolone (1 pg ml-I)
Progesterone (20 ng ml-I)
Testosterone (20 ng ml-I)
1
Isolate 1
Isolate 2
Isolate 3
1*4 f 0.7
0.8 f 0.5*
0.9 f 0*6*
1.2 f 1.1
1.2 f 0.9
1.2 & 0.9
1-3 1-1
1-4 f 0.6
0.9 f 0*5*
1.1 f0.4
1.0 f 0.9
1.1 f 0 . 7
1-0 f 0.7
1.3 f 0.7
1.7 f 1.2
1.3 f 1.2*
1.5 f 1.6
1.6 f 0.9
1.6 f 1.7
1.6 f 1.5
1.5 f 1.2
* Significantly different from control (P< 0.05).
Table 2. Efect of oestrogens and an anti-oestrogen (tamoxifen) on the growth and adherence of
T . vaginalis isolates
Growth was measured after 3 d as described for Table 1. Adherence to McCoy cells was measured in a
45 min assay (see Methods) after 1 h exposure of the trichomonads to the test solution (except for the
results in line 3, where the trichomonads were exposed to oestradiol immediately before the assay). The
results are means f SD (n = 6).
Growth
r
PBS (control)
17fl-Oestradiol(1 ng ml-I)
17fl-Oestradiol(1 ng ml-l)
without preincubation
Tamoxifen
(10 ng m1-I)
(100 ng ml-I)
17mestradiol (1 ng ml-l)
and tamoxifen (10 ng ml-I)
x no. of cells ml-l)
Isolate 1
Isolate 2
Isolate 3
0.50 f 0.19
0.40 f 0.14*
0.51 f 0.21
0.39 f 0.17*
0.47 f0.17
0.48 & 0.19
0.55 f 0.20
0.51 f 0.18
0.52 f 0-21
0.60 f 0.18
-
-
Adherence
~r
x no. of cells per well)
1
Isolate 1
Isolate 2
Isolate 3
0.83 f 0.40
0.58 f 0.41*
-
0.53 f 0.04
0.15 f O W *
0.52 f 0.04
0.54 f 0.09
0.21 f 0.04*
0.54 f 0.09
0.43 f 0.07
0-21f 0-05*
0.43 f 0.08
0.84 f 0.25
0.84 f 0.26
0.77 f 0.28
0.52 f0.04
0.52 f0.04
0.52 f0.04
0.50 f 049
0.50 f 0.08
0.54 f 0.09
0.44 f 0.07
0.45 f0.06
0.44 f 0.07
* Significantly different from the appropriate control (Pc 0.05).
apparent after 28 h of oestrogen exposure. We consistently noted no effect on growth until 2-3 d
of incubation for any of numerous isolates tested.
Exposure to oestrogens during growth had no effect on the morphology and motility of
trichomonads as observed by 1000 x phase-contrast light microscopy. Scanning electron
microscopy of glutaraldehyde-fixed trichomonads similarly failed to show any morphological
effects of 3 d exposure to oestrogen (data not shown).
Adherence of T. vaginalis
The adherence of radiolabelled trichomonads to cultured mammalian cells was maximal after
30-45 min incubation with a ratio of trichomonads to mammalian cells of 10: 1 (data not
shown); therefore, these conditions were used for the remainder of the reported experiments.
Incubation of the trichomonads with oestradiol or oestriol at 1 ng ml-l (3.7 and 3-5x
M,
respectively) for 1 h prior to incubation with mammalian receptor cells was associated with a
statistically significant decrease in adherence to fibroblastic McCoy cells and epithelioid HeLa
cells. Other hormones had no such effect (Table 3). Momentary exposure of the trichomonads to
oestrogens was not associated with decreased attachment (Table 2) and more prolonged (up to
4 h) incubation of trichomonads with oestrogens prior to adherence was not associated with
further decreases in attachment compared with 1 h incubation (data not shown). The anti-
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Thu, 15 Jun 2017 15:45:22
1626
B . S U G A R M A N A N D N . MUMMAW
Table 3. Eflect of hormones on adherence of T. vaginalis (isolate 1) to McCoy and HeLa cells
propagated in 24-well culture plates
Adherence was measured in a 45 min assay (see Methods) after exposure of the trichomonads to the test
solution for 1 h. The results are means f SD (n = 6).
Adherence
x no. of
cells per well)
Hormone
None (PBS control)
17fl-Oestradiol(1 ng ml-l)
Oestriol (1 ng ml-l)
Hydrocortisone (0.4 pg ml-l)
Prednisolone (1 pg ml-l)
Progesterone (20 ng ml-I)
Testosterone (20 ng ml-l)
f
1
McCoy
HeLa
0.37 f 0.02
0.22 f 0.01*
0.22 f 0.01*
0.35 f 0-01
0.36 f 0.01
0.38 f 0.01
0.37 f 0.01
0.32 2 0.03
0.20 f 0*02*
0.21 f 0*02*
0.32 f 0.03
0.32 f 0.03
0.32 f 0.03
0.32 f 0.03
* Significantly different from control (Pc 0.05).
Table 4. Chemotaxis of T. vaginalis (isolate I ) in response to hormones
The trichomonads were placed below the filter, and the test solution was placed above or below the
filter, to measure attraction or repulsion respectively (see Methods for details). The results are
means f SD (n = 7).
x No. of cells in filter
f
1
Hormone
Attrac tiont
Repulsion:
None (media control)
17B-Oestradiol (10 ng ml-l)
17B-Oestradiol (100 ng ml-l)
Testosterone (200 ng ml-l)
24.7 f 9.0
16.9 f 2*1*
16.7 f 8*6*
23.3 f 0.1
24.7 f 9.0
34.6 f 2*2*
35.2 2-3*
24.4 f 0.1
* Significantly different from control (P< 0.05).
Test solution placed above filter.
$ Test solution placed below filter.
oestrogen tamoxifen incubated with trichomonads at concentrations up to 100 ng ml-l for 1 h
did not alter subsequent adherence; however, addition of tamoxifen with oestrogens in the
incubation medium negated subsequent effects of oestradiol (Table 2). Analogous results were
obtained with McCoy or HeLa cells as receptor cells (Table 3).
Modest increases in the subsequent adherence of many bacteria to mammalian cells have
been reported after overnight exposure of the mammalian cells to oestrogens (Sugarman &
Epps, 1982; Sugarman & Agbor, 1986). Martinotti et al. (1986) showed that in vitro attachment
of one strain of T. vaginalis to cultured epithelioid cells was moderately increased if the receptor
cells were exposed to certain oestrogens for 18 h prior to exposure to the trichomonads. We
repeated this experiment with our assay system and also noted a modest (mean = 12%) increase
in attachment after 18 h incubation of the receptor cells with oestrogens; but no increase was
found with 1 h incubation. This verifies an effect of oestrogens on T. vaginalis infection caused
by oestrogen-mammalian cell interactions.
Chemotaxis of T. vaginalis
The use of a membrane filter separating media with trichomonads on one side and various
media components on each side allowed the assay of chemoattraction as well as chemorepulsion.
Random migration was assayed with the same medium on both sides. The chemotaxis of
trichomonads towards nutrients was verified by the increased movement (and subsequent
membrane filter trapping) of organisms towards nutrient medium away from PBS (data not
shown).
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Thu, 15 Jun 2017 15:45:22
Hormones and Trichomonas vaginalis
1627
Using this assay system, the organisms moved away from their area of containment if
supplemental oestrogens were present in the incubation medium, and more were trapped in the
membrane filter than for control medium (Table 4). Similarly, if supplemental oestrogens were
present in the medium on the other side of the filter, fewer organisms were trapped in the
membrane filter than in control filters (chemorepulsion). Testosterone caused no chemotactic
response (Table 4).
Our results show that oestrogens directly inhibit the growth and adherence of T. vaginalis to
mammalian cells, and that they act as chemorepellents. These findings might suggest that
exposure to oestrogens would reduce susceptibility to T. vaginalis infection. This contrasts with
other studies that have demonstrated that prolonged exposure of mammals or mammalian cells
to oestrogens may increase susceptibility to T. vaginalis infection.
This work was supported, in part, by a grant from Michigan State University. Electron microscopy was
performed at the Michigan State University Center for Electron optics. Dr James B. Jensen provided expert
manuscript review. We thank Diane Bannerman for secretarial assistance.
REFERENCES
ALDERETE,
J. F. & GARZA,G. E. (1985). Specific
nature of Trichomonas vaginalis parasitism of host
cell surfaces. Infection and Immunity 50, 701-708.
ARNOLD,
J. D., BERGER,
A. E. & MARTIN,D. C. (1969).
Role of endocrine system in controlling growth and
division synchrony of Plasmodium berghei in mice.
Journal of Parasitology 55, 956-962.
T. (1968). A study of the parasitizing condition
AZUMA,
of Trichomonas vaginalis with special reference to the
relationship between estrogen and the growth of
Trichomonas vaginalis. Journal of the Japanese Obstetrics and Gynecology Society 15, 168-172.
BOTTA,G. A. (1979). Hormonal and type-dependent
adhesion of group B streptococci to human vaginal
cells. Infection and Immunity 25, 1084-1086.
CAPPUCCINELLI,
P., LATTES,C., CAGLIANI,
I. & PONZI,
A. N. (1974), Features of intravaginal Trichomonas
vaginalis infection in the mouse and the effect of
estrogen treatment and immunodepression. Giornale
di batteriologia, virologia ed immunologia 67, 31-40.
CHRISTOW,
P. (1971). Experiments on the influence of
metabolites and antimetabolites on the model of
Trichomonas vaginalis. VIII. Communication : experiments with hormones. Zentralblatt fur Bakteriologie, Parasitenkunde, Infektionskrankheiten und
Hygiene (I. Abt. Orig. A) 218, 542-552.
CSABA,G., LANTOS,T., NAGY,S. U., ARANYI,P. &
N m Y , A. (1978). Effects of steroids on Tetrahymena. Acta biologica et medica germanica 37,13771380.
A. & FEHER,
T. (1985).
CSABA,G., INCZEFI-GONDA,
Induction of steroid binding sites (receptors) and
presence of steroid hormones in the unicellular
Tetrahymena pyriformis. Comparative Biochemistry
and Physiology 82A, 567-570.
L. S. (1983). Lumen dwelling protozoa:
DIAMOND,
Entamoeba, trichomonads and Giardia. In In Vitro
Cultivationof Protozoan Parasites, pp. 65-1 09. Edited
by J. B. Jensen. Boca Raton: CRC Press.
EL-BOULAQI,
H. A., EL-REFAIE,S. A., BASSIOUNY,
G. A. & AMIN,F. M. A. (1984). The relation between
Trichomonas vaginalis and contraceptive measures.
Journal of the Egyptian Society of Parasitology 14,
495-499.
FORSLIN,L., DANIELSSON,
D. & FALK, V. (1979).
Variations in attachment of Neisseria gonorrhoeae to
vaginal epithelial cells during the menstrual cycle
and early pregnancy. Medical Microbiology and
Immunology 167, 231-238.
HAMANA,
K. & IWAI, K. (1971). Effects of steroid
hormones on the growth of Tetrahymena. Journal of
Biochemistry 69, 463-469.
S.
KITA, E., TAKAHASHI, S., YASUI,K. & KASHIBA,
(1985). Effect of estrogen (17b-estradiol) on the
susceptibility of mice to disseminated gonococcal
infection. Infection and Immunity 49, 238-243.
LINSTEAD,
D. (1981). New defined and semi-defined
media for cultivation of the flagellate Trichomonas
vaginalis. Parasitology 83, 125-1 37.
D. (1982). Characterization
LOOSE,D. S. & FELDMAN,
of unique corticosterone-binding protein in Candida
albicans. Journal of Biological Chemistry 257, 49254930.
D. (1985). Influenza di
MARTINOTTI,
M. G. & SAVOIA,
alcuni ormoni steroidei sulla crescita di Trichomonas
vaginalis.Giornale di batteriologia,virologia ed immunologia 78, 52-59.
MARTINOTTI,
M. G., MARTINETTO,
P. & SAVOIA,
D.
(1986). Adherence of Trichomonas vaginalis to cell
culture monolayers. European Journal of Clinical
Microbiology 5, 320-323.
POWELL,B. L., DRUTZ,D. J., HUPPERT,M. & SUN,
S. H. (1983). Relationship of progesterone- and
estradiol-binding proteins in Coccidioides immitis to
coccidioidal dissemination in pregnancy. Infection
and Immunity 40,478-485.
M. I., HAYES,H. T. & RADER,J.
PUNG,0. J., LUSTER,
(1 984). Influence of steroidal and nonsteroidal sex
hormones on host resistance in mice: increased
susceptibility to Listeria monocytogenes after
exposure to estrogenic hormones. Infection and
Immunity 46, 301-307.
PUNG,0. J., TUCKER,A. N., VORE, S. J. & LUSTER,
M. I. (1985). Influence of estrogen on host resistance : increased susceptibility of mice to Listeria
monocytogenes correlates with depressed production
of interleukin 2. Infection and Immunity 50, 9196.
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Thu, 15 Jun 2017 15:45:22
1628
B . SUGARMAN A N D N . MUMMAW
S c m , G . , STOVER,
E. P., CLEMONS,
K. V., FELDMAN,
Chlamydia trachomatis. Proceedings of the Societyfor
D. & STEVENS,
D. A. (1986). Progesterone binding
Experimental Biology and Medicine 183, 125-1 3 1.
and inhibition of growth in Trichophyton mentagroSUGARMAN,
B. & EPPS,L. R. (1982). Effect of estrogens
phytes. Infection and Immunity 52, 163-161.
on bacterial adherence to HeLa cells. Infection and
K. V., STEVENS, Immunity 35, 633-638.
STOVER,
E. P., ScHXR, G . , CLEMONS,
D. A. & FELDMAN,
D. (1986). Estradiol-binding WEISS,M., INGBAR, S. H., WINBLAD,
S., & KASPER,
proteins from mycelial and yeast-form cultures of
D. L. (1983). Demonstration of a saturable binding
Paracoccidioides brasiliensis. Infection and Immunity
site for thyrotropin in Yersinia enterocolitica. Science
51, 199-203.
219, 1331-1333.
SUGARMAN,
B. & AGBOR,P. (1986), Estrogens and
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Thu, 15 Jun 2017 15:45:22