MERS-CoV: alpacapalooza...

In the search for the animals that may be another reservoir for, or just support infection by, Middle East respiratory syndrome coronavirus (MERS-CoV), a few studies have looked at furry little alpacas (Vicugna pacos). 

We already know from Eckerle and colleagues' work that cells derived from alpacas have the required receptor molecule used by MERS-COV (DPP4) and that they can support replication of MERS-CoV in the lab,[1] But are cells in flasks different from furry animals in the wild?

Three articles came out in the June 2016 issue of Emerging Infectious Diseases looking at alpacas and MERS-CoV. There's no way to tell who submitted when in this journal (ongoing grr).

A Colorado State University team sought an easier - but still relevant - animal model than camels to work with.[2] They infected 3 (called A1, A2 and A3) alpacas with 10⁷ plaque forming units (PFU; a cell-specific measure of the amount of infectious virus in a diluted sample) of the HCoV-EMC/2012 variant of MERS-CoV via 3ml of diluted virus per nostril. 3 days later, they housed 3 more uninfected alpacas (A4-A6) with the infected ones.  

After 70 days, A1-A6 were infected again ("challenged"), in the same way. Three other alpacas were infected the same way but euthanized 5 days after infection and their tissues collected for analysis. Nasal swabs were collected before infection and then daily from all living animals for 5 days post-infection and on day 10. A4-A6 were also swabbed 3 times per week to day 19. This could all have been a bit more clearly demonstrated using a timeline by the way.

None of the infected animals had a fever or observable nasal discharge and their appetites and activity remained constant; they didn't seem to be affected by infection with MERS-CoV. Infectious MERS-CoV was shed by A1-A3 to day 5 and transmission occurred to the A4-A6 arrivals:

• A4 was found to shed virus for 1 day, 
• A5 shed no virus and 
• A6 shed across 8 days. 

Lots of variation even in a controlled environment like this.

When furries A1-A6 were challenged with a fresh infection, A1-A3 developed antibodies which protected them from infection (because no virus was shed). A4-A6 shed a little infectious virus for at most 2 days. A4-A6 also developed neutralizing antibodies but took longer than A1-A3 to do so. 

A wild transmission occurred from inoculated A1-A3 to naive A4-A6. This was a lower dose than the original inoculum, and seemed to elicit a milder antibody response in animals A4-A6 as well. 

Virus was found in the nose, larynx and trachea of A7-A9 but not the lungs.

The Australian/Singapore team sought an animal model of MERS with a better temperament and more manageable size than the camel.[3] 

Under biosafety level 3 conditions, they used a camel MERS-CoV variant (Al-Hasa_KFU-HKU13/2013) and exposed each alpaca to 10⁶ 50% tissue culture infectious doses (TCID50; another cell-specific measure of infectious virus quantity), monitored for 21 days then challenged as described above.

Blood as well as nasal, oral, rectal and urogenital swabs were collected over time and tested by sensitive RT-PCR, culture and for the presence of neutralizing antibodies.

The furries once again did not develop a fever (animal No.2 had a raised temperature though) or a respiratory illness.
Not a great model of disease, perhaps good for transmission?

Infectious virus was isolated from oral upper and deep nasal swabs but not from urogenital or rectal swabs. RNA detection by RT-PCR followed this same pattern. After challenge and in the presence of antibodies which had appeared from day 10-12, viral RNA could not be detected anywhere in any animal.

Neutralizing antibody did not appear until 21 days in animal No. 1, 10 days in No. 2 and still hadn't appeared at day 35 in No. 3. But, it was apparently unnecessary for protection from reinfection in this study of alpacas.

These 2 studies used 10⁶-10⁷ cell-specific quantities of MERS-CoV to infect the alpacas, but, when sought, less was produced by the newly infected animals - except in one of the Australian/Singapore animals where the peak of 10⁶ TCID50 detected equalled the input dose. This finding suggests the laboratory inoculum may not be relatable to real-world amounts of virus produced by an infected animal source. It may however, just be how much is needed to get a model system infected.

The authors all agreed that alpacas could be a good model for MERS-CoV in camels and that animal infections supported the finding of the initial alpaca cell culture work. But that culture link is not quite so straightforward.

Deliberately infected goats, sheep and horses (using the HCoV-EMC/2012 variant of MERS-CoV) - also animals whose cells had supported MERS-CoV in the laboratory - showed little or no sign of viral replication and the animals mostly remained healthy (some nasal discharge was seen from 2 of 4 horses).[5]

Despite some signs of neutralizing antibody developing in goats and a sheep, the same Colorado team were not convinced that any of these animals would be likely hosts for MERS-CoV in the wild.

Cells in a flask are not always the most realistic model for animal transmission I guess.

The final alpaca article was from a Qatar/Netherlands team who tested alpacas in a region of Qatar where MERS-CoV is found to be enzootic among camels (spreading naturally among the animals).[4] 

Hobby alpaca and camel herds were the subjects of this study. They had been kept about 200m apart in the same farm and cared for by the same animal workers.

Blood samples were tested from 15 alpacas and 10 dromedary camels; nasal swabs were also collected from the 10 camels. Nasal, rectal and oral samples were only collected from a subset of the alpacas for antibody testing and sensitive RT-PCR.

MERS-CoV neutralizing antibody was present in 15 of 15 alpacas and 9 of the 10 camels according to a 90% plaque-reduction neutralization test. This indicated past natural infections had occurred.

Antibodies were also detected that suggested past infection by dromedary betacoronaviruses and camelid alphacoronaviruses but this was not unusual nor unexpected.

No swabs were positive by RT-PCR indicating that no animals were infected at the time of sampling. The authors did not know when, how or how often MERS-CoV may have naturally infected the alpacas.

So we can add alpacas to camels on the short list of animals that can host MERS-CoV infection.

References...

1. Replicative Capacity of MERS Coronavirus in Livestock Cell Lines
   https://wwwnc.cdc.gov/eid/article/20/2/pdfs/13-1182.pdf
2. Infection, Replication, and Transmission of Middle East Respiratory Syndrome Coronavirus in Alpacas
   http://wwwnc.cdc.gov/eid/article/22/6/pdfs/16-0192.pdf
3. Experimental Infection and Response to Rechallenge of Alpacas with Middle East Respiratory Syndrome Coronavirus
   http://wwwnc.cdc.gov/eid/article/22/6/pdfs/16-0007.pdf
4. MERS-CoV Infection of Alpaca in a Region Where MERS-CoV is Endemic
   http://wwwnc.cdc.gov/eid/article/22/6/pdfs/15-2113.pdf
5. Inoculation of Goats, Sheep, and Horses with MERS-CoV Does Not Result in Productive Viral Shedding
   http://www.mdpi.com/1999-4915/8/8/230

A negative viral culture result is not the end of the story...just a negative result

Lately, for both Ebola virus and Middle East respiratory syndrome coronavirus, there have been instances where I've been reminded that one must not rely on the growth of an infectious virus from a sample to be sure that there is virus in that sample.

Past diagnostic methods have failed to isolate many newly identified viruses (NIVs), which is not surprising considering that those culture-based methods can be over 100-fold less sensitive than current molecular (PCR-based) tests [1,2,3,4] and that many new viruses do not grow in traditional culture at all.

So when an RT-PCR or PCR result cannot be confirmed by the culture of a virus from the same sample, that really doesn't mean more than...that. 

Virus may be present, but our relatively insensitive culture techniques, which haven't really advanced in a long time, may just fail to get it growing. 

Viral isolation by cell culture is really a dying art form. And it is a very lengthy, demanding and sometimes subjective art form at that requiring particularly skilled artistes. 

References...

1. Templeton,K.E. et al. Improved diagnosis of the etiology of community-acquired pneumonia with real-time polymerase chain reaction. Clin Infect Dis 41, 345-351 (2005).
2. van Kraaij,M.G.J. et al. Frequent detection of respiratory viruses in adult recipients of stem cell transplants with the use of real-time polymerase chain reaction, compared with viral culture. Clin Infect Dis 40, (2005).
3. Garbino,J. et al. Lower respiratory viral illnesses: Improved diagnosis by molecular methods and clinical impact. Am J Resp Crit Care Med 170, (2004).
4. Gunson,R.N., Collins,T.C., & Carman,W.F. Real-time RT-PCR detection of 12 respiratory viral infections in four triplex reactions. Journal of Clinical Virology(2005).

The impact of cell culture on virus as highlighted by deep sequencing..[UPDATED]

Alignment of complete or near complete MERS-CoV
genomes made using Geneious v6.1.7.
The Neighbor-Joining tree was made using MEGA with
1000 bootstraps. Red stars indicate virus from same patient. 
Vertical bars to the right indicate Clade A (dark blue)
and B (pale blue). Sequences from the the 2013 Al-Ahsa
hospital outbreak are boxed in pink. GenBank
accession numbers are indicated at the end of
each sequence name which also includes region
of detection, host and date of sample collection.
Click on image to enlarge.

Just a quick post to note the difference that "passage" in cell culture (isolation of an infectious virus using lab cells inoculated with the original virus-positive patient material, some of which is taken off and added to a new flask of cells and this process repeated as needed)  can do to the virus as it changes to grow most effectively in the new environment...this is akin to the adaptive changes seen when a virus first jumps to a an entirely different species.

Apparently the 2 starred (red stars in the figure) virus genomes are from the same 60-year old male patient [1,2,3; ] but the original variant, EMC/2012, was sequenced from material after 6 rounds of cell culture [3] while the Bisha_1 variant was not [1]; it was subjected to deep sequencing directly after nucleic acid preparation using an original respiratory sample aliquot (nasopharyngeal swab)[1]. 

Given that cell culture passage seems to be related to the positioning of EMC/2012 in Clade A versus Bisha_1 in Clade B (indicated by a pale blue line), does this mean there is no Clade A (dark blue vertical bar) and that it's just an artefact??

Probably not. Why? Because the Jordan-N3/2012 virus that is also found in Clade A and it also originates from a human specimen collected in 2012. It is listed as having been sequenced from a bronchial sample. There is no mention of cell culture on its GenBank record - which does not mean there was no culture. But when that sample was passaged through culture and sequenced (N3/2012 MG167; sequence not shown in this tree) it remained 99.95% identical to the original sequence; just 2 nucleotide differences out of 30,028nt. 1 difference in the spike gene and 1 Open Reading Frame 1a). These are unlikely to be enough to switch its clade  but I'm realigning with this sequence included just to be sure about that!

I thought that was kinda interesting.

NB. There may be some concern over the specimen labelling used to identify samples for sequencing of EMC/2012 or Bisha_1. I'm attempting to sort our by following this up with the lead author.[1]

Reference...

1. Transmission and evolution of the Middle East respiratory syndrome coronavirus in Saudi Arabia: a descriptive genomic study. Cotten et al. Lancet 2012;382:1993-2002
2. 60-year old man from Bisha who died in a Jeddah hospital (EMC/2012 variant)
   Isolation of a Novel Coronavirus from a Man with Pneumonia in Saudi Arabia. Zaki et al. N Engl J Med 2012; 367:1814-1820
3. Genomic workup on EMC/2012
   Genomic Characterization of a Newly Discovered Coronavirus Associated with Acute Respiratory Distress Syndrome in Humans. van Boheemen. mBio 2002; 3(6): e00473-12

Monkey magic: Vero cells make more MERS-CoV RNA than any other animal's...

Apart from camel, goat, bat and human cells which draw the eye in studies on the source of MERS-CoV, did anyone notice the Vero cells? 

These cells are derived in the dim dark ages from African green monkey kidney (Cercopithecus aethiops). After infection, these monkey cells were shown to make more MERS-CoV RNA than any other cell line tested and the second greatest quantity of viral particles (after goat cells).

Check out Eckerle et al. over at the CDC's Emerging Infectious Diseases journal for a very nice graphic.

This may be an artefact of the adapted cell-line....but let's not forget to test those baboons hmm?

Does MERS-CoV delay the antiviral response against it?

Lau and colleagues from the University of Hong King recently wrote about MERS-CoV's ability to delay the induction of proinflammatory cytokines in human cells. These are potent defensive chemicals that cause and accelerate an inflammatory response to viral infection.

While inflammation is part of kicking out the virus, it can get out of control, especially with more "foreign" viruses, like those that are still zoonotic and not as co-evolved with humans. Excessive inflammation can lead to tissue remodelling and damage.

In their paper in the Journal of General Virology, the authors compared the results of growing viruses in 2 different cell systems, which was need to account for differences in tropism by the different viruses. Key findings include:

• The lower airway cell-line, CaLu-3 was used for MERS-CoV and SARS-CoV because they produce good titres
• IL-1β, IL-6, IL-8, TNF-α, IFN-β & IP-10 mRNA levels were increased by MERS-CoV and SARS-CoV compared to uninfected cells
• Proinflammatory cytokines shown in bold above were induced more by MERS-CoV infection than by SARS-CoV
• Innate antiviral cytokines TNF-α, IFN-β & IP-10  were induced more by SARS-CoV infection
• MCP-1 (a chemokine) and TGF-β (anti-inflammatory cytokine) remained unimpressed by infection
• At 48-hours, MERS-CoV infection induced less IL-8 or IFN-β protein than did SARS-CoV

• Embryonal lung fibroblasts (HFL) to grow MERS-CoV and compare it to HCoV-229E (which does not grow well in CaLu-3 cells), an infrequently identified but well characterised "common cold" CoV.

• The CaLu-3 results suggested MERS-CoV generates an attenuated innate immune response  the response which induces inflammation. SO HCoV-229E was added as it produces a strong innate response through IFN-β
• IL-1β, IL-6, IL-8, TNF-α, IFN-β & IP-10 mRNA levels were increased by MERS-CoV and HCoV-229E
• HCoV-229E was a stronger inducer than MERS-CoV of all but TNF-α, which triggered more by MERS-CoV
• MCP-1 and TGF-β again remained unchanged

The authors conclude a delayed innate immune response by MERS-CoV infection compared to SARS-CoV. 

This study contrasted with that by Kindler and colleagues which I reviewed earlier. The reason may be because Kindler only sampled for immune analyses up to 12-hours, rather than the 30-hours used by Lau. This would certainly lend weight to claims of a "delayed" induction by MERS-CoV infection since in this study no IFN-β was produced either virus at 12-hour, the first rise in mRNA was apparent at 24-hour post-infection (Lau's protein data suggest there may have been a tiny amount of IFN-β translation at 12-hours, but Kindler did not measure protein).