0021-972X/01/$03.00/0
The Journal of Clinical Endocrinology & Metabolism
Copyright © 2001 by The Endocrine Society
Vol. 86, No. 6
Printed in U.S.A.
Interleukin-8 in the Human Fallopian Tube
STEVEN F. PALTER, NACIYE MULAYIM, LEVENT SENTURK,
AND
AYDIN ARICI
Division of Reproductive Endocrinology, Department of Obstetrics and Gynecology, and Division of
Reproductive Endocrinology, Yale University School of Medicine, New Haven, Connecticut 06520
ABSTRACT
The human fallopian tube is a dynamic structure that undergoes
cyclic variation in its functional epithelium. This epithelium contains
both secretory and ciliated cells. The mechanisms regulating the
growth and function of the tubal epithelium are not fully understood.
Interleukin-8 (IL-8) is one potential local regulatory factor. We therefore characterized the IL-8 system, which includes IL-8, its receptors
A and B, and its degradative enzyme aminopeptidase N, in the human
fallopian tube by immunohistochemistry. Immunohistochemistry
was performed on isthmic, ampullary, and fimbrial fallopian tubal
T
HE FALLOPIAN TUBE, a dynamic muscular tube with
an epithelial lining, is the site of gamete transport,
maturation, and fertilization as well as early embryo development and transport. Far from functioning as a passive
conduit to gametes and embryos, the fallopian tube actively
secretes multiple substances, has an active ciliated epithelium, and undergoes muscular contractions (1).
Diseased tubes are associated with decreased fecundity
and may be the site of ectopic gestations. The tubal epithelial
lining is composed of two populations of differentiated cells:
one nonciliated and secretory, and one ciliated. Distinction
between the cell types is not absolute, with occasional cells
demonstrating both cilia and secretory capacities in nonhuman studies (2). The secretory function of the tube remains
incompletely characterized. The tube is commonly divided
into four anatomical segments (3). The most proximal intramural segment courses through the wall of the uterus. The
isthmic segment lies immediately outside of the uterus and
extends to the point where the tube is seen to widen externally, with thick muscular layers and a relatively narrow
lumen. The distal two thirds of the fallopian tube is visibly
wider than the isthmus and comprises the ampulla, which
has a wider lumen and thinner muscular layers than the
isthmus. The most distal portion of the tube is the fimbrae,
which are the folds that contact the ovary.
The entire length of tubal epithelium undergoes cyclic
variation across the menstrual cycle (1, 4 – 6). Cell height and
cellular population percentages as well as mitotic activity
vary significantly across the cycle. Circulating ovarian sex
steroids may regulate these cyclic changes; however, the
mechanism of action of these steroids (direct or indirect via
Received August 24, 2000. Revision received January 11, 2001. Rerevision received January 19, 2001.
Address all correspondence and requests for reprints to: Steven F.
Palter, M.D., Department of Obstetrics and Gynecology, Division of
Reproductive Endocrinology, Yale University School of Medicine, 333
Cedar Street, P.O. Box 208063, New Haven, Connecticut 06520. E-mail:
[email protected].
segments obtained from women undergoing gynecological surgical
procedures for benign conditions (n ⫽ 52). IL-8 was found in the
human fallopian tube predominantly in the epithelial cells. It was
present in greater amounts in the distal compared with the proximal
tube. IL-8 receptors A and B localized in the tube in similar patterns.
The degradative enzyme aminopeptidase N is found in tubal stromal
tissue at the epithelial stromal border and perivascularly and may
limit the systemic effects of epithelial IL-8. The IL-8 system seems to
be an active component of tubal physiology. (J Clin Endocrinol Metab
86: 2660 –2667, 2001)
growth factors and cytokines) remains largely unknown.
Studies of other active epithelia (such as in the gastrointestinal tract and endometrium) have demonstrated central
roles for local regulatory substances governing similar functions (7, 8).
Chemokines are a large family of chemotactic cytokines
with molecular masses between 8 –10 kDa. As soluble chemoattractant molecules, they attract leukocytes (8, 9). Interleukin-8 (IL-8) is a member of the ␣, or CXC, subfamily, so
classified based on the spacing of the second and fourth
cysteine residues (C-C) with an intervening amino acid (X)
in the amino-terminus region (9). The CXC chemokines particularly attract and activate neutrophils and include IL-8,
melanoma growth-stimulating activity (MGSA), GRO-␣, and
platelet factor-4 (10). In addition to its proinflammatory effects, IL-8 is mitogenic for endometrial cells as well as for
epidermal, melanoma, and vascular smooth muscle cells (11–
14). IL-8 also causes angiogenesis via chemotaxis of endothelial cells (11).
Peripheral blood monocytes, endothelial cells, fibroblasts,
neutrophils, keratinocytes, synovial cells, mesothelial cells,
and endometrial glandular cells all produce IL-8 (15–19). In
the human endometrium IL-8 is present by immunohistochemistry (IHC) staining in the glandular and surface epithelium and is secreted by endometrial stromal and epithelial
cells in culture (15). IL-1 and tumor necrosis factor-␣ regulate
the production of IL-8 from these cells in vitro, and secretion
in vivo varies across the menstrual cycle (20). The presence of
IL-8 in human fallopian tube tissue has not previously been
described.
Two specific membranous receptors mediate the effects of
IL-8 (10, 21–24). These G protein-coupled receptors are classified as IL-8 receptor A (IL-8RA) and IL-8 receptor B (IL8RB). These receptors share 77% amino acid sequence identity, but differ in their ligand specificities (10, 22). IL-8RA
preferentially binds IL-8, whereas IL-8RB also binds other
chemokines, such as MGSA and neutrophil-activating protein-2, with equal affinity (10, 23).
2660
IL-8 IN THE HUMAN FALLOPIAN TUBE
Aminopeptidase N (APN; arylaminidase), also known as
CD13, is a single chain 150-kDa membrane glycoprotein with
metalloproteinase activity that functions as the primary degradative enzyme for IL-8 (25–27). APN is expressed on the
cell surface of many cell types and cleaves IL-8, thereby
inactivating its chemotactic activity in vitro. Inactivation of
IL-8 is one mechanism controlling the chemotactic activity of
IL-8. To date, the presence of the IL-8 receptors and APN has
not been reported in the fallopian tube.
We hypothesized that the human fallopian tube produces
and secretes IL-8, which may have autocrine growth-regulating and/or chemokine activity. We first conducted IHC of
human fallopian tube segments throughout the menstrual
cycle and identified IL-8 and its receptors, IL-8RA and IL8RB. Next, we assessed the presence of APN by IHC. We
finally identified macrophages in the tube via IHC staining
with CD68, because they also produce IL-8 and may participate in this system.
Materials and Methods
Tissue collections and experimental subjects
Fallopian tubes were obtained from cycling women undergoing tubal
ligations or hysterectomy conducted for benign gynecological conditions (excluding tubal pathology) at Yale-New Haven Hospital. Isthmic
tubal segments were obtained 2 cm from the cornua, ampullary segments were obtained 2 cm from the fimbrial segment, and frimbrial
segments were obtained from the distal 1 cm of the tube. Tubes with
gross pathology were not collected for this study. Tissue was immediately collected into Hanks’ Balanced Salt Solution on ice in the operating
room and then frozen in OCT (Tissue-Tek, Sakura, Torrance, CA) in
isopentane and liquid nitrogen within 1 h of collection. Informed consent
was obtained, and the human investigation committee of Yale University School of Medicine approved the protocol. Endometrial tissue from
the subjects was submitted for histological examination, and menstrual
cycle day was established using the criteria reported by Noyes et al. (28).
Dating characterized the samples as early proliferative (days 1–5 of the
cycle), midproliferative (days 6 –10 of the cycle), late proliferative (days
11–14 of the cycle), early secretory (days 15–18), midsecretory (days
19 –23), and late secretory (days 24 –28) phases of the menstrual cycle. In
many cases all three tubal segments were not available from a given
subject.
IHC staining
Tissues were sectioned in 6-␮m sections on a cryostat and placed on
poly-l-lysine-coated glass microscope slides for IHC. Tissue sections
were fixed in acetone at 4 C for 5 min and rinsed twice in PBS (pH 7.4)
for 5 min each time and then in PBS-BSA (0.1%, wt/vol) for 5 min.
Nonspecific staining was reduced via incubation with 4% blocking normal horse serum (Vector Laboratories, Inc., Burlingame, CA) for 1 h at
room temperature in a humidified chamber. Excess serum was drained,
and the primary antibodies were applied. The tissue was incubated
overnight at 4 C in a humidified chamber with specific monoclonal
murine IgG1 antibodies.
For IL-8 IHC, a specific monoclonal antibody directed against human
IL-8 was used (murine monoclonal antihuman IL-8, clone Nap II, IgG1;
10 ␮g/mL; Bender Med Systems, Vienna, Austria). For chemokine receptor IHC, specific monoclonal murine IgG1 antibodies directed
against human IL-8-RA and IL-8-RB were used as follows: 1) murine
monoclonal antihuman CXCR-1 (IL-8 RA) antibody IgG2B, clone 5A12
(PharMingen, San Diego, CA), 50 ␮g/mL, 1:300 dilution in PBS-BSA; and
2) murine monoclonal antihuman CXCR-2 (IL-8 RB) antibody IgG2A
clone 8311.211, 500 ␮g/mL, 1:300 dilution in PBS-BSA (R&D Systems,
Inc., Minneapolis, MN). For APN, a specific monoclonal murine IgG1
antibody directed against human APN/CD13 [murine monoclonal antihuman myeloid cell antibody, clone VS5E, 1:50 dilution (Novocastra
Laboratories, Newcastle Upon Tyne, UK)] was used. To identify macrophages within the fallopian tube, IHC staining was performed with
2661
CD68 (murine monoclonal antihuman macrophage antibody, IgG1,
clone EBM11; DAKO Corp., Carpinteria, CA; 395 ␮g/mL; 1:100 dilution
in PBS-0.1% BSA).
Initial experiments were performed to determine the optimal primary
antibody concentrations for each antibody in human fallopian tube
sections. For the negative control, nonspecific mouse IgG was used at the
same concentrations. Endogenous peroxidase activity was quenched
with 0.6% H2O2 in PBS (vol/vol) for 15 min. Sections were rinsed and
then incubated with biotinylated horse antimouse IgG (1.5 mg/mL;
Vector Laboratories, Inc.) for 45 min in humidified chambers at room
temperature. The antigen-antibody complex was detected using an avidin-biotin peroxidase kit (Vector Laboratories, Inc.). Freshly diluted
filtered diaminobenzidine (3,3-diaminobenzidine tetrahydrochloride
dihydrate, Aldrich Chemical Co., Inc., Milwaukee, WI)/hydrogen peroxide (0.5 mg diaminobenzidine in 0.03% H2O2 in PBS) was used as the
chromogen, and sections were counterstained with hematoxylin and
mounted with Permount (Fisher Scientific, Springfield, NJ) on glass
slides.
IHC staining was scored by two evaluators blinded to the sample’s
menstrual phase and the subject’s clinical history in a semiquantitative
fashion incorporating both intensity and the distribution of staining (29).
The evaluations were recorded as percentages of positively stained
target cells in each of four intensity categories, which were denoted as
0 (no staining), 1⫹ (weak but detectable above negative control), 2⫹
(distinct), and 3⫹ (intense). Epithelial and stromal cells were separately
scored. Where applicable, staining intensity was evaluated for the overall epithelium, apical surface of the epithelium, basal surface of the
epithelium, stroma, perivascular area, and tubal muscular layer. For
each tissue, an HSCORE value was derived by summing the percentages
of cells that stained at each intensity multiplied by the weighted intensity
of the staining [HSCORE ⫽ Pi(I ⫹ 1), where i is the intensity score, and
Pi is the corresponding percentage of the cells]. Macrophages stained
with CD68 antibody were counted using an Olympus Corp. microscope
(New Hyde Park, NY) with a special ocular grid.
Statistical analyses
Epithelial and stromal HSCOREs were compared using KruskalWallis one-way ANOVA on ranks or one-way ANOVA for staining
scores as applicable. One-way ANOVA was used when scores were
normally distributed and with equal variance, as tested by the Kolmogorov-Smirnov test. When significant differences between groups were
found, post-hoc multiple comparison tests were used to isolate these
differences. Multiple comparisons were performed with Tukey’s test or
Dunn’s (when group sizes were unequal). All statistical analyses were
performed using SigmaStat version 3.02 (SPSS, Inc., Chicago, IL), with
significance at the P ⬍ 0.05 level.
Results
IL-8 immunohistochemistry in the fallopian tube
We evaluated 37 fallopian tube samples by IHC for IL-8.
Samples were from the isthmic (n ⫽ 10), ampullary (n ⫽ 16),
and fimbrial (n ⫽ 11) portions of the tube. Samples were
taken from women in various phases of the menstrual phase
as judged by endometrial histological dating. We obtained
samples from women in the early proliferative (n ⫽ 10),
midproliferative (n ⫽ 9), late proliferative (n ⫽ 4), early
secretory (n ⫽ 5), midsecretory (n ⫽ 6), and late secretory
(n ⫽ 3) phases of the menstrual cycle. There was diffuse IHC
staining for IL-8 of the surface epithelium throughout the
fallopian tube (Fig. 1, A–D). This staining was primarily
membranous, with lesser cytoplasmic staining visible. The
membranous staining was most intense at the apical membranes. In contrast, the stroma was predominantly negative
for staining. Isolated cells with the morphological characteristics of macrophages stained positively in the stroma of
selected sections. Distinct staining was also visible in the
tubular muscular wall as well as, to a lesser extent, in the
2662
PALTER ET AL.
JCE & M • 2001
Vol. 86 • No. 6
FIG. 1. Immunohistochemistry staining for IL-8 (A–D) and IL-8RA (E–H) throughout the fallopian tube. A section of isthmic tube (⫻200
magnification) shows predominantly epithelial staining in A. A section of ampullary tube demonstrating predominantly epithelial staining is
shown in B (⫻400). Higher intensity fimbrial epithelial staining is shown in C (⫻400). Isolated positively staining cells with the morphological
characteristics of macrophages are also seen in the stroma. A second fimbrial section is shown in D (⫻400), demonstrating lower intensity
staining of vessel endothelium. A section of proximal tube shows predominantly epithelial IL-8 RA staining in E (⫻200). A section of distal tube
demonstrating muscular, but not epithelial, staining of a vessel wall is shown in F. Fimbrial epithelial and stromal staining is shown in G (⫻400)
and at higher magnification in H (⫻600).
IL-8 IN THE HUMAN FALLOPIAN TUBE
endothelia of blood vessels. There was no significant variation in the staining intensity in the vascular structures or the
muscle across the menstrual cycle. The negative controls had
no distinct staining.
We compared epithelial IHC staining HSCOREs among
the different tubal segments and across the phases of the
menstrual cycle. Throughout the menstrual cycle we found
greater HSCOREs in the distal than in the proximal tube, and
HSCOREs were significantly higher in the fimbria than in the
isthmic segment (P ⫽ 0.009; Fig. 2). Apical membrane staining was significantly greater in the distal than in the proximal
segments, but the difference was not significant for basal
membrane staining along the different tubal segments. There
was no significant variation in total HSCOREs or apical
HSCOREs among different phases of the menstrual cycle. In
contrast, there was significantly higher basal membrane
staining in the late proliferative vs. late secretory phase samples (P ⫽ 0.007; Fig. 3).
IL-8RA and IL-8RB IHC in the fallopian tube
We evaluated 33 fallopian tube samples by IHC for IL8RA. The samples were from the isthmic (n ⫽ 11), ampullary
(n ⫽ 11), and fimbrial (n ⫽ 11) portions of the tube. Endometrial histological dating of the samples was early proliferative (n ⫽ 15), midproliferative (n ⫽ 4), late proliferative
(n ⫽ 3), early secretory (n ⫽ 4), midsecretory (n ⫽ 3), and late
secretory (n ⫽ 4) phases of the menstrual cycle. There was
diffuse IL-8RA IHC staining of the surface epithelium
throughout the tube (Fig. 1, E–H). This staining was membranous and was most intense at the apical membranes.
Epithelial staining was observed in all three tubal segments.
Membranous stromal staining was also observed, but at a
greatly reduced intensity than that of epithelial staining.
Distinct staining was also visible in the muscular cells of the
tubal wall and vascular structures. The vessel endothelium
did not exhibit IHC staining. No distinct staining was observed in the negative controls.
Similar to IL-8, when HSCOREs were compared among
the different tubal segments there was significantly greater
2663
FIG. 3. IL-8 IHC staining HSCOREs in the different phases of the
menstrual cycle. Bars represent the mean ⫾ SEM (P ⫽ 0.007, late
proliferative vs. late secretory). P1, Early proliferative; P2, midproliferative; P3, late proliferative; S1, early secretory; S2, midsecretory;
S3, late secretory phases of the menstrual cycle.
FIG. 4. IL-8 RA IHC staining HSCOREs along the length of the tube.
Bars represent the mean ⫾ SEM (P ⫽ 0.029, fimbria vs. isthmic).
FIG. 2. IL-8 IHC staining HSCOREs along the length of the tube.
Bars represent the mean ⫾ SEM (P ⫽ 0.009, fimbria vs. isthmic).
staining in the distal (fimbria) vs. proximal (isthmic) tube
(P ⫽ 0.029; Fig. 4). We observed no significant variation in
epithelial staining HSCOREs in tubal samples obtained from
different phases of the menstrual phase. There was no significant variation in stromal staining HSCOREs across either
tubal segments or phases of the menstrual cycle. There was
no staining observed in the negative controls.
We evaluated 34 fallopian tube samples by IHC for IL-8
RB. Samples comprised the isthmic (n ⫽ 13), ampullary (n ⫽
10), and fimbrial (n ⫽ 11) portions of the tube. The distribution of menstrual cycle phase samples obtained was early
proliferative (n ⫽ 15), midproliferative (n ⫽ 6), late proliferative (n ⫽ 2), early secretory (n ⫽ 5), midsecretory (n ⫽ 3),
and late secretory (n ⫽ 3). We observed IHC for IL-8RB
predominantly in the tubal epithelium, and it was greatest at
the apical membranes (Fig. 5, A and B). In contrast to IL-8 and
IL-8 RA, epithelial staining for IL-8RB was in a patchy, rather
than a continuous, pattern. The distribution of staining along
2664
PALTER ET AL.
JCE & M • 2001
Vol. 86 • No. 6
FIG. 5. IHC staining for IL-8RB (A and B), APN (C and D), and macrophage marker (CD68; E and F) throughout the fallopian tube. A section
of proximal tube is shown in A (⫻600) and demonstrates predominantly epithelial staining for IL-8RB with a patchy distribution. A section
of distal fimbrial tube demonstrating predominantly epithelial staining is shown in B (⫻600). A section of proximal tube is shown in C (⫻600)
and demonstrates predominantly stromal APN staining. Arrows denote the area of maximal stromal staining at the epithelial-stromal junction.
Perivascular stromal staining is shown in D (⫻400). A section of proximal tube is shown in E (⫻600) demonstrating cells with the morphological
characteristics of macrophages stained with CD68 in the stroma. Sections of the tubal fimbria are seen in F (⫻400).
the length of the fallopian tube for IL-8RB was similar to that
of IL-8 and IL-8RA, with significantly greater staining in the
distal than in the proximal tube (P ⫽ 0.045; Fig. 6). We
observed a trend toward a variation in the staining intensity
among the phases of the menstrual cycle with a midcycle
nadir; however, this did not reach statistical significance (Fig.
7). There was only minimal stromal staining observed without cycle or segment variation. There was no IHC staining
observed in the blood vessels or muscular wall of the tube or
in the negative controls.
IHC of APN in the fallopian tube
Thirty fallopian tube samples were evaluated by IHC for
APN. The samples included the isthmic (n ⫽ 10), ampullary
(n ⫽ 9), and fimbrial (n ⫽ 11) portions of the fallopian tube.
Samples were from subjects in the early proliferative (n ⫽ 10),
midproliferative (n ⫽ 5), late proliferative (n ⫽ 3), early
secretory (n ⫽ 3), midsecretory (n ⫽ 5), and late secretory
(n ⫽ 4) phases of the menstrual cycle. We observed APN IHC
staining in all segments of the fallopian tube. The staining
was present in both epithelium and stroma of the tube, but
was preferentially located in a distinct linear stromal band at
the epithelial-stromal junction (Fig. 5C). Stromal staining
was also observed to a lesser extent in the perivascular stromal tissue (Fig. 5D). Stromal staining was uniformly present
in tubal segments, whereas the vast majority of tubal segments did not demonstrate epithelial staining. In cases where
epithelial staining was present, it was seen in a small focal
IL-8 IN THE HUMAN FALLOPIAN TUBE
2665
stromal tissue of the different tubal segments (Fig. 5, E and
F). The majority of the samples examined contained relatively few macrophages, whereas selected samples had a
higher concentration. We did not observe significant differences in the number of macrophages present in the different
tubal segments or in tubal segments from different phases of
the menstrual cycle. The distribution of macrophages observed did not correspond to the majority of the IL-8 or IL-8R
staining.
Discussion
FIG. 6. IL-8RB IHC staining HSCOREs along the length of the tube.
Box plots show the data range with outliers. Lines indicate the median. P ⫽ 0.045, distal vs. proximal tube.
FIG. 7. IL-8RB IHC staining HSCOREs in the different phases of the
menstrual cycle. Box plots show the data range with outliers. Lines
indicate median. P ⫽ NS. P, Proliferative; S1, early secretory; S2,
midsecretory; S3, late secretory.
patchy distribution. No variation in epithelial or stromal
HSCOREs was observed between tubal segments or menstrual cycle phases, and there was no staining of the negative
controls.
IHC localization of macrophages in the fallopian tube
We evaluated 52 fallopian tube samples by IHC for the
presence of macrophages via staining with CD68. The tubal
samples were taken from the isthmic (n ⫽ 16), ampullary
(n ⫽ 18), and fimbrial (n ⫽ 18) portions of the fallopian tube.
Samples were from women in the early proliferative (n ⫽ 16),
midproliferative (n ⫽ 6), late proliferative (n ⫽ 4), early
secretory (n ⫽ 6), midsecretory (n ⫽ 2), and late secretory
(n ⫽ 5) phases of the menstrual cycle. We obtained an additional 13 samples from subjects for whom endometrial
dating was not available. We identified macrophages in most
tubal segments by CD68 staining, and they localized to the
The fallopian tube is far from a hollow passive conduit for
gametes. Rather, the tube plays an active role in the reproductive system and undergoes continual cyclic epithelial changes.
Despite a central role in natural fertilization, relatively little is
known about regulation of the changes observed in the tubal
epithelium. To date, the changes described include cell populations, ciliation, cell height, secretion, and muscular contractions (1). Although it has been postulated that circulating ovarian sex steroids regulate these cyclic changes, direct effects have
yet to be demonstrated. In addition, ciliated tubal epithelial cells
do not express classical estrogen receptors (1). For these reasons,
a putative role of local regulatory factors has been postulated.
IL-8 displays potent chemoattractant and cellular activating properties and is therefore classified as a chemokine. It
is produced locally in the human female reproductive tract
and has been localized to endometrial epithelial and stromal
cells (8, 14, 15). In the human endometrium, IL-8 has been
localized by IHC to the glandular and surface epithelium and
is produced by endometrial stromal and epithelial cells in
culture (20). A menstrual cycle-related variation in IL-8 messenger ribonucleic acid expression peaks in the late secretory
and early to midproliferative phases (15). IL-8 potentially
regulates leukocyte recruitment into the endometrium, and
its up-regulation at the end of the menstrual cycle may regulate the influx of neutrophils before menstruation. The rise
in IL-8 in the early proliferative phase may contribute to
neovascularization of the endometrium (11, 15).
We observed IL-8 in the human fallopian tube via IHC
staining, and it was predominantly localized to tubal epithelial cells. In addition, there was a significant increase in
staining in the distal compared with the proximal tube. Previously, IL-8 was demonstrated via enzyme-linked immunosorbent assay in human tubal fluid, probably secreted by
tubal epithelial cells (30).
The segmental variation observed may represent differences in the populations of epithelial cells present in each
location. Some reports have suggested that secretory activity
is maximal in the isthmus, where a higher proportion of
epithelial cells may be secretory, whereas ciliation is probably maximal in the distal tube (4, 6). Exact identification of
ciliated vs. secretory cells was not possible in our study via
light microscopy and would require either immunoelectron
microscopy or IHC staining with additional specific cellular
subtype markers.
We also observed a variation in IL-8 IHC staining across
the phases of the menstrual cycle. The preovulatory peak
coincides with the peaks in cell height, secretion, and ciliation
observed at this time. Notably, the epithelium reaches max-
2666
PALTER ET AL.
imal height and secretory activity in the late preovulatory
proliferative phase, whereas secretion decreases after ovulation in the luteal phase (4, 31). Donnez et al. demonstrated
that ciliation is maximal in the late follicular period, followed
by low levels of ciliation in the luteal phase of the cycle (6).
At the end of the luteal phase epithelial cell height diminishes
to its lowest, and ciliation and secretion are both maximally
suppressed (4, 5, 32, 33).
Although the epithelial changes occur across the menstrual cycle coincident with variations in estrogen and progesterone, several investigators suggested that growth factors might mediate the observed changes, especially as
estrogen receptors have not been identified in ciliary cells (1,
5, 6, 34, 35). Smotrich et al. localized transforming growth
factor-␣ to tubal epithelial cells via IHC, with weak staining
observed in stromal tissues (35). In contrast, epidermal
growth factor receptors did not localize to the tubal cells.
Others have identified insulin-like growth factor I, its binding proteins, and its receptor in the fallopian tube (34, 36, 37).
IL-8’s distribution, namely predominantly epithelial apical
membranous IHC positivity, resembles the distribution of
epidermal growth factor previously described (38). This
study suggests that IL-8 may serve as one such local regulatory cytokine.
Several fundamental physiological differences exist between the endometrium and the fallopian tube. Specifically,
the endometrium contains a population of resident and migratory lymphocytes. After ovulation, an influx of large granular lymphocytes occurs that may participate in the IL-8
system (8). We performed IHC staining for macrophages
using CD68 as a marker to determine whether macrophages
were involved in the expression of IL-8 in the tube. Our
observations did not reveal such an influx in the tube. Although macrophages were on occasion observed in the tubal
segments, their distribution was different than that of IL-8
and its receptors.
Several groups described increased cleavage rates and improved outcomes when they cocultured human in vitro fertilized embryos on cellular monolayers, including tubal epithelial cells (39). IL-8 may play an interactive role with
gametes or embryos in addition to a local regulatory role in
the tubal epithelium. In contrast, clinical in vitro fertilization
outcomes may be reduced in cases where hydrosalpinges are
present (40). As IL-8 may play a role in inflammatory conditions, tubal infections and subsequent hydrosalpinx formation may lead to altered local or secreted levels of IL-8,
which may be embryotoxic.
Tubal scarring and adhesions are other sequellae of pelvic
inflammatory disease. These are most commonly associated
with infections by Neisseria gonorrhea or Chlamydia trachomatis
in women. Tubal epithelial IL-8 may play an additional role
in this setting as a barrier to ip infection from ascending
cervical pathogens. The host response to a primary chlamydial infection of a mucosal surface is acute inflammation
and is characterized by infiltration of neutrophils and monocytes (41). In animal models of ophthalmological chlamydial
infection, the host’s inflammatory reaction can eliminate the
infectious organism, whereas in reinfection the host’s inflammatory response occurs more rapidly and with greater
T cell infiltration (41). In the gastrointestinal tract it has been
JCE & M • 2001
Vol. 86 • No. 6
postulated that the epithelial cells, which secrete proinflammatory cytokines, play a defensive role against infection (42).
Rasmussen et al. demonstrated in vitro that infection and
intracellular growth of Chlamydia trachomatis in cervical epithelial cells triggered a sustained production and secretion
of cytokines, including IL-8 from the epithelial cells (41). In
this regard, IL-8 in the tube may play a similar defensive role.
CXCR-1 and CXCR-2 were the first two chemokine receptors cloned in 1991 and were first identified on neutrophils
(10). CXCR-1 (IL-8R-A) is more specific for IL-8 and will bind
other cytokines with much lesser affinity. CXCR-2 (IL-8RB)
is less specific for IL-8 and, in addition, binds neutrophilactivating polypeptide-2, 78-amino acid epithelial neutrophil
activating factor, and MGSA/GRO-␣ as well as GRO-␤ and
-␥ (8, 23, 24). The presence of IL-8RA and IL-8RB in the
epithelial layer of the fallopian tube suggests that IL-8 may
have an autocrine effect on these cells. A similar growthpromoting effect of IL-8 was described in other cells types,
such as endometrium (14).
APN (CD13) is the primary degradative enzyme for IL-8
(27). It is a glycoprotein ectoenzyme, referring to its catalytic
site that is expressed on the external cell surface of many cell
types, including macrophages, neutrophils, renal tubular
cells, small intestinal epithelial cells, neuronal synaptic membranes, and endometrial stromal cells (25–27, 43, 44). APN
cleaves IL-8 and inactivates its chemotactic activity in vitro
(25–27, 43, 44). APN also inhibits autocrine stimulation of
neutrophils by IL-8 (25–27). In the human endometrium,
APN is localized to the perijunctional stroma (i.e. the glandular/stromal interface) by IHC (45).
In our study we observed APN by IHC in the human
fallopian tube. The IHC staining pattern localized predominantly in a focal band at the junction of the epithelial and
stromal cells, with a distribution similar to that seen in the
endometrium. Unlike the endometrium, in the fallopian tube
APN is also present in the perivascular stromal tissue. This
may represent a control mechanism to limit systemic absorption of IL-8, as IL-8 produced by the tubal epithelial cells
would be limited to either the tubal epithelium and/or secretion into the tubal lumen. In this regard, IL-8 has been
identified in human fallopian tubal secretions (30). As IL-8 is
a chemoattractant for neutrophils, APN in this location may
degrade epithelial IL-8 before diffusion into the stromal tissue. Similarly, the perivascular stromal APN may serve as a
second barrier against systemic absorption and prevent an
influx of neutrophils and proinflammatory effects of IL-8 in
the normal tube.
Further studies are warranted to define the role of the IL-8
system in the human fallopian tube. Experiments may evaluate the protein production and secretion both in vivo and in
vitro in fallopian tube cells. There is also a need to study the
distribution and expression of IL-8 and its receptors in postmenopausal women with and without hormonal replacement therapy and in reproductive-aged women undergoing
hormonal treatments. Studies should address the IL-8 system
in hydrosalpinges and hydrosalpinx fluid. The role of the
IL-8 system in disease states of chlamydial infection and
ectopic pregnancy should also be investigated. As IL-8RA
and IL-8RB localized to the tubal epithelium, the growth
effects of in vitro treatment of epithelial cells with IL-8 should
IL-8 IN THE HUMAN FALLOPIAN TUBE
be assessed. Further confirmation of the role of IL-8 may be
investigated in vitro and in vivo with regulatory experiments
using blocking antibodies, antisense techniques, and receptor homolog gene deletion (knockout) mice.
References
1. Brenner R, Slayden O. 1996 The fallopian tube cycle. In: Adashi E, Rock J,
Rosenwaks Z, eds. Reproductive endocrinology: surgery and technology. Philadelphia: Lippincott-Raven; vol 1:325–339.
2. Jansen R, Bajpai V. 1982 Oviduct acid mucus glycoprotein in the estrous
rabbit: ultastructure and histochemistry. Biol Reprod. 26:155–168.
3. Hoffman J. 1999 Anatomy and physiology of the fallopian tube. In: Hunt R,
ed. Text and atlas of female infertility surgery. St. Louis: Mosby; 23–30.
4. Jansen R. 1984 Endocrine response of the fallopian tube. Endocr Rev.
5:525–551.
5. Verhage H, Bareither M, Jaffee R, Akbar M. 1979 Cyclic changes in ciliation,
secretion, and cell height of the oviductal epithelium in women. Am J Anat.
156:505–521.
6. Donnez J, Casanas-Roux F, Caprasse J, Ferin J, Thomas K. 1985 Cyclic
changes in ciliation, cell height, and mitotic activity in human tubal epithelium
during reproductive life. Fertil Steril. 43:554 –559.
7. Welsh M, Smith P, Fromm M, Frizzell R. 1982 Crypts are the site of intestinal
fluid and electrolyte secretion. Science. 218:1219.
8. Garcia-Velasco JA, Arici A. 1999 Chemokines and human reproduction. Fertil
Steril. 71:983–993.
9. Baggiolini M, Dewald B, Moser B. 1994 Interleukin-8 and related chemotactic
cytokines-CXC and CC chemokines. Adv Immunol. 55:97–179.
10. Horuk R. 1994 The interleukin-8-receptor family: from chemokines to malaria.
Immunol Today. 15:169 –174.
11. Koch A, Polverini P, Kunkel S, et al. 1992 Interleulin-8 as a macrophage
derived mediator of angiogenesis. Science. 258:1798 –1801.
12. Schadendorf D, Moller A, Algermissen B, Worm M, Sticherling M,
Czarnetzki B. 1993 IL-8 produced by human malignant melanoma cell in vitro
is an essential autocrine growth factor. J Immunol. 151:2667–2675.
13. Tuschil A, Lam C, Haslberger A, Lindley I. 1992 Interleukin-8 stimulates
calcium transients and promotes epidermal cell proliferation. J Invest Dermatol. 99:294 –298.
14. Arici A, Seli E, Zeyneloglu HB, Senturk LM, Oral E, Olive DL. 1998 Interleukin-8 induces proliferation of endometrial stromal cells: a potential autocrine growth factor. J Clin Endocrinol Metab. 83:1201–1205.
15. Arici A, Seli E, Senturk LM, Gutierrez LS, Oral E, Taylor HS. 1998 Interleukin-8 in the human endometrium. J Clin Endocrinol Metab. 83:1783–1787.
16. Yoshimura T, Matsushima K, Oppenheim J, Leonard E. 1987 Neutrophil
chemotactic factor produced by lipopolysaccharide (LPS)-stimulated human
blood mononuclear leukocytes: partial characterization and separation from
interleukin-1 (IL-1). J Immunol. 139:788 –793.
17. Streiter R, Kunkel S, Showell H, et al. 1989 Endothelial cell gene expression
of a neutrophil chemotactic factor by TNF-␣, LPS, and IL-1␤. Science.
243:1467–1469.
18. Larsen C, Anderson A, Oppenheim J, Matsushima K. 1989 Production of
interleukin-8 by human dermal fibroblasts and keratinocytes in response to
interleukin-1 or tumour necrosis factor. Immunology. 68:31–36.
19. Streiter R, Kasahara K, Allen R, Showell H, Standiford T, Kunkel S. 1990
Human neutrophils exhibit disparate chemotactic factor gene expression. Biochem Biophys Res Commun. 173:725–730.
20. Arici A, Head JR, MacDonald PC, Casey ML. 1993 Regulation of interleukin-8
gene expression in human endometrial cells in culture. Mol Cell Endocrinol.
94:195–204.
21. Hoch RC, Schraufstatter IU, Cochrane CG. 1996 In vivo, in vitro, and mo-
22.
23.
24.
25.
26.
27.
28.
29.
30.
31.
32.
33.
34.
35.
36.
37.
38.
39.
40.
41.
42.
43.
44.
45.
2667
lecular aspects of interleukin-8 and the interleukin-8 receptors. J Lab Clin Med.
128:134 –145.
Murphy PM. 1997 Neutrophil receptors for interleukin-8 and related CXC
chemokines. Semin Hematol. 34:311–318.
Lee J, Horuk R, Rice G, Bennett G, Cmaerato T, Wood W. 1992 Characterization of two high affinity human interleukin-8 receptors. J Biol Chem.
267:16283–16287.
Wells T, Power C, Lusti-Narasimhan M, et al. 1995 Selectivity and antagonism
of chemokine receptors. J Leukocyte Biol. 59:53– 60.
Imai K, Maeda M, Fujiwara H, et al. 1992 Human endometrial stromal cells
and decidual cells express cluster of differentiation (CD) 13 antigen/aminopeptidase N and CD 10/neutral endopeptidase. Biol Reprod. 46:328 –334.
Imai K, Kanzaki H, Mori T. 1996 Cell surface peptidases in human endometrium. Mol Hum Reprod. 2:425– 431.
Kanayama N, Kajiwara Y, Goto J, et al. 1995 Inactivation of interleukin-8 by
aminopeptidase N (CD13). Leukocyte Biol. 57:129 –134.
Noyes R, Hertig A, Rock J. 1950 Dating the endometrial biopsy. Fertil Steril.
1:3–25.
Bacus S, Flowers J, Press M, JW B, McCarty K. 1988 The evaluation of estrogen
receptor in primary breast carcinoma by computer-assisted image analysis.
Am J Clin Pathol. 90:233–239.
Srivastava MD, Lippes J, Srivastava BI. 1996 Cytokines of the human reproductive tract. Am J Reprod Immunol (Copenh). 36:157–166.
Jansen R. 1980 Cyclic changes in the human fallopian tube isthmus and their
functional importance. Am J Obstet Gynecol. 136:292.
Novak E, Everett H. 1928 Cyclical and other variations in the tubal epithelium.
Am J Obstet Gynecol. 16:499.
Clyman M. 1966 Electron microscopy of the human fallopian tube. Fertil Steril.
17:281–301.
Guidice L, Dsupin B, Irwin J, Eckert R. 1992 Identification of insulin-like
growth factor binding proteins in human oviduct. Fertil Steril. 57:294 –301.
Smotrich D, Stillman R, Widra E, et al. 1996 Immunocytochemical localization
of growth factors and their receptors in human pre-embryos and fallopian
tubes. Hum Reprod. 11:184 –190.
Pfeifer T, Chegini N. 1994 Immunocytochemical localization of insulin-like
growth factor (IGF-1), IGF-1 receptor, and IGF-1 binding proteins 1– 4 in
human fallopian tube at various reproductive stages. Biol Reprod. 50:281–289.
Wegang B, Nap M, Bittl A, Jaeger W. 1994 Immunohistochemical localization
of insulin-like growth factor 1 receptors in benign and malignant tissues of he
female reproductive tract. Tumor Biol. 15:236 –246.
El-Danasouri I, Frances I, Westhal L. 1993 Immunocytochemical localization
of transforming growth factor-␣ and epidermal growth factor receptor in
human uterine tubes and cumulus cells. Am J Reprod Immunol. 30:82– 87.
Wiemar K, Cohen J, Tucker M, Godke R. 1998 The application of co-culture
in assisted reproduction: 10 years of experience with human embryos. Hum
Reprod. 13(Suppl 4):226 –238.
Zeyneloglu HB, Arici A, Olive DL. 1998 Adverse effects of hydrosalpinx on
pregnancy rates after in vitro fertilization-embryo transfer. Fertil Steril.
70:492– 499.
Rasmussen S, Eckmann L, Quayle A, et al. 1997 Secretion of proinflammatory
cytokines by epithelial cells in response to chlamydia infection suggestion a
central role for epithelial cells in chlamydial pathogenesis. J Clin Invest.
99:77– 87.
Eckmann L, Kagnoff MF, Fierer J. 1993 Epithelial cells secrete the chemokine
interleukin-8 in response to bacterial invasion. J Clin Invest. 95:4569 – 4574.
Bradshaw RA, Brickey WW, Walker KW. 1998 N-terminal processing: the
methionine aminopeptidase N and ␣-acetyl transferase families. Trends Biochem Sci. 23:263–267.
Kenney A, Maroux S. 1982 Topology of microvilar membrane hydrolases of
kidney and intestine. Physiol Rev. 62:91–128.
Seli E, Bahtiyar M, Senturk L, Zeyneloglu H, Olive D, Arici A. Aminopeptidase N activity in the human endometrium. J Soc Gynecol Invest. 1998:5.:
110A.