Introduction 
Welcome to the Zika experimental science team (ZEST) data portal. Given the urgency of the ongoing Zika virus epidemic, we are making our study results available in real-time. Each study and its available data are shown below. For example, the first study performed by the ZEST team is ZIKV-001. If there are data you would like but are not available, please contact us. We are also happy to answer questions about the data as best as possible, but we apologize in advance if we do not have time to answer each and every question.

Questions or comments can be directed to Dave O'Connor dhoconno@wisc.edu. You can also follow the O'Connor lab (@dho_lab) or Dave (@dho) on Twitter.

LabKey (long-time technology partners of the O'Connor Lab) has launched the Zika Open-Research Portal to help facilitate collaborative research. This portal provides researchers with a platform to share raw data, study commentary and results with the public in real-time. Projects are freely available to investigators.

If you are interested in making your research publicly available through the portal, please contact LabKey to get started.

Natural history of ZIKV infection

Asian lineage infection (ZIKV-FP) African lineage infection (MR766)

ZIKV infection during pregnancy

1st trimester infection 3rd trimester infection Dose Titration of ZIKV in pregnancy

Mucosal challenges

Tonsillar challenge with 10e6 ZIKV-FP Subcutaneous infection of saliva donor Saliva recipients (mucosal challenge: conjunctiva, intranasal, or tonsils)

Interactions between DENV and ZIKV

ZIKV infection in rhesus previously exposed to DENV-3 DENV-2 infection in rhesus previously infected and re-challenged with ZIKV-FP ZIKV infection in cynomolgus macaques previously exposed to different serotypes of DENV.

Tissue tropism in pregnancy

Tissue tropism at 30dpi Tissue tropism at 65dpi Tissue tropism at 105dpi

Tissue tropism in fetal macaques

Infections with PR-ZIKV, clone, and barcoded virus

Mosquito inoculation

Prevalence of ZIKV in wild NHP populations

Mock challenges


Direct Study Links


Messages 
Showing: all messages
Whole virion ELISA results from Tulika Singh
mkoenig5 2017-07-24
Tulika conducted a whole virion ELISA to assess the binding of rhesus macaque plasma to Zika virus during infection and secondary challenge. Her summary of the experiment and results is included below and the results are included in the attached PDF


I did a whole virion ELISA in order to assess the binding of rhesus macaque plasma at sequential timepoints to Zika virus. We hypothesize that prior to exposure to Zika virus, plasma is not specific to Zika virus and thus will not bind the virion; but as the humoral immune response develops after challenge with Zika virus, we expect that antibodies specific to Zika will expand and display stronger binding to Zika virus.

To do test this, I ran the following ELISA: I coated flat-bottomed 96 well plates with 4G2 anti-flavivirus mAb, which allows us to immobilize Zika virion on the plate. Then, I incubated this with serial dilutions of plasma, starting at 1:12.5 and diluting 4-fold 11 times. The plasma binding to Zika virions was detected with a goat anti-human mAb conjugated to horseradish peroxidase. This enzyme catalyzes a substrate (commercially known as TrueBlue/SureBlue) and generates a colorimetric luminescence at different intensities, which can be measured. The plasma serial dilution generates a curve – where lower dilutions have more plasma antibodies and bind more with the virus, and higher dilutions will have less plasma and thus display weaker binding (curves shown on slides 3–5). From each curve we can calculate an ED50: the dilution at which the plasma displays half as much binding as its maximum binding.

To evaluate the plasma binding profile along the timeline of ZIKV primary and secondary challenge, I tested plasma at multiple time-points from the baseline of Day 0 (day of Zika challenge) through the re-challenge on Day 70 (shown on slide 2). Based on the data, specificity and binding to Zika virus increases dramatically, to near-maximal threshold, between days 5 and 15 post primary infection. High binding to Zika is maintained for the next 60 days, indicative of sterilizing immunity. Upon re-challenge, binding to Zika virus does not increase significantly in the plasma, indicating that the pre-existing humoral response was protective. This is consistent with the lack of viremia upon re-challenge, as well as lack of plasmablast expansion upon re-challenge.

Tulika Singh Duke Molecular Genetics and Microbiology Graduate Student Permar Lab Duke University School of Medicine Durham, N.C.

 Tulika_Singh_whole_viron_ELISA.pdf 
view message
Detection of antisense ZIKV RNA following vertical transmission
david h oconnor 2017-04-29
In the ZIKV-019 study, a fetus infected with Zika virus died in utero. Subsequent analyses showed that Zika virus RNA was detected in many tissues, far more than we've detected at term in any pregnancy studied to date. A major outstanding question is whether this viral RNA indicates replication-competent virus. There is increasing evidence that not all tissues where viral RNA is detected contain replicating virus. To address this question, we sent tissues to Xiankun (Kevin) Zeng from USAMRIID who has developed an assay to detect negative sense RNA. Briefly, our viral RNA qRT-PCR assay detects positive-sense RNA which can exist in the presence or absence of replicating virus. Negative sense RNA, in contrast, is produced only as a replication intermediate, which means that its detection is evidence of active virus replication.

As shown in this PDF from Dr. Zeng, negative sense RNA was detected in:

  • fetal colon
  • fetal amniotic/chorionic membrane
  • fetal lung
Here is an example of the negative sense RNA staining:

While we were unable to examine all fetal tissues with this method, this data provides unequivocal support for a systemic, disseminated Zika virus infection in this fetus.

This method will also be extremely valuable for assessing whether the detection of Zika virus RNA in other tissues (e.g., lymph nodes, semen, red blood cells) correlates with the presence of actively replicating virus, which, at the present time, is not established.

--dave

 colon-neg-sense.png 
view message
Puerto Rican Harvest stock virus
soconnor 2017-02-17
We wanted to determine what sort of variants were present in the Puerto Rican stock virus that was prepared at UW-Madison. We used the short amplicon method developed by Quick et. al., and we were able to amplify and sequence two replicates of the harvest virus called ‘PR-ABC59-Harvestvirus’ .

For each replicate, we used the following number of vRNA templates as input for the cDNA synthesis reaction:

PR-ABC59 harvest virus: 1e6
animal 566628: 5,598
animal 634675: 6787
animal 311413: 23,123


The data was analyzed using the Sequencer pipeline (see the tools attached as a .zip file below), using the following processes:

1. Trims and merges the paired reads from the FASTQ data.

2. Extracts 1000 reads (if they are present in the sequence data) spanning each of the 35 amplicons that were generated by the amplification protocol.

3. Maps the 1000 reads x 35 amplicons to a full reference genome — KU50125

4. Calls the SNP positions using SNPeff, and generates a VCF file.

5. Generates a BAM file that can be viewed in a program like Geneious.

Once a BAM file is created, then the VCF file and BAM file are opened in Geneious. An annotation table is created with the following columns:

a. Track Name
b. Minimum
c. Maximum
d. Change
e. FREQ (this is the variant frequency)

f. You can export other columns, such as Sample depth.

The exported CSV files are concatenated and a pivot table is created so that you can see the frequency of variation at each of the positions in the genome.

The pivot table is attached and labeled ‘PR_Stock&Animals_Wiki.xlsx.’

There are four samples on the table, each generated in 2 replicates:
PR-ABC-Harvest Virus
634675 - animal 1 infected with PR-ABC-Harvest Virus
566628 - animal 2 infected with PR-ABC-Harvest Virus
311413 - animal 3 infected with PR-ABC-Harvest Virus

Here are some key points to observe:

1. First, the less interesting information. There are three sites where the variants are likely an artifact of the sequencing method. I put double asterisks next to their position in the excel table.

Site 118 — this is a string of A’s, and it looks like there is a low level insertion that could be a consequence of some slippage in PCR or sequencing

Site 405 — this is in a region that is high in G/A content. There is an occasional deletion that is in one amplicon, but not the overlapping amplicon. So, this seems to be an artifact of one of the amplicons generated during the process.

Site 9343 — this is also an artifact present in one amplicon, but not the overlapping amplicon. So this is also likely not real.

2. Second, we have the more interesting sites:

Site 1964 — This site is a real difference from the reference, but the frequencies may be inaccurate. It is present at the end of amplicon 6, but the middle of amplicon 7. The nucleotide gets deleted in some sequences found in amplicon 6. So, to get an accurate count of frequency, we need to look solely at amplicon 7.

Site 2780 — This looks to be a real variant whose frequency increases in the animals, relative to the stock.

Site 3147 — Looks real, but not too variable across animals.

Site 3281 — There is a variant at >5% in some of the animals, but it does not show up in the stock. This is a case where we made our cutoff to call a SNP at 5%. The stock does have variation in this site, but it is just less than 5%.

Site 5679 — Looks real, and frequency changes in 2 animals

Site 7915 — Looks real, but not too variable across animals.

Those are the interesting sites to consider. What does this mean?

This Puerto Rican Zika virus stock has a few sites of variation greater than 5%, and these are maintained in the animals. There are now new fixed changes in the animals, consistent with the idea that the stock virus is competent to grow in animals. Since we only looked at day 3 post infection, we don’t know whether there are other changes as the virus replicates in the animals. Future studies would need to address that.

For your viewing pleasure, I’ve attached the following:

1. The excel file ‘PR_Stock&Animals_Wiki’ that has the information about variation at the individual sites.

2. The Zequencer.zip file that was used to generate the mapped reads.

3. The BAM and VCF files needed to view this data set.
 PR_Stock&Animals_Wiki.xlsx  Zequencer.zip  311413.zip  566628.zip  634675.zip  PR_ABC9-Harvest virus.zip 
view message
Zika Virus barcode in vivo data
soconnor 2017-02-10
We obtained a barcode Zika virus that had been generated by Greg Ebel at Colorado state. We used a Zika barcode that had degenerate bases at 8 positions between positions 4007 and 4030 in the genome, with reference to the KU501215 genbank sequence. These are located in the NS2A gene.

See: barcodeposition.png below


We sequenced the barcode stock using the protocol that was pioneered by Quick et. al.:

http://www.biorxiv.org/content/early/2017/01/09/098913

For each replicate, here is the number of input templates:

pjW232-WT: this is the Wild type clone stock: 1 x 10^6 templates

pjW236-C1_p0: this is the barcode virus stock: 1 x 10^6 templates

514982 day 2: 146 templates (the animal simply had a low viral load)

715132 day 3: 2865 templates

688387 day 3: 667 templates

776301 day 3: 13, 153 templates

The sequence data was analyzed using the Zequencer pipeline that was engineered by Dave and is attached as a compressed zip file below.


Specifically, this pipeline performs the following steps:

1. Trims and merges the paired reads from the FASTQ data.

2. Extracts 1000 reads (if they are present in the sequence data) spanning each of the 35 amplicons that were generated by the amplification protocol.

3. Maps the 1000 reads to a full reference genome.

4. Calls the SNP positions using SNPeff, and generates a VCF file.

5. Generates a BAM file that can be viewed in a program like Geneious.


Once a BAM file is created, it is opened in Geneious. The barcode region can be extracted and then duplicates identified in Geneious. A FASTA file containing the information about the number of each sequence is generated and then converted using a geneiousFASTAtoTSV.py script. The TSV files can be merged and a pivot table generated to assess how frequent each barcode sequence is present in the stocks, and then in samples isolated from animals.

In this experiment, we sequenced the barcode stock virus pjW236_C1, from passage 0 and passage 1. We also sequenced the day 2 time point from animal 514982, and the day 3 time point from animals 715132 and 688387 in ZIKV-022. We tried to sequence the day 3 time point from animal 514982, but the viral loads were too low to get consistent data across replicates. Lastly we sequenced the day 3 time point from the pregnant animal 776301 from ZIKV-021. All samples were sequenced in duplicate, where possible. For one of the replicates in animal 514982, we were only able to generate data from half the genome.

We generated a pivot table containing the information about each barcode in the stock and in the animal samples. The excel file is called Zikv-022&021_barcode+inocula.xlsx (see attachment). There is one tab labeled ‘ConditionalFormat’ which has the data colored on a scale of 0% to 100% in shades of red, and one tab labeled ‘0.1%orgreater’ which has every sample at 0.1% or greater labeled in red. Also note that the samples are sorted. They are first sorted based on the pjW232WT_p0_RepA clone and then on the pjW236_C1_p0_RepA inocula. Both are listed as most frequent to least frequent barcode.

The key points are as follows:

1. The most common barcodes are fairly similar in the stock and in the animals at this early time point.

2. There are some rarer barcodes that are detectable. We do not know how or if these will persist.

3. The barcode virus replicated as a population and the barcodes were maintained, rather than reverting back to wild type.

4. The barcodes that commonly appear in the animals infected with the barcode virus do not naturally develop in the animals infected with the WT clone.

5. In the animal with the lowest number of virus templates used in the experiment (Animal #514982), the frequencies of each barcode between replicates was not particularly consistent. This is a consequence of starting with such a low viral load. Barcode consistency improves when the number of input virus templates are higher.
 barcodeposition.png  Zequencer.zip  geneiousFastaToTSV.py  Zikv-022&021_barcodes+inocula.xlsx 
view message
Establishing a Puerto Rican Zika Virus stock
mmohns 2017-02-03
Our lab has turned to using a ZIKV stock grown from a Puerto Rican isolate for our latest in vivo experiments (ZIKV-021, ZIKV-022, and ZIKV-023).

ZIKV strain PRVABC59 (GenBank:KU501215), was originally isolated from a traveler to Puerto Rico. The seed stock passaged with three rounds of amplification on Vero cells was obtained from Brandy Russell (CDC, Ft. Collins, CO). Virus stocks were prepared by inoculation onto a confluent monolayer and passaged two times on C6/36 mosquito cells.

Indian rhesus macaques challenged with 104 PFU PRVABC59 achieve peak viral load within the acute timeframe previously seen in inoculations with the French Polynesian ZIKV strain, albeit at a slightly lower peak (Figure 1a). Prolonged viremia has been detectable in pregnant animals, and is a consistent finding in an animal inoculated with PRVABC59 (Figure 1b).

Additionally, 4 IFNAR-/- mice were inoculated in the left hind footpad with 106 PFU (in 50ul) PRVABC59. At 2 days post-infection, the mice were all successfully infected with viremias of 5.72, 5.51, 5.51, and 5.81 log10 PFU/ml (Figure 2). Mosquitoes were allowed to feed on the mice at this time point and then will be allowed to feed on macaques for a future transmission study.

Sequencing data will be uploaded soon.

 ZIKV_PR_titer.png  ZIKV_PR_Fig1.png  ZIKV_PR_Fig2.png 
view message
Further investigation of saliva transmission in rhesus macaques
mmohns 2016-10-17
Following the experimental design of the ZIKV-007 through ZIKV-012 studies, today we challenged two more rhesus macaques to continue our studies on the potential for saliva transmission of Zika virus.

ZIKV-015 was challenged via tonsils with a high dose 10e6 PFU of the Asian lineage Zika virus to confirm that mucosal infection is possible. ZIKV-016 was inoculated with 10e4 PFU Asian lineage Zika virus subcutaneously as we have demonstrated the outcome of infections in previous Zika studies. Throughout the course of infection, we will collect saliva/drool samples which will be used to challenge two rhesus macaques via the tonsillar route on 10/20/2016 (ZIKV-017).

In this set of studies, we will be analyzing viral loads daily through 10dpi, then weekly through 28dpi. We will also perform assays to identify neutralizing antibody titers using samples from the later timepoints post-infection.

view message
Preliminary data for ZIKV infection in DENV3-exposed macaques
mmohns 2016-09-16
A primary infection with one dengue serotype elicits an immune response resulting in the development of specific antibodies. A secondary infection with a different dengue serotype tricks the immune system into developing a response -- the antibodies will attack the virus, but since they are serotype-specific, they do not inactivate the virus but rather help facilitate infection. This results in an acute infection due to antibody-dependent enhancement (ADE). Given the similarities between dengue virus and Zika virus, infection due to ADE is a concern.

On September 6th, 2016, we challenged three dengue-exposed Indian rhesus macaques with 10e4 PFU of the French Polynesian Zika virus isolate (ZIKV-013). These dengue-exposed animals were previously challenged a year ago with 600,000 PFU of dengue serotype-3 (DENV-3).

We were able to run dengue PRNT assays on these dengue-exposed animals and comparison of DENV-3 PRNT titers with Zika PRNT titers shows that there is no cross-neutralization. Open bars are DENV-3 PRNT titers. Open circles are Zika PRNT titers.

As of 7 days post-infection, viral load kinetics appear consistent with typical Zika virus infection as observed in our previous challenge studies. Animal 850585 had a much higher peak than the other two animals, and as of days 6 and 7, viremia has leveled off instead of the typical decline. Animal 489988 had a negative viral load on day 7.

We will continue to monitor these animals weekly through 28dpi.

 DENV_ZIKV_PRNT.png  ZIKV013_VL.png 
view message
ZIKV infection via mucosal challenges
mmohns 2016-09-13
Given that Zika virus RNA is present in saliva of infected people and monkeys, we sought to determine if Zika virus could be transmitted by contact with saliva.

First, to assess whether or not mucosal infection was even possible, we challenged two animals (ZIKV-007 and ZIKV-010) with a high dose (10e6 PFU) of the Asian lineage virus via tonsillar challenge. In both challenges, we observed a 1-2 day delay of infection, followed by peak viremia around 5-6 days post infection. Viremia resolved to below the limit of detection by 14dpi similar to what we’ve seen in previous Zika challenges.

These experiments suggest that Zika infection is theoretically possible to transmit via a mucosal route given a high enough titer.

*gray lines represent historical viral loads of Zika infection in rhesus macaques

To follow these experiments, we subcutaneously challenged two animals (ZIKV-008 and ZIKV-011) with 10e4 PFU of the Asian lineage virus to use as a saliva donor for mucosal challenges in 5 other animals (ZIKV-009 and ZIKV-012).

The subcutaneously challenged animals displayed a viral trajectory consistent with previous Zika challenges.

Oral swabs from the ZIKV-008 donor animal were eluted in saline to use for saliva challenges in three additional animals (ZIKV-009), with saliva applied to either the tonsils, conjunctiva, or intranasal mucosa. The highest titer of ZIKV RNA isolated from oral swabs was just over 2,000 copies/ml.

A very low level blip was detected 2dpi in the ZIKV-009 animal that received donor saliva to the tonsils. However, a retest of this timepoint was negative for viral RNA. Additionally, none of the animals seroconverted (PRNT data pending later timepoints).

Alternatively, saliva was obtained via drool in the ZIKV-011 donor animal. This saliva/drool was used to directly infect the tonsillar mucosa of two animals (ZIKV-012). The highest titer of ZIKV RNA isolated from saliva/drool was just over 10,000 copies/ml.

ZIKV RNA was undetectable in both of the ZIKV-012 recipients and the animals did not seroconvert.

It is likely that the saliva viral loads in the donor animals was not nearly sufficient to infect the recipients.

While the theoretical transmission via a mucosal route is a possibility, it is likely that certain conditions need to be met, including high titer of virus in body fluids.

 Screen Shot 2016-09-13 at 4.39.49 PM.png 
view message
ZIKV infection in dengue-exposed macaques at Caribbean Primate Research Center
mmohns 2016-09-02
Researchers at CPRC in Puerto Rico have shared preliminary data online, challenging animals with pre-existing DENV immunity (DENV-1 and DENV-2) with ZIKV. Secondary ZIKV infection does not appear to support evidence of antibody-dependent enhancement (ADE) in vivo using the rhesus macaque model.

Data can be found here: http://nprcresearch.org/primate/hot-topics/CPRC-Zika-Virus-Research-Page.pdf

On the heels of the work at CPRC, our group will be starting our next set of Zika studies next Tuesday, infecting 3 rhesus macaques with pre-existing immunity to serotype-3 of DENV. Results will be posted as soon as possible here on our web portal.
view message
Histopathology tissues
mmohns 2016-08-26
Approximately 60 tissues were fixed and analyzed following fetal necropsy for ZIKV-003, ZIKV-005, and ZIKV-006. Tissues were graded using the severity values below. Each positive tissue has a link for a description of histopathological findings from our pathology team at WNPRC. Comparative tissue images are pending and will be added to these links shortly.

ValueSeverity
0
Normal
1
Minimal
2
Mild
3
Moderate
4
Severe


Tissue
Animal 598248
Animal 357676
Animal 827577
Animal 660875
Maternal Spleen
0
3
Materna Liver
0
0
0
0
Materna Lymph Node
0
0
0
0
Placenta
0
3
Decidua
1
3
Placental Bed
0
0
0
0
Umbilical cord
0
0
0
0
Amniotic Membrane
0
Chorionic Membrane
1
Mammary Gland
0
0
0
0
Axillary Lymph Node
0
0
0
0
Forearm - skin, subq, vessels, muscle
0
0
0
0
Skin Thighs
0
0
0
0
Submandibular LN
0
0
0
0
Submucosal Salivary Gland
0
0
0
0
Buccal Mucosa
0
0
0
0
Sternum
0
0
0
0
Thymus
0
0
0
0
Pericardium
0
0
0
Heart
0
0
0
0
Aorta - Thoracic
0
0
0
0
Tongue
0
0
0
0
Tonsils
0
0
0
0
Thyroids
0
0
0
0
Esophagus
0
0
0
0
Trachea
0
0
0
0
Lung
3
Spleen
0
2
Pancreas
0
0
0
0
Pancreatic Lymph Node
0
0
0
Mesocolonic Lymph Node
0
0
0
0
Mesenteric Lymph Node
0
0
0
0
Liver
0
0
1
Kidney
0
0
0
0
Adrenal Glands
0
0
0
0
Urinary Bladder
0
0
0
0
Testis/Ovary
0
0
0
0
Seminal Vesicle/Prostate/Uterus
0
0
0
h
0
Gastric Lymph Node
0
0
0
Stomach
0
0
0
0
Duodenum
0
0
0
0
Jejunum
0
0
0
0
Ileum
0
0
0
0
Colon
0
0
0
0
Internal Iliac artery & Vein
0
0
0
0
Aorta - Abdominal
0
0
0
0
Dura Mater
0
0
0
0
Brain
0
0
0
0
Pituitary
0
0
0
0
Eye
0
0
0
3
Spinal Cord - Cervical
0
0
0
0
Spinal Cord - Thoracic
0
0
0
0
Spinal Cord - Lumbar
0
0
0
0
Saphenous Vein
0
0
0
0
Quadriceps
0
0
0
0
Femoral Bone
0
0
0
0
Bone Marrow
0
0
0
0
Femoral Head
0
0
0
0



view message
More complete Zika virus antibody analysis by peptide array
david h oconnor 2016-08-24

Comparison of SIV-specific and ZIKV-specific antibody responses in macaques

2016-08-24

O'Connor experiment 17912

Background correction

Peptide array data was generated by Roche/Nimblegen as part of an early access program. Ten macaque plasma and serum samples were run on each of three plates. Two negative controls (secondary antibody only) were also run on each plate. For each peptide on the array in each of the two plates, we calculated the mean and standard deviation of the two negative controls. The sum of the mean and standard deviation of the negative controls were subtracted from the signal intensity of the experimental samples tested against the corresponding peptide in the same plate. These corrected signal intensities are used as the basis for this analysis.

In [2]:
# import Python3 modules needed for analysis
# use Cufflinks and Plotly to produce interactive heatmaps
import numpy
import pandas
import cufflinks
import plotly.graph_objs as go
import plotly.tools as tls
import plotly.plotly as py

Data import

A CSV file containing signal intensities for all thirty samples will be used for this analysis. Since the peptide array also contains peptides from hundreds of other viral proteins, we will simplify the analysis by only sharing the SIV and ZIKV peptide data. If you would like to collaborate with us on analyses of other targets, or wish to get the data for other purposes, please contact us. Use of the data may require MTAs with either Roche/Nimblegen or UW-Madison, but we can cross that bridge if we come to it.

In [3]:
# import CSV file containing all samples
df = pandas.read_csv('/Users/dho/Downloads/peptidearray_results_backgroun_2016-08-19_19-59-01.csv')

Prepare data

Simplify sample descriptions

The original sample descriptors used are complex. In the next step, these complex identifiers will be replaced with simple identifiers. To reduce confusion, public IDs from the Zika Open portal will be used whenever possible. The exact transformation is masked because some of the sample identifiers have not been cleared for public distribution. Only samples relevant to a particular analysis are shown (e.g., samples from SIV-infected animals are shown in the HIV/SIV comparisons)

In [18]:
renamed_df = df.replace({
       'XXX': '28-Baboon 001 - post-arterivirus - 2014-04-15',
       'XXX': '27-Baboon 001 - pre-arterivirus - 2007-07-23',
       'XXX': '29-Baboon 002 - pre-arterivirus - 2013-12-11',
       'XXX': '30-Baboon 002 - post-arterivirus - 2014-04-15',
       'XXX':  '23-Cynomolgus 001 - pre-SIV - 2012-06-28',
       'XXX': '24-Cynomolgus 001 - 52 weeks post-SIV - 2013-09-10',
       'XXX': '21-Cynomolgus 002 - pre-GBV-C - 2014-03-10',
       'XXX': '22-Cynomolgus 002 - post-GBV-C - 2014-07-29',
       'XXX': '19-Cynomolgus 002 - pre-SIV - 2012-06-25',
       'XXX': '20-Cynomolgus 002 - 52 weeks post-SIV - 2013-09-24',
       'XXX': '25-Cynomolgus 003 - pre-GBV-C - 2014-05-27',
       'XXX': '26-Cynomolgus 003 - post-GBV-C - 2014-10-07',
       'XXX': '01-295022 ZIKV-002 0 days post-ZIKV',
       'XXX': '02-295022 ZIKV-002 67 days post-ZIKV',
       'XXX': '03-295022 ZIKV-002 98 days post-ZIKV',
       'XXX': '04-562876 ZIKV-002 0 days post-ZIKV',
       'XXX': '05-562876 ZIKV-002 67 days post-ZIKV',
       'XXX': '06-562876 ZIKV-002 98 days post-ZIKV',
       'XXX': '07-405734 ZIKV-002 0 days post-ZIKV',
       'XXX': '08-405734 ZIKV-002 67 days post-ZIKV',
       'XXX': '09-405734 ZIKV-002 98 days post-ZIKV',
       'XXX': '10-610107 ZIKV-004 0 days post-ZIKV',
       'XXX': '11-610107 ZIKV-004 67 days post-ZIKV',
       'XXX': '12-610107 ZIKV-004 99 days post-ZIKV',
       'XXX': '13-181856 ZIKV-004 0 days post-ZIKV',
       'XXX': '14-181856 ZIKV-004 67 days post-ZIKV',
       'XXX': '15-181856 ZIKV-004 99 days post-ZIKV',
       'XXX': '16-411359 ZIKV-004 0 days post-ZIKV',
       'XXX': '17-411359 ZIKV-004 67 days post-ZIKV',
       'XXX': '18-411359 ZIKV-004 99 days post-ZIKV'
})

Visualize responses to SIV and HIV Env

In the SIV+ samples, we expect robust antibody responses directed against epitopes in SIVmac239, the challenge strain. Weaker responses would be expected against heterologous SIV isolates, with very weak responses against HIV Env, if responses are detected at all.

In [19]:
# select pre- and post-infection SIV samples

siv_samples_df = renamed_df[renamed_df['sample_description'].isin([
        '23-Cynomolgus 001 - pre-SIV - 2012-06-28',
        '24-Cynomolgus 001 - 52 weeks post-SIV - 2013-09-10',
        '19-Cynomolgus 002 - pre-SIV - 2012-06-25',
        '20-Cynomolgus 002 - 52 weeks post-SIV - 2013-09-24',
        '21-Cynomolgus 002 - pre-GBV-C - 2014-03-10',
        '22-Cynomolgus 002 - post-GBV-C - 2014-07-29'
        ])]

Antibody responses to homologous SIVmac239

First, examine responses against SIVmac239. Set minimum value for plotting to 0 (so any signal intensities below the negative control threshold will all be plotted as 0). The maxiumum value for plotting will be the maximum observed intensity or 10,000 units, whichever is lower. This prevents extremely high values from making it difficult to visualize lower values.

In [51]:
# plot responses to SIVmac239 Env probes

# select rows containing data with SIVmac239 Env probes
sid_df = siv_samples_df[siv_samples_df['seq_id'] == 'Q88018']

# make pivot table
pivoted = sid_df.pivot_table(index='position_id', columns='sample_description', values='subtract_mean_background_plus_1_stddev_signal')

# set title
title = sid_df.seq_id.unique()[0] + ' - ' + sid_df.description.unique()[0]

# set layout options
layout = go.Layout(
title=title,
autosize=False,
width=1000,
height=600,
margin=go.Margin(
    l=350,
    r=200,
    b=100,
    t=100,
    pad=4
))

# make heatmap and save as figure
dataframe_fig = pivoted.iplot(kind='heatmap',
          layout=layout,
          colorscale='rdylbu',
          world_readable=False,
          asFigure=True)

# set zmin = 0 to not plot negative values
dataframe_fig['data'][0]['zmin'] = 0

# set zmax to 10000 or maximum observed value in figure, whichever is lower
zmax_obs = numpy.amax(dataframe_fig['data'][0]['z'])
if zmax_obs > 10000:
    dataframe_fig['data'][0]['zmax'] = 10000
elif zmax_obs > 10000:
    dataframe_fig['data'][0]['zmax'] = zmax_obs

# set filename
filename = 'cufflinks/17912/ZikaOpen' + '/' + title

py.iplot(dataframe_fig, filename=filename)

Out[51]:

SIVmac239 Env-specific responses in the two animals after SIV infection are robust. Pre-infection responses are very low. This demonstrates that the technology platform is able to detect homologous responses that we anticipate based on these animals' infection histories.

Antibody responses to SIVsmE660

SIVsmE660 is more distantly related to SIVmac239. We would expect more modest antibody responses to SIVsmE660 peptides. Below the responses to SIVsmE660 are plotted:

In [52]:
# plot responses to SIVsmE660 Env probes

# select rows containing data with SIVsmE660 Env probes
sid_df = siv_samples_df[siv_samples_df['seq_id'] == 'I6TPL2']

# make pivot table
pivoted = sid_df.pivot_table(index='position_id', columns='sample_description', values='subtract_mean_background_plus_1_stddev_signal')

# set title
title = sid_df.seq_id.unique()[0] + ' - ' + sid_df.description.unique()[0]

# set layout options
layout = go.Layout(
title=title,
autosize=False,
width=1000,
height=600,
margin=go.Margin(
    l=350,
    r=200,
    b=100,
    t=100,
    pad=4
))

# make heatmap and save as figure
dataframe_fig = pivoted.iplot(kind='heatmap',
          layout=layout,
          colorscale='rdylbu',
          world_readable=False,
          asFigure=True)

# set zmin = 0 to not plot negative values
dataframe_fig['data'][0]['zmin'] = 0

# set zmax to 10000 or maximum observed value in figure, whichever is lower
zmax_obs = numpy.amax(dataframe_fig['data'][0]['z'])
if zmax_obs > 10000:
    dataframe_fig['data'][0]['zmax'] = 10000
elif zmax_obs > 10000:
    dataframe_fig['data'][0]['zmax'] = zmax_obs

# set filename
filename = 'cufflinks/17912/ZikaOpen' + '/' + title

py.iplot(dataframe_fig, filename=filename)

Out[52]:

Some responses are still detected, but note that the response directed against ~aa 500 in SIVmac239 is no longer detected against SIVsmE660 peptides

Antibody responses to HIV-2

HIV-2 Env sequences are divergent from SIVmac239, so antibody responses against SIVmac239 peptides are unlikely to cross-react extensively. Yet SIV and HIV-2 are both originally derived from sooty mangabeys. What do these responses look like?

In [53]:
# plot responses to HIV-2 Env probes

# select rows containing data with HIV-2 Env probes
sid_df = siv_samples_df[siv_samples_df['seq_id'] == 'P18094']

# make pivot table
pivoted = sid_df.pivot_table(index='position_id', columns='sample_description', values='subtract_mean_background_plus_1_stddev_signal')

# set title
title = sid_df.seq_id.unique()[0] + ' - ' + sid_df.description.unique()[0]

# set layout options
layout = go.Layout(
title=title,
autosize=False,
width=1000,
height=600,
margin=go.Margin(
    l=350,
    r=200,
    b=100,
    t=100,
    pad=4
))

# make heatmap and save as figure
dataframe_fig = pivoted.iplot(kind='heatmap',
          layout=layout,
          colorscale='rdylbu',
          world_readable=False,
          asFigure=True)

# set zmin = 0 to not plot negative values
dataframe_fig['data'][0]['zmin'] = 0

# set zmax to 10000 or maximum observed value in figure, whichever is lower
zmax_obs = numpy.amax(dataframe_fig['data'][0]['z'])
if zmax_obs > 10000:
    dataframe_fig['data'][0]['zmax'] = 10000
elif zmax_obs > 10000:
    dataframe_fig['data'][0]['zmax'] = zmax_obs

# set filename
filename = 'cufflinks/17912/ZikaOpen' + '/' + title

py.iplot(dataframe_fig, filename=filename)

Out[53]:

Some responses are preserved, but others are lost. Interestingly, there is a response in Cynomolgus 002 around amino acid 487 that isn't seem against either of the two SIV peptide sets. Perhaps this is a response against an escape variant that arose during SIVmac239 infection.

Antibody responses to HIV-1

HIV-1 is a fundamentally different from from HIV-2/SIV. It originated in chimpanzees, not sooty mangabeys, and has extensive sequence divergence. Antibody responses would not expect to cross-react.

In [54]:
# plot responses to HIV-1 Env probes

# select rows containing data with HIV-1 Env probes
sid_df = siv_samples_df[siv_samples_df['seq_id'] == 'P04578']

# make pivot table
pivoted = sid_df.pivot_table(index='position_id', columns='sample_description', values='subtract_mean_background_plus_1_stddev_signal')

# set title
title = sid_df.seq_id.unique()[0] + ' - ' + sid_df.description.unique()[0]

# set layout options
layout = go.Layout(
title=title,
autosize=False,
width=1000,
height=600,
margin=go.Margin(
    l=350,
    r=200,
    b=100,
    t=100,
    pad=4
))

# make heatmap and save as figure
dataframe_fig = pivoted.iplot(kind='heatmap',
          layout=layout,
          colorscale='rdylbu',
          world_readable=False,
          asFigure=True)

# set zmin = 0 to not plot negative values
dataframe_fig['data'][0]['zmin'] = 0

# set zmax to 10000 or maximum observed value in figure, whichever is lower
zmax_obs = numpy.amax(dataframe_fig['data'][0]['z'])
if zmax_obs > 10000:
    dataframe_fig['data'][0]['zmax'] = 10000
elif zmax_obs > 10000:
    dataframe_fig['data'][0]['zmax'] = zmax_obs

# set filename
filename = 'cufflinks/17912/ZikaOpen' + '/' + title

py.iplot(dataframe_fig, filename=filename)

Out[54]:

As anticipated, there are no strong responses directed against HIV-1 peptides, even in SIV+ samples. This provides evidence that positive reactivity is observed against homologous viral peptides, as expected, but not against divergent viral peptides. Demonstrating that the peptide array platform performs well against HIV/SIV provides confidence that the results we observed in ZIKV-infected animals using the same platform are beleivable.

Visualize responses to Zika virus

Samples from two different Zika virus studies were included in this analysis. ZIKV-002 macaques were infected with African-lineage Zika virus and rechallenged with Asian-lineage Zika virus 70 days later. ZIKV-004 macaques were infected with Asian-lineage Zika virus and rechallenged with the same Asian-lineage Zika virus 70 days later. The three timepoints from each animal correspond to pre-ZIKV infection, immediately before rechallenge, and approximately one month following rechallenge.

In [34]:
# select pre- and post-infection ZIKV samples

zikv_samples_df = renamed_df[renamed_df['sample_description'].isin([
'01-295022 ZIKV-002 0 days post-ZIKV',
'02-295022 ZIKV-002 67 days post-ZIKV',
'03-295022 ZIKV-002 98 days post-ZIKV',
'04-562876 ZIKV-002 0 days post-ZIKV',
'05-562876 ZIKV-002 67 days post-ZIKV',
'06-562876 ZIKV-002 98 days post-ZIKV',
'07-405734 ZIKV-002 0 days post-ZIKV',
'08-405734 ZIKV-002 67 days post-ZIKV',
'09-405734 ZIKV-002 98 days post-ZIKV',
'10-610107 ZIKV-004 0 days post-ZIKV',
'11-610107 ZIKV-004 67 days post-ZIKV',
'12-610107 ZIKV-004 99 days post-ZIKV',
'13-181856 ZIKV-004 0 days post-ZIKV',
'14-181856 ZIKV-004 67 days post-ZIKV',
'15-181856 ZIKV-004 99 days post-ZIKV',
'16-411359 ZIKV-004 0 days post-ZIKV',
'17-411359 ZIKV-004 67 days post-ZIKV',
'18-411359 ZIKV-004 99 days post-ZIKV'
])]

Antibody responses to African-lineage Zika virus

ZIKV-002 animals were infected with Zika virus/R.macaque-tc/UGA/1947/MR766-3329. Responses to the entire polyprotein are shown.

In [62]:
# plot responses to ZIKV MR766 probes

# select rows containing data with ZIKV MR766 probes
sid_df = zikv_samples_df[zikv_samples_df['seq_id'] == 'A0A140D2T1']

# make pivot table
pivoted = sid_df.pivot_table(index='position_id', columns='sample_description', values='subtract_mean_background_plus_1_stddev_signal')

# set title
title = sid_df.seq_id.unique()[0] + ' - ' + sid_df.description.unique()[0]

# set layout options
layout = go.Layout(
title=title,
autosize=False,
width=1000,
height=900,
margin=go.Margin(
    l=350,
    r=200,
    b=100,
    t=100,
    pad=4
))

# make heatmap and save as figure
dataframe_fig = pivoted.iplot(kind='heatmap',
          layout=layout,
          colorscale='rdylbu',
          world_readable=False,
          asFigure=True)

# set zmin = 0 to not plot negative values
dataframe_fig['data'][0]['zmin'] = 0

# set zmax to 10000 or maximum observed value in figure, whichever is lower
zmax_obs = numpy.amax(dataframe_fig['data'][0]['z'])
if zmax_obs > 10000:
    dataframe_fig['data'][0]['zmax'] = 10000
elif zmax_obs > 10000:
    dataframe_fig['data'][0]['zmax'] = zmax_obs

# set filename
filename = 'cufflinks/17912/ZikaOpen' + '/' + title

py.iplot(dataframe_fig, filename=filename)

Out[62]:

The responses vary between macaques. ZIKV-002 animal 295022 shows nice reactivity against aa ~1420 and ~3060 only in the post-infection timepoints. Responses in the other two ZIKV-002 macaques are generally weak. Note that these animals were protected from heterologous rechallenge, so it could be that non-linear epitopes (which will likely be missed by this peptide array) are important for protection. Alternately, weak responses relative to the maximal responses may be relevant to protection. View the same data again, but this time limiting the maximum color intensity to 1000 signal units.

In [63]:
# plot responses to ZIKV MR766 probes

# select rows containing data with ZIKV MR766 probes
sid_df = zikv_samples_df[zikv_samples_df['seq_id'] == 'A0A140D2T1']

# make pivot table
pivoted = sid_df.pivot_table(index='position_id', columns='sample_description', values='subtract_mean_background_plus_1_stddev_signal')

# set title
title = sid_df.seq_id.unique()[0] + ' - ' + sid_df.description.unique()[0]

# set layout options
layout = go.Layout(
title=title,
autosize=False,
width=1000,
height=900,
margin=go.Margin(
    l=350,
    r=200,
    b=100,
    t=100,
    pad=4
))

# make heatmap and save as figure
dataframe_fig = pivoted.iplot(kind='heatmap',
          layout=layout,
          colorscale='rdylbu',
          world_readable=False,
          asFigure=True)

# set zmin = 0 to not plot negative values
dataframe_fig['data'][0]['zmin'] = 0

# set zmax to 1000 to amplify modest responses
dataframe_fig['data'][0]['zmax'] = 1000

# set filename
filename = 'cufflinks/17912/ZikaOpen' + '/' + title + 'zmax1000'

py.iplot(dataframe_fig, filename=filename)

Out[63]:

Low-level responses in the other two animal after Zika virus infection are observed, but additional work would be necessary to validate these responses. Interestingly, all three ZIKV-004 animals have responses at d67 and 99 against aa ~420 and ~530. This region also shows a subtle reactivity in 562876 and 295022. Though there is no evidence for reactivity in the third ZIKV-002 animal.

Antibody responses to Asian-lineage Zika virus

ZIKV-004 animals were infected with Zika virus/H.sapiens-tc/FRA/2013/FrenchPolynesia-01_v1c1. Responses to the entire polyprotein are shown.

In [57]:
# plot responses to ZIKV FP probes

# select rows containing data with ZIKV FP probes
sid_df = zikv_samples_df[zikv_samples_df['seq_id'] == 'A0A024B7W1']

# make pivot table
pivoted = sid_df.pivot_table(index='position_id', columns='sample_description', values='subtract_mean_background_plus_1_stddev_signal')

# set title
title = sid_df.seq_id.unique()[0] + ' - ' + sid_df.description.unique()[0]

# set layout options
layout = go.Layout(
title=title,
autosize=False,
width=1000,
height=900,
margin=go.Margin(
    l=350,
    r=200,
    b=100,
    t=100,
    pad=4
))

# make heatmap and save as figure
dataframe_fig = pivoted.iplot(kind='heatmap',
          layout=layout,
          colorscale='rdylbu',
          world_readable=False,
          asFigure=True)

# set zmin = 0 to not plot negative values
dataframe_fig['data'][0]['zmin'] = 0

# set zmax to 10000 or maximum observed value in figure, whichever is lower
zmax_obs = numpy.amax(dataframe_fig['data'][0]['z'])
if zmax_obs > 10000:
    dataframe_fig['data'][0]['zmax'] = 10000
elif zmax_obs > 10000:
    dataframe_fig['data'][0]['zmax'] = zmax_obs

# set filename
filename = 'cufflinks/17912/ZikaOpen' + '/' + title

py.iplot(dataframe_fig, filename=filename)

Out[57]:

Robust post-challenge antibody responses are detected in two of the three ZIKV-004 animals, as well as ZIKV-002 animal 295022. Regions where reactivity is detected in two or more animals may be good targets for assessing antibody responses using focused, more sensitive tools. In particular, the regions from ~aa 450-460 and 515-530 are intriguing because responses are seen in all three ZIKV-004 animals. For these responses to be useful for diagnostics, they need to be specific for Zika virus. What about cross-reactivity against other flaviviruses?

Antibody responses to Dengue virus

Responses against D1/PF/FP1104/2001 French Polynesia are shown

In [58]:
# plot responses to D1/PF/FP1104/2001 French Polynesia probes

# select rows containing data with D1/PF/FP1104/2001 French Polynesia probes
sid_df = zikv_samples_df[zikv_samples_df['seq_id'] == 'A6XDI1']

# make pivot table
pivoted = sid_df.pivot_table(index='position_id', columns='sample_description', values='subtract_mean_background_plus_1_stddev_signal')

# set title
title = sid_df.seq_id.unique()[0] + ' - ' + sid_df.description.unique()[0]

# set layout options
layout = go.Layout(
title=title,
autosize=False,
width=1000,
height=900,
margin=go.Margin(
    l=350,
    r=200,
    b=100,
    t=100,
    pad=4
))

# make heatmap and save as figure
dataframe_fig = pivoted.iplot(kind='heatmap',
          layout=layout,
          colorscale='rdylbu',
          world_readable=False,
          asFigure=True)

# set zmin = 0 to not plot negative values
dataframe_fig['data'][0]['zmin'] = 0

# set zmax to 10000 or maximum observed value in figure, whichever is lower
zmax_obs = numpy.amax(dataframe_fig['data'][0]['z'])
if zmax_obs > 10000:
    dataframe_fig['data'][0]['zmax'] = 10000
elif zmax_obs > 10000:
    dataframe_fig['data'][0]['zmax'] = zmax_obs

# set filename
filename = 'cufflinks/17912/ZikaOpen' + '/' + title

py.iplot(dataframe_fig, filename=filename)

Out[58]:

The numbering between ZIKV and DENV isn't perfectly concordant, but there is no clear evidence for post-challenge responses against the corresponding region of DENV-1. Next, repeat for DENV-2, DENV-3, and DENV-4.

In [59]:
# plot responses to DENV 2 98900665 DF DV-2 Indonesia 1998 human probes

# select rows containing data with DENV 2 98900665 DF DV-2 Indonesia 1998 human probes
sid_df = zikv_samples_df[zikv_samples_df['seq_id'] == 'Q689G0']

# make pivot table
pivoted = sid_df.pivot_table(index='position_id', columns='sample_description', values='subtract_mean_background_plus_1_stddev_signal')

# set title
title = sid_df.seq_id.unique()[0] + ' - ' + sid_df.description.unique()[0]

# set layout options
layout = go.Layout(
title=title,
autosize=False,
width=1000,
height=900,
margin=go.Margin(
    l=350,
    r=200,
    b=100,
    t=100,
    pad=4
))

# make heatmap and save as figure
dataframe_fig = pivoted.iplot(kind='heatmap',
          layout=layout,
          colorscale='rdylbu',
          world_readable=False,
          asFigure=True)

# set zmin = 0 to not plot negative values
dataframe_fig['data'][0]['zmin'] = 0

# set zmax to 10000 or maximum observed value in figure, whichever is lower
zmax_obs = numpy.amax(dataframe_fig['data'][0]['z'])
if zmax_obs > 10000:
    dataframe_fig['data'][0]['zmax'] = 10000
elif zmax_obs > 10000:
    dataframe_fig['data'][0]['zmax'] = zmax_obs

# set filename
filename = 'cufflinks/17912/ZikaOpen' + '/' + title

py.iplot(dataframe_fig, filename=filename)

Out[59]:
In [60]:
# plot responses to DENV 3 PF89/27643 French Polynesia 1989 human probes

# select rows containing data with DENV 3 PF89/27643 French Polynesia 1989 human human probes
sid_df = zikv_samples_df[zikv_samples_df['seq_id'] == 'Q3L302']

# make pivot table
pivoted = sid_df.pivot_table(index='position_id', columns='sample_description', values='subtract_mean_background_plus_1_stddev_signal')

# set title
title = sid_df.seq_id.unique()[0] + ' - ' + sid_df.description.unique()[0]

# set layout options
layout = go.Layout(
title=title,
autosize=False,
width=1000,
height=900,
margin=go.Margin(
    l=350,
    r=200,
    b=100,
    t=100,
    pad=4
))

# make heatmap and save as figure
dataframe_fig = pivoted.iplot(kind='heatmap',
          layout=layout,
          colorscale='rdylbu',
          world_readable=False,
          asFigure=True)

# set zmin = 0 to not plot negative values
dataframe_fig['data'][0]['zmin'] = 0

# set zmax to 10000 or maximum observed value in figure, whichever is lower
zmax_obs = numpy.amax(dataframe_fig['data'][0]['z'])
if zmax_obs > 10000:
    dataframe_fig['data'][0]['zmax'] = 10000
elif zmax_obs > 10000:
    dataframe_fig['data'][0]['zmax'] = zmax_obs

# set filename
filename = 'cufflinks/17912/ZikaOpen' + '/' + title

py.iplot(dataframe_fig, filename=filename)

Out[60]:
In [61]:
# plot responses to DENV 4 strain H241 Phillipines 1956 human probes

# select rows containing data with DENV 4 strain H241 Phillipines 1956 human probes
sid_df = zikv_samples_df[zikv_samples_df['seq_id'] == 'A0A0M4C4I5']

# make pivot table
pivoted = sid_df.pivot_table(index='position_id', columns='sample_description', values='subtract_mean_background_plus_1_stddev_signal')

# set title
title = sid_df.seq_id.unique()[0] + ' - ' + sid_df.description.unique()[0]

# set layout options
layout = go.Layout(
title=title,
autosize=False,
width=1000,
height=900,
margin=go.Margin(
    l=350,
    r=200,
    b=100,
    t=100,
    pad=4
))

# make heatmap and save as figure
dataframe_fig = pivoted.iplot(kind='heatmap',
          layout=layout,
          colorscale='rdylbu',
          world_readable=False,
          asFigure=True)

# set zmin = 0 to not plot negative values
dataframe_fig['data'][0]['zmin'] = 0

# set zmax to 10000 or maximum observed value in figure, whichever is lower
zmax_obs = numpy.amax(dataframe_fig['data'][0]['z'])
if zmax_obs > 10000:
    dataframe_fig['data'][0]['zmax'] = 10000
elif zmax_obs > 10000:
    dataframe_fig['data'][0]['zmax'] = zmax_obs

# set filename
filename = 'cufflinks/17912/ZikaOpen' + '/' + title

py.iplot(dataframe_fig, filename=filename)

Out[61]:

I don't know why DENV-4 isn't respecting the zmin=0 setting, but it doesn't really matter. It is still clear there is not much evidence for cross-reactivity against these interesting ZIKV regions in any of the DENV subtypes.

Conclusion

The Roche/Nimblegen peptide array detected strong antibody responses, as expected, in SIV+ plasma samples. Responses were weaker, in general, in samples from Zika virus-infected macaques. Strong post-infection responses were observed in some animals. Modest to strong responses against the same region of the virus proteome were observed in many Zika virus-infected macaques; this region could be attractive for future reagent development because there does not appear to be cross-reactivity against the same region of dengue viruses. There was also no increase in the magnitude of responses 28 days following Zika virus rechallenge, supporting the concept that immunity against the primary infection protected from acquistion of virus upon reinfection, rather than partially containing viral replication below thresholds detectably by qRT-PCR. The peptide array technology holds promise for quickly characterizing antibody responses against viruses where little is known about the immune response.

In [ ]:
 

view message
Fine-mapping antibody epitope specificity
david h oconnor 2016-08-03
It is still preliminary, but we recently screened serum samples from Zika-infected macaques for antibody reactivity using a high-density peptide array. The array contains 894 proteins from viruses including Zika virus (ZIKV), but in our initial analysis we are focusing on reactivity against ZIKV peptides. We show data from macaque 295022 sera at days 0 (pre-infection), 67 (immediately before rechallenge), and 98 (28 days after rechallenge) against ZIKV MR766 peptides (accession: KU963573.1). The signal intensities correlate with antibody binding to peptides immobilized to the array. The higher the number, the greater the intensity of antibody binding.

Two regions have very strong binding at days 67 and 98 relative to background:

451-472: AKVEVTNSPRAEATLGGFGSL

Reactivity against peptides in this region is shown below (red cells are those with reactivity in the top 2% of all peptides tested in these samples)

Below is an alignment of this region in multiple ZIKV strains as well as other flaviviruses. Note that this region is proximal to the four amino acid in-frame deletion that has been described in some ZIKV MR766 isolates. As we showed previously, this in-frame deletion is selected against rapidly in vivo. The identification of antibody reactivity against this region raises the intriguing possibility that the insertions and deletions can be involved in antibody escape, though testing this hypothesis will require additional studies.

The second region is located at:

1423-1448: RAGDITWEKDAEVTGNSPRLDVALDE

The signal intensities for recognition of peptides in this region are:

This sequence is conserved between Asian and African ZIKV and may be an attractive specific target for therapeutic and diagnostic design. The strength of this response may also suggest that it is an immunodominant response. Conservation of this region relative to other flaviviruses is shown below:

The full table of results from this animal can be downloaded here. Note that we have a wealth of additional data from other samples and a large panel of viral proteins that will take time to analyze.

This peptide array data was provided by Roche NimbleGen through an early technology access program. If you are interested in learning more about the peptide array technology, contact: madison.peptidesales@roche.com

 Screen Shot 2016-08-03 at 4.40.35 PM.png  Screen Shot 2016-08-03 at 4.54.36 PM.png  Screen Shot 2016-08-03 at 4.57.18 PM.png  295022-ZIKV-MR766-reactivity.xlsx  Screen Shot 2016-08-03 at 5.09.16 PM.png  Screen Shot 2016-08-03 at 5.13.50 PM.png 
view message
More ZIKV-003 tissues positive for ZIKV vRNA
mmohns 2016-07-22
Further testing of tissues isolated from ZIKV-003 mother and fetus show that the maternal spleen and maternal liver also have trace amounts of vRNA detectable.

Maternal liver: 2.15e2 copies/mg

Maternal spleen: 3.23e2 copies/mg

These biopsies were taken at the time of C-section -- 120 days post-infection. The mother had prolonged viremia through day 70, but was then free of virus through the end of her pregnancy. Go here for the current list of raw data for tissues we have processed so far. We are still waiting on histopathology results from fixated tissues.

view message
A few additional thoughts on yesterday's detection of viral RNA in fetal tissues
david h oconnor 2016-07-08
Yesterday Mariel posted viral load data from the fetus born to the mother who had the longest sustained viremia. If the duration of maternal viremia correlates with fetal outcome (as we speculate it may), this would be the fetus where we might expect the most significant adverse outcomes.

Indeed, while we are still catching up on getting the data online, we did not detect viral RNA in any of the tissues we examined from fetuses born to mothers infected with Zika virus in the third trimester of pregnancy.

What does this mean in terms of fetal pathology? We do not know -- yet. Pathologists on our team are currently fixing the tissues, a process that takes about two weeks. After the tissues are fixed, we will know more about pathologic abnormalities.

--dave
view message
ZIKV-003 Fetectomy Tissues positive for Zika vRNA
mmohns 2016-07-07
This week we performed a scheduled C-section for the ZIKV-003 animal followed by fetal necropsy. This animal was challenged with Asian lineage virus in her 1st trimester, and had prolonged plasma viremia for 71 days post-infection.

After delivery, there were no observed clinical signs of microcephaly in the fetus. We biopsied over 60 tissues for investigative sampling, including looking for viral RNA in several tissues. In our first set of tissue samples for the ZIKV-003 fetus, we found viral RNA in the axillary lymph node, bone marrow, and the optic nerve of the eye. Other tissues run in this assay were negative for virus, including tissue samples from the brain.

Raw data can be found here, and will be updated as we continue to analyze further tissues from this fetectomy.

(See attached for high resolution graph)

 ZIKV003_fetectomy_screenshot.png  ZIKV003_fetectomy_2016.07.06.pdf 
view message
Monitoring fetal development via ultrasound
mmohns 2016-06-24
In all our pregnant macaques, we have been monitoring fetal development by ultrasound imaging at weekly timepoints. While the image files and raw data have been available in each study page here on the portal, it is a bit difficult to wade through all of the files. Here we provide a longitudinal visualization of fetal growth using the values for biparietal diameter (BD), head circumference (HC), and femur length (FL).

The clinical definition of microcephaly is anything greater than 3 standard deviations (SD) below the average. Overall, all fetuses appear to be smaller than average, but none are significantly decreased from the average. One limitation with this analysis is that the gestational ages of the fetuses are estimated. If the gestational age is off by a week or two, it may appear that the fetus is smaller than it actually should be at the correct calculated day.

We have also performed MRI imaging on the ZIKV-006 animals. Since we lack the expertise in this type of imaging, we are working with several collaborators to analyze and interpret the results from the scans.

Click on the attachments for high-resolution images of the charts below.

 fetal_growth_screenshot.png  BPD_2016.06.23.pdf  HC_2016.06.23.pdf  FL_2016.06.23.pdf 
view message
Tissue biopsies positive for ZIKV RNA
mmohns 2016-06-17
Last week, we posted about the successful C-section delivery and fetal necropsy for animal #598248 (ZIKV-006). We collected over 60 tissues for investigative sampling, including looking for viral RNA in several tissues. At that time, we also took a few tissue biopsies from the mother for sampling.

So far, the fetal tissues have been negative for ZIKV RNA, which is not particularly surprising given that animal #598248's plasma viral load was undetectable by 10 days post-infection. It's likely that the mother cleared her infection before the virus could pass the placental barrier.

That being said, two maternal tissues had positive yields for ZIKV RNA at the time of the C-section, 37 days post-infection:
#598248 mesenteric lymph node - 2.57E+03 vRNA copies/mg of tissue
#598248 spleen - 3.06E+03 vRNA copies/mg of tissue

We wanted to further investigate these findings in non-pregnant animals, so we performed lymph node biopsies from the ZIKV-004 animals 70 days post-infection just prior to homologous rechallenge with ZIKV-FP.

The left inguinal lymph node from animal #610107 was positive for ZIKV RNA at 3.40E+03 vRNA copies/mg of tissue.

While these two independent events demonstrate that viral nucleic acid is present in these tissues despite aviremic plasma, it does not necessarily mean that the virus is actively replicating in these tissues. For those that are using other animal models (i.e. mice), given these results it may be useful to pursue lymph node biopsies post-infection to see if similar results occur.

We also performed a C-section delivery and fetal necropsy for animal #357676 yesterday. Given that this animal had prolonged plasma viremia and a positive amniotic fluid viral load, we may expect to see viral RNA in the fetal tissues. We hope to process and collect data from these samples early next week, but given the extensive amounts of tissues to process and analyze, it may take a bit longer to upload the results.
view message
Confirmed protection from homologous reinfection and pregnancy updates
mmohns 2016-06-11
This past week, we reinfected our second cohort of macaques (ZIKV-004) primarily infected with the Asian lineage Zika virus. All three animals received a challenge dose of 10e4 PFU. After primary challenge, peak viremia was observed around 3-4 days post-infection. After 3-4 days post-rechallenge, virus remains undetectable in plasma and urine. We plan to perform viral loads on saliva swabs and vaginal swabs collected at the same time points in the near future.

Additionally, last week we performed a successful C-section delivery for pregnant animal #598248 (ZIKV-006) and performed a fetal necropsy at that time. We collected over 60 tissues for investigative sampling. We plan to analyze these samples for histopathology, cytokines, viral RNA, and live virus replication. Due to the exhaustive number of tissues, analysis will take some time, but we will post results as soon as possible.

Finally, uisng the ELISA kit from Xpressbio that we previously tested, we tested several samples of amniotic fluid from our pregnant animals. This suggests that IgG antibodies cross the placental barrier. In animal #598248, it is most likely that we did not have a time point far enough post-infection for antibodies to be detectable.

 ZIKV004_rechallenge.jpg  ELISA_results.png 
view message
Southern Research Begins Zika Open-Data Sharing
lankage22 2016-05-24
Southern Research has initiated experiments to better understand the course of ZIKV infection in non-human primates and develop animal models that may be used for the evaluation of candidate vaccines and therapies.

Southern Research Data Sharing Portal

view message
Macaques are protected from reinfection with heterologous Zika virus
mmohns 2016-05-24
The ZIKV-002 animals were initially challenged with different doses of African lineage Zika virus (Uganda 1947) on March 7th, 2016. We rechallenged these animals with 10e4 PFU of the Asian lineage virus stock 10 weeks later on May 16th, 2016. Virus is undetectable in all animals at 77dpi (one week post-rechallenge), demonstrating that macaques are protected from reinfection with heterologous Zika virus. Pan urine viral loads remain negative as well. This would suggest that any Asian-origin antigen used for vaccine design may cross-protect against Zika virus of other lineages.

Additionally, the ZIKV-001 viral loads remain negative at 98dpi (28 days post-homologous rechallenge). Together, these results may warrant future studies to evaluate long-term immunity either against subsequent Zika virus reinfections or potentially infections from other flaviviruses.

 ZIKV_002_VL_77dpi.png 
view message
Heterologous re-challenge of ZIKV-002 animals
mmohns 2016-05-18
ZIKV-002 animals were initially challenged with different doses of African lineage Zika virus (Uganda 1947) and viral trajectory was consistent with other acute infections from our studies, reaching a peak around 3-4dpi and undetectable around day 10. On Monday (70dpi), we rechallenged these animals with 10e4 PFU of the Asian lineage virus stock to determine if protection occurs following heterologous rechallenge. So far, all animals have negative plasma viral loads. We will continue to monitor them daily for 10 days and then weekly thereafter. If protection is observed, this would suggest that any Asian-origin antigen used for vaccine design may cross-protect against any virus of other lineages.

 ZIKV_002_PlasmaVL.png 
view message
Sustained viremia in pregnant macaques
mmohns 2016-05-18
The ZIKV-003 pregnant macaque that was infected in her 1st trimester still has low levels of virus detectable in plasma at 71dpi.

Last week we also detected virus in amniotic fluid for the first time in one of the ZIKV-006 (3rd trimester) animals. As of 22dpi, a positive viral load is still detectable in both plasma and amniotic fluid. We plan to continue weekly amniocentesis testing (as the health and safety of the animals allows) through the end of term.

It will be interesting to investigate why prolonged viremia is observed in some pregnant macaques, but not all.

 ZIKV_003_PlasmaVL.png  ZIKV_006_PlasmaVL.png  ZIKV_006_AmnioVL.png 
view message
Positive viral load in amniotic fluid
david h oconnor 2016-05-11
While we were unable to perform amniocentesis in the ZIKV-003 and ZIKV-005 animals soon after infection (it was too early in pregnancy), we collected amniotic fluid from the two ZIKV-006 macaques infected in the third trimester of pregnancy. One of these animals did not have detectable Zika virus RNA in amniotic fluid at either day 8 or day 15 post-infection, but the other had 12,300 viral RNA copies / mL of amniotic fluid at day 15. [amniotic fluid viral load chart]

--dave

view message
Updated manuscript published on biorxiv
david h oconnor 2016-05-10
An updated manuscript describing ZIKV-001, ZIKV--003, ZIK-004, and ZIKV-005 experiments is now available on the biorxiv preprint server: http://biorxiv.org/content/early/2016/05/10/046334

Let us know if you have any questions!

--dave
view message
Updates on pregnant macaques
david h oconnor 2016-05-10
Thanks to Sydney Skopos, Ted Golos, and Michael Graham, updated ultrasound images are available from ZIKV-003 (1st trimester infection), ZIKV-005 (1st trimester infection), and ZIKV-006 (3rd trimester infections).

The two fetuses from mothers who were infected during the first trimester are smaller than normal (values below are expressed as standard deviations SD below normal projected growth), but there is not enough data nor are the differences relative to normal large enough to confidently attribute this to Zika virus infection. As these infections progress, we anticipate that these differences, if they are due to Zika virus infection, will become more apparent.

--dave

 Screen Shot 2016-05-10 at 12.47.03 PM.png 
view message
Macaques are protected from reinfection with homologous Zika virus
david h oconnor 2016-05-10
Three ZIKV-001 animals infected with Asian lineage Zika virus were rechallenged with 10e4 PFU of the same virus 15 days ago. The key question was whether the immune responses elicited by primary infection provide protection from reacquisition of Zika virus. In all three animals, there is no evidence of plasma viremia at any timepoint following rechallenge.

On May 16, we will rechallenge the ZIKV-002 macaques with the same Asian lineage Zika virus. The key difference is that these animals were originally infected with African lineage Zika virus, so we will assess the degree to which exposure to African lineage virus confers protection against a virus that is somewhat genetically distinct. If we observe similar protection in this study, it will provide support that any Asian lineage vaccine immunogen will likely elicit immunity that would be cross-protective against other Asian lineage viruses.

--dave

 Screen Shot 2016-05-10 at 12.37.06 PM.png 
view message
Macaque variation in the ZIKV candidate receptor AXL
david h oconnor 2016-05-10
The protein AXL is a candidate cell surface receptor for Zika virus (see "Expression Analysis Highlights AXL as a Candidate Zika Virus Entry Receptor in Neural Stem Cells"). If AXL is important for Zika virus entry, polymorphism in the AXL gene may impact the biology of Zika virus infection.

We have curated exon variation in 174 Indian rhesus macaques and Mauritian cynomolgus macaques (MCM). The animals that we have infected with Zika virus so far are all Indian rhesus macaques, but there may be good reasons to infect MCM in the future. Michael Graham assessed variation in AXL within this cohort (see attached file). There are five nucleotide sites that would be predicted to encode variant amino acids.

If the role of AXL as the cell surface receptor for Zika virus is confirmed, it might be worth considering using AXL-identical macaques for Zika virus studies to eliminate the possibility that variation in this gene could influence experimental outcomes.

--dave

 AXLsummary.pdf 
view message
50 days and counting
david h oconnor 2016-05-01
The ZIKV-003 macaque infected with Asian-lineage Zika virus in the first trimester of pregnancy remains continuously viremic to 50 days post-infection.

The ZIKV-005 macaque, which was infected several weeks later chronologically but also during the first pregnancy trimester, also remains continuously viremic at 29 days post-infection.

--dave

 Screen Shot 2016-05-01 at 12.44.25 PM.png  Screen Shot 2016-05-01 at 12.46.27 PM.png 
view message
Two macaques are successfully infected in the third trimester of pregnancy
david h oconnor 2016-05-01
In ZIKV-006, we infected two third trimester rhesus macaques with Asian-lineage Zika virus. Both of these animals have characteristic plasma viral loads between 10E4-10E6 copies of viral RNA per mL. A key question for the upcoming week is whether these animals resolve their plasma viremia like non-pregnant animals or develop sustained viremia like the ZIKV-003 and ZIKV-005 macaques infected during the first trimester of pregnancy.

--dave

 Screen Shot 2016-05-01 at 12.41.04 PM.png 
view message
ZIKV-001 animals are protected from rechallenge
david h oconnor 2016-05-01
The initial objective of ZIKV-001 was to determine if macaques could be infected with Asian-lineage Zika virus. We quickly learned that they could. A secondary objective was to learn whether immune responses elicited by primary infection protect against subsequent reinfection. This asks the question - once you've been infected with Zika virus, can you be infected again? So if a woman is infected as a child, can she be reinfected during pregnancy?

Last week we rechallenged all three ZIKV-001 macaques to ask this question. None of the animals have had detectable plasma viremia in the four days since rechallenge. This indicates that immunity to primary Zika virus infection protects from reinfection.

We do not know yet how long it takes for a protective immune response to develop or whether this happens in every case. But at least in the three animals tested here, there is apparent sterilizing immunity from reinfection.

--dave

 Screen Shot 2016-05-01 at 12.34.23 PM.png 
view message
ELISPOT Data
mmohns 2016-04-26
ELISPOT data for ZIKV-002 and ZIKV-004 is now uploaded on each respective project page. We performed ELISPOT assays at 4dpi, 10dpi, and 14dpi using peptide pools spanning the NS5 peptide for either the African (ZIKV-002) or Asian (ZIKV-004) lineage zika virus. Each pool was comprised of 10 overlapping 15mer peptides. Several peptide pools elicited positive T cell responses in multiple animals with shared MHC haplotypes, suggesting that T cell responses may be restricted by common MHC alleles.
 ZIKV_002_ELISPOT_SUMMARY.png  ZIKV_004_ELISPOT_SUMMARY.png 
view message
Small fetal head size in ZIKV-003
david h oconnor 2016-04-22
As we noted previously, we are carefully monitoring the fetuses of mothers infected with Zika virus for evidence of abnormalities. Thus far, the most notable finding of Zika virus infection during pregnancy has been unusually lengthy plasma viremia. This week, Ted Golos, in collaboration with other maternal-fetal specialists, obtained new measurements from the ZIKV-003 fetus. Previous head size measurements of this fetus have been within the normal range. The most recent measurements are also within the normal range but are approximately 2 standard deviations below average. We also observed placental calcification. This is sometimes observed in normal pregnancies, but has also been noted in the pregnancies of women infected with Zika virus. A more detailed description of the results from Ted Golos is pasted below, and we certainly intend to continue following this fetus carefully.

"Team Placenta has been reviewing the measurements from ultrasound on ZIKV-003 Tuesday 4/19/16 to determine where the fetus lies on growth curves, such as are available for the rhesus. Because it is not precisely certain when pregnancy started, we are considering her gestational age on 4/19 as ~d76-81. There are two sets of reference data we have reviewed- one provides means +/- SD so I’ll refer to that data, the reference (Nyland et al) although the closest data it  provides is for d77 and d84. We compared 003 fetal biparietal diameter, head circumference, and femur length with data sets from Nyland. The fetus measurements are attached as is the reference paper. The mean +/- SD for BPD in Nyland et al on d77 is 24-28 mm, and 003 had 24.7 mm.  The femur length reference range is 12-18 mm, and 003 had 13.5 mm.  The reference head circumference range for mean +/- TWO SD is 90-112 mm. ZIKV-003 had 91.9 mm, so her head circumference is nearly 2 SD below the mean.  Katie Antony initiated the review of this data, and in humans, the Society of Maternal Fetal Medicine is recommending a neurosonogram for fetuses with HC >2SD below the mean, and defines microcephaly as >= 3 SD below the mean. So, to summarize, this fetus currently has a bpd at the low end of normal, and a rather low head circumference that would prompt further evaluation in the human clinic. She also has calcifications in the placenta (which admittedly occurs in normal pregnancies) and the fetal brain ultrasounds are under review.  So we can’t say there is a clear deficit, but continued surveillance is warranted."

--dave
view message
Zika virus detected in vaginal swabs
aweiler 2016-04-21
We tested vaginal swabs from animals in the ZIKV-004 study, from a few time points. At one week post infection we were able to detect virus in samples from all 3 animals. By two weeks post infection the virus was not detected.
view message
Sustained viremia in pregnancy update
david h oconnor 2016-04-20
Both pregnant animals remain viremic, now at 42 and 21 days post-infection.

Interestingly, we performed amniocentesis yesterday on the animal that is 42 days post-infection but did not detect virus in the amniotic fluid by qRT-PCR. Plasma viremia was also slightly lower at this timepoint than it has been the past several weeks. Next week, we are planning an ultrasound on the pregnant animal that will be 28 days post-infection.

We are also challenging two additional pregnant animals during the third trimester of pregnancy next Monday, April 25.

--dave
 Screen Shot 2016-04-20 at 3.18.36 PM.png  Screen Shot 2016-04-20 at 3.19.12 PM.png 
view message
Luminex Data
mrasheed 2016-04-14
Milliplex 23 plex NHP Immunology Panel (PCYTMG-40K-PX23) was used to perform Luminex on two ZIKV-002 animals’ CSF and 393422’s plasma from ZIKV-001. Miliplex’s protocol was used with overnight incubation at 4 degrees Celsius. An additional dilution was added for the CSF standards as well to account for possible lower analyte levels. Standards were created per protocol for plasma. The panel was run on a Bio-Plex 200 system and analyzed using the Bio-Plex Manager Software.
        405734 had no copies of ZIKV in the CSF at day 4 while 295022 did. At day 14, 405734 had detectable copies in the CSF while 295022 did not. In the graphs, 14 analytes are shown for each animal. Levels of analytes, according to the the data, increase during Zika Virus Infection.
    Animal 393422 plasma cytokine and chemokine profile is shown relative to 0 dpi. There is a general increase of concentrations around day 2 where we see peak viral loads.

-Mustafa
 295022 CSF Luminex.png  405734 CSF Luminex.png  393422 Plasma Luminex.pdf 
view message
Unconventional learning about science in the 21st century
david h oconnor 2016-04-13
The other day I posted a message about sequence differences in various "MR766" isolates, including a 12 nt in-frame deletion in some isolates. To illustrate how journals aren't the only way scientific knowledge can be communicated today, I thought I would share two anecdotes from recent days.

First, soon after writing that post, Nick Loman (@pathogenomeick) tweeted at me that he had just sequenced a mosquito-derived ZIKV isolate (MP1751) and found it nearly identical to the sequence of Genbank DQ859059.1, which is in Genbank as a MR766 strain. This provides a second data point suggesting that the strain label of DQ859059 in Genbank is likely incorrect.

Second, this morning I listened to the TWiV podcast #383 (http://www.virology.ws/2016/04/03/twiv-383-a-zillion-zika-papers-and-a-brazilian/) featuring my friend and colleague Esper Kallas. During the podcast, they mentioned a comprehensive review on Zika virus that I immediately downloaded and began reading (http://cmr.asm.org/content/29/3/487.abstract). In this review, they noted two previous studies that had looked at the specific deletion in MR766 and correlated it with a potential glycosylation site that could be relevant to pathogenesis. A description of differences among MR766 isolates, as well as a note about the possible impact of the deletion on glycosylation, is described here: http://journals.plos.org/plosntds/article?id=10.1371/journal.pntd.0001477#pntd-0001477-g002

That viruses with the deletion are not detected two days post-infection suggests that it is selected against in vivo in macaques, which builds on these previous observations. It also underscores that there is currently a lot of confusion in the naming of Zika virus isolates, which is something that will bedevil the field unless better standards for naming are agreed to and enforced by the community.

That Twitter and a podcast provide important information isn't and shouldn't be surprising. A huge part of the job of being a scientist is collecting information and synthesizing the observations of others. When I started as a scientist, in-person conversations, textbooks, journals, and to a lesser extent magazines and newspapers were the ways I learned about new information. It's a very different world today.

--dave
view message
Sustained viremia in pregnant macaques update
david h oconnor 2016-04-12
We now have two pregnant macaques (ZIKV-003 and ZIKV-005 studies) infected with Asian-lineage ZIKV. Both have had detectable plasma viremia at every timepoint tested. The animal infected first remains viremic 35 days post-infection. The second animal remains viremic at day 14 post-infection (both of these samples were collected yesterday, April 12, 2016). This is different from non-pregnant animals where viremia is typically cleared by approximately 10 days post-infection.

The obvious next question is whether prolonged detection of viremia correlates with fetal abnormalities.

--dave

 Screen Shot 2016-04-12 at 10.06.51 AM.png  Screen Shot 2016-04-12 at 10.07.14 AM.png 
view message
Source of prolonged viremia in pregnant macaque
david h oconnor 2016-04-11
We just completed a preliminary analysis comparing viruses at different timepoints in the pregnant female in ZIKV-003 who is exhibiting prolonged viremia. There are two lines of evidence that lead us to believe the virus we continue to detect in maternal plasma is coming from the fetus. First, we only see virus in plasma, not in urine, which suggests that the maternal immune system was able to adequately contain virus replication. Second, and perhaps more persuasively, viral variants detected near the end of the "typical" viremia (day 7) are completely distinct from those observed during the "prolonged" viremia (day 21). This would be extremely unlikely during a continuous infection of the mother, but is exactly what one would expect if the fetus is infected with a small number of transmitted/founder viruses from the mother during the initial burst of virus replication.

The complete analysis is here: https://zika.labkey.com/wiki/OConnor/ZIKV-003/page.view?name=prolonged_viremia_origin

--dave and shelby
view message
What's in a name?
david h oconnor 2016-04-10
I just completed analyzing deep sequencing data from three African-linage challenge stocks, all called "MR766." [https://zika.labkey.com/wiki/OConnor/ZIKV-002/page.view?name=17301_mr766_challenge_comparison] One of these was used to challenge macaques in ZIKV-002, while one was the seed stock used to generate this stock and a third was from an independent source. In the course of analyzing this data, I discovered Genbank has at least six sequences all called some form of "MR766" -- but they aren't necessarily the same sequence!

I show that the viruses we used to infect the ZIKV-002 animals are most similar to a virus from UTMB (Genbank accession: HQ234498) but that the majority of sequences in the challenge stock harbor a deletion in the envelope peptide that are not in the UTMB Genbank sequence. Within two days of infection, though, sequences with this deletion cannot be found suggesting strong selection against deletion in vivo. At the amino acid level, viruses circulating two days after infection have an identical sequence to the polyprotein sequence of HQ234498.

Since MR766 is one of the most widely available ZIKV viruses, this analysis shows how important it is to sequence specific viruses used in analyses, since these sequences differences could have unpredictable biological consequences. It also shows that as a community virologists need to be very careful when comparing results across studies where the same strain name is used. In HIV/SIV, we learned this lesson the hard way, with many studies using viruses like "SIVmac251" that do not necessarily have the same sequence composition.

--dave
view message
Prolonged viremia in pregnant macaque (ZIKV-003)
david h oconnor 2016-04-06
This morning we received viral loads from ZIKV-003, the first study of a pregnant macaque infected with Asian-lineage Zika virus. In stark contrast to other animals, this animal has maintained viremia for 28 days after infection. Yesterday (day 28) the viral load was a robust 11,500 copies / mL. This is similar to the prolonged viremia reported in a pregnant woman last week in the New England Journal of Medicine. We concur with the hypothesis put forward in the manuscript that prolonged viremia may represent shedding of virus from the fetus back into the mother. The next obvious question - will the fetus show any developmental effects?

--dave

 Screen Shot 2016-04-06 at 2.43.26 PM.png 
view message
Plasmablast responses in ZIKV-002
david h oconnor 2016-04-05
Josh Eudailey and Sallie Permar (Duke University) assessed plasmablast (PB) responses in the ZIKV-002 animals using the same approach they used to study the ZIKV-001 animals. A question is whether the lower viral loads in ZIKV-002 corresponded with a less robust PB response. Here is the data and analysis per Josh (data files are attached):

---
The findings are as follows:
• For the lowest dose of virus, the greatest # of PBs are seen on Day 14 in both studies, however the increase is much more gradual in the ZIKV002 study.
• For the intermediate dose, the PBs are at their peak at Day 11/10 in both studies, however they seem to reach this peak by Day 7 in the ZIKV001 study.
• For the highest dose, the PBs peak at Day 7 before tapering off in ZIKV001, but don’t peak until Day 10 and then taper off for ZIKV002.
In short, the kinetics are similar for the lowest dose, however the PB response is greater in ZIKV001. The response is similar for the intermediate dose, but the kinetics are slightly faster for the ZIKV001 NHPs. Both the kinetics and PB peak response are increased for ZIKV001 at the highest dose.
---

--dave
 ZIKV001 vs ZIKV002 Plasmablast Comparison.pptx  ZIKV002 Plasmablast Absolute Numbers.xlsx 
view message
CSF viral loads
david h oconnor 2016-04-04
During a discussion this morning, we talked quite a bit about CSF sampling. One caveat that we hadn't fully recognized is that a very small amount of blood contamination in the needle used for lumbar puncture could contribute to measured day 4 viral loads. Soon enough, we hope to have direct data on whether virus is in the CNS.

--dave
view message
NHP placental biology
david h oconnor 2016-03-31
In light of the ZIKV-003 and ZIKV-005 challenges of pregnant females, we realized some viewers might want more information about the reproductive biology of these animals. Especially for those of us (like me!) coming from an infectious disease background, pregnancy is a mystery. Here are some slides courtesy of Ted Golos describing placental morphology.

--dave
 Placental Morphology.ppt 
view message
ZIKV-004 and ZIKV-005 infections
david h oconnor 2016-03-31
All animals in ZIKV-004 and ZIKV-005 have detectable viremia one day post-challenge. As of now, 11/11 macaques challenged with Zika virus have become productively infected.

--dave
view message
Natural history of Asian lineage Zika virus infection in macaques
david h oconnor 2016-03-31
We have completed our first manuscript describing the ZIKV-001 study. It is under submission at a peer-reviewed journal but we have also published it to BioRxiv [link]

Please let us know if you have any questions about the manuscript.

--dave

 046334.full.pdf 
view message
Persistent viremia in a pregnant macaque?
(1 response) david h oconnor 2016-03-30
Though the number of animals infected remains small, the first pregnant animal we infected with Zika virus (in ZIKV-003) remains viremic at 21 days post-infection, which is already 7 days longer than any of the non-pregnant animals. This could have implications for both persistence and fetal effects.

--dave
view message
Positive selection in NS1 peptide
david h oconnor 2016-03-26
Louise Moncla and Tom Friedrich calculated measures of synonymous (πS) and nonsynonymous (πN) nucleotide diversity in each coding region of the genome relative to the inoculum for each animal at 4 dpi. The ratio of πS to πN can indicate how selection is acting on a gene. πS > πN suggests that purifying selection is acting to remove new variation, πN > πS suggests that diversifying selection is favoring new mutations, and πN = πS indicates no strong selection. While most coding regions revealed signatures of purifying selection, the region encoding NS1 exhibited elevated levels of nonsynonymous diversity in all three animals.
 zika_piNpiS_bar_2.5x8.pdf 
view message
Zika Virus Antibody ELISA
lankage22 2016-03-23
We tested an ELISA kit from XpressBio, the NHP Zika serology test kit. Xpressbio Elisa Kit This kit uses a Ugandan ZIKV NS1 as the capture antigen. There is about 97.5% amino acid identity (NS1 region) between the Ugandan virus and the French Polynesian virus, which we tested here.

For our initial test of this kit we used serum from 3 rhesus macaques infected with a French Polynesian strain of Zika virus. To test for cross-reactivity we also tested serum from an animal that had been infected with dengue and one that was vaccinated for yellow fever. All three of the Zika virus positive animals tested positive by ELISA while the other 2 animals did not.

Here are the results from the ELISA. Results are an average of 2 tests per sample.

 + Antigen strips- Antigen strips
Positive control2.4950.139
Negative control0.1030.113
Dengue+0.1650.105
Yellow Fever+ 0.1110.123
ZIKV +2.9550.108
ZIKV +2.7870.118
ZIKV +2.9930.120
Empty0.1010.109
view message
Zequencer: a Geneious Pro workflow for analyzing Zika virus sequencing data
lankage22 2016-03-22
Over the last few weeks I have helped several groups analyze Zika virus sequencing data, as well as working with Shelby to analyze data collected in ZIKV-001 and ZIKV-002. I developed a straightforward and simple workflow in Geneious Pro that can be used for push-button analysis of Zika virus datasets.

Detailed information about the workflow is here [Bitbucket] and the workflow file can be downloaded here [ Geneious Pro]

Let me know if you find it useful!

--dave

view message
Infectious virus quantification from ZIKV-infected macaque plasma
lankage22 2016-03-22
I have attempted to titrate ZIKV from macaque plasma to quantify the amount of infectious virus present in infected animals. To date this has been unsuccessful. The method I use is a plaque assay on Vero cell culture. Briefly, Duplicate wells are infected with 0.1 ml aliquots from serial 10-fold dilutions in maintenance media and virus is adsorbed for one hour. Following incubation, the inoculum is removed, and monolayers are overlaid with 3 ml containing a 1:1 mixture of 1.2% oxoid agar and 2X DMEM (Gibco, Carlsbad, CA) with 10% (vol/vol) FBS and 2% (vol/vol) penicillin/streptomycin. Cells are incubated at 37°C in 5% CO2 for four days for plaque development. Cell monolayers then are stained with 3 ml of overlay containing a 1:1 mixture of 1.2% oxoid agar and 2X DMEM with 2% (vol/vol) FBS, 2% (vol/vol) penicillin/streptomycin, and 0.33% neutral red (Gibco). Cells are incubated overnight at 37°C and plaques were counted. This method has been successful for titration of ZIKV challenge virus (straight from tissue culture), ZIKV virus recovered from mouse serum, and ZIKV virus recovered from mosquitoes. However, it has been problematic for macaque plasma. Initially, I attempted to titer all plasma samples from ZIKV-001 that were positive via qRT-PCR using 10-fold dilutions from 10E-1 to 10E-3. All samples were negative for infectious virus. Next, I attempted to titer undiluted plasma. All but three samples were negative for infectious virus. The three positive samples had very low levels of virus present 2.00-2.50 log10 PFU/ml. This begs the question, is there replicating virus in our animals? I think so, but we will be doing further testing to determine whether or not these data are accurate or if for whatever reason macaque plasma causes interference in bioassays (common in the dengue virus world).
--Matt Aliota
I think there are several reasons to believe the virus is replicating --
1. We see viral loads increase and persist for variable periods of time in different animals.
2. We observe adaptive immune responses following challenge
3. The virus we recover from animals has nucleotide variation relative to the input virus
But the point remains...we have had a difficult time getting infectious virus out of macaque plasma. We intend to try several other approaches (diluting plasma and then concentrating virus, spiking known amounts of purified virus into plasma to define the relative degree of inhibition). But I wanted to make this information available for others who may be trying to grow virus directly from macaque plasma/serum.
--Dave O'Connor
view message
Asian-lineage Zika virus sequencing with 5 overlapping PCR amplicons
lankage22 2016-03-22
AIDS Vaccine Research Laboratory, UW-Madison Dawn Dudley, Shelby O’Connor, Dane Gellerup

Here is the protocol we are using to prepare viral RNA, reverse transcribe, and PCR amplify with five overlapping PCR primer pairs to generate amplicons for Illumina deep sequencing. These primer pairs might be useful for others interested in PCR amplifying Asian-lineage Zika virus. Similar primers for African-lineage Zika virus PCR amplification are under development.

Isolate viral RNA.

  1. Using a QIAamp MinElute Virus Spin Kit, elute vRNA in 25µL Elution Buffer.

Dilute 10µL of 100µM primer into 90µL Nuclease-Free H2O. End result required is 10µM primer.

  1. Forward Primer 1: GCGACAGTTCGAGTTTGAAGCG
  2. Reverse Primer 1: ATCCAAAGTCCCAGGCTGTG
  3. Forward Primer 2: AGATCCCGGCTGAAACACTG
  4. Reverse Primer 2: CCCATGTGATGTCACCTGCT
  5. Forward Primer 3: TACTCACAGCTGTTGGCCTG
  6. Reverse Primer 3: CACCTCGGTTTGAGCACTCT
  7. Forward Primer 4: TGTTTGGCTGGCCTATCAGG
  8. Reverse Primer 4: CTGCGGATCCTTTCAATGCG
  9. Forward Primer 5: TATGGGGGAGGACTGGTCAG
  10. Reverse Primer 5: ACTAGCAGGCCTGACAACAC

Using an Invitrogen SuperScript III One-Step RT-PCR System with Platinum Taq High Fidelity kit, make an rtPCR master mix for all samples. All values are per sample:

  1. 2x Reaction Mix:...………………………………….12.5µL
  2. SuperScript III RT/Platinum Taq Enzyme Mix:…...0.5 µL
  3. Ultra Clean PCR Water:…………………………….8.5 µL
  4. MgSO4 (1.5mM):…………………………………….1.5 µL

Aliquot 23µL Master Mix into PCR strip cap tubes.

Add 0.5µL forward and 0.5µL reverse primer to each tube.

Add 3µL viral RNA to each tube. Mix and spin down.

On a thermocycler, run a program:

  1. 55C for 30 minutes
  2. 94C 2 minutes

  1. 35 cycles of the following:
  2. 94C 15 seconds
  3. 56C 30 seconds
  4. 68C 3.5 minutes

  1. 68C 10 minutes
  2. 10C forever
view message

ZIKV-001: Infection of three rhesus macaques with French Polynesian Zika virus 

Primary objectives

  • Assess the infectivity of a Zika virus isolate from French Polynesia at different doses in rhesus macaques
  • Measure concentration of Zika virus RNA in plasma, urine, CSF, saliva, and feces
  • Determine whether immunity elicited by Zika virus infection protects from subsequent re-infection with genetically similar, Asian-lineage Zika viruses

Study design

Three Indian-origin rhesus macaques will be challenged subcutaneously with different doses of French Polynesian Zika virus. For each of the first 10 days, samples will be collected daily for intensive virologic analyses. From days 11-28, samples will be collected 2-3x per week. After 28 days, the animals will be rested for approximately 6 weeks before being re-challenged with 10E6 PFU of French Polynesian Zika virus.

Result summary

All three macaques were successfully infected with Zika virus. Plasma viremia peaked at more than 1e6 viral RNA copies / mL in two of the three animals. Plasma viremia resolved by approximately 10 days post-infection. In two of three animals, viral RNA was detected in urine slightly longer than in blood. Viral RNA was also detected in salvia from all three animals.

Real-time study results


ZIKV-002: Infection of three rhesus macaques with Uganda 1947 Zika virus 

Primary objectives

  • Assess the infectivity of a Zika virus isolate from Africa at different doses in rhesus macaques
  • Measure concentration of Zika virus RNA in plasma, urine, CSF, saliva, and feces
  • Determine whether immunity elicited by Zika virus infection protects from subsequent re-infection with genetically heterologous, Asian-lineage Zika viruses

Study design

Three Indian-origin rhesus macaques will be challenged subcutaneously with different doses of African lineage Zika virus (Uganda 1947). For each of the first 10 days, samples will be collected daily for intensive virologic analyses. From days 11-28, samples will be collected 2-3x per week. After 28 days, the animals will be rested for approximately 6 weeks before being re-challenged with 10E6 PFU of French Polynesian Zika virus.

Result summary

This study started Monday, March 7, 2016 and is under way.

Real-time study results


ZIKV-003: Infection with French Polynesian Zika virus during the first pregnancy trimester 

Primary objectives

  • Assess whether fetal development is impacted by maternal infection with French Polynesian Zika virus during the first pregnancy trimester
  • Measure concentration of Zika virus RNA in amniotic fluid and virus transmission to the fetus
  • Pilot methods for studying Zika virus infection during pregnancy

Study design

A single pregnant Indian-origin rhesus macaque (approximately 25-30 days post-conception) will be challenged subcutaneously with 10E4 PFU French Polynesian Zika virus. Natural history of Zika virus will be evaluated as in ZIKV-001. Additionally, we will quantify virus in amniotic fluid at multiple timepoints and monitor fetal development by ultrasound imaging.

Result summary

This study started Monday, March 7, 2016 and is under way.

Real-time study results


ZIKV-004: Infection of three rhesus macaques with French Polynesian Zika virus 

Primary Objectives

  • Assess the infectivity of a Zika virus isolate from French Polynesia at different doses in rhesus macaques.

Study design

Three female rhesus macaques will be challenged with different doses of French Polynesian Zika virus.

Real-time study results

This study began 3/28/16 and is under way.


ZIKV-005: Infection of a single pregnant rhesus macaque with Asian lineage Zika virus 

Primary objectives

  • Assess whether fetal development is impacted by maternal infection with Asian lineage Zika virus during the first pregnancy trimester
  • Measure concentration of Zika virus RNA in amniotic fluid and virus transmission to the fetus
  • Pilot methods for studying Zika virus infection during pregnancy

Study design

A single pregnant rhesus macaque will be challenged subcutaneously with 10E4 PFU Asian lineage Zika virus. We will quantify virus in amniotic fluid at multiple timepoints and monitor fetal development by ultrasound imaging.

Result summary

This study started Monday, March 28, 2016 and is under way.

Real-time study results


ZIKV-006 

Primary objectives

  • Assess whether fetal development is impacted by maternal infection with Asian lineage Zika virus during the third pregnancy trimester
  • Pilot methods for studying Zika virus infection during pregnancy

Study design

Two pregnant rhesus macaques will be challenged subcutaneously with 10E4 PFU Asian lineage Zika virus. We will monitor fetal development by ultrasound imaging.

Result summary

This study began April 25, 2016.

Real-time study results


California National Primate Research Center Zika virus projects 
I'm delighted to report that the California National Primate Research Center Zika virus studies, led by Koen Van Rompay, are also sharing their data in real-time. Who's next? Contact me (dhoconno@wisc.edu) or LabKey Software if you are interested in having your Zika virus data hosted on a new LabKey Server that has been setup for this purpose.

California National Primate Research Center Zika virus studies

--dave


Oregon National Primate Research Center Zika virus projects 

Our colleagues at the Oregon National Primate Research Center are also making their Zika virus study results available in real-time. [link]


Rationale for using nonhuman primates to study Zika virus 
There has long been a reluctance, born out of experience, that making data from experiments on animals (particularly nonhuman primates) publicly available is inadvisable because of the potential for information to be taken out of context by those who oppose animal research. We have argued vigorously that the public health emergency of Zika virus demands transparent, open data sharing. We ask that those of you who consider using information we provide here to slow research on Zika virus to reconsider. We know nearly nothing about the virology, immunology, and pathogenesis of this virus. Zika virus infection of pregnant women has been associated with severe birth defects, but it is still far from clear how, or how often, Zika virus infection may induce these defects in developing fetuses. There have also been anecdotal reports that Zika virus can be transmitted sexually, and that it is shed in saliva. With the minimal data available, we cannot yet determine how severe the risks of sexual or other non-mosquito transmission of Zika might be. Understanding these critical biological features in animals where we control the dose, route, and timing of infection will accelerate the accumulation of knowledge and hopefully alleviate the suffering of people who are impacted by this virus. Some may not agree that this is sufficient justification for animal research; we can respectfully disagree.


Project ZEST funding 
We are grateful for the financial support of:
  1. NIH P51 5P51OD011106-54 to the Wisconsin National Primate Research Center
  2. NIH R01 supplement 3R01AI116382-01A1S1 to O'Connor


Project ZEST investigators 
The Project Zika experimental science team (ZEST) is comprised on interdisciplinary researchers and clinicians interested in understanding why Zika virus has recently been associated with significant disease. This information will aid in the development and evaluation of interventions to minimize future Zika virus-associated disease.

Project leaders
Dave O'Connor
Jorge Osorio
Ted Golos
Tom Friedrich

Immunology and pathogenesis
Dave O'Connor
Emma Mohr
Dawn Dudley
Christina Newman
Laurel Stewart
Adam Bailey
Mariel Mohhs
Meghan Breitbach
Michelle Koenig
Mustafa Rasheed
Sallie Permar (Duke University)

Vector biology
Jorge Osorio
Matt Aliota
Bruce Christensen

Pregnancy and fetal development
Ted Golos
Greg Wiepz
Igor Iruretagoyena
Ei Terasawa

Virology
Tom Friedrich
Andrea Weiler
Gabrielle Lehrer-Brey
Dave O'Connor
Jorge Osorio
Matt Aliota
Shelby O'Connor
Dane Gellerup

Neurobiology
Jon Levine
Shahriar Salamat

Nonhuman primate core
Saverio "Buddy" Capuano
Nancy Schultz-Darken
Jennifer Post

Administrative core
Kristi Hall
Sandra Boehm

Data management
Michelle Koenig
Michael Graham
Mustafa Rasheed
Laurel Stewart
Luis Gomes
Kirsten Wingate
Josh Eckels (LabKey Software)

Translational core
Emma Mohr
Esper Kallas
Amilcar Tanur
Renato Aguiar
Carlos Moreira