Transepithelial Projections from Basal
Cells Are Luminal Sensors
in Pseudostratified Epithelia
Winnie Wai Chi Shum,1,2Nicolas Da Silva,1,2Mary McKee,1Peter J.S. Smith,3Dennis Brown,1,2and Sylvie Breton1,2,*
1Center for Systems Biology, Program in Membrane Biology/Nephrology Division, Massachusetts General Hospital, Boston, MA 02114, USA
2Harvard Medical School, Boston, MA 02115, USA
3BioCurrents Research Center, Molecular Physiology Program, Marine Biological Laboratory, Woods Hole, MA 02543, USA
side of several epithelia, and they have never been
observed reaching the lumen. Using high-resolution
3D confocal imaging, we report that basal cells ex-
tend long and slender cytoplasmic projections that
not only reach toward the lumen but can cross the
tight junction barrier in some epithelia of the male
reproductive and respiratory tracts. In this way, the
basal cell plasma membrane is exposed to the lumi-
nal environment. In the epididymis, in which luminal
acidification is crucial for sperm maturation and stor-
age, these projections contain the angiotensin II type
2 receptor (AGTR2). Activation of AGTR2 by luminal
clear cells, which are devoid of AGTR2. We propose
a paradigm in which basal cells scan and sense the
luminal environment of pseudostratified epithelia
and modulate epithelial function by a mechanism
involving crosstalk with other epithelial cells.
Epithelial cells have developed complex mechanisms that allow
them to detect both apical and basolateral stimuli, and modulate
their function in response to physiological demands. The differ-
certed manner to coordinate their barrier function. Previous
studies have largely focused on the morphologically dominant
epithelial cells in several tissues, whereas basal cells that are
nestled beneath these epithelial cells have remained mostly
enigmatic. Thesecells were believed to berestricted to the basal
region of pseudostratified epithelia where they may function as
et al., 2002; Lavker et al., 2004; Leung et al., 2007; Rizzo et al.,
2005) and participate in basolateral signaling (Evans et al.,
2001; Hermo and Robaire, 2002; Ihrler et al., 2002; Leung
et al., 2004; van Leenders and Schalken, 2003). We now show
that, in contrast to the prevailing view, basal cells of the upper
respiratory tract and the male reproductive tract (epididymis
and coagulating gland) extend slender cytoplasmic projections
that cross the tight junction (TJ) barrier and reach the epithelial
The epididymal epithelium, which connects the testis to the
vas deferens, forms a tight blood/epididymis barrier and estab-
lishes an optimal luminal environment for the maturation and
storage of spermatozoa (Hermo and Robaire, 2002; Hinton and
Palladino, 1995). Male fertility is partially regulated via the re-
nin-angiotensin system (RAS) located in the tubule lumen (Haga-
man et al., 1998; Krege et al., 1995; Leung and Sernia, 2003).
Both angiotensin II (ANGII) type 1 and type 2 receptors (AGTR1
and AGTR2) are expressed in the epididymal epithelium (Leung
ney collecting duct, which bears a striking functional and cellular
resemblance to the epididymal tubule, ANGII increases proton
secretion in specialized intercalated cells (Pech et al., 2008;
Rothenberger et al., 2007). Similar cells, called clear cells, are
also present in the epididymis where they are responsible for
luminal acidification (Breton et al., 1996; Brown et al., 1992),
which is essential for keeping sperm dormant during maturation
and storage (Hinton and Palladino, 1995; Pastor-Soler et al.,
2005). In this study, we found that basal cells are the only cells
lium where they interact with ANGII. We provide evidence that
they then report their findings to neighboring clear cells, which
nal sampling by so-called basal cells is a novel mechanism for
hormonal signaling that might also be generally applicable to
other pseudostratified epithelia, including the respiratory tract.
Basal Cells Send Long, Slender Cytoplasmic Projections
toward the Lumen
Epididymis sections from rat (16 mm) were labeled for COX1,
a marker of basal cells (Leung et al., 2004). While a dense
network of basal cells is located at the base of the epithelium
confirming previous reports (Clermont and Flannery, 1970; Veri
body extension that infiltrates between other epithelial cells
toward the lumen (Figure 1A, arrows). This was confirmed using
1108 Cell 135, 1108–1117, December 12, 2008 ª2008 Elsevier Inc.
and Movie S1 available with this article online). The probability of
observing these slender structures in thinner sections, which are
more commonly used for staining, and in ultrathin sections used
for electron microscopy is low, probably explaining why they
have not been described extensively in previous publications.
Figure 1C shows an oblique section stained for claudin-1
(Cldn1, green), another marker of basal cells (Gregory et al.,
2001). Numerous projections, positive for Cldn1, are seen
between epithelial cells (arrows). Cldn1 is also present at lower
levels in the lateral membrane of principal cells (Gregory et al.,
study. We did, however, detect basolateral Cldn1 in principal
cells using higher concentrations of antibodies (data not shown).
The panel C inset shows two basal cells double-stained for
COX1 (red) and Cldn1 (green) that extend their narrow body to-
ward the lumen. This result was confirmed by 3D reconstruction
(Figure 1D and Movie S2).
A quantitative analysis was performed to determine the num-
ber of basal cells reaching the apical pole of the epithelium (de-
fined as the region located above the nuclei of adjacent epithelial
the epithelium were also counted (Figures 1F). These numbers
were normalized for the total number of basal cell nuclei
(Figure 1E). Individual epididymis regions and the proximal vas
deferens (VD) were analyzed separately. While very few basal
cells reached the lumen in the proximal regions, the frequency
of events progressively increased toward the distal regions,
reaching a maximum in the distal corpus and proximal cauda.
Figure 1. Basal Cells Send Projections toward the Lumen
(A) Rat corpus epididymidis stained for COX1 (green). Higher magnification is
shown in inset. Arrows indicate basal cells that extend toward the lumen. The
scale bars represent 50 mm and 5 mm (inset).
(B) 3D reconstruction of cauda epididymidis labeled for COX1 (green) showing
two basal cells reaching toward the lumen (arrows). See Movie S1. The scale
bar represents 8 mm.
(C) Oblique section of cauda epididymidis stained for Cldn1 (green). Basal cell
body projections infiltrating between epithelial cells are seen as small dots (ar-
rows). The inset shows two basal cells with intracellular COX1 (red) and mem-
brane-bound Cldn1 (green). The scale bars represent 20 mm and 5 mm (inset).
(D) 3D reconstruction of corpus epididymidis double-stained for COX1 (red)
and Cldn1 (green) showing several basal cell extensions reaching out to the
lumen. See Movie S2. The scale bar represents 8 mm.
(E) Quantification of the total number of basal cells in different regions of the
epididymis and the proximal vas deferens. Data are represented as mean ±
SEM. No significant differences were detected between the regions. pIS,
proximal initial segment; dIS, distal initial segment; Inter-zone, intermediate
zone; pCPT, proximal caput; dCPT, distal caput; pCPS, proximal corpus;
mCPS, middle corpus; dCPS, distal corpus; pCD, proximal cauda; mCD,
middle cauda; dCD, distal cauda; VD, proximal vas deferens.
(F) Open bars: percentage of basal cells detected with their body projection
reaching the apical pole of the epithelium. Data are represented as mean ±
SEM. Number of cells reaching the apical pole/total number of basal cells
are indicated above the bars. Solid bars: percentage of basal cells detected
at the apical border.
(G) Rat trachea stained for COX1 (red) and tubulin (green). Arrow shows
a COX1-positive basal cell that extends toward the lumen (arrow). The cilia
of adjacent ciliated cells are labeled for tubulin. Some unidentified COX1-
positive cells are also detected. The scale bar represents 15 mm.
(H) 3D reconstruction of a trachea section double-stained for COX1 (red) and
ZO1 (green) showing a basal cell reaching the apical border of the epithelium
(arrow). Unidentified COX1-positivecells arealso detected. See MovieS3.The
scale bar represents 5 mm.
(I) Rat coagulating gland stained for COX1 (green). Several basal cells extend
toward the lumen (arrows). The inset shows a COX1-positive basal cell (green)
visualized by DIC (arrows). The scale bars represent 15 mm and 5 mm (inset).
(J) Human epididymis 5 mm section stained for COX1 (green). Numerous basal
cells are seen in the basal region of the epithelium. Some basal cells are also
detected, even on this thinner section, with their body projections reaching the
apical region of the epithelium (arrows).
Lu, lumen; IT, interstitium. In somepanels, nucleiand spermatozoa are stained
in blue with DAPI.
Cell 135, 1108–1117, December 12, 2008 ª2008 Elsevier Inc. 1109
In the distal cauda and in the vas deferens, fewer basal cells
reached the lumen.
Rat trachea sections (16 mm) were labeled for COX1 (red) and
tubulin (green), a marker of airway ciliated cells (Figure 1G). Sim-
ilarly to the epididymis, some COX1-positive basal cells exhibit
a slender projection that extends toward the lumen (arrow).
This was confirmed by 3D reconstruction of sections double-
stained for ZO1 (green) and COX1 (red) (Figure 1H and Movie
S3). While the ZO1-labeled tight junctions (TJs) located at the
corner between three epithelial cells (tricellular corners) are
closed, the TJ located adjacent to the basal cell apical region
is partially open. Basal cells with long cytoplasmic projections
in contact with the epithelial apical border were also detected
in the larynx (data not shown).
Figure 1I shows that the rat coagulating gland, a tissue mor-
phologically and physiologically analogous to the middle lobe
of the human prostate (Price, 1963; Wei et al., 1997), also con-
tains numerous basal cells (stained for COX1 in green; arrows)
that send a narrow body projection toward the lumen. The inset
is a higher magnification differential interference contrast (DIC)
image of a COX1-stained basal cell (green) reaching the luminal
border between adjacent epithelial cells (arrows). A dense net-
work of basal cells was also detected in human epididymis
stained for Cldn1 (Figure 1J). Importantly, some basal cell exten-
sions were detected even on ‘‘thin’’ 5 mm sections, indicating
that these cells have the capacity to reach the lumen in humans
Figure 2. Basal Cells Cross TJs
(A) Three different rotations of a 3D reconstruction of an epi-
didymis section stained for Cldn1 (red) and ZO1 (green). Basal
cells reach the TJs at the intersection between three epithelial
cells (arrows; see Movie S4). The scale bar represents 10 mm.
(B) Conventional microscopy image of one basal cell (stained
for Cldn1 in red) forming a tight junction (stained for ZO1 in
green) with adjacent principal cells (arrow). A clear cell
expressing apical V-ATPase (blue) is seen (arrowhead). The
nuclei are also detected in blue (DAPI). The scale bar repre-
sents 5 mm.
of interaction with TJs. (C) no colocalization between Cldn1
and ZO1 (arrow; Movie S5); (D) partial colocalization of Cldn1
with ZO1 (arrows; Movie S6); (E) basal cell that penetrates the
TJ (arrow; Movie S7); (E) basal cell showing ZO1-stained TJ
(green) with adjacent principal cells (arrows; Movie S8).
(G) Rotations of a 3D reconstruction of epididymis stained for
Cldn1 (green) and F-actin (red). A Cldn1-positive basal cell
reaches the luminal side (arrow) between F-actin-labeled prin-
cipal cells (see also Movie S9). The scale bar represents 5 mm.
(H) Enface view visualized by DIC and immunofluorescence
Cldn1 labeling (green). The dotted lines in H’’ indicate the
junctions between epithelial cells. Arrows show the tricellular
corners between epithelial cells. One corner is occupied by
a Cldn1-positive basal cell.
Basal Cells Cross TJs to Reach the Lumen
While basal cells were shown to extend processes
between epithelial cells, they were never described
to have direct access to the lumen (Evans et al.,
2001; Hermo and Robaire, 2002; Robaire and
Viger, 1995; van Leenders and Schalken, 2003; Veri et al.,
1993). To determine whether basal cells can cross the TJ barrier,
double labeling for Cldn1 and ZO1 was performed on rat epidid-
ymis. Basal cells preferentially reach TJs at the tripartite junction
between other epithelial cells (Figure 2A, arrows; Movie S4).
Various patterns of interaction between basal cells and TJs
were seen. Figure 2B shows a basal cell (arrow) that has crossed
the TJ barrier and established contact (labeled with ZO1) with
adjacent principal cells. The arrowhead indicates a clear cell
with apical V-ATPase labeling (blue). Figures 2C–2F are 3D re-
constructions of basal cells showing various patterns of interac-
tions with the TJs. Figure 2C shows one cell underneath the TJ
(arrow) but showing no colocalization between Cldn1 and ZO1
(see Movie S5). The yellow staining in Figure 2D (arrows) indi-
cates partial colocalization between the Cldn1-positive basal
cell and the ZO1-labeled TJ. The tripartite junction adjacent to
this basal cell is partially open (see Movie S6). Figure 2E shows
one cell that crosses the TJ barrier (arrow; see also Movie S7).
Figure 2F shows one cell that has penetrated the epithelium
beyond the TJ barrier and has established contact with adjacent
principal cells (arrows and Movie S8; this cell is also shown
in Figure 2B). In the trachea, similar patterns of interaction
between basal cells and TJs were also seen (see Figure 1H
and Movie S3).
The contact of basal cells with the epididymal lumen was fur-
ther demonstrated with a marker of principal cell apical stereoci-
lia (F-actin). 3D reconstruction clearly showed that basal cells,
1110 Cell 135, 1108–1117, December 12, 2008 ª2008 Elsevier Inc.
stained for Cldn1 (green) but negative for F-actin (arrow), reach
for F-actin (Figure 2G; see also Movie S9). The luminal contact of
this basal cell is apparent only on panels showing rotations
around the x axis (Figures 2G’’ and 2G’’’), and is not visible on
the XY image shown in Figure 2G’. This is due to the presence
of long stereocilia in adjacent principal cells, which mask the
small apical pole of the basal cell. The apical surface of a cut-
open epididymal tubule was visualized by DIC coupled to
Cldn1 labeling (green) (Figure 2H). While most tripartite cell junc-
tions are closed (arrows), one corner is occupied by a Cldn1-
Functional Role of Epididymal Basal Cells
We then examined the potential role of the basal cell extensions.
Can these structures scan the lumen for the presence of biolog-
ical factors? To test this hypothesis, we examined the expres-
sion of hormone receptors in these cells. Because luminally
located RAS is an important contributor to male fertility, we
examined the expression of ANGII receptors in the epididymal
Basal Cells Express AGTR2
Double-labeling for the proton-pumping V-ATPase (red), located
in the apical pole of clear cells (Figure 3A; arrowheads), and
AGTR2 (green) showed that AGTR2 is exclusively expressed in
basal cells (arrows). This is in agreement with previous studies
showing AGTR2 in the basal region of the epithelium, although
the cell type expressing AGTR2 was not identified (Leung
et al., 1997). AGTR2 staining was abolished when the antibody
was preabsorbed with the immunizing peptide (10-fold excess;
Figure 3B). Western blots of rat epididymis showed two bands,
one at about 44 kDa, the expected molecular weight of
AGTR2, and a second at about 88 kDa (Figure 3C; arrows).
Both bands were abolished by preincubation of antibody with
its peptide (data not shown). The higher molecular weight band
is twice the molecular weight of AGTR2, indicating potential
dimerization of AGTR2, as reported for other G protein coupled
receptors (Bulenger et al., 2005; Parnot and Kobilka, 2004; Skra-
banek et al., 2007). 3D reconstruction (Figure 3D) confirmed that
AGTR2 is expressed in basal cells (arrows and Movie S10). Two
clear cells stained at their apical pole for the V-ATPase (arrow-
heads), but negative for AGTR2, are located close to basal cells.
The absence of AGTR2 from clear cells was further confirmed by
RT-PCR using B1-EGFP transgenic mice in which enhanced
GFP is expressed only in clear cells (Miller et al., 2005). Clear
cells isolated by fluorescence activated cell sorting (FACS)
were compared to GFP-negative cells, i.e., all other cell types
in the epididymis. Whereas a positive signal was obtained using
primer sets spanning the coding region of agtr2 in nonclear cells
(Figure 3E: GFP-) no signal was detected in clear cells (GFP+).
The identity of the PCR product was confirmed by direct
sequencing (data not shown).
Basal Cells Sense Luminal ANGII and Regulate Clear
Cells via AGTR2
In the kidney, ANGII stimulates proton secretion by intercalated
cells (Pech et al., 2008; Rothenberger et al., 2007), which resem-
ble epididymal clear cells (Breton and Brown, 2007). The expres-
sionof AGTR2exclusivelyin basalcellsraisedthe possibilitythat
these cells might regulate proton secretion by clear cells, follow-
ing sampling of luminal ANGII. We previously showed that
V-ATPase apical membrane accumulation and extension of
microvilli in clear cells correlate with proton secretion (Beaulieu
Figure 3. Expression of AGTR2 in Basal Cells
(A) Three examples of AGTR2 (green) and V-ATPase (red) labeling in cauda ep-
ididymidis. Arrows indicate AGTR2-labeled basal cells that send projections
toward the lumen. Arrowheads show nearby V-ATPase-labeled clear cells.
Nuclei are visualized with DAPI (blue). The scale bar represents 5 mm.
(B) Epididymis stained using anti-AGTR2 antibody with (+ peptide) and without
(AGTR2) preincubation with the immunizing peptide. The scale bar represents
(C) Western blot detection of AGTR2.180 mg of epididymalhomogenates were
loaded onto the gel. Two bands at around 44 and 88 kDa were detected
(D) 3D reconstruction showing AGTR2-positive basal cells (green; arrows).
One basal cell sends a projection between principal cells. Two clear cells,
stained apically for the V-ATPase (red), are visible (arrowheads). See also
Movie S10. The scale bar represents 5 mm.
(E)RT-PCR analysisofAgtr2mRNA expressioninclearcells, isolated byFACS
from B1-EGFP mouse epididymides (GFP+), and in all other epididymal cell
types (GFP-). While a positive signal was detected in the GFP negative cell
population, no Agtr2 mRNA expression was observed in GFP-positive clear
Lu, lumen; SMC, smooth muscle cells.
Cell 135, 1108–1117, December 12, 2008 ª2008 Elsevier Inc. 1111
et al., 2005; Pastor-Soler et al., 2003). Here, we examined the
effect of ANGII on the extension of V-ATPase-labeled microvilli.
Rat cauda epididymides were perfused luminally in vivo with
phosphate-buffered saline (pH 6.6). Under these control condi-
tions, clear cell V-ATPases are distributed between short micro-
villiand subapical vesicles(Figure 4A). Addition ofANGII (0.1and
1 mM) to the luminal perfusate significantly increased the exten-
sion of V-ATPase-labeled microvilli to 141 ± 4% and 153 ± 7%
versus control, respectively (Figures 4B–4D). Immunogold elec-
tron microscopy confirmed this accumulation of V-ATPase in
apical microvilli (Figures 4E–4G). ANGII induced a significant
increase in the density of V-ATPase molecules in microvilli
(G: Gold/mm apical membrane). Because numerous and longer
microvilli were observedin clear cellsexposed to ANGII,the total
number of V-ATPase molecules located at the cell surface was
further amplified compared to control (G: Gold/mm cell width).
The effect of ANGII on proton secretion was examined in cut-
open proximal VD (Figure 4H), a tissue that also contains clear
and basal cells, using an extracellular proton-selective micro-
electrode (Beaulieu et al., 2005; Breton et al., 1996). After a con-
trol period during which stable proton secretion was recorded,
ANGII was added to the bath. After a rapid and transient rise
due to disturbance of the proton gradient, proton secretion
showed a sustained increase. Addition of the V-ATPase inhibitor
concanamycin A markedly inhibited proton secretion. For each
VD, both the control value (prior to addition of ANGII) and the
ANGII value (30 min after its addition) were corrected for the
value measured after addition of concanamycin A. On average,
ANGII caused a significant increase of concanamycin-sensitive
proton secretion of 68% compared to control (Figure 4I). Prein-
cubation of the tissue with concanamycinA for 10 minprevented
the response to ANGII (data not shown).
Losartan, an AGTR1 antagonist, had no inhibitory effect on
ANGII-induced V-ATPase apical accumulation (Figures 5C and
5D). However the AGTR2 antagonist, PD123319, prevented the
stimulatory effect of ANGII on clear cells (Figures 5B and 5D).
These results are consistent with participation of AGTR2 in the
regulation of clear cell-dependent luminal acidification. Nitric
oxide (NO) is the downstream effector of AGTR2 activation
(Carey, 2005; Toda et al., 2007). p-cpt-cGMP, a cell-permeable
analog of cGMP, or sodium nitroprusside (SNP), a NO-donor,
induced a significant elongation of V-ATPase-rich microvilli,
compared to control (Figures 6A and 6C). Pretreatment with
the soluble guanylate cyclase (sGC) inhibitor ODQ, or the NO
ANGII-induced V-ATPase apical accumulation (Figures 6B and
6C). Immunofluorescence labeling showed a strong staining for
the b1subunit of sGC (b1-sGC) in the basolateral membrane
and apical region of clear cells (Figure 6D: green), identified by
apical staining for the V-ATPase (red; arrows). Specificity of the
antibody was confirmed by Western blot and immunofluores-
cence using antibody that had been preabsorbed with b1-sGC
peptide (Figures 6E and 6F).
Figure 4. Luminal ANGII Induces V-ATPase Apical Accumulation in
(A–C) Confocal microscopy images of V-ATPase-labeled clear cells (green)
luminally perfused in vivo under control conditions (A) or in the presence of
0.1 mM (B) or 1 mM ANGII (C) for 20 min. The arrows show the border between
the base of the apical microvilli and the apical pole of the cell. The scale bar
represents 5 mm.
(D) Quantitative analysis of the dose-dependent effect of ANGII on the elonga-
tion of V-ATPase-labeled microvilli normalized for the apical width of the cell.
Valuesaremean ±SEMobtained fromatleast10cellsper epididymis from‘‘n’’
number of epididymis per group. **p < 0.001 versus control.
(E and F) V-ATPase immunogold labeling of the apical pole of a clear cell per-
and a few short V-ATPase-labeled microvilli are detected. In the presence of
ANGII, longer and more numerous V-ATPase-labeled microvilli are detected.
The scale bar represents 500 nm.
(G) Apical accumulation of V-ATPase by luminal ANGII (1 mM) in clear cells.
The left axis shows the density of V-ATPase-associated gold particles in
the apical membrane including microvilli (Gold/mm apical membrane). The
right axis shows the total number of gold particles located in the apical
membrane of clear cells normalized for the width of the cell (Gold/mm cell
width). 28 cells were analyzed in each group. Data are expressed as means ±
SEM *p < 0.0005.
(H) Effect of ANGII (1 mM) on proton secretion in cut-open proximal VD using
a proton-selective electrode. After an initial spike due to disturbance of the
proton gradient, a sustained increase in proton secretion (expressed as mV)
was induced by ANGII. A marked inhibition was then observed following addi-
tion of concanamycin A (1 mM).
(I) Mean effect of ANGII (1 mM) on concanamycin-sensitive proton secretion
(mean ± SEM, n = 7) measured 30 min after addition of ANGII. *p < 0.05.
1112 Cell 135, 1108–1117, December 12, 2008 ª2008 Elsevier Inc.
The present study provides evidence that narrow projections
nal side of an epithelium. This previously unrecognized property
of basal cells was observed in several tissues located in the male
reproductive tract and upper respiratory tract, indicating that it is
a widespread phenomenon that could have general significance
to the biology of pseudostratified epithelia.
so-called basal cells, which appear to nestle beneath columnar
epithelial cells, and tall basal cells, which extend processes
between other epithelial cells (Evans et al., 2001). Further studies
will be required to determine whether these two morphological
features are associated with different degrees of body extension
in the same population of basal cells, or whether they truly repre-
sent two distinct cell populations. Nevertheless, the property of
substances that constantly invade the upper respiratory tract.
Two distinct tissues of the male reproductive tract have basal
cells that contact the luminal milieu: the coagulating gland and
the epididymis. Interestingly, the coagulating gland in rodents
is analogous to the middle lobe of the human prostate (Price,
1963; Wei et al., 1997). The notion that basal cells can reach
the prostate lumen will have significant repercussions for our
understanding of the (patho)physiology of this important organ.
Numerous basal cells reaching toward the lumen were also ob-
served in the rat and human epididymis, indicating that luminal
epididymal sampling by basal cells occurs across species.
Basal Cells Cross the TJ Barrier
In the epididymis, we showed that basal cells can cross the
blood/epididymis barrier while preserving its integrity. They do
so by establishing a new TJ between themselves and adjacent
epithelial cells. While sending their body projections toward the
lumen, basal cells often seem to ‘‘stop short’’ just beneath the
TJs of the epithelium. While virtually no such cells were detected
in the proximal regions, the number of basal cells reaching the
luminal border dramatically increased in the distal corpus and
proximal cauda. This indicates that the capacity of basal cells
to reach the lumen is a property that is locally regulated in differ-
ent regions of the epididymis. A fraction of the total number of
basal cells could be seen crossing the TJs at one given time in
still images, indicating a potential dynamic interplay between
these cells and the epithelium. Thus, basal cells may ‘‘come and
go’’ to and from the lumen, and the establishment of a new TJ
between basal cells and epithelial cells, in addition to being
dynamic, may be temporary. The time required for a leukocyte
to cross the TJ barrier of endothelia is less than 2 min (Stein
etal.,1997),and it isconceivable thatbasalcellsmight modulate
the epididymal barrier in a shorter time frame. The signal respon-
sible for inducing basal cells to interact with and cross the
TJ barrier remains unclear.
Interestingly, basal cells always reached TJs at the regions
also described for neutrophils crossing endothelial barriers
(Burns et al., 2000). Most remarkably, basal cells can actually
open up and cross these TJs. The continuous ZO1 labeling in
the region of contact between these cells and adjacent epithelial
cells suggests that new TJs had been established. High expres-
sion of Cldn1 in epididymal basal cells (this study and (Gregory
et al., 2001)), as well as in principal cells, at lower levels (Gregory
et al., 2001) might provide a molecular ‘‘grip’’ by which basal
cells extend projections toward the lumen. Cldn1 forms pairs
not only with itself, but also with other claudins, including
Cldn3 and Cldn4 (Schneeberger and Lynch, 2004), which are ex-
pressed in epididymal TJs (Gregory and Cyr, 2006). This might
contribute to the formation of a new TJ between the penetrating
basal cell and adjacent epithelial cells. TJ strands constantly
form and reform, without disturbing their barrier function
(Schneeberger and Lynch, 2004). This remodeling allows migra-
tion of leukocytes across endothelia (Burns et al., 2000), as well
as penetration of dendritic cells, which also express Cldn1,
across the intestinal epithelium (Niess et al., 2005; Rescigno
et al., 2001) and the upper respiratory tract (Takano et al., 2005).
Basal Cells Are Luminal Hormone Sensors
The present study also shows that activation of AGTR2 by lumi-
nal ANGII stimulates proton secretion by epididymal clear cells
Figure 5. AGTR2 Mediates ANGII-Induced V-ATPase Apical Accu-
mulation and Microvilli Elongation in Clear Cells
(A–C) Confocal images showing clear cells perfused for 20 min with 1 mM
ANGII (A), or preincubated for 10 min with PD123319 (1 mM; [B]) or losartan
(1 mM; [C]), before addition of ANGII, still in the presence of antagonists.
PD123319, but not losartan, prevented the ANGII-induced microvilli elonga-
tion. Arrows show the frontier between the base of apical microvilli and the
cytoplasm of the cell.
(D) Mean effects of PD123319 and losartan on ANGII-mediated microvilli elon-
gation. PD123319 inhibited the effect of ANGII at both 0.1 and 1 mM concen-
trations. Values were obtained from at least 10 cells per epididymis. Data
ined. *p < 0.001 versus control; ns: no significant difference versus control.
Cell 135, 1108–1117, December 12, 2008 ª2008 Elsevier Inc. 1113
via activation of the NO/cGMP pathway. NO is a downstream
effector of AGTR2 and because basal cells are the only cell
type in which this receptor is expressed, they are the likely site
for ANGII-induced NO production. However, determination of
the exact cellular origin of NO following AGTR2 activation will
require novel tools and animal models for the measurement of
NO in single basal cells. A schematic view of our current cell-
cell crosstalk model for activation of proton secretion in clear
cells following AGTR2 stimulation in basal cells is illustrated in
Figure 7. Consistent with this model, endothelial NO synthases
(eNOS) have been detected in basal-like cells in human and bo-
vine epididymis (Mewe et al., 2006; Zini et al., 1996). Sampling of
luminal ANGII by basal cells followed by activation of proton
secretion by clear cells would ensure that the luminal fluid is
maintained at its physiological acidic pH. Crosstalk between
basal and principal cells has also been proposed to modulate
Figure 6. The NO-sGC-cGMP Pathway Mediates
ANGII-Induced V-ATPase Apical Accumulation and
Microvilli Elongation in Clear Cells
(A) Confocal images of V-ATPase-labeled clear cells (green)
perfused under control conditions (control), or in the presence
of 1 mM p-cpt-cGMP, or 1 mM SNP for 20 min. A marked
elongation of V-ATPase-labeled microvilli is observed in the
presence of p-cpt-cGMP and SNP, compared to control.
The arrows indicate the border between the base of apical
microvilli and the cytoplasm. The scale bars represent 5 mm.
(B) Effect of ODQ or L-NAME on the ANGII-induced response.
Pretreatment for 10 min with ODQ (3 mM) or L-NAME (100 mM),
followed by ANGII still in the presence of inhibitors for 20 min,
abolished the V-ATPase apical accumulation and microvilli
elongation induced by ANGII alone. The scales bar represent
(C) Mean microvilli elongation in clear cells. Values were
obtained from at least 10 cells per epididymis. Data are repre-
sented as mean ± SEM, and ‘‘n’’ is the number of epididymi-
des examined. *p < 0.01 versus control (CTL), **p < 0.001
versus CTL, and #p < 0.001 versus ANGII.
(D) Localization of b1-sGC (green) and V-ATPase (red) in
epididymis. V-ATPase-positive clear cells (arrows) show
abundant b1-sGC staining in their basolateral membrane and
apical pole. Weaker and more uniform staining is also de-
tected in principal, basal, and smooth muscle cells. Nuclei
are visualized with DAPI (blue) in the merged panel. The scale
bar represents 10 mm.
(E) Western blot detection of b1-sGC in rat (RE) and mouse ep-
ididymis (ME) (120 mg/lane). In both samples, a band at about
70 kDa was detected corresponding to the molecular weight
of b1-sGC. Additional bands at around 35 kDa and 50 kDa
were also detected in rat and mouse epididymis, respectively,
possibly indicating degradation products in these tissues. All
bands were absent after preincubation of the antibody with
the immunizing peptide.
(F) Inhibition of immunofluorescence staining using the anti-
body preincubated with the immunizing peptide (+ peptide).
The scale bars represent 50 mm.
anion secretion by principal cells in response to
basolateral lysylbradykinin (Cheung et al., 2005;
Leung et al., 2004).
Basolateral stimulation of AGTR1 by ANGII acti-
vates anion secretion in cultured principal cells
(Leung et al., 1997; Leung and Sernia, 2003), and we now
mentwithour study,recentreportsshowed thatANGIIincreases
V-ATPase-dependent proton extrusion by renal intercalated
cells (Pech et al., 2008; Rothenberger et al., 2007), which are
analogous to clear cells (Breton and Brown, 2007). We have
previously shown that cAMP elevation following activation of
the bicarbonate sensitive soluble adenylyl cyclase (sAC) and
PKA, induced apical accumulation of the V-ATPase into well-
developed microvilli in clear cells (Pastor-Soler et al., 2003,
2008). The present study shows that cGMP can also activate
V-ATPase-dependent luminal acidification; the mechanism by
which this occurs is the subject of ongoing work.
Spermatozoa require an acidic environment to prevent their
premature activation during maturation and storage in the epi-
didymis (Hinton and Palladino, 1995; Jones and Murdoch,
1114 Cell 135, 1108–1117, December 12, 2008 ª2008 Elsevier Inc.
1996; Pastor-Soler et al., 2005). However, the mechanisms by
which spermatozoa interact with epithelial cells of the epididy-
mal tubule remain, for the most part, unknown. In ACE KO
mice, absence of the germinal form of ACE (gACE) induces
a marked reduction in the quality of sperm, which are unable
to fertilize an egg (Esther et al., 1996; Hagaman et al., 1998;
Krege et al., 1995). gACE, which is linked to the sperm mem-
mature in the epididymis (Gatti et al., 1999), providing a potential
impair the acidifying capacity of the epididymis with detrimental
consequences on sperm quality. Indeed, FOXI-1 KO male mice,
which have impaired luminal acidification, are also infertile due
to sperm inability to fertilize an egg (Blomqvist et al., 2006).
Thus, the concerted interaction between sperm, basal cells and
clear cells might represent a complex process by which the
luminal environment for the maturation and storage of sperm.
While we have focused on the regulation of luminal acidifica-
tion in the male reproductive tract, we propose that sensing
and signaling by transepithelial basal cells represents a novel
mechanism of cellular crosstalk and functional regulation in
Here, we show that basal cells project slender body extensions
that reach the luminal border of pseudostratified epithelia. In
the male reproductive tract, basal cells cross the blood/epididy-
mis barrier to monitor luminal factors. We also provide evidence
for the presence of a novel crosstalk between basal cells and
clear cells to control V-ATPase-dependent proton secretion,
a process that is crucial for maintaining sperm quiescent during
their maturation and storage in theepididymis. Luminal sampling
by basal cells has not been recognized previously and will pro-
Figure 7. Schematic Representation of Cell-to-Cell
Crosstalk in the Epididymal Epithelium
Basal cells extend narrow projections between epithelial cells
to reach the lumen. A new TJ is formed between the basal cell
and adjacent epithelial cells. Basal cells express AGTR2 and
luminal ANGII triggers the production of NO in these cells.
NO then acts locally on clear cells to produce cGMP via
activation of the sGC, which is enriched in these cells.
cGMP induces the accumulation of V-ATPase in well devel-
oped apical microvilli in clear cells, which results in the
increase of proton secretion.
vide a new framework for future studies aimed
at unraveling the cellular mechanisms by which
epithelia respond to luminal stimuli.
Tissue Fixation and Preparation
Adult male Sprague Dawleyrats (Charles RiverLabs, Wilming-
ton, MA)wereanesthetizedwithnembutal(60mg/kg, i.p.). The
male reproductive and upper respiratory tracts were fixed by
perfusion through the left ventricle with paraformaldehyde-lysine-periodate
(PLP) fixative, as described previously (Pastor-Soler et al., 2003). All proce-
dures were approved by the Massachusetts General Hospital Institutional
Committee on Research Animal Use.
In Vivo Microperfusion
Rats were anesthetized and the cauda epididymidis was luminally perfused
in vivo, harvested and fixed in PLP, as described previously (Beaulieu et al.,
2005; Pastor-Soler et al., 2003; Wong and Yeung, 1978).
Immunofluorescence labeling was performed on cryostat sections, as de-
scribed previously (Beaulieu et al., 2005; Pastor-Soler et al., 2003). Primary
antibodies, the AGTR2 peptide and secondary antibodies used are listed in
supplementary material. Slides were mounted in Vectashield (Vector Labs,
Burlingame, CA) with or without DAPI. For confocal microscopy, nuclei were
stained using TOPRO-3 iodide (Invitrogen, Carlsbad, CA). Immunostained
sections were examined using a Nikon E800 microscope (Nikon Instruments,
Melville, NY). Digital images were acquired with IPLab Spectrum software
(Scanalytics, Fairfax, VA) and imported into Adobe Photoshop. Sections
were also examined using a Zeiss Radiance 2000 confocal microscope (Zeiss
Laboratories). Z-series (0.1 mm interval) were imported into Volocity software
(Improvision Inc., version 4.1) for 3D reconstruction and final animations
were exported as Quicktime movies.
Quantification of V-ATPase Apical Membrane Accumulation
in Clear Cells
The level of accumulation of V-ATPase in clear cell microvilli was quantified
using IPLab software as described previously (Beaulieu et al., 2005; Pastor-
Soler et al., 2003). 10 mm sections of microperfused cauda epididymidis
were immunostained under identical conditions, and confocal images were
acquired using the same parameters. The segmentation procedure of IPLab
was used to measure the area of V-ATPase-positive microvilli, which was
normalized against the length of apical pole of each cell (Beaulieu et al.,
2005; Pastor-Soler et al., 2003). At least three epididymides from different
animals were perfused for each condition, and a minimum of 10 cells/tissue
were examined for a total of at least 30 cells/condition.
Immunogold Electron Microscopy and Quantification
of Gold Labeling
Pieces of PLP-fixed epididymis were embedded at ?45?C using HM20 resin
(Electron Microscopy Sciences, Hatfield, PA) in a Leica EM AFS, and ultrathin
Cell 135, 1108–1117, December 12, 2008 ª2008 Elsevier Inc. 1115
sections were cut, as described previously (Da Silva et al., 2007; Pastor-Soler
et al., 2003). Sections were immunostained for the V-ATPase A subunit, fol-
lowed by goat anti-rabbit IgG coupled to 15 nm gold (Ted Pella, Reading,
CA). Grids were examined in a JEOL 1011 electron microscope. Images
were acquired using an AMT digital imaging system.
The number of V-ATPase associated gold particles on the apical membrane
and microvilli was counted for each clear cell (Da Silva et al., 2007; Pastor-
Soler et al., 2003). To determine the density of V-ATPase molecules along
the apical membrane, the number of gold particles was divided by the length
of apical membrane, including microvilli, of each cell. This value is referred to
as ‘‘gold/mm apical membrane.’’ To determine the relative density of the
V-ATPase at the cell surface, the number of gold particles was normalized for
the width of the cell, measured at the baseof the microvilli (gold/mm cell width).
Protein extracts from rat and mouse epididymis were subjected to electropho-
resis and western blotting, as described previously (Beaulieu et al., 2005;
Pastor-Soler et al., 2003).
Isolation of Clear Cells, RNA Extraction, and RT-PCR
with trypsin and collagenase. Fluorescence-activated cell sorting (FACS) was
used to separate clear cells (GFP-positive) from other cell types (GFP-nega-
tive). Total RNA was isolated using the PicoPure RNA Isolation kit (Molecular
Devices, Sunnyvale, CA),and RT-PCR wasperformed as described previously
(Isnard-Bagnis et al., 2003). The primers amplifying a 674 bp fragment of the
mouse Agtr2 coding sequence are: attggctttttggacctgtg (MAGTR2-F3) and
aaacacactgcggagcttct (MAGTR2-R2). The PCR product was purified with
the Qiaquick PCR kit and sequenced by the MGH sequencing core.
Detection of Proton Secretion
The proximal VD was cut open to expose the apical surface of the epithelium
and anchored onto a custom-made chamber. Proton secretion was measured
using a self-referencing proton-selective electrode, as described previously
(Beaulieu et al., 2005; Breton et al., 1996; Smith et al., 2007).
paired Student’s t test when appropriate. Comparisons between multigroups
were determined by one-way ANOVA with Bonferroni’s post hoc test. All tests
were two-tailed and the limit of statistical significance was set at p = 0.05.
Supplemental Data include Supplemental Experimental Procedures and
ten movies and can be found with this article online at http://www.cell.com/
DK38452 (S.B. and D.B.), DK42956 (D.B.), and NCRR P41 RR001395 grant
(P.J.S. Smith). TheMicroscopy Core facility of theMGH Program inMembrane
Biology receives support from the Boston Area Diabetes and Endocrinology
Research Center (DK57521) and the Center for the Study of Inflammatory
Bowel Disease (DK43341).
Received: March 13, 2008
Revised: July 23, 2008
Accepted: October 8, 2008
Published: December 11, 2008
Beaulieu, V., Da Silva, N., Pastor-Soler, N., Brown, C.R., Smith, P.J., Brown,
D., and Breton, S. (2005). Modulation of the actin cytoskeleton via gelsolin
regulates vacuolar H+-ATPase recycling. J. Biol. Chem. 280, 8452–8463.
Blomqvist,S.R.,Vidarsson, H., Soder,O.,and Enerback, S.(2006). Epididymal
expression of the forkhead transcription factor Foxi1 is required for male fertil-
ity. EMBO J. 25, 4131–4141.
dependent proton secretion. Am. J. Physiol. Renal Physiol. 292, F1–F10.
Breton, S., Smith, P.J., Lui, B., and Brown, D. (1996). Acidification of the male
reproductive tract by a proton pumping (H+)-ATPase. Nat. Med. 2, 470–472.
ATPase in specialized cells of rat epididymis. Am. J. Physiol. 263, C913–C916.
Bulenger, S., Marullo, S., and Bouvier, M. (2005). Emerging role of homo- and
heterodimerization in G-protein-coupled receptor biosynthesis and matura-
tion. Trends Pharmacol. Sci. 26, 131–137.
Burns, A.R., Bowden, R.A., MacDonell, S.D., Walker, D.C., Odebunmi, T.O.,
Donnachie, E.M., Simon, S.I., Entman, M.L., and Smith, C.W. (2000). Analysis
of tight junctions during neutrophil transendothelial migration. J. Cell Sci. 113,
Carey, R.M. (2005). Update on the role of the AT2 receptor. Curr. Opin. Neph-
rol. Hypertens. 14, 67–71.
Cheung, K.H., Leung, G.P., Leung, M.C., Shum, W.W., Zhou, W.L., and Wong,
P.Y. (2005). Cell-cell interaction underlies formation of fluid in the male repro-
ductive tract of the rat. J. Gen. Physiol. 125, 443–454.
Clermont, Y., and Flannery, J. (1970). Mitotic activity in the epithelium of the
epididymis in young and old adult rats. Biol. Reprod. 3, 283–292.
Da Silva, N., Shum, W.W., El-Annan, J., Paunescu, T.G., McKee, M., Smith,
P.J., Brown, D., and Breton, S. (2007). Relocalization of the V-ATPase B2 sub-
subunit. Am. J. Physiol. Cell Physiol. 293, C199–C210.
Esther, C.R., Jr., Howard, T.E., Marino, E.M., Goddard, J.M., Capecchi, M.R.,
and Bernstein, K.E. (1996). Mice lacking angiotensin-converting enzyme have
low blood pressure, renal pathology, and reduced male fertility. Lab. Invest.
Evans, M.J., Van Winkle, L.S., Fanucchi, M.V., and Plopper, C.G. (2001).
Cellular and molecular characteristics of basal cells in airway epithelium.
Exp. Lung Res. 27, 401–415.
Ford, J.R., and Terzaghi-Howe, M. (1992). Basal cells are the progenitors of
primary tracheal epithelial cell cultures. Exp. Cell Res. 198, 69–77.
Gatti, J.L., Druart, X., Guerin, Y.,Dacheux, F.,and Dacheux, J.L. (1999). A 105-
tensin I-converting enzyme (ACE); evidence that sperm are the source of this
ACE. Biol. Reprod. 60, 937–945.
Gregory, M., and Cyr, D.G. (2006). Identification of multiple claudins in the rat
epididymis. Mol. Reprod. Dev. 73, 580–588.
Gregory, M., Dufresne, J., Hermo, L., and Cyr, D. (2001). Claudin-1 is not
restricted to tight junctions in the rat epididymis. Endocrinology 142, 854–863.
Hagaman, J.R., Moyer, J.S., Bachman, E.S., Sibony, M., Magyar, P.L., Welch,
J.E., Smithies, O., Krege, J.H., and O’Brien, D.A. (1998). Angiotensin-convert-
ing enzyme and male fertility. Proc. Natl. Acad. Sci. USA 95, 2552–2557.
Hajj, R., Baranek, T., Le Naour, R., Lesimple, P., Puchelle, E., and Coraux, C.
(2007). Basal cells of the human adult airway surface epithelium retain transit-
amplifying cell properties. Stem Cells 25, 139–148.
Hermo, L., and Robaire, B. (2002). Epididymal cell types and their functions. In
The Epididymis: From Molecules to Clinical Practice, B. Robaire and B.T.
Hinton, eds. (New York: Kluwer Acedemic/Plenum Publishers), pp. 81–102.
Hinton, B.T., and Palladino, M.A. (1995). Epididymal epithelium: its contribu-
tion to the formation of a luminal fluid microenvironment. Microsc. Res.
Tech. 30, 67–81.
Ihrler, S., Zietz, C., Sendelhofert, A., Lang, S., Blasenbreu-Vogt, S., and Lohrs,
U. (2002). A morphogenetic concept of salivary duct regeneration and meta-
plasia. Virchows Arch. 440, 519–526.
Isnard-Bagnis, C., Da Silva, N., Beaulieu, V., Yu, A.S., Brown, D., and Breton,
S. (2003). Detection of ClC-3 and ClC-5 in epididymal epithelium:
1116 Cell 135, 1108–1117, December 12, 2008 ª2008 Elsevier Inc.
immunofluorescenceandRT-PCRafterLCM.Am.J.Physiol.CellPhysiol.284, Download full-text
Jones, R.C., and Murdoch, R.N. (1996). Regulation of the motility and metab-
olism of spermatozoa for storage in the epididymis of eutherian and marsupial
mammals. Reprod. Fertil. Dev. 8, 553–568.
Kondoh, G., Tojo, H., Nakatani, Y., Komazawa, N., Murata, C., Yamagata, K.,
tensin-converting enzyme is a GPI-anchored protein releasing factor crucial
for fertilization. Nat. Med. 11, 160–166.
Krege, J.H., John, S.W., Langenbach, L.L., Hodgin, J.B., Hagaman, J.R.,
male differences in fertility and blood pressure in ACE-deficient mice. Nature
Lavker, R.M., Tseng, S.C., and Sun,T.T. (2004). Cornealepithelialstem cells at
the limbus: looking at some old problems from a new angle. Exp. Eye Res. 78,
Leung, C.T., Coulombe, P.A., and Reed, R.R. (2007). Contribution of olfactory
neural stem cells to tissue maintenance and regeneration. Nat. Neurosci. 10,
Leung, G.P., Cheung, K.H., Leung, C.T., Tsang, M.W., and Wong, P.Y. (2004).
Regulation ofepididymalprincipalcellfunctions bybasal cells:roleoftransient
receptor potential (Trp) proteins and cyclooxygenase-1 (COX-1). Mol. Cell. En-
docrinol. 216, 5–13.
Leung, P.S., Chan, H.C., Fu, L.X., Zhou, W.L., and Wong, P.Y. (1997). Angio-
tensin II receptors, AT1 and AT2 in the rat epididymis. Immunocytochemical
and electrophysiological studies. Biochim. Biophys. Acta 1357, 65–72.
Leung, P.S., and Sernia, C. (2003). The renin-angiotensin system and male re-
production: new functions for old hormones. J. Mol. Endocrinol. 30, 263–270.
Mewe, M., Bauer, C.K., Muller, D., and Middendorff, R. (2006). Regulation of
spontaneous contractile activity in the bovine epididymal duct by cyclic
guanosine 50-monophosphate-dependent pathways. Endocrinology 147,
Miller, R.L., Zhang, P., Smith, M., Beaulieu, V., Paunescu, T.G., Brown, D.,
Breton, S., and Nelson, R.D. (2005). V-ATPase B1-subunit promoter drives ex-
pression of EGFP in intercalated cells of kidney, clear cells of epididymis and
airway cells of lung in transgenic mice. Am. J. Physiol. Cell Physiol. 288,
Niess, J.H., Brand, S., Gu, X., Landsman, L., Jung, S., McCormick, B.A., Vyas,
J.M., Boes, M., Ploegh, H.L., Fox, J.G., et al. (2005). CX3CR1-mediated den-
dritic cell access to the intestinal lumen and bacterial clearance. Science 307,
Parnot, C., and Kobilka, B. (2004). Toward understanding GPCR dimers. Nat.
Struct. Mol. Biol. 11, 691–692.
Pastor-Soler, N., Beaulieu, V., Litvin, T.N., Da Silva, N., Chen, Y., Brown, D.,
Buck, J., Levin, L.R., and Breton, S. (2003). Bicarbonate-regulated adenylyl
cyclase (sAC) is a sensor that regulates pH-dependent V-ATPase recycling.
J. Biol. Chem. 278, 49523–49529.
Pastor-Soler, N., Hallows, K.R., Smolak, C., Gong, F., Brown, D., and Breton,
is mediated by protein kinase A in epididymal clear cells. Am. J. Physiol. Cell
Physiol. 294, C488–C494.
Pastor-Soler, N., Pietrement, C., and Breton, S. (2005). Role of acid/base
malfunction. Physiology (Bethesda) 20, 417–428.
Pech, V., Zheng, W., Pham, T.D., Verlander, J.W., and Wall, S.M. (2008).
Angiotensin II activates H+-ATPase in type A intercalated cells. J. Am. Soc.
Nephrol. 19, 84–91.
Price, D. (1963). Comparative Aspects Of Development And Structure In The
Prostate. Natl. Cancer Inst. Monogr. 12, 1–27.
Rescigno, M., Urbano, M., Valzasina, B., Francolini, M., Rotta, G., Bonasio, R.,
Granucci, F., Kraehenbuhl, J.P., and Ricciardi-Castagnoli, P. (2001). Dendritic
cells express tight junction proteins and penetrate gut epithelial monolayers to
sample bacteria. Nat. Immunol. 2, 361–367.
Rizzo, S., Attard, G., and Hudson, D.L. (2005). Prostate epithelial stem cells.
Cell Prolif. 38, 363–374.
Robaire, B., and Viger, R.S. (1995). Regulation of epididymal epithelial cell
functions. Biol. Reprod. 52, 226–236.
Rothenberger, F., Velic, A., Stehberger, P.A., Kovacikova, J., and Wagner,
C.A. (2007). Angiotensin II stimulates vacuolar H+ -ATPase activity in renal
acid-secretory intercalated cells from the outer medullary collecting duct. J.
Am. Soc. Nephrol. 18, 2085–2093.
Saez, F., Legare, C., Laflamme, J., and Sullivan, R. (2004). Vasectomy-depen-
dent dysregulation of a local renin-angiotensin system in the epididymis of the
cynomolgus monkey (Macaca fascicularis). J. Androl. 25, 784–796.
Schneeberger, E.E., and Lynch, R.D. (2004). The tight junction: a multifunc-
tional complex. Am. J. Physiol. Cell Physiol. 286, C1213–C1228.
Skrabanek, L., Murcia, M., Bouvier, M., Devi, L., George, S.R., Lohse, M.J.,
Milligan, G., Neubig, R., Palczewski, K., Parmentier, M., et al. (2007). Require-
ments and ontology for a G protein-coupled receptor oligomerization knowl-
edge base. BMC Bioinformatics 8, 177.
Smith, P.J.S., Sanger, R.S., and Messerli, M.A. (2007). Principles, develop-
ment and applications of self-referencing electrochemical microelectrodes
tothe determinationof fluxes at cell membranes. InMethods and new frontiers
in neuroscience, A.C. Michael, ed. (CRC Press), pp. 373–406.
Stein, B., Khew-Goodall, Y., Gamble, J., and Vadas, M.A. (1997). Transmigra-
tion of leukocytes. In Endothelium in Clinical Practice: Source and Target of
NovelTherapies, G.M. Rubanyi and V.J. Dzau, eds. (New York: Marcel Dekker,
Inc.), pp. 147–202.
Takano, K., Kojima, T., Go, M., Murata, M., Ichimiya, S., Himi, T., and Sawada,
N. (2005). HLA-DR- and CD11c-positive dendritic cells penetrate beyond well-
developed epithelial tight junctions in human nasal mucosa of allergic rhinitis.
J. Histochem. Cytochem. 53, 611–619.
Toda, N., Ayajiki, K., and Okamura, T. (2007). Interaction of endothelial nitric
oxide and angiotensin in the circulation. Pharmacol. Rev. 59, 54–87.
van Leenders, G.J., and Schalken, J.A. (2003). Epithelial cell differentiation in
the human prostate epithelium: implications for the pathogenesis and therapy
of prostate cancer. Crit. Rev. Oncol. Hematol. Suppl. 46, S3–S10.
Veri, J.P., Hermo, L., and Robaire, B. (1993). Immunocytochemical localization
of the Yf subunit of glutathione S-transferase P shows regional variation in the
staining of epithelial cells of the testis, efferent ducts, and epididymis of the
male rat. J. Androl. 14, 23–44.
Wei, C., Willis, R.A., Tilton, B.R., Looney, R.J., Lord, E.M., Barth, R.K., and Fre-
linger, J.G. (1997). Tissue-specific expression of the human prostate-specific
antigen gene in transgenic mice: implications for tolerance and immunother-
apy. Proc. Natl. Acad. Sci. USA 94, 6369–6374.
Wong, P.Y., and Yeung, C.H. (1978). Absorptive and secretory functions of the
perfused rat cauda epididymidis. J. Physiol. 275, 13–26.
Yeung, C.H., Nashan, D., Sorg, C., Oberpenning, F., Schulze, H., Nieschlag,
E., and Cooper, T.G. (1994). Basal cells of the human epididymis–antigenic
and ultrastructural similarities to tissue-fixed macrophages. Biol. Reprod. 50,
Zini, A., O’Bryan, M.K., Magid, M.S., and Schlegel, P.N. (1996). Immunohisto-
chemical localization of endothelial nitric oxide synthase in human testis,
epididymis, and vas deferens suggests a possible role for nitric oxide in sper-
matogenesis, sperm maturation, and programmed cell death. Biol. Reprod.
Cell 135, 1108–1117, December 12, 2008 ª2008 Elsevier Inc. 1117