Zika Virus Infection – Lark Coffey

Zika Virus Infection – Lark Coffey


– Thanks and thank you for the
invitation to be here today. So as I’m sure you’ve heard by now, Zika is an emerging mosquito-borne virus that in its urban cycle is transmitted between Aedes species
mosquitoes and humans. Aedes can live in close contact with us. They preferentially feed on human blood. They can live in our house and leave to lay their eggs in standing water outside human residences. As you also know they’re now understood to be other less common ways
that Zika can be transmitted, vertically from mother to
child during pregnancy, through sexual contact, and through transfusion transmission. So Zika historically is an African virus that was first isolated in Uganda in 1947. And then it caused sporadic and poorly-described
human disease starting, the first human disease was recognized in west Africa in the 1960s. So from the ’60s until about 10 years ago, Zika caused small and focal
outbreaks in Africa and Asia. But as you know starting in 2014, Zika was introduced into the Americas. This is the current state
of Zika transmission where all of the countries
that are highlighted in orange are places where there’s
active Zika virus transmission. The virus has been introduced
into 48 new countries and territories in the Americas. So you might be looking
at the orange color of all the continental US
and scratching your head. That is I think a somewhat
artificial representation, and it just reflects CDC labeling scheme where there have only
been local transmission in mosquitoes in people that
don’t have a travel history to an endemic area in two
places in the continental US, South Florida and Miami last summer where there were 218
clinically-confirmed cases, and then in Brownsville,
Texas also last fall. Looking more close to home, CDPH has recorded that there have been 541 confirmed human cases in travelers returning to California
from a Zika-endemic area since 2015. You’ll notice that the volume of cases parallels the major population centers. The concerning feature here is that the counties that I’ve highlighted in the red outlines are where the invasive mosquito vectors, Aedes aegypti and Aedes albopictus that have been historically implicated in Zika transmission cycles elsewhere are also reported. So CDPH and the Mosquito
and Vector Control districts of California with whom we work closely are concerned about
setting up the situation for the perfect storm which is the establishment of local mosquito-borne transmission in the state. So here are maps from last summer in southern California on the left is LA and on the right is San Diego. What you’re looking at is
the co-occurrence in space of either of the two mosquito vectors shown in the color circles, or Zika cases in travelers
shown as the diamonds. So what happens when there is a Zika case that’s reported to CDPH, the Mosquito and Vector Control districts go out to the case’s residence, try to kill adult mosquitoes in the area and then remove the
standing water containers to eliminate breeding sites. So at UC Davis, my colleagues are involved in statewide surveillance of arboviruses that are in mosquitoes. So Mosquito Control districts send their speciated
mosquitoes to UC Davis where they’re pooled and then we test them by an RT-PCR for the three viruses that are endemic in the state that are human and animal pathogens Those include West Nile, Western equine encephalitis virus and Saint Louis encephalitis virus. And now we also test in the Aedes pools for the exotic viruses
that we’re worried about being introduced into the state, those include Dengue,
chikungunya and now Zika. So speaking of the media, which I think the panel
was really informative and related to this. So a year ago, the vector community was thrown into controversy when a press release was
issued from a Brazilian group based on data that was not published, so not available to the
scientific community, but that indicated that
it wasn’t Aedes mosquitoes that were being found and infected in outbreak settings in Brazil. Although some were infected, what they found was that there is actually a higher infection rate
in Culex mosquitoes. Culex are a different genus
of mosquitoes than Aedes, so this was sort of unprecedented. Even though by arboviruses can switch their use of vectors, such a large switch it
was sort of unexpected. What is called for in the vector community was to ask which mosquitoes are actually responsible for transmitting Zika? And this is obviously really
epidemiologically-relevant because if you want to
target vector control without a licensed vaccine, you have to know which species
are transmitting the virus, because mosquitoes have
different ecologies and different feeding behaviors and they also have
different distributions. So on the left I have some
stats for the two Culex species that are most abundant in North America. And then on the right, I have the two established Zika vectors in other areas of the world. Those mosquitoes also vector
Dengue and chikungunya. So you see Aedes are more
active in the daytime. They preferentially feed on people. Whereas the two Culex vectors tend to feed more at dawn and dusk, and they have a more
general host preference. One reason why it’s important to know whether North American
Culex can transmit Zika is you’ll notice that
their range extends much, especially for Culex pipiens much further north than the Aedes species. So when that press release came out, a lot of the the Mosquito
and Vector Control districts around the United States
expressed a lot of concern in understanding whether
there was a big host switch. So how do we know which mosquito species are competent vectors? As you know there are hundreds of different mosquito
species around the world. And to the naked eye
they look very similar, but they are very different. And they’re not all
capable of transmitting human and animal pathogens. So there are four criteria
that a mosquito species has to fulfill in order to be considered a competent vector. They have to be abundant, they have to survive
long enough to transmit, they have to feed frequently
on competent vertebrate hosts. So these are criteria
that we can establish in field entomological surveys. The fourth criterion is the
one that we’ve been looking at which is that you can show
in a laboratory setting that certain species are capable
of transmitting the virus. This is a cross-section of a mosquito. She ingests an infectious blood meal that goes into the mosquito
stomach or the midgut. The virus has to infect
the midgut epithelium and then disseminate onto
the other side of the midgut into the open circulatory
cavity of the mosquito and be in the liquid that bathes it, which is called the hemolith. So that then it can infect
the secondary target organs which in the case of a
transition-competent mosquito would include the salivary glands. So that the virus will then be excreted into the saliva so that when the mosquito refeeds, she transmits into a new host. This period takes a certain
amount of time in a mosquito. So the way we do this experimentally is we take viremic mice that
have been inoculated with Zika, and then are at their peak of viremia. We anesthetize them and
then we present them to cohorts of female mosquitoes
of the different species that we’re interested in studying. Then after the period
of time that the virus moves through the mosquito body, we tear off the legs and the wings of the still living mosquito and we inject the proboscis
into a tube filled with liquid which stimulates salivation. So we’re essentially
milking the mosquitoes. If we detect Zika virus
or RNA in the expectorant, we know that that species
is transmission-competent. There’s another interesting phenomenon that’s been observed which is that within a given mosquito species, there can be regional variations
or geographic variations in the capacity for those
mosquitoes to transmit. So there’s also been an
interest in understanding whether Aedes aegypti from one place is also as transmission-competent as Aedes aegypti from another place. This data synthesizes the field of vector transmission studies where you can see groups
from various different places have been collecting
their local mosquitoes and then looking at the capacity of them to transmit. You can see this data really confirms that Aedes aegypti from all over the world are capable of transmitting Zika as is Aedes albopictus from
the one study in Germany. Looking at the Culex data, you can see almost uniformly that not a single mosquito that was tested and unlike invertebrate studies, we have lots more mosquitoes, so our cohort sizes
are often in 50 to 100. We can say out of the thousands of Culex that have been tested
from around the world, they’re not transmission-competent, save for this one outlier study which I think could be
potentially really interesting. So barring any methodological
issues with the paper, this could be a case that
there is something different about these Chinese Culex
quinquefasciatus mosquitoes that enables them to transmit Zika. Alternately there could
be something different about the Zika virus strain that was used in that Chinese study. It was an Asian lineage genotype virus, the same that’s circuiting in the Americas that was in a traveler returning to China. But overall, the burden of evidence shows that Zika has not shifted to use Culex mosquitoes as vectors. So most likely, what my theory for the data in that
press release was that in the Brazilian outbreak, people were surveying all the mosquitoes that were around houses
and they were by chance just picking up Culex that had recently fed on viremic people, but they were actually just
measuring the whole body, so they were looking at virus
that was still in the midgut of the mosquito or viral RNA there. Another question that
we’ve been interested in, as you know there’s
been a lot of curiosity as to what and what factors have promoted the emergence of Zika now. On the mosquitoes side, one question is whether the spread of
the Asian lineage virus has been promoted by
enhanced transmissibility by the primary vector in most places, Aedes aegypti. To address this question, we took an ancestral
virus in the Asian lineage from 1966 in Malaysia that was low passage and we compared that
in the vector-competent experiment I described to you, side-by-side with the
contemporary Puerto Rican strain. And in related studies
that I’ll also show you from a group at Colorado State, they were interested in the
relative transmissibility of the some of the African lineage viruses with respect to the same Puerto Rican genotype strain that we’re using. They used a Uganda strain from 1947 and a Senegalese strain from 1997. Looking at our data
that’s on the left graph, so these are Aedes aegypti
that we collected in LA, brought into the lab in Davis and tested for their
transmission capacity. What you can see is
that between 40% and 60% of the mosquitoes that fed on the mouse were capable of transmitting between seven and 14 days post-feed, but there was no significant
strain-specific difference. This really refutes the idea that the emerging Asian lineage
Zika is more transmissible than its Asian lineage progenitor strain. To go even further if you look on the right side of the graph where the CSU group was
testing Aedes aegypti collected from Mexico, they actually saw that the
Puerto Rican virus shown in red was significantly less transmissible than either of the two
African lineage viruses. So in fact the emerging viruses are doing worse in aegyptia than the African lineage strains. I want to devote the
rest of the presentation to talking about nonhuman
primate model of Zika that we’ve been developing. So the reasons for having animal models of infectious diseases, I think are pretty obvious to this crowd, but we feel that the placental and neurological development
of nonhuman primates is closer to humans than that of mice. So we’ve been developing a model so that we can study
pathogenesis, transmission, timing of infection which you obviously can’t do in human studies, clinical outcome and then
ultimately downstream look at viral and host genetic
determinants of pathogenesis. All together with a goal
of developing interventions like therapies or vaccines, so that we could either prevent infection or mitigate disease. So even in the last year and half, the Zika non-human primate
field has rapidly advanced and we know a lot already. So I’ve divided these into, this is a summary of the published studies focusing mostly on rhesus macaques. Already this list is not exhaustive because I’ve excluded
cynomolgus macaque studies. But on the top I’m showing you data from the nonpregnant animals and on the bottom the developing studies in pregnant animals. If you inoculate a rhesus
macaque with a Zika, they’ll get a viremia
that endures in the plasma for about three weeks, sometimes shorter, neutralizing antibodies
starting on day five. Like most humans, they won’t
to show clinical disease, but they’ll shed viral RNA
in various different fluids and viral RNA will be detectable
in many different tissues. And in our studies where we looked at, in our preliminary studies where we looked two weeks post-inoculation, there is a hemolymphatic tropism. We also saw that in some animals, there’s Zika RNA in whole
blood longer than in plasma. So the second batch of studies were using rhesus macaques for testing various
different vaccine platforms. And they all found that
each of these vaccines are protective and immunogenic
and relative and safe, and that they conferred
sterilizing immunity. They also did some
passive transfer studies and found that that
confers protection as well. So moving on to the pregnancy studies which are more related to the data I’ll present you today, there was an N of one study out of an Oregon primate research center where they used a pigtail macaque which is a different species, but also an old world macaque. And they found that when they inoculated subcutaneously Zika virus into a GD119 macaque, which means gestation day 119, which is equivalent to the third trimester in a macaque pregnancy, they observed that the fetus developed fetal brain lesions, including ependymal cell
injury which is relevant to, ’cause we found similar
findings in our studies. And then in a publication
that’s not yet out, but is available via preprint, there’s data from the Wisconsin
National Primate Center where they took four pregnant animals in their first and second trimesters and subcutaneously
inoculated them with Zika. They observed vertical transmission in four out of four, but they didn’t see
any fetal brain lesions and they didn’t see
reduced fetal head growth, characteristic of the human microcephaly that we’re trying to replicate. So when we designed our pregnancy studies a year and a half ago, the efficacy of vertical
transmission really wasn’t known, and we figured that in a study where we only had four animals, we wanted to ensure
infection of the fetus. So we decided, and I submit it’s an artificial route, to inoculate the dam intravenously, but also at the same time intra-amniotically inoculate the fetus. Following that approach, we obtained pilot funding from the California Primate Center, and we inoculated two
first trimester animals and two second trimester animals. The full term in rhesus
macaque is 165 days or five and half months. Our plan was at near-term GD155, so 10 days pre-full-term, we would necropsy the animals for extensive optimal sampling collection. And in the interim, we would sample more
regularly in the acute disease or acute infection and then weekly thereafter
until the near-term. So the first thing we noticed
in the first trimester animal inoculated at gestation day 41 was that the fetus died, seven days post-inoculation. So the kinetics of RNA in various different
of the maternal fluids were really unremarkable compared to what we had seen in non-pregnant animals. The viremia was pretty short, but the notable finding
in this animal was that the viral RNA levels and
the infectious virus levels in the amniotic fluid went
up from days two to seven. We also saw very high viral RNA levels in all of the placental
tissues that we collected, as well as all of the fetal tissues that could be differentiated at that time, because in a GD41 or 47 fetus, it was only about three centimeters long. We found corresponding high
levels of infectious virus in all of the placental tissues, but none in any of the fetal tissues. So this is an anomaly compared
to what we normally see, because in all the other fluids in our non-pregnancy studies, we see that there’s generally a genome to PFU ratio
of about 1000 to one. So this was obviously much greater, the fact that we didn’t
see any infectious virus above the detection limit
of our plaque essay. so my speculation for that is that, talking to the pathologists, they said that the fetus was in a severe state of
degradation with autolysis. And the last ultrasound had been performed at five days post-inoculation, so it’s possible if the
fetus died soon after that that it had been decaying for two days which would have a
disproportionate maybe effect on the virus infectivity, but not as much of an effect
on degrading the viral RNA which we still detected. But overall we think that in this animal the fetal death was precipitated by high viral RNA levels
in the amniotic fluid and placental and fetal tissues. So next in these two animals, they progressed to near-term and they were euthanized
according to schedule. But in this GD65 animal, she developed vaginal bleeding at GD103, so about a month after the inoculation. She birthed a viable but preterm neonate, two weeks preterm. So we tracked the head sizes by ultrasound and we saw that the animal
that had the preterm neonate tracked along the two
standard deviation line below the colony average, meeting the clinical
definition for microcephaly. But at birth when we did measurements of the head with respect
to the rest of the body, this animal was not abnormally small. So we think that this was
just a smaller animal, but not a microcephalic infant. We tracked viral RNA levels
in the maternal tissues and we found that they were not detectable by about three weeks, or excuse me, a month. This parallels with what was
seen in the Wisconsin study where they did the peripheral inoculation where they saw that the
maternal viremia was prolonged compared to nonpregnant people, animals, it’s also been seen in people. We saw little blip in one of animals 43 days after the inoculation. We saw the development
of neutralizing antibody starting on day five, IgM at a week, waning at six weeks and IgG from two weeks. Looking at viral RNA levels
in the amniotic fluid, there was this spike in the
animal with the fetus that died, but then the other three animals, the viral RNA decayed over time below the limit of detection in one animal and just above it in this one. This is the animal that
had the vaginal bleeding and so the amniocentesis was stopped because we didn’t want to exacerbate the risk for fetal loss. So I should also mention that we had GD-matched control animals
that were anesthetized and sampled on the same
schedule as the infected animals to control for any non-virus
related manipulation effects. So we did an extensive
tissue sampling and testing from all of the fetuses and
all the dams at necropsy. And I don’t have time
to tell you the data, but I want to highlight what we saw in the fetal and neonatal brains. These are different of peripheral nervous or central nervous system sections and we saw in all of the animals there was viral RNA detectable in some and sometimes many of the sections. We also did in-situ hybridization with Patty Pesavento, Pathologist. She found that there’s labeling of neurons especially in this animal. So the fetal tropism actually is different across the tissue systems in
the fetus compared to the dam where in the mothers, we don’t see any viral RNA in the CNS. So I’m not pathologist, but I want to show you two H&E slides. This is the brain section from the control and the GD65 fetus where there’s a loss of
ependymal cell lining in the inoculated animals. There were foci of mineralization
and increased cellularity that was not observed in the controls that were blindly read
by the pathologists. They also noted that there was the absence of neural precursor cells in all three of the animals that were also blindly scored
compared to the controls. And so what we have here
in just an N of four study is a spectrum of fetal outcomes. In the earliest trimester,
first trimester animal, we see fetal death. In the other three we see
no clinical gross signs of reduced fetal head growth, but we do see histopathological changes. And these parallel what have
been described in humans including calcification, neural progenitor cell death and a similar CNS tropism. So we feel that although
we don’t represent in this small study, the severe outcome of microcephaly, we do have lesser neuropathologic changes consistent with what’s
been seen in animals, in humans, excuse me. So that pilot study has now spawned many more studies that
are currently ongoing. We’re further developing
the fetal macaque model with R21 monies to Koen Van Rompay, the primary collaborator
at the Primate Center. We’re presently testing the NIH vaccine that’s a DNA platform
that’s currently going into phase two trials
in nonpregnant people in Zika-endemic areas. And our preliminary results
in the pregnant macaque show that this vaccine is
going to be efficacious. And then together with the Blood
Systems Research Institute, we have a transfusion transmission study in nonpregnant animals where we want to know how little Zika is sufficient to infect macaques. Are the current RNA detection
methods sensitive enough? And how can pathogen reduction technologies prevent transmission? I should also mention because these animals are precious, we want to maximize the use of them every time we do a rhesus macaque study, so if any of you in the audience are working on Zika and have some macaque samples, tissues, fluids that would be potentially
furthering your research, we’re amenable to sharing. Especially if you tell us in advance so that we can make sure that
the right kind of samples are collected for your needs at necropsy. So I’ve shown you here that
laboratory vector competence implicates Aedes aegypti
and Aedes albopictus in Zika virus cycles in urban settings. This means that we can
target vector control to those species and also hopefully reduce chikungunya,
Dengue at the same time since they use the same vectors. And then the macaque work shows that these pregnant rhesus macaques that are IV- and IA-inoculated produce a similar fetal neural
tropism and neuropathology to what has been seen in humans. So we think this could be a viable model for understanding human fetal outcomes. So last, I’d like to
thank all of the people who participated in the study, especially the group
at the Primate Center. Anil Singapuri who’s my technician who’s been doing most of the Zika testing. And then the two medical entomologists working on the project, Brad Main who’s here
today and has a poster, and Jay Nicholson, as well as a PhD student, Cody Steiner. Our collaborators at BSRI
and Hologic who’s been, who has a very sensitive
RNA detection assay, so they’ve been testing a
lot of these macaque samples, as well as our funding sources. So thanks. (audience applauds) – [Moderator] Thank you very much. We have time for questions. – [Audience Member] Thanks,
that was a great talk. So one question about the
first part of your talk regarding transmissibility in mosquitoes, there was a paper I think
published two days ago in Nature implicating an alanine residue in NS1 in transmission with the
more recent outbreak. Can you comment on whether that alanine is present in the strain that you tested? – I haven’t seen the paper, so what locus? I’m probably not gonna know but, where was the strain from? – [Audience Member] They tested a number of different strains, but they compared it to
the Cambodia 2010 strain which did not have increased transmission. – Well, I’ll check out the paper. – I’ll send it to you.
– I haven’t seen it, so. – [Audience Member] Okay. One more
– That’s the thing with Zika you have to read about three papers a day just to keep on top of it.
(audience laughs) – [Audience Member] One more question. Are there any local species
of Aedes in California that you found to transmit Zika? – Yes, so technically, I know this is probably semantics, but we call Aedes aegypti
and Aedes albopictus, exotic, even though they’ve been in the state and increasing in their geographic spread for up to 20 years. So I would I would argue that those are probably no longer exotic. But both of those occur in
multiple counties of California. I can go back. So even albopictus was
reported two summers ago as far north as Silicon Valley. So all of the counties
that are circled in the red are places where one or
both of the Aedes species that are Zika transmission-competent
occur in California. And in the vector competence
data I talked to you about, we were testing low field
generation mosquitoes from LA. – [Moderator] Okay, question? – [Audience Member] Yes,
thanks for the talk. I guess I’m really curious to know if in your studies with monkeys if you have any insight into the impact of postnatal
infection with Zika. So that’s an area where there isn’t a lot of definite data about, and I think people understand now that prenatal infection will
cause birth defects, but what happens to infants or children who are infected with Zika? And even the CDC doesn’t have a clear, they’re not saying there’s
going to be any impact, but I’ve talked to them, specifically to Eric Zubian and he said within the CDC there are a lot of debates about this. So what is your research
or have you found about what may impact that question? – So to my knowledge, none of the primate research centers of which there are seven in the country have done a study to infect
a neonate or an infant, a non-human primate. But I imagine that’s going to forthcoming. What we are doing in our study and then I know some of the
other primate centers are doing is if the microcephaly is
the tip of the iceberg, and there are all these other
lesser neurologic defects that could have long-term neurological development implications, we are following the design
where we inoculate the dam while the fetus is in utero, and then instead of
euthanizing pre-delivery, we allow the infant to be born and then we’ll follow
them for a period of time. And then there are specialists that can look at macaque behavior and associate that with
neurologic defects. But that really, that still addresses the question of what happens when the mom is infected, not the neonate. So I imagine once people
start figuring out the most I think pressing
question of the pregnancy, then they’ll start
looking at infant studies. And maybe somebody in the audience knows if there’s ongoing human clinical studies in pediatric cohorts? I don’t know, I imagine so. – [Audience Member] I guess
as a follow-up question, is there any reason to believe that it would not have an impact? Like is there some fundamental
basis to sort of say, Well it probably, the virus would not make
it into the postnatal, you know the baby, the baby’s brain? – You mean the congenital or the infant gets fed on by a mosquito? – [Audience Member] Yeah,
like is there any reason to believe that the infant should be fine? You know like– – Well if Zika becomes like Dengue in lots of places where
Dengue is hyperendemic, most children get infected with multiple serotypes of Dengue by the time they go to school. So if the transmission
intensity of Zika’s high, we could expect infants or
children to get infected. You’re asking whether in
immunological development, they would be refractory or? – [Audience Member] No,
I’m just asking like, is there any reason to believe that the virus would, that the children would not be affected if they were you know, they’ll be infected but not, that there wouldn’t be any
neurological brain damage. And clearly when Zika infects the fetus, we see all these effects, but why is it that we don’t
see any effects when the– – The supposition is that we will, but we haven’t had a long
enough time to study it. – [Audience Member] Okay. – [Moderator] Okay, any more questions? – [Audience Member] Professor Lark, you probably know one
of the big controversies is basically what is the source of, how does infection get
maintained in pregnant women? And there’s been the hypothesis that it’s really the placental infection that drives basically
periodic exposure of the fetus so the fetus eventually becomes infected. And that’s been shown by
some of the clinical data suggesting persistent
viremia in pregnant women, but once they deliver the fetus the viremia disappears as if, but once the placenta’s gone. So the question is, is
it the infected fetus or is it infected placenta that’s driving the persistent
infection in women? – Yeah, so I think that’s
an interesting idea. So our studies can’t
directly address that, because we obviously had
infection on both sides. But in the studies that
we’re doing right now, we’re infecting only the mother side. And then I would love to do a study where we infect only IA. So then we can really see, to me that would be definitive, if the dam’s never been infected and we see spill back into the damn, then we know it’s coming
from the fetal replication. But on the placenta question, so when we titrated all the virus in the viral RNA-positive
tissues from the dam, even at 100 days post-inoculation, the only tissue that we
found infectious virus in in two of the three
animals was the placenta. – [Moderator] Okay, well thank you very much for a great talk. (audience applauds)
We’ll move on.

Leave a Reply

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