Factors in Semen Change the Dynamics of Sexually Transmitted Viral Infections

Factors in Semen Change the Dynamics of Sexually Transmitted Viral Infections


[MUSIC PLAYING] LUCIA VOJTECH: Thank you very much for the
invitation to talk today about my work looking at how semen is immunomodulatory in the recipient
genital tract, with a special attention to how this changes the dynamics of sexually
transmitted viral infections. So I have no conflicts of interest to disclose. I’ll start with some goals for today’s talk. My main goal is just to get everyone to think
broadly about how semen impacts immunity in viral transmission in the recipient genital
tract. Because this has important implications for
a number of different biological systems. So this has implications for the design of
therapeutics against sexually transmitted infections like HIV, where it’s really important
to consider the role of semen. So in sexually transmitted infections, semen
is almost always present during transmission of the pathogen. And it’s a very immunoactive biofluid. So we need to consider what it might be doing
when we’re thinking about how to protect against STIs. It has important implications for the detection
of infectious pathogens in biological fluids. So for example, if you can pick up something
by, say, an RNA assay in semen, that doesn’t necessarily mean it’s going to be infectious. And that’s partly because the man has a really
strong effect on the infectious potential of viruses. And this has important implications for fertility
and pregnancy complications, including things like unexplained infertility, recurrent miscarriages,
preeclampsia, and pre-term birth. So in general, what I’m going to go over today
is the components in semen, and think about what the job of semen really is. Most of my work focuses on the extracellular
vesicle fraction of semen. So I’m going to introduce what those are,
and why I’m particularly interested in studying them when I’m thinking about immunoregulatory
factors in semen. I’m going to present some data showing that
Semen Extracellular Vesicles, which I call SEV, impair memory immune responses by affecting
antigen-presenting cell function. I’m going to go over a little bit of the RNA
cargo carried by our semen extracellular vesicles. And I’m going to show some data about the
direct effects of SEV on sexually transmitted viruses, including HIV and Zika virus. So it’s been recognized for quite a long time
that semen impacts immunity. And this makes a lot of sense in light of
an evolutionary pressure to support conception, and to decrease any potential immune responses
against semen or fetal antigens. And it’s true that semen is overall tolerizing
against co-delivered antigens, like paternal antigens, and any conceptions that result
from that specific partner. So the Robertson group in Australia has done
a lot of elegant mouse studies showing that exposure to semen induces a small pool of
antigen-specific Tregs in the periphery in mice. And this is without the need for conception. And it’s thought that this small pool of antigen-specific
Tregs is then poised to expand upon successful conception, and really helps support the development
of pregnancies. And there’s a lot of epidemiological evidence
in humans that exposure to semen is important in reproductive health. So there is a condition called preeclampsia
that can arise during pregnancy. And it’s thought to arise primarily due to
an aberrant immune response to fetal cells during placentation. And a number of studies have demonstrated
that motor exposure to a specific partner’s semen– so longer-term exposure during unprotected
sex to a specific partner’s semen is very protective against developing preeclampsia. This is all suggesting that the immunotolerizing
affects of semen are very important for fertility and conception. Now semen is a complex fluid. It contains a cellular fraction. And we’re all familiar with sperm cells in
semen. There are also germ cells and leukocytes,
and the concentration of leukocytes in semen can vary a lot according to men and their
health status. There is a soluble protein fraction in semen,
and this contains things like cytokines, chemokines, other bioactive proteins. There are bioactive lipids, like prostaglandins,
that are present in very high concentration in semen. There are RNA protein complexes, and there
is an extremely high concentration of extracellular vessels, which themselves contain all of these
different types of cargo as well. Effects of each of these different fractions
are very multifaceted. They probably target different cell types
in the recipient mucosa, and they do a lot of different things. They may impact on multiple different outcomes
in the recipient mucosa. So we believe that it’s important to study
these fractions sort of separately to understand mechanistically what they’re doing, and then
we can put together a whole holistic picture of what semen does in the recipient mucosa. So despite the fact that semen is immunosuppressive
and tolerizing, studies on how this impacts immunity and immune responses in the recipient
mucosa are few and far between. A few studies have looked at unfractionated
seminal plasma and isolated extracellular vesicles– which are sometimes called prostasomes
in the literature– and shown that they have some sort of general immunosuppressive properties,
like inhibiting NK cell function, or phagocytosis, or inhibiting neutrophil degranulation. But in general, these studies are older, few
and far between, and mechanisms aren’t well-studied. So this immunosuppressive property of semen
might also be really important in infectious disease. For example, in couples with an active human
papilloma virus infections, if you can get them to start using condoms, thereby reducing
exposure to semen, you can get a regression of infection, and even regression of early
cervical lesions in the female partner, implying that the immune response to HPV functions
better in the absence of exposure to semen. This also might help to explain why we have
HIV vaccine candidates that seem to elicit strong HIV-specific T-cell responses when
we measure them in the blood, but of yet, these vaccines have failed to protect against
sexually transmitted HIV. So as I mentioned, my work is primarily on
the extracellular vesicle fraction of semen. So I’d like to give a quick overview of what
this is. So extracellular vesicles is a broad term
that encompasses both exosomes and microvesicles. Exosome and microvesicles in general are lipid-bound
little subcellular particles that are around on the order of 40 to 200 nanometers typically. The difference between exosomes and microvesicles
is their origin in the cell. So exosomes are thought to arise from an endosomal
pathway. So you get formation of endosomes by budding
of the plasma membrane, and then inward budding of this endosomal membrane forms an exosome. At this point, the exosomes can be loaded
with contents from the cell cytoplasm or transmembrane proteins from the endosomal membrane. In formation of these multivesicular endosomes,
some of these go on to fuse with lysosomes, and this is just degraded contents that the
cell can recycle. But many of these multi-vesicular endosomes
can go on to fuse with the plasma membrane, and release the burst of exosomes into the
extracellular space. Microvesicles, on the other hand, are thought
to be a little bit larger, and thought to arise primarily from outward budding of the
plasma membrane. Now as I mentioned, these vesicles– both
types, exosomes and microvesicles– contain cargo that’s loaded in their cell of origin. So they can have transmembrane proteins. They have an interesting lipid makeup that’s
different from the plasma membrane of the cell. For example, they have exposed phosphatidylserine. They can carry mRNAs, including long noncoding
mRNAs, and long noncoding RNAs and mRNAs that are competent for translation in cells that
take up these exosomes. And they contain– they’re highly enriched
for microRNAs, and they’re enriched for specific types of microRNAs. So if you survey the microRNA content of a
cell versus a vesicle, you’ll typically find differences implying that the loading of these
micro RNAs into vesicles is a very controlled process. So why are we studying the EV component of
semen? The original reason came many years ago from
my mentor, Florian Hladik, who was studying HIV transmission, and doing a lot of transmission
electron microscopy studies. And he observed often that HIV variance–
so here, with the caps, and there are caps that you can see– were very often found bound
to, surrounded by, or surrounding these vesicles in these TEM pictures. So that made him start to think about, are
there EVs in semen when HIV is sexually transmitted, and what might they be doing? So we started this project by just sort of
looking at what is the EV content in semen. Out of 23 different samples, we found that
they’re on the order of 10 to the 11th to 10 to the 14th extracellular vesicles per
mL. And EVs are made by every cell type, and you
can find them in all body fluids. So they’re in tears, urine, saliva, blood
plasma, cerebrospinal fluid, all sorts of different body fluids. But the concentration of EVs is about the
highest of any body fluid ever tested. Plasma is pretty close to this concentration,
but it seems that semen has the highest concentration of vesicles in any biofluid. And this is really interesting, because when
we think about viral transmission, we can detect viruses in semen RNA copies. And if we assume that an RNA copy of a virus
corresponds to a variant, in samples of semen with a low viral loads, it’s on the order
of hundreds of copies of RNA. And for HIV and Zika, with semen samples that
have the highest detected viral loads, it’s usually on the order of 10 to the sixth, 10
to the seventh, 10 to the eight variants per mL. And that means that in all semen samples,
these extracellular vesicles are outnumbering variants on the order of 10 to the fourth
to 10 to the eighth. So there are a lot of them in there, and they’re
really outnumbering viruses. So we thought what they might be doing is
really important. And then vesicles in general, or extracellular
vesicles in general, are probably most well-studied in the cancer field. And this is because extracellular vesicles
are very important to cancer– tumor and cancer cells make a lot more vesicles than regular
cells, and they’re very important for remodeling the tumor microenvironment. This figure comes from a review of extracellular
vesicles in the tumor microenvironment. And I just put it up here to point out that
there is a lot of precedent for EVs being important, and being immunosuppressive or
immunomodulatory messengers. So it’s already been demonstrated that they
can induce dendritic cells to become tolerogenic [INAUDIBLE] and induce antigen-specific T-cell
tolerance. They can affect dendritic cell cross-presentation,
and thereby, activation of memory T-cells. They’ve been shown to induce CD4 T-cells to
turn into FOXP3-positive Tregs. They can decrease proliferation of T-cells
and inhibit cytocoxic functions of CD8-positive T-cells and NK cells. So I came into the project with the hypothesis
that EVs in semen impact immunoregulation in the recipient genital tract. And we were particularly interested in understanding
this in context of what are they doing to memory immune responses that we might be able
to induce with anti-HIV vaccines. So I asked the question, how do SEV alter
the function of leukocytes? The main leukocytes I’m interested in are
the antigen-presenting cells, so dendritic cells of the genital mucosa. Dendritic cells in the skin are called Langerhans
cells, and they reside in the epithelial layer. And then there are T-cells in the tissue that
typically reside a bit more deeply, and might not be directly accessed by exosomes from
semen. So we are interested in looking at whether
SEV impede memory T-cell immune responses, because as I mentioned, we’re interested in
the context of vaccines against HIV. But people who have had HIV vaccines are very
few and far between. So we use model antigens in the study, CMV
or EBV peptides or proteins. So these are viruses that the majority of
people have been exposed to, and will have memory T-cell immune responses. So we can just take typical blood donors,
add in these antigens, and then look for a memory T-cell immune response by production
of cytokines by a T-cell. So we just ask the question, what happens
when we throw SEV into this mix? And what we saw was that SEV significantly
decreased cytokine production by both CD4 and CD8 memory T-cells. So here we’re plotting on the y-axis. This is percent of cells responding to the
antigen, either CMV or EBV antigen. And it’s cultured along with antigen here,
and with SEV. And we saw in every case that the presence
of SEV inhibits the production of cytokines. OK, so it seems that SEV do impair memory
T-cell immune responses, but that was a read-out of T-cell function. But we had a lot of reason to suspect that
it wasn’t actually T-cell function that was being affected directly by SEV. But that was partially from some studies we
did where we were trying to look at what sorts of cell types were binding to or taking up
our SEV. So for these experiments, we have dendritic
cells labeled with a DC marker that are derived from blood. Or we have ex vivo vaginal tissues that we’ve
chopped up, and then we isolated the cells that migrate out of those tissues, which are
primarily these Langerhans cells, the specialized dendritic cells of the skin. And they’ve migrated out of tissue in conjunction
with a T-cell in this case. So you can see the LC bodies labeled with
our fluorescent marker here, and we see the nucleus of the T-cell, but not the cell body. So these cells were exposed to SEV that were
labeled with a fluorescent dye called the DII. And what we see is that the DCs and the LCs
readily took up the EVs, and turned red. But we never saw any evidence that a T-cell
was taking up an EV. So we can also see this by flow cytometry,
where we do the same sorts of experiments, putting these fluorescently-labeled EVs into
PBMC cultures. And what we saw is that antigen-presenting
cells, monocytes, and DCs rapidly and readily take up exosomes, whereas the T-cells and
the B-cells never did. So that suggests that the SEV are affecting
primarily antigen-presenting cells. And to test this directly, I did this experiment
where I isolated antigen-presenting cells, and loaded them with antigen and SEV, and
then washed so that the only SEV in the system were those that were already taken up by the
antigen-presenting cell. And then I added responder T-cells, some of
which had also been pre-exposed to SEV and then washed, and then again looked for the
production of cytokines as a read-out of memory T-cell immune response. These are the results from these experiments. Down here, I’m plotting which fraction of
cells was exposed to SEV– so either T-cells alone, dendritic cells alone, or both fractions
together. And for our CD8-positive T-cells, we saw what
we expected, that adding SEV to DCs alone impaired responses just as much as it did
in mixed PBMC cultures, so about 30% impairment in production of cytokines when we exposed
DCs alone. And adding SEV to T-cells alone, or to DCs
and T-cells, didn’t really change this impairment. Interestingly, we did not see this for CD4
T-cells. Although in the mixed PBMC cultures we saw
about the same level of impairment, about 34% less cells making cytokines in response
to antigens in the presence of SEV, we never recapitulated this when we add the SEV to
different fractions, implying that they have to be present during the interaction of APCs
and T-cells in order to inhibit CD4 and T-cell responses. So cytokine production isn’t the only important
function of a memory T-cell. It’s also very important that memory T-cells
are able to kill virally-infected cells. So we measured that in a few ways. We did measure direct killing of CD8 cells,
but those experiments are really complicated, and I’m not going to go over them today, in
the interest of time. But CD107a is a marker of degranulation of
cytotoxic T-cells. And the expression on the surface of the cells
means that the cells have a cytotoxic response. So for these experiments, the cells were stimulated
with SEB, a super-antigen that causes between about 5% and 20% of cells to respond with
cytokines or degranulation. And adding a CV here, again, we’re plotting
the percent reduction from SEB alone in SEV-exposed cells. And again, we see a strong impairment in cytokine
production in SEB-stimulated cells. And we see a strong impairment in the expression
of CD107a. And again, for CD8-positive T-cells, we could
recapitulate this response when DCs alone were exposed to SEB. So it seemed clear that SEV are affecting
antigen-presenting cell function. But what exactly are they doing to these DCs? So the first thing I looked for was expression
of classical co-stimulatory markers, which are important to activating a memory T-cell. For these experiments, we have DCs that were
cultured either alone, in the dark gray, or in the presence of SEV, or were given a maturation
cocktail called MCM– either alone, in the light gray, or with SEV, in the gray lines. What we saw is, with the MCM cocktail, as
expected, these markers are upregulated. But the presence of SEV didn’t seem to change
expression in either case. So next, I looked for the key imunoregulatory
enzyme expressed by tolerogenic DCs, called IDO. IDO is a molecule that basically is a great
marker for tolerance in dendritic cells. It catabolizes tryptophan. So in the presence of high expression of IDO,
tryptophan is depleted and T-cells are pushed towards a regulatory phenotype. And what we’ve seen is that exposure to SEV
does cause marked upregulation of IDO in DCs. So here, at the RNA level, we looked at six
hours of culture with SEV, or 20 hours, and we saw 10 to over 100-fold upregulation of
IDO at the RNA level. But we’ve also seen this at the level that
protein expression. So this is looking for IDO expression by intracellular
staining in DCs, either mock or exposed to SEV, and we see high expression of IDO in
SEV-exposed DCs. So another pathway of tolerance that we’re
looking at in SEV-exposed DCs is metabolism. So one key difference between tolerogenic
and immunostimulatory DCs is their metabolic phenotype. In dendritic cells, activation or maturation
shifts metabolism to glycolysis. Whereas in contrast, tolerogenic DCs don’t
shift to glycolysis, and they preferentially utilize oxidative phosphorylation or fatty
acid oxidation for energy production. So to look at if exposure to SEV is shifting
metabolism in our DCs, we used a Seahorse XF analyzer that measures the ECAR– extracellular
acidification rate– that’s plotted on the y-axis here. It’s a proxy for glycolysis. So we have our DCs alone here, and our DC
is treated with a maturation stimulus. In this case, we used a polyIC, a TLR ligand
that will activate DCs. And as expected, we see this upregulation
in glycolysis. But in the DCs that were treated with SEV
before adding polyIC, they were completely unable to upregulate their glycolysis rate. And we also see this over here in this experiment,
where we’re looking at DCs that were cultured alone or in the presence of SEV. And this is baseline ECAR rate. So first off, the DCs with SEV had a lower
baseline glycolysis rate. And then when we inhibit mitochondrial ATP
synthesis with this chemical called FCCP, it causes cells to shift entirely to glycolysis. It can’t make ATP with mitochondria anymore. So this is a measure of maximal respiration
rate. Cells that were treated with SEV were completely
unable to upregulate glycolysis to compensate for this. So to summarize this part of the talk, extracellular
vesicles are present at very high concentrations in semen and rapidly enter antigen-presenting
cells. SEV impair the stimulation and activation
of T-cells primarily by impairing antigen-presenting cell function. And I have some other lines of evidence that
I didn’t go over today that demonstrate that for CD8 T-cells, this is likely by inhibiting
cross-presentation of antigens. And I’m actively looking at what the phenotypes
in DCs other than IDO expression or metabolic function are altered by exposure to SEV. And I’d also like to look at whether changes
in DCs affect functions other than stimulating memory T-cell immune responses. OK, so what I’ve shown you so far is that
SEV impede DC function. But what is it about SEV that might be mediating
this effect? So as I mentioned, like every other extracellular
vesicle, they carry a lot of proteins, lipids, and nucleic acids. And perhaps the most-studied cargo in EVs
in general is RNA. And partly, this is because RNA has a great
potential for use as a biomarker in EVs, because it’s amplifiable, and you can detect it at
very low levels. So there’s a lot of interest in using EVs
in circulation as markers for cancers, because cancers release extracellular vesicles into
circulation. You can isolate extracellular vesicles, and
find differentially-expressed RNAs to detect cancer. But whether or not these RNAs are actually
functional in cells that take up EVs was a little unclear for a while. But I would say that in recent years, a number
of papers– very convincing papers in many different fields– have come out, demonstrating
that these RNAs actually are very functional. So these are just some random examples that
I pulled from literature in completely different systems– stem cell, cardiovascular biology,
glioblastomas, and even plants use RNAs and some extracellular vesicles as functional
messengers. So we decided to sequence the small RNA cargo
in our semen extracellular vesicles. And this is a bioanalyzer trace of the total
RNA we’ve isolated from SEV. And like other vesicles, it’s highly enriched
for small RNAs. So we have some here around the 20-nucleotide
region that are probably microRNAs. And then we had a large peak around 75 that
we suspected was probably tRNA. So we sequenced these two different-sized
fractions from 15 to 40 nucleotides and 40 to 100 nucleotides in six different semen
donors. And this is a snapshot of the sequencing results,
the most broad-range, so what sort of RNAs are present. And as we expected, there are full-length
tRNAs here in the 40 to 100 nucleotide library. There were also a lot of yRNAs, which is a
regulatory RNA. It’s about 90 nucleotides, and it’s important
for DNA replication, but it’s really not studied very much. And very little is known about yRNA. In our smaller libraries, we also found a
lot of degradation products of these yRNAs. As expected, we found mature microRNAs, too. And I’ll just zoom in on these a little bit,
because the microRNAs are an interesting cargo because they’re very regulatory in target
cells. When we look at the top 15 most abundant microRNAs
that are present in our SEV, they account for over 90% of the reads. We believe that we can just look at these
top 15 most abundant, and those are the ones that are actually present in enough volume
to affect targets in a recipient cell. And we see that many of these microRNAs have
validated immune-related microRNAs or messenger RNAs that they do indeed target. So it could be that these microRNAs delivered
by a SEV are regulating targets in the cells, but I haven’t yet followed up on proving that
point. I’d like to just go over one other type of
small RNA we found in our sequencing study, because I thought it was really interesting. And that is that we found a lot of these tRNA
fragments in our smaller libraries. It was responsible for an even higher percentage
of the sequencing reads than the mature microRNAs, which I found very surprising. So I started to read more about tRNAs, and
discovered that transfer RNA-derived fragments is actually– generation of these tRNA fragments
is a highly-regulated process. So stress cells will activate enzymes that
can specifically cleave specific tRNAs at specific locations. And it seems to be important for a stress
response of the cell. And when I look at the sorts of distribution
of tRNAs that we saw in our two different libraries, it seems that there is an interesting
pattern of loading into the vesicles. So if we expected that it was just random
degradation products getting packaged into an EV for, say, just cellular garbage or something,
we would expect that the distribution of isoacceptor types would be the same between the full-length
tRNAs, the fragmented tRNAs. And that wasn’t the case at all. We saw a quite different distribution of the
types of tRNAs that were fragmented or full-length that are carried by our vesicles. And when we map back the tRNAs to their full-length
precursors, there was this interesting pattern where the tRNA fragments that were most prevalent–
so the glycine and alanine isoacceptor types– were entirely the five prime ends of these
tRNAs. 99 percent of the reads map to either a 19-nucleotide
fraction for tRNA alanine or a 28-nucleotide fraction with the tRNA glycine, whereas tRNA
fragments that were less prevalent in the EVs, some of them had these patterns that
were more indicative of sort of random degradation products, implying that these are kind of
specifically loaded into the EVS. And interestingly, a few papers have shown
that these five prime tRNA halves of exactly the same lengths that I’ve seen inhibit protein
translation in cells. So I’ve done one experiment to look at this. And for this study, I used THP cells, which
is a monocytic cell line. It’s very happy to take up SEV. And they were transfected with a rinella luciferase
expression plasmid, and then treated with SEV. And we looked for the presence of the expression
plasmid by QPCR, and all the cells got about the same amount of plasma transfection. But when we looked for the production of protein
by reading out luciferase expression, we saw that the cells treated with a high dose of
the SEV were very impaired in producing rinella luciferase, implying that protein production
is indeed impaired in cells that take up a lot of EVs. OK, so to summarize this part of the talk,
SEV, like other extracellular vesicles types, are highly-enriched for small, potentially
regulatory RNAs, including mature microRNAs known to target immune-related mRNAs. We have not yet proven that these cargo RNAs
are delivered to the cytosol of important cells in sufficient quantity to actually regulate
cellular targets. tRNA and yRNA fragments are commonly found
in extracellular vesicles. So our study was one of the earlier ones,
but many, many sequencing studies of EVs have subsequently come out, and found very similar
things that tRNA fragments and yRNA fragments seem to be generated in a very controlled
way, and loaded into vesicles specifically. And what the biological meaning of that is
remains to be seen. And whether or not they’re packaged into EVs
to functionally impair protein production recipient cells, or if it’s just a cell getting
rid of garbage, we don’t yet know. So SEV impair memory immunity. This has important implications for vaccines
against HIV and other STIs, but it’s also interested in looking at how SEV might be
directly impacting viral infections. So I started doing a little bit of work in
HIV. And for these experiments, I use the TZM indicator
cell line. So this is a cell line that expresses HIV
receptors. So it can be infected by HIV, and when infection
with HIV happens, it turns on luciferase. So it’s a very easy way to detect infection. And I’ve done experiments with SEV and HIV
in these cell lines in a couple of different ways. So one way that sort of mimics what would
be biologically relevant is pre-incubating SEV and virus together before adding to the
cells. I’ve also done these experiments where I incubate
SEV and cells together before adding virus, either with washing off or not washing off
the SEV before adding the virus. And these are the results from those experiments. So in a few different experiments, we saw
that increasing the dose of SEV, we get increasing impairment of HIV infection in the TZM indicator
cells. This was also true when EVs were added to
the cells before adding virus, if they were left there. If the extracellular vesicles were washed
off before adding virus, the impaired effect is kind of only evident at the highest dose
of SEV. So the greatest level of impairment occurs
when the SEV and virus are incubated together before adding to the cells. So that’s fine, but there is another group
who is very actively working on HIV and SEV infection, and they’ve made a lot more progress
in terms of discerning the mechanism of this and things. And if you’re interested, I will point you
to their publications. And I became more interested in studying how
SEV might be impacting Zika virus at the start of the Zika epidemic a few years ago. So I had a project looking at modeling sexual
transmission of Zika virus in the female genital mucosa. And when I started, although there was really
a clear and convincing evidence that Zika can be transmitted sexually, what cell types
were initially infected in replicating virus in the female mucosa was not yet known. So just to convince you that Zika does infect
human epithelial– vaginal epithelial cells– I’ll tell you about our model system. So we use primary human epithelial cells in
ex vivo vaginal tissues. So we get surgical scraps from vaginal repair
surgeries, and from hysterectomies, so we can get cervical tissues as well. And we can use these tissues either as biopsy-sized
pieces to do experiments, or we digest them, and do some cell culture tricks to turn them
into primary and transformed epithelial cell lines from vaginal epithelium, endocervical,
or ectocervical. And all these three tissue types have distinctly
different epithelial cell phenotypes, so it’s important to look at all three. So my first experiments we’re just looking
at whether these cells are infectable by Zika virus. Here, I’m going to present data just looking
at RNA viral load in these cell types. And we saw indeed that Zika does replicate
in genital epithelial cells. It takes about two days for replication to
really occur, but it occurs in all three of the cell types we’ve looked at, and to varying
levels in different donors. I looked at infection in a number of other
ways as well, like immunofluorescent staining for viral proteins, and production of progeny
variants and [? supernines, ?] and those all correlate really well with RNA levels in the
cells. So I’m only presenting RNA today. So again, we did the same sorts of experiments
as HIV with SEV and Zika, so pre-incubation variants with SEV, and then adding to cells,
and looking for productive infection by RNA viral load. And here’s an example from one experiment. So we’re reading viral load by digital droplet
PCR in the cells. And what we saw here is, Zika alone, we got
good infection level in the cells. When the variants were pre-incubated with
SEV, we got strong impairment, and when they were pre-incubated with 100-fold less SEV,
there was no impairment. And this is a summary of all the experiments
we’ve done thus far with SEV and Zika infection. So the e6 is 10 to the sixth SEV per PFU of
Zika, and in every cell line we’ve studied thus far, we get over 50% inhibition of infection,
so strongly-impaired infection in the presence of this many SEV. As we decrease the level the ratio of SEV
to [INAUDIBLE], we see decreasing levels of impairment. And we plan to continue studying lower and
lower doses. And this is really interesting in the context
of this paper that came out earlier this year that was a systematic survey of Zika virus
detection in the semen of symptomatic infected men. So they collected a lot of symptomatic men,
and looked at semen shedding over months and months. And they found that finding Zika iris RNA
was extremely common in these men. And it persisted for a long time, and sometimes
more than six months. And the novelty of this study is that they
actually took a lot of these semen samples and tried to culture on infectious virus. And interestingly, in only three of 78 samples
with detectable Zika virus RNA were they able to get transmission of infection. And these three were all samples that had
more than 7 log copies of RNA per mL of semen. So it’s suggesting that you needed an extremely
high viral load in semen in order to get sexual transmission. And potentially this is because SEV are so
impairing to infection. So to summarize this part of the talk, SEV
impairs HIV and Zika virus infection. And pre-incubating variants in SEV is the
most effective way to inhibit infection. And this is the best mimic of what would be
a true biological situation. And it suggests the effect isn’t entirely
on the cell, that there is some level of occupying receptors or hiding a virus from being able
to bind its receptors on a cell. Some preliminary evidence suggests that there
is this dose threshold SEV per virion, below which– as in the case of samples with very
high viral loads– the virus can overcome this SEV-mediated inhibition and actually
sexually transmit. What we don’t know is the mechanism of this
effect. Are the SEV preventing binding in initial
infection? Are they redirecting trafficking of virus? Are they somehow altering production of progeny
variants? And these are questions that we’re very actively
researching right now. Other questions I’d like to address as we
continue is whether SEV changes the immune responses to virus, and whether other components
of semen like the SEV-depleted seminal supernatant have similar or different sorts of effects
on viral infection. OK, so for the last part of my talk, I would
just like to introduce an idea that I’m thinking about, and haven’t done a lot of active research
in yet, but I think it’s really interesting. And that’s the idea that variation in immunosuppression
mediated by semen might help to explain some fertility and pregnancy complications. And this arises from sort of looking at all
of the experiments we did that I presented first that were looking at impairment of T-cell
responses in the presence of SEV. So here we’re plotting, just in general, the
percent reduction from the no-SEV exposure conditions for all the different antigens
we’ve tested, and for all the different assays– so cytokine production, CD107a expression,
or direct killing by cytotoxic T-cells. And what we saw was this interesting effect
where donors– a PBMC donor or SEV recipient, and we observed that there were certain recipients
who had less impairment in every assay we tested over here, in certain recipients who
were highly impaired in every assay we tested. And then there were some who it seemed sort
of assay dependent, like this person’s killing was highly impaired, but maybe cytokine production
was much less impaired. And that’s really interesting to think about,
because when you think about it, exposure to semen and to alloantigentic fetuses really
represents a remarkable immune tolerance. In any other context, exposure to cells from
another person would be highly immune-inducing, but it’s not for semen. Allergies to semen are extremely rare. So something about semen and the female genital
mucosa is just exquisitely fine-tuned for tolerance, specifically to paternal antigens. However, it appears that there is some recipient-level
variability in susceptibility to SEV-mediated immunosuppression. And thinking about this is really interesting,
because roughly a third of cases of infertility are currently unexplained by a diagnosable
medical cause. The risk of pregnancy complications increase
in pregnancies resulting from assisted reproductive technologies where semen isn’t present, or
with less exposure to a specific partner’s semen. And most studies have implicated soluble factors,
which include SEV in semen, as essential to generating alloantigenetic-specific Tregs,
but the mechanism of this is not well understood. So we really think that by studying the mechanism
of how semen induces specific tolerance in recipient cells, it could really inform maternal
fetal medicine, and maybe reveal some new strategies to address these issues. It has important implications for infectious
disease, of course. But it also might just generally reveal new
strategies and pathways that we could target to therapeutically manipulate cells like APCs
to turn them tolerogenic, and treat diseases like autoimmune diseases. OK, so with that, I will give my overall summary. My work focuses on how semen impacts immunity
and viral pathogenesis. And again, this has important implications
for vaccines against STIs. We need to think about overcoming impaired
antigen-presenting cell function when semen is present. So we need to think about therapies, strategies,
vaccines that will overcome this immunosuppressive effect. Semen has direct effects on viral infection. And it’s important to remember that a simple
analysis of an RNA viral load in semen, a very bioactive fluid, will not tell the whole
story of the infectious potential in the context of sexual transmission of that virus. And this has important implications for fertility
and pregnancy. Does variation in semen-mediated immunosuppression
in particular partner pairs contribute to unexplained infertility, recurrent miscarriage,
preeclampsia, pre-term birth, et cetera. And this is a research direction I hope to
pursue in the future. And with that, I’ll make my acknowledgments. And I’d like to especially acknowledge my
mentor, Florian Hladik, who originated a lot of these projects, and has been a wonderful
mentor and colleague. And we hope to continue to work closely together
in the future. All the people in his group, current and past,
who have helped with these projects. And if you want to see what they’re up to,
they have a pretty neat website, hlab.science. Any questions or comments– and if anyone
is working in EVs or interested in working in EVs, I used to run a little EV interest
group that we would love to get going again. We just talk about techniques, and how to
work with these vesicles, because some of the experimental approaches are a little bit
difficult, and it’s really nice to talk about with others working in this field. I’d like to acknowledge the people in my department,
and other collaborators who have helped with this. And these are the surgeons who provide us
with tissues from vaginal surgeries and hysterectomies, which is very, very important to our work. We couldn’t do it without them. And of course my funding sources. And I would be happy to take any questions. [APPLAUSE] Yeah. SUBJECT 1: Where do you think– [INAUDIBLE]
where do you think the extracellular vesicles are coming from within the male anatomy? Do you think that vesicles coming from different
parts of the male anatomy could lend different immuno– LUCIA VOJTECH: Definitely. I think that’s definitely true. So they’re largely coming from the prostate
gland. And in literature, and the papers from the
’70s and ’80s, people who originally described vesicles in semen, they called them prostasomes
because they thought they were arising entirely from the prostate gland. But you can look at RNA markers and things
and find them from all tissues in the male genital tract, including MHC II, implying
that there are some leukocyte-derived vesicles in there as well. And I think it’s entirely possible that they
have different functions. I mean, one way we could study that is isolating
vesicles from isolated tissues or cell types, and seeing if they do the same thing. SUBJECT 1: Are you able to sort– LUCIA VOJTECH: It’s really– no, basically. Flow is really– these are smaller than light,
so it’s not trivial to do the flow cytometry. It takes a dedicated machine, and a lot of
expertise, and flow technology is not there. You can do some sort of immunoselection sorting
if you have a specific marker, like a protein marker. You can use antibodies. But that’s about the level of sorting right
now. SUBJECT 2: So my question is about in the
presence of infection. Do you think that the number or the cargo
within the vesicles is changed, and that that– LUCIA VOJTECH: That’s a great question. So we haven’t looked at that. We’ve been only looking at healthy men. We do have a project looking at opioid users
and semen. We’re going to look at extracellular RNA profiles
in vesicles and non-vesicular fractions in opioid users. And I think that will be really interesting
to see if there are big differences there. But we don’t know. I haven’t looked. I would suspect in infection, you probably
would find a lot more leukocyte-derived vesicles. SUBJECT 3: Thank you for your talk. This is super-interesting. There’s been a lot of work recently in resident
memory cells. And what this talk makes me think about is
that in a young woman versus a sexually-active woman, there’s a window in which to vaccinate,
but might the presence of semen later in life undo any good that could have been done with
a tissue-specific vaccine? So with HPV, or herpes, or whatever, you could
vaccinate a young woman. Do you have any way to figure out whether,
if we make a good vaccine and give it to teenagers or whatnot, that it will persist into adulthood? LUCIA VOJTECH: No, but I would say– I mean,
the HPV vaccine, which is mostly antibody-based does work. So there must be ways to induce immunity that
it does protect against viral infections. I just think we need to think about it. And You know, I think if you had a really
strong, robust, tissue-resonant immune response, you probably would be overcoming the effect
of semen. Because it’s probably relatively transient. These things are probably cleared within 24
hours or something of sexual exposure, and the virus would be there much longer. So– SUBJECT 4: Do you think that this effect could
be used to counsel patients, either with infections or with infertility, as to kind of stack the
odds in their favor for clearing infection, or for increased fertility? LUCIA VOJTECH: Yeah, I think so. I don’t know if our knowledge is at the point
to make clinical recommendations like that. But I think for, like, HPV, it is clear that
if you have active HPV infections, using condoms really helps. That’s partially probably due to just less
exposure to virus if you’re trading virus back and forth, but also increasing good immune
responses. In terms of infertility, that’s just all hypothesis
right now. So actually a lot of– some types of assisted
reproductive technologies actually will add back components from semen because they have
seen that it does have a positive effect. And so that is being done. [MUSIC PLAYING]

Leave a Reply

Your email address will not be published. Required fields are marked *