Content uploaded by Terje Dokland
Author content
All content in this area was uploaded by Terje Dokland on Oct 13, 2017
Content may be subject to copyright.
The Prostate
Structural Heterogeneity and Protein Composition of
Exosome-LikeVesicles (Prostasomes) in
Human Semen
Anton Poliakov,
1
Michael Spilman,
2
Terje Dokland,
2
Christopher L. Amling,
1
and James A. Mobley
1
*
1
Department of Surgery/Urology, University of Alabama at Birmingham, Birmingham, Alabama
2
Department of Microbiology, University of Alabama at Birmingham, Birmingham, Alabama
BACKGROUND. Human seminal fluid contains small exosome-like vesicles called prosta-
somes. Prostasomes have been reported previously to play an important role in the process of
fertilization by boosting survivability and motility of spermatozoa, in addition to modulating
acrosomal reactivity. Prostasomes have also been reported to present with sizes varying from
50 to 500 nm and to have multilayered lipid membranes; however, the fine morphology of
prostasomes has never been studied in detail.
METHODS. Sucrose gradient-purified prostasomes were visualized by cryo-electron micro-
scopy (EM). Protein composition was studied by trypsin in-gel digestion and liquid
chromatography/mass spectrometry.
RESULTS. Here we report for the first time the detailed structure of seminal prostasomes by
cryo-EM. There are at least three distinct dominant structural types of vesicles present. In
parallel with the structural analysis, we have carried out a detailed proteomic analysis of
prostasomes, which led to the identification of 440 proteins. This is nearly triple the number of
proteins identified to date for these unique particles and a number of the proteins identified
previously were cross-validated in our study.
CONCLUSION. From the data reported herein, we hypothesize that the structural hetero-
geneity of the exosome-like particles in human semen reflects their functional diversity. Our
detailed proteomic analysis provided a list of candidate proteins for future structural and
functional studies. Prostate #2008 Wiley-Liss, Inc.
KEY WORDS: prostasomes; cryo-electron microscopy; proteomics
INTRODUC TION
Prostasomes are small lipid membrane-confined
vesicles present in large quantities in human semen
[1] and semen from male animals of several species
[2,3]. Prostasomes have been shown to play an
important role in the process of fertilization. They
increase sperm motility [4–6] and delay acrosomal
reaction so that it would not occur prior to reaching the
egg [7]. Prostasomes have also been reported to be
immunosuppressive, which is important in preventing
sperm destruction by the female’s immune system.
These functions are most likely achieved by direct
interaction of prostasomes with spermatozoa [8,9] and
immune system components present in female reproduc-
tive organs. Several complement regulation molecules
have also been reported to be present on prostasomes:
CD59 (membrane attacking complex inhibitor [10]),
CD46 (membrane cofactor protein, cofactor of Factor I-
mediated proteolysis of C3b [11,12]) and CD55 (decay
accelerating factor [12]). Additionally, complement
factor C3 along with fibrinogen has been reported to
Additional supporting information may be found in the online
version of this article.
*Correspondence to: Dr. James A. Mobley, 1900 University Boule-
vard, 513, Birmingham, AL, 35294.
E-mail: jim.mobley@ccc.uab.edu
Received 11 August 2008; Accepted 2 September 2008
DOI 10.1002/pros.20860
Published online in Wiley InterScience
(www.interscience.wiley.com).
2008Wiley-Liss,Inc.
be phosphorylated by protein kinases present on
prostasomes [13], resulting in altered activation of
complement factor C3 [14]. Prostasomes are also
known to inhibit mitogen-induced lymphoprolifera-
tion [15], phagocytosis and creation of free radicals by
neutrophils and macrophages [16,17] and the migration
of macrophages [18]. Finally, prostasomes are also known
to contain components of coagulation system [19].
Previously, prostasomes and other exosome-like
particles were studied by thin-section electron micro-
scopy (EM) [10,15,20–23], scanning EM [12,24– 27] and
negative staining EM [28]. Prostasomes were shown to
range in size from 50 to 500 nm and to have multi-
layered membranes [21,22]. Based on the electron
density of the particles, prostasomes were described
as either ‘‘light’’ or ‘‘dark’’ [22]. However, the fine
morphology of prostasomes was never studied in
detail, since the fixation, embedding and dehydration
steps of traditional EM preparation methods are
strongly detrimental to the integrity of membranous
structures like prostasomes.
Unlike thin-section EM, cryo-EM preserves the
sample in its native state, in the absence of dehydration
and staining artifacts [29]. Furthermore, cryo-EM
produces a projection through the entire three-dimen-
sional (3D) sample volume with which the 3D structure
of the specimen can be deduced.
We have studied the structure of sucrose gradient-
purified seminal prostasomes by cryo-EM and were
able to observe for the first time the fine details of
prostasomes. We show that the prostasomes possess
sub-vesicular internal and external structures in the
form of daughter vesicles and depositions of dark
substance that was not visible in glutaraldehyde fixed
thin-section preparations. We found that prostasomes
display significant morphological heterogeneity with
several distinct structural types present.
In parallel with the structural analysis, we have
performed proteomic analysis of the prostasomes and
identified 440 proteins with high probability. Proteins
known to be specific to both prostate and epididymis
were detected, which, together with the structural data,
suggested composite origin of exosome-like particles in
human semen.
We postulate that the structural heterogeneity of
prostasomes must translate into functional heteroge-
neity, with different types of prostasomes reflecting
varied and specific functions. It is also possible that
these structurally specific prostasomes originate from
several male sexual organs (i.e., seminal vesicles,
testicles, prostate). Our detailed proteomic analysis
has provided a list of the milieu of proteins to be
studied for a more detailed future structural and
functional analysis of prostasomes, and the origin of
these very interesting vesicles.
MATERIAL S AND METHODS
Preparation of Prostasomes
Semen from three healthy volunteers was collected
and left to liquefy at room temperature for 20 min. Cells
and large pieces of undissolved seminal gel were
removed by centrifugation at 1,000gfor 10 min, and the
supernatant was further centrifuged at 10,000gfor
30 min to remove cellular debris and smaller pieces of
undissolved seminal gel. The resulting supernatant
was centrifuged at 100,000gfor 2 hr, and the pelleted
prostasomes were washed and redissolved in TBS buffer.
Prostasomes were further purified by sucrose
gradient using step gradients of 0.3, 0.6, 0.9, and 1.3 M
sucrose in TBS. The prostasome suspension was layered
on the top of the sucrose gradient and centrifuged at
75,000gfor 8 hr. An intense opaque prostasome band was
collected by puncturing the tube from the side. The
prostasome suspension was slowly diluted 6with ice-
cold TBS and pelleted at 100,000gfor 30 min. The
resulting pellet was used for all further experiments.
Electron Microscopy
For cryo-EM 3 ml of prostasomes was applied to
C-flat holey film (Electron Microscopy Sciences),
blotted briefly before plunging into liquid ethane
and transferred to a Gatan cryo-sample holder. All
samples were visualized in an FEI Tecnai F20 electron
microscope operated at 200 kV, and images were
captured on a 4 k 4 k Gatan Ultrascan CCD camera
at a magnification of 29,000.
For obtaining tilt series, sample holder was tilted
incrementally by 18in both directions 608. The image s
were taken at each angle at reduced beam intensity so
that the total electron dose on the sample was below
100 e
/A
˚
2
.
For thin-section microscopy 2.5% glutaraldehyde in
phosphate-buffered saline (pH 7.2). After fixation, the
specimens were trimmed into a few smaller pieces and
embedded in Epon to form plastic blocks according to
conventional techniques. The plastic blocks were cut in
2 mm sections for light microscopy to find areas with
neoplastic tissue. These areas were trimmed to be cut in
50 nm sections for ultrastructural examinations.
The sections were put on slot grids with 0.5% Formvar
film, contrasted with lead citrate and uranyl acetate,
and examined in an electron microscope (FEI Tecnai
F20).
Determination of Protein Composition
of Prostasomes
A total of 100 mg of prostasomes (by protein) were
loaded on a 12% SDS–PAGE gel and separated. The gel
was stained with Coomassie and the gel lane was cut
The Prostate
2Poliakovetal.
into 15 strips. The proteins were reduced and alkylated
with iodoacetamide. The strips were destained in 60%
methanol, 0.1% trifluoroacetic acid (TFA) and dried
with pure acetonitrile. The acetonitrile was removed by
evaporation in a Speedvac centrifugal evaporator and
protein digestion was performed by the addition of
10 mg/ml Trypsin Gold (Promega) solution in 100 mM
ammonium bicarbonate and incubated for 8 hr at 378C.
After extraction with 10% acetonitrile, the peptides
were loaded on a 100 nm 10 cm capillary column in-
house packed with C18 Monitor 100 A-spherical silica
beads and eluted by a 1 hr gradient of 10–100%
acetonitrile, 0.1% TFA. Mass spectrometric analysis
was performed on an LTQ XL (Thermo Finnigan)
spectrometer. The search for matching peptide sequen-
ces was performed using the SEQUEST search engine
with UniProt database including mouse entries. The
SEQUEST results were processed by Trans Proteomic
Pipeline software and peptides with a probability of
>0.6 were reported [30].
Classif|cation of Protein Hits
UniProt protein IDs were enriched for GO terms using
The Protein Information and Property Explorer (http://
pipe.systemsbiology.net/pipe/#summary). Search and
calculation of the number of proteins falling into different
categories according to their cellular location and bio-
logical processes they take part in was performed using
Excel spreadsheet.
RESULTS
Sucrose Gradient Purif|cation of Prostasomes
Coarse prostasome preparations were purified from
human semen by differential centrifugation. This
preparation was shown to be contaminated by the
previously termed ‘‘amorphous substance’’ [21,22],
which is most likely residues of partially cleaved
seminal gel (Poliakov A, unpublished results). In order
to remove this contamination and to separate prosta-
somes from other undissolvable material, we separated
the prostasomes further on sucrose gradients. A very
intense, opaque band was observed at the 0.9/1.3 M
sucrose interface and some precipitated material was
also evident at the bottom of the tube. Material collected
from the band was slowly diluted fourfold with
isotonic TBS buffer and the particles were spun down
at 70,000gfor 1 hr. It was important to dilute the sample
in high sucrose concentration slowly as it appeared
that prostasomes are prone to disruption upon quick
dilution, probably due to osmotic shock, as was
witnessed by hard-to-resuspend pellet after harvesting
prostasomes by ultracentrifugation. Microsocpic
assessment of a thin section of glutaraldehyde-fixed
pellet revealed large, aggregated, elongated membra-
nous structures (not shown) with few intact prosta-
somes. The washing procedure was repeated twice in
order to remove any trace of sucrose that would
produce strong background signal in cryo-EM.
Prostasomes Display Signif|cant
Morphological Diversity
The sucrose gradient-purified prostasomes were
imaged by cryo-EM. In total, images of 301 vesicles
were analyzed (Supplementary Figure 1A –F). Selected
particles are presented in Figures 1 –3. The particles
generally varied in size from 50 to 200 nm, and only on
occasion, were found to present as slightly larger. The
overwhelming majority of particles were intact (Sup-
plementary Figure 1D). All particles were confined by a
trilamellar membrane. We classified the particles
according to the four morphological criteria listed
below:
(1) Multiplicity of vesicles: Most vesicles contained
secondary, and occasionally, tertiary vesicles
of smaller size (Fig. 1A), while a few vesicles
appeared as singular entities (Fig. 2).
(2) Vesicle shape: Most vesicles were close to round in
shape, with some being completely round and
The Prostate
Fig. 1. Clos e-up view of high electron density vesicles. Carto on
representationisontop, showinggeneralarrangementof struc tural
elements in the vesicle.D ashed white arrows are showing dark dep-
ositions, solidwhite arrows areshowing secondary vesicles.A:Pro-
jections of the vesicles where arrangement of secondary vesicles
and darkdepo sitionsis clearly seen. B:Projections where secondary
vesicles apparently overlay withdark depositions.
Prostasomes Structure and Composition 3
others egg-shaped (Figs. 1 and 2). However, few
vesicles were distinctly elongated (Fig. 3), yet fully
intact, meaning that they are not a result of vesicle
rupture and emptying. Inspection of tilted images
series showed that the elongated vesicles are
sausage-shaped, rather than being side projections
of erythrocyte-shaped particles (see below).
(3) External Features: Cauliflower-like protrusions
measuring about 5 nm in diameter were clearly
evident on some of the vesicles (Fig. 2), while other
vesicles were devoid of them (Figs. 1 and 3). The
globular protrusions were separated from the
membrane by about 4 nm presumably via a stalk
domain, and most likely correspond to membrane
proteins. The density of the protrusions on the
vesicle surface varied, with some vesicles being
completely surrounded by the protrusions and
some having large gaps in between them. How-
ever, the protrusions were rather uniform in
appearance, suggesting that they correspond pre-
dominantly to a single type of protein. There
was no apparent correlation between the presence
of protrusions and the presence of secondary
vesicles. However, larger vesicles tended to con-
tain fewer protrusions and at lower density, and
the largest vesicles were devoid of the protrusions
all together.
(4) Overall Density of Vesicles: The electron density of
the particles varied significantly. Inside some
particles, a markedly electron dense substance
was apparent (Fig. 1). This substance was usually
generally organized into a sickle shape at one of
the poles of the vesicle, being confined by daughter
vesicles at another pole.
Using the above defined morphological criteria, we
were able to distinguish at least three major types of
particles as noted below
Type 1 (dark vesicles): High electron density vesicles
with asymmetric deposition of dark substance sur-
rounding one or more secondary vesicles (Fig. 1A).
These particles are invariably smooth on the outside,
lacking the cauliflower-like protrusions, and are
nearly round or slightly elongated. Their sizes range
from 100 to 200 nm. The fraction of these vesicles in the
sample is around 50%. An exact count is complicated
since the morphology of this type of particle is only
obvious when it is seen in a projection depicted in
Figure 1A. However, when viewed perpendicular to
the horizontal axis of the view in Figure 1A, these
particles will be much harder to recognize, as dark
substance and the daughter vesicle will overlay.
Particles depicted in Figure 1B are most likely of this
type, with daughter vesicles barely visible due to the
high density of the dark matter depositions.
Type 2 (light vesicles): Low electron density vesicles
with abundant cauliflower-like protrusions (Fig. 2).
These vesicles do not have internal depositions of dark
substance, but sometimes contain daughter vesicles,
and are nearly round or only slightly elongated. Their
sizes range from 50 to 200 nm. This class represents
approximately 25–30% of all vesicles.
The Prostate
Fig. 2. Close- up viewo f low electron den sity particle s with cauli-
flower-like pro trusions. Car toon to the left shows pres ence of sur-
face depositions (pro trusions) on thevesiclemembrane.Cartoon to
theright shows approximate dis tances between membrane and the
blobs a nd the siz e of the later.
Fig. 3. Close-up view of selec ted elongated low electron density
particles without cauliflower-like protrusions.Whitebar is 20 0 nm.
4Poliakovetal.
Type 3: Low electron density vesicles without
cauliflower-like features, presenting as elongated, or
pear-shaped (Fig. 3), with no dark depositions, and
smooth outer surface. They can be as long as 400 nm
and about 50–100 nm wide. This class constitutes the
remaining 20–25% of the vesicles.
In order to empirically determine if the above-
described morphological features would be observ-
able with thin-section EM we have further pelleted the
purified fraction of prostasomes by ultracentrifugation
and fixed the resulting pellet with glutaraldehyde.
Thin sections were analyzed with transmission EM
(Fig. 4). Both dark and light vesicles were observed in
these sections, which has been described previously
[22]. However, no protrusions on the outside surface
of the vesicles were visible, and the arrangement of the
dark substance in the vesicles was not obvious, though
the particles indicated by the arrows in Figure 4 are
likely the same dark vesicles that we have noted in our
previous experiments. This offers the explanation for
why the morphological features described in the
present work were not described before with the use
of thin-section EM.
We have noted that light vesicles remotely resemble
mycoplasma (ureaplasma) cells [31]. According to
the World Health Organization some 40–80% of
all people are carriers of Ureaplasma urealyticum so
it was important to establish that light vesicles
are not ureaplasma cells. Proteomic analysis indi-
cated that mycoplasmal protein is below detection
(Supplementary Table II), which indicates that light
vesicles cannot be ureaplasma or other mycoplasma
cells.
Conf|rmation of the Observed Structural Types of
Prostasomes by Cryo-EM Tilt Series
In order to obtain information on the spatial
arrangement of morphological elements and to further
confirm the morphological types of prostasomes, we
have collected a tilt series of images. To carry out these
experiments, the sample holder with the frozen vesicles
was incrementally tilted in 1–28steps over an angular
range of 608and images were taken at each angle. An
animated view of the tilt series gives an idea of the 3D
structure of the prostasome particles and is presented
in Movies 1–3 that can be viewed at:
http://www.urology.uab.edu/urologyresearch/files/
Movie1.avi
http://www.urology.uab.edu/urologyresearch/files/
Movie2.avi
http://www.urology.uab.edu/urologyresearch/files/
Movie3.avi
For better orientation in the animations dark and
light prostasomes are indicated by dark and white
arrows correspondingly on images taken at 08from
each animation (Supplementary MovieSnapshot1.tif—
MovieSnapshot3.tif). The high electron density
vesicles have clearly asymmetric deposition of the
dark substance (Supplementary Movie 1 and Movie 3),
which surrounds, rather than pushes, the daughter
vesicles. It is clear from the tilt series that from some
projections these particles look just like those in
Figure 1B, and would not be distinguishable at these
projections. The spacing between the dark substance
and the membranes of the parent and daughter
vesicles is constant and is about 4–5 nm.
Light vesicles were clearly visible and it was clear that
the cauliflower-like protrusions completely cover the
whole surface of the particles (Supplementary Mov.1–3).
It was also obvious that there are no dark depositions in
these particles as they were invariably transparent.
Elongated vesicles were clearly sausage-like and
not side-projections of erythrocyte-like shapes
(Movies 1–3). It was also clear that these vesicles are
intact, ruling out a possibility of their shape being a
result of rupture and partial emptying.
From animations of tilt series it is very clear
that secondary and tertiary vesicles are indeed, inside
of their parent particles and not simply overlay with
them in the images (Supplementary Mov.1 and Mov.3).
The daughter vesicles did not have any connection to
the membrane of the parent vesicle and appeared to be
‘‘free-floating’’ and unconstrained inside them.
Protein Composition of Prostasomes
In order to study the protein composition of the
prostasomes we have separated prostasomal proteins
The Prostate
Fig. 4. Thin-sec tion transmission EM image of prostasomes.
White arrows indicatevesicles that lo ok similar to the darkvesicles
imaged in Figure1B.Da shed arrows indicate presumable dark depo -
sitions,solid arrows indicate secondary vesicle.
Prostasomes Structure and Composition 5
by SDS–PAGE, cut the lane into 15 bands and
performed trypsin in-gel digestion with each band.
The resulting peptides were extracted and analyzed by
nano-LC ESI MS/MS on an LTQ XL mass spectrometer
(Thermo Finnigan). The resultant spectra were ana-
lyzed by three matching algorithms against a subset of
the UniRef database specific for human proteins using
SEQUEST (Thermo Finnigan), followed by the Pipeline
software (ISB). Proteins identified with high confidence
are listed in Supplementary Table. We have performed
similar searches against a mycoplasma-specific data-
base, and in stark contrast with the human database
searches, no significant hits were present (Supplemen-
tary Table II), which strongly suggested that myco-
plasmal contamination in the sample is minimal if
present at all.
All matching algorithm output files were combined
and analyzed with Pipeline software as mentioned,
which reduces assignments of single spectrum to
peptides from multiple proteins and calculates a proba-
bility for each peptide and protein. Using an empirically
determined protein probability cutoff of 0.6 we obtained
440 protein hits, of which 304 were assigned with two or
more peptides (Supplementary Table I). At this proba-
bility cutoff, only 15 proteins with probability <0.98 were
predicted to be incorrectly assigned.
The data obtained here are not quantitative, that is,
we cannot reliably tell the relative amounts of proteins
present in prostasomes. However, the coverage and the
number of successfully identified peptides (otherwise
termed ‘‘ion counting’’) can give us rough measure of
protein abundance. From Supplementary Table II it is
clear that, for example, lactoferrin, aminopeptidase N,
dipeptidyl peptidase IV, protein-glutamine gamma-
glutamyltransferase 4, neprilysin, and various other
proteins are highly abundant, which is witnessed by
the high protein sequence coverage and high proba-
bility of the identified peptides.
Using The Protein Information and Property Explorer
(http://pipe.systemsbiology net/pipe/#summary) (ISB),
we have classified the identified proteins into groups
according to several criteria, including cellular location
(Supplementary Table I), molecular function and bio-
logical process they take part in. Of 440 proteins
73 proteins were non-annotated.
Most of the 440 identified proteins are assigned as
intracellular. However, 32 proteins are known to be
secreted. Most of these proteins are unlikely to be pro-
stasome preparation contaminants, as all soluble proteins
must have been efficiently excluded at ultracentrifugation
and sucrose gradient steps. For example, prostate-specific
antigen (present in normal human semen at concentration
of approx. 1.2 mg/ml [32]) is not detected, which confirms
that the purification protocol is efficient in removing
soluble components of semen.
The presence of various secreted proteins including
protease inhibitors is more easy to make sense of since
they must be bound to their corresponding membrane
proteases/peptidases identified in prostasomes, such
as UniRef100_Q16651 (transmembrane protein), Uni-
Ref100_Q96KP4, UniRef100_P08473 (membrane met-
allo protease), UniRef100_P14384 (membrane protein),
UniRef100_P15144 (membrane protein). Thus the
protease inhibitors must be truly associated with
prostasomes, albeit in an indirect way.
However, the presence of other proteins is not so
easily explainable. Proteins that form seminal gel,
semenogelin I and II and fibronectin, were detected
with high confidence. These proteins constitute the
bulk of the protein in ejaculate and are present at very
high concentrations, so it is quite probable that they are
a contamination, rather than a component of prosta-
somes. Soluble proteins should be very efficiently
removed on sucrose gradient step; however, both
semenogelins and fibronectin form a precipitate right
after ejaculation, which slowly gets dissolved by
prostate-specific antigen [33,34]. Our observations
suggest that the dissolution is never fully complete,
which means that semenogelin precipitates of different
sizes must be present in the liquefied semen and some
of them might co-purify with prostasomes.
DISCUSSION
Our work has provided the first glimpse at the fine
structure of the exosome-like particles otherwise
termed ‘‘prostasomes’’ and established cryo-EM as
the method of choice for studying exosomes. Morpho-
logical diversity observed in prostasomes raises several
important questions about the origin of prostasomes
and their function.
The name prostasomes suggests that the vesicles are
produced exclusively in the prostate and some claim this
to be the case [1,24,35] but in other works the origin of
prostasomes is suggested to be composite [36], which
might be the case since semen is a product of five
different organs: testes, epididymis, prostate, seminal
vesicles, and Cowper gland. Indeed, exosome-like
vesicles were observed in epididymis of men [18] and
several other mammal species [37–43]. However, the
relative abundance of the epididymal vesicles is likely
low and, to our knowledge, their microscopic structure
has never been studied. Additionally, no difference was
observed in prostasomes size and number between
semen from vasectomized and non-vasectomized
individuals [44], suggesting that epididymosomes are
present in insignificant quantities. So it is not possible
to assign the structural types observed in this work
to either prostate or epididymis. Proteomic analysis
has revealed cystatin SA (UniRef100_P01034) and
The Prostate
6Poliakovetal.
dicarbonyl/L-xylulose reductase (UniRef100_Q7Z4W1)
in the vesicle preparation: both proteins are abundantly
expressed in epididymis, but not in prostate. This
also suggests that some vesicles might indeed be of
epididymal origin. But since proteomic data presented
here are not quantitative it does not rule out relative
scarcity of epididymosomes in semen.
In previous work [20] prostasome-like particles were
observed in prostate tissue of men. They appeared very
similar to the dark vesicles observed in the present
work, with some vesicles clearly having secondary
vesicles surrounded by dark substance (Figs. 2 and 8
in [20]), suggesting that at least dark vesicles have
prostate origin. Other vesicles might still be a product
of prostate too, reflecting excretion of mixed popul ation
of vesicles by prostate epithelial cells or functional
heterogeneity of those cells. Study of pure prostate
secretion is required to answer those questions.
Prostasomes were assumed to be produced in an
exosome-like manner: first intracellular membrane
budding into a multi-vesicular body and then fusion of
the later with the cellular membrane and release of
prostasomes. This mechanism can hardly explain how
dark prostasomes are formed; specifically, how and why
daughter vesicles are produced and how dark deposi-
tions occupy only just the space left outside of the
daughter vesicles. It is plausible upon first inspection to
assume that thedark depositions are of higher molecular
density in addition to electron density, and therefore of
higher rigidity than the daughter vesicles. In that respect
it is rather odd that a presumably ‘‘softer’’ daughter
vesicle pushes aside the presumed more rigid dark
deposition as depicted in Figure 1, and this observation
is clearly evident in tilt series animations. Rather, one
may more feasibly conclude that the dark depositions
are more fluid with a lower molecular density, poten-
tially made up of more dilute high electron density
material, thereby allowing them to easily form to any
shape, especially if these features are produced after the
daughter vesicles are formed.
The discovery of sphingomyelin phosphodiesterase
(UniRef100_Q5T0Y8) in the prostasomes might explain
how secondary vesicles are formed. This enzyme has
recently been implicated in the ESCRT-independent
inward budding of exosomes into multivesicular
exosomes originating from murine oligodendroglial
cells [45]. Sphingomyelinase cleaves sphingomyelin
into phosphocholine and ceramide, which has been
shown to cause formation of secondary vesicles in
giant unilamellar vesicles [45]. Prostasomes were
shown to be rich in sphingomyelin [46,47] and
sphingomyelinase activity might result in formation
of secondary vesicles.
Equally enigmatic is how a second type of vesicle
gets produced, that is, whether the noted features on
the outside of the vesicles are acquired after the vesicles
are formed or prior to budding into a multi-vesicular
body. One plausible explanation is that this may occur
following the liquefaction of semen with specific
attachment of various protein(s) to the surface of light
prostasomes. Of special interest is what these features
are made of, as this would potentially provide a clue as
to the function of the light prostasomes. The apparent
size of these features (approx. 4–6 nm) suggests that
this might be a single large protein molecule. For
example, aminopeptidase N (UniRef100_P15144) fits
well into this description, being a large membrane-
bound protein [48] with dimensions of 85 A
˚56 A
˚
(PDB entry 2DQM). Aminopeptidase N attaches to the
lipid membrane at the N-terminus, with the main body
of the protein presenting approximately 5 nm from the
membrane [48], which fits well with what we have
observed here (Fig. 2). Additionally, aminopeptidase N
is clearly a very abundant protein in prostasomes
(Supplementary Table I), being identified with 55
unique peptides covering 32.4% of its sequence. This
The Prostate
Fig. 5. Graphicalrepresentation of the pro tein groups present in prostasomes.A: Spatial distribu tion of the proteins. B:Distributionofthe
biological processes the proteins take par t in.
Prostasomes Structure and Composition 7
protein was also previously described in prostasomes
[8,49].
If the cauliflower-like features are indeed made up of
aminopeptidase N (a cell-surface enzyme), then light
vesicles might not be exosome-like, but rather being
directly shed from the surface of the cell membrane
(ectosomes).
Another important question is what the functions of
the different structural types of prostasomes are as it
seems quite probable that vesicles with such differing
morphologies likely present with very different prop-
erties and functions, a point we hope to answer more
definitively in future studies of this kind.
Proteomic analysis performed here has provided a
large list of proteins that will be of great use for future
studies on structure and function of prostasomes and,
likely, for the discovery of new prostate cancer
biomarkers. Discovery of two known candidate bio-
markers of prostate cancer (prostate acid phosphatase
(PAP) and prostate-specific membrane antigen
(PSMA)) clearly demonstrates that prostasomes might
be a valuable source of the biomarkers. Elucidation of
spatial arrangement of PSMA and PAP (and other
candidate biomarkers) in prostasomes and their asso-
ciation with one of the structural types of prostasomes
would be of great practical value for usage of those
biomarkers and understanding of their biology.
CONCLUSIONS
From the data reported herein, we hypothesize that
the structural heterogeneity of the exosome-like par-
ticles in human semen reflects their functional diversity
and, possibly, origin from several organs in addition to
the prostate. Our detailed proteomic analysis provided
a list of candidate proteins for future structural and
functional studies.
REFERENCES
1. Ronquist G, Brody I, Gottfries A, Stegmayr B. An Mg2þand
Ca2þ-stimulated adenosine triphosphatase in human prostatic
fluid: Part I. Andrologia 1978;10(4):261– 272.
2. Arienti G, Carlini E, De Cosmo AM, Di Profio P, Palmerini CA.
Prostasome-like particles in stallion semen. Biol Reprod 1998;
59(2):309– 313.
3. Piehl LL, Cisale H, Torres N, Capani F, Sterin-Speziale N, Hager
A. Biochemical characterization and membrane fluidity of
membranous vesicles isolated from boar seminal plasma. Anim
Reprod Sci 2006;92(3– 4):401–410.
4. Fabiani R, Johansson L, Lundkvist O, Ulmsten U, Ronquist G.
Promotive effect by prostasomes on normal human spermatozoa
exhibiting no forward motility due to buffer washings. Eur J
Obstet Gynecol Reprod Biol 1994;57(3):181– 188.
5. Fabiani R, Johansson L, Lundkvist O, Ronquist G. Enhanced
recruitment of motile spermatozoa by prostasome inclusion in
swim-up medium. Hum Reprod 1994;9(8):1485– 1489.
6. Wang J, Lundqvist M, Carlsson L, Nilsson O, Lundkvist O,
Ronquist G. Prostasome-like granules from the PC-3 prostate
cancer cell line increase the motility of washed human
spermatozoa and adhere to the sperm. Eur J Obstet Gynecol
Reprod Biol 2001;96(1):88– 97.
7. Cross NL, Mahasreshti P. Prostasome fraction of human seminal
plasma prevents sperm from becoming acrosomally responsive
to the agonist progesterone. Arch Androl 1997;39(1):39– 44.
8. Arienti G, Carlini E, Verdacchi R, Cosmi EV, Palmerini CA.
Prostasome to sperm transfer of CD13/aminopeptidase N (EC
3.4.11.2). Biochim Biophys Acta 1997;1336(3):533– 538.
9. Arienti G, Polci A, Carlini E, Palmerini CA. Transfer of CD26/
dipeptidyl peptidase IV (E.C. 3.5.4.4) from prostasomes to
sperm. FEBS Lett 1997;410(2– 3):343–346.
10. Rooney IA, Atkinson JP, Krul ES, Schonfeld G, Polakoski K,
Saffitz JE, Morgan BP. Physiologic relevance of the membrane
attack complex inhibitory protein CD59 in human seminal
plasma: CD59 is present on extracellular organelles (prosta-
somes), binds cell membranes, and inhibits complement-
mediated lysis. J Exp Med 1993;177(5):1409– 1420.
11. Kitamura M, Namiki M, Matsumiya K, Tanaka K, Matsumoto M,
Hara T, Kiyohara H, Okabe M, Okuyama A, Seya T. Membrane
cofactor protein (CD46) in seminal plasma is a prostasome-
bound form with complement regulatory activity and measles
virus neutralizing activity. Immunology 1995;84(4):626– 632.
12. Rooney IA, Heuser JE, Atkinson JP. GPI-anchored complement
regulatory proteins in seminal plasma. An analysis of their
physical condition and the mechanisms of their binding to
exogenous cells. J Clin Invest 1996;97(7):1675–1686.
13. Babiker AA, Ronquist G, Nilsson B, Ekdahl KN. Overexpression
of ecto-protein kinases in prostasomes of metastatic cell origin.
Prostate 2006;66(7):675– 686.
14. Ekdahl KN, Nilsson B. Alterations in C3 activation and binding
caused by phosphorylation by a casein kinase released
from activated human platelets. J Immunol 1999;162(12):7426–
7433.
15. Kelly RW, Holland P, Skibinski G, Harrison C, McMillan L,
Hargreave T, James K. Extracellular organelles (prostasomes)
are immunosuppressive components of human semen. Clin Exp
Immunol 1991;86(3):550– 556.
16. Skibinski G, Kelly RW, Harkiss D, James K. Immunosuppression
by human seminal plasma– extracellular organelles (prosta-
somes) modulate activity of phagocytic cells. Am J Reprod
Immunol 1992;28(2):97– 103.
17. Saez F, Motta C, Boucher D, Grizard G. Prostasomes inhibit the
NADPH oxidase activity of human neutrophils. Mol Hum
Reprod 2000;6(10):883– 891.
18. Frenette G, Legare C, Saez F, Sullivan R. Macrophage migration
inhibitory factor in the human epididymis and semen. Mol Hum
Reprod 2005;11(8):575– 582.
19. Lwaleed BA, Greenfield RS, Birch BR, Cooper AJ. Does human
semen contain a functional haemostatic system? A possible role
for tissue factor pathway inhibitor in fertility through semen
liquefaction. Thromb Haemost 2005;93(5):847– 852.
20. Brody I, Ronquist G, Gottfries A. Ultrastructural localization of
the prostasome—an organelle in human seminal plasma. Ups J
Med Sci 1983;88(2):63– 80.
21. Stridsberg M, Fabiani R, Lukinius A, Ronquist G. Prostasomes
are neuroendocrine-like vesicles in human semen. Prostate
1996;29(5):287– 295.
22. Brody I, Ronquist G, Gottfries A, Stegmayr B. Abnormal
deficiency of both Mg2þand Ca2þ-dependent adenosine
The Prostate
8Poliakovetal.
triphosphatase and secretory granules and vesicles in human
seminal plasma. Scand J Urol Nephrol 1981;15(2):85– 90.
23. Arienti G, Carlini E, Polci A, Cosmi EV, Palmerini CA. Fatty acid
pattern of human prostasome lipid. Arch Biochem Biophys 1998;
358(2):391– 395.
24. Nilsson BO, Jin M, Einarsson B, Persson BE, Ronquist G.
Monoclonal antibodies against human prostasomes. Prostate
1998;35(3):178– 184.
25. Bordi F, Cametti C, De Luca F, Carlini E, Palmerini CA, Arienti G.
Hydrodynamic radii and lipid transfer in prostasome self-
fusion. Arch Biochem Biophys 2001;396(1):10– 15.
26. Vivacqua A, Siciliano L, Sabato M, Palma A, Carpino A.
Prostasomes as zinc ligands in human seminal plasma. Int J
Androl 2004;27(1):27– 31.
27. Ronquist G, Nilsson BO, Hjerten S. Interaction between
prostasomes and spermatozoa from human semen. Arch Androl
1990;24(2):147– 157.
28. Llorente A, van Deurs B, Sandvig K. Cholesterol regulates
prostasome release from secretory lysosomes in PC-3 human
prostate cancer cells. Eur J Cell Biol 2007.
29. Dubochet J, Adrian M, Chang JJ, Homo JC, Lepault J, McDowall
AW, Schultz P. Cryo-electron microscopy of vitrified specimens.
Q Rev Biophys 1988;21(2):129– 228.
30. Nesvizhskii AI, Keller A, Kolker E, Aebersold R. A statistical
model for identifying proteins by tandem mass spectrometry.
Anal Chem 2003;75(17):4646– 4658.
31. Seybert A, Herrmann R, Frangakis AS. Structural analysis of
Mycoplasma pneumoniae by cryo-electron tomography. J Struct
Biol 2006;156(2):342– 354.
32. Lynne CM, Aballa TC, Wang TJ, Rittenhouse HG, Ferrell SM,
Brackett NL. Serum and semen prostate specific antigen
concentrations are different in young spinal cord injured men
compared to normal controls. J Urol 1999;162(1):89– 91.
33. Robert M, Gagnon C. Semenogelin I: A coagulum forming,
multifunctional seminal vesicle protein. Cell Mol Life Sci 1999;
55(6– 7):944– 960.
34. Lwaleed BA, Greenfield R, Stewart A, Birch B, Cooper AJ.
Seminal clotting and fibrinolytic balance: A possible physio-
logical role in the male reproductive system. Thromb Haemost
2004;92(4):752– 766.
35. Ronquist G, Brody I, Gottfries A, Stegmayr B. An Mg2þand
Ca2þ-stimulated adenosine triphosphatase in human prostatic
fluid– part II. Andrologia 1978;10(6):427–433.
36. Renneberg H, Konrad L, Dammshauser I, Seitz J, Aumuller G.
Immunohistochemistry of prostasomes from human semen.
Prostate 1997;30(2):98– 106.
37. Legare C, Berube B, Boue F, Lefievre L, Morales CR, El-Alfy M,
Sullivan R. Hamster sperm antigen P26h is a phosphatidylino-
sitol-anchored protein. Mol Reprod Dev 1999;52(2):225– 233.
38. Frenette G, Sullivan R. Prostasome-like particles are involved in
the transfer of P25b from the bovine epididymal fluid to the
sperm surface. Mol Reprod Dev 2001;59(1):115– 121.
39. Eickhoff R, Wilhelm B, Renneberg H, Wennemuth G, Bacher M,
Linder D, Bucala R, Seitz J, Meinhardt A. Purification and
characterization of macrophage migration inhibitory factor as
a secretory protein from rat epididymis: Evidences for alter-
native release and transfer to spermatozoa. Mol Med 2001;
7(1):27– 35.
40. Yanagimachi R, Kamiguchi Y, Mikamo K, Suzuki F, Yanagima-
chi H. Maturation of spermatozoa in the epididymis of the
Chinese hamster. Am J Anat 1985;172(4):317– 330.
41. Frenette G, Lessard C, Sullivan R. Selected proteins of
‘‘prostasome-like particles’’ from epididymal cauda fluid are
transferred to epididymal caput spermatozoa in bull. Biol
Reprod 2002;67(1):308– 313.
42. Rejraji H, Sion B, Prensier G, Carreras M, Motta C, Frenoux JM,
Vericel E, Grizard G, Vernet P, Drevet JR. Lipid remodeling of
murine epididymosomes and spermatozoa during epididymal
maturation. Biol Reprod 2006;74(6):1104– 1113.
43. Fornes MW, Barbieri A, Cavicchia JC. Morphological and
enzymatic study of membrane-bound vesicles from the lumen
of the rat epididymis. Andrologia 1995;27(1):1– 5.
44. Carlsson L, Ronquist G, Eliasson R, Egberg N, Larsson A. Flow
cytometric technique for determination of prostasomal quantity,
size and expression of CD10, CD13, CD26 and CD59 in human
seminal plasma. Int J Androl 2006;29(2):331– 338.
45. Trajkovic K, Hsu C, Chiantia S, Rajendran L, Wenzel D, Wieland
F, Schwille P, Brugger B, Simons M. Ceramide triggers budding
of exosome vesicles into multivesicular endosomes. Science
2008;319(5867):1244– 1247.
46. Arvidson G, Ronquist G, Wikander G, Ojteg AC. Human
prostasome membranes exhibit very high cholesterol/phospho-
lipid ratios yielding high molecular ordering. Biochim Biophys
Acta 1989;984(2):167– 173.
47. Carlini E, Palmerini CA, Cosmi EV, Arienti G. Fusion of sperm
with prostasomes: Effects on membrane fluidity. Arch Biochem
Biophys 1997;343(1):6– 12.
48. Bauvois B, Dauzonne D. Aminopeptidase-N/CD13 (EC 3.4.11.2)
inhibitors: Chemistry, biological evaluations, and therapeutic
prospects. Med Res Rev 2006;26(1):88– 130.
49. Utleg AG, Yi EC, Xie T, Shannon P, White JT, Goodlett DR, Hood
L, Lin B. Proteomic analysis of human prostasomes. Prostate
2003;56(2):150– 161.
The Prostate
Prostasomes Structure and Composition 9
