Schmallenberg Virus RNA in Culicoides midges

There have generally been two views regarding the use of RT-PCR as a way to say whether Culicoides midges are vectors for arboviruses. One side argues that, if you find viral RNA, then the midges are competent. In both the bluetongue virus (BTV) and Schmallenberg virus (SBV) outbreaks in Northern Europe there were thousands of midges mashed and tested for the presence of viral RNA. Of course, midges were found containing RNA. The other side historically replies with something along the lines of, whilst a good indication, this is not evidence of virus replication and spread to the salivary glands, a necessity for transmission to a naive host during the next blood-feed. The hesitation in the latter camp is on the basis that, if you test the entire midge for viral RNA, then there is no way to determine whether any RNA detected is a result of replication, or whether it merely represents virus taken in during the original blood-meal, i.e. it is possible for a midge to be positive, without virus replication taking place.

This all relates to the very first post I wrote - having a sound understanding about what you are detecting and what can be concluded from the result. In this case, it's the fact that RNA does not equate to live virus.

There are ways to gain confidence that RNA presence = competent midge. The simplest is to just argue that the level of RNA is too high to just be a blood-meal, indicating that replication has occurred; this was the case in a recent paper regarding SBV positive midges in Denmark. In this study they also gained confidence that there was viral replication by testing for host (i.e. cow or sheep) RNA; the absence of host RNA implies that the blood-meal had been digested, and therefore the presence of SBV RNA is as a result of virus replication.

Another way is to only test the heads, as an indication that the salivary glands have been infected (a necessity for transmission). The blood-meal is in the abdomen, so any RNA detected in the head will be as a result of replicated and disseminated virus. Dissecting individual midges is a huge undertaking though, and is largely going to be impractical.

In this study by Veronesi et al, midges were either injected with virus, or allowed to feed on blood spiked with virus. After 10 days incubation to allow the virus to replicate and spread, the surviving midges were homogenised and tested by real-time 'semi-quantitative' RT-PCR for the presence of viral RNA. Quantitative RT-PCR would obviously have been preferable, but Cq values (lower value = more RNA) at least give an indication of how intense the infection is.

(A) in the figure above represents RNA levels in midges which had been injected with virus, and shows that RNA was present in the heads and even saliva (8/10 midges tested) of some midges, which may indicate that these midges would be competent to transmit the virus. The lowest values were found in the abdomen/thorax (where the virus was injected), indicating that either the replication was local or the assay was detecting the blood-meal. (B) and (C) are more interesting as they represent lines of midges that are either competent (B, C. sonorensis), or incompetent (C, C. nebeculosus) for BTV, that have been allowed to feed on blood containing virus, thus imitating more closely the 'natural' situation. The C. sonorensis infections resulted in a conspicuous bimodal distribution of Cq values, something which allows the midges to be divided into either transmissible (low Cq values) or non-transmissible (high Cq values) infections. For C. nubeculosus, this distribution was absent, indicating that this species of midge is, like for other arboviruses, non-competent.

Working out what copy numbers of RNA will equate to an infectious midge will be the next step, and this will require the adoption of quantitative RT-PCR.

Culicoides midges have proved themselves to be important vectors of arboviruses, most famously in recent years bluetongue and schmallenberg. The Pirbright lab, originally driven by the great Professor Philip Mellor, have done much work towards unpicking the precise role midges play in virus transmission, particularly in a European situation. Now Philip's one time understudy, Simon Carpenter, and his team are taking things forward, as he explains in the following video about Bluetongue virus.

For a paper looking to implicate midges as vectors of Schmallenberg though, there does seem to be something rather obvious that's missing. Pirbright are in the rare situation of having competent colonies of Culicoides - why not do the key experiment of attempting to infect sheep or cattle with blood-fed midges? RNA, or indeed virus, may be in the saliva in the midges tested for this paper, but that's not evidence that it would be sufficiently infectious to initiate infection in an animal (although the likelihood is it would). 

Overall though, the paper has a slightly different focus - more about the idea of whether or not real-time RT-PCR can be used to indicate whether or not midges are competent vectors. A study tackling this shady area of RNA = infection has needed to be done for a long time; finally it has.


Rasmussen, L., Kristensen, B., Kirkeby, C., Rasmussen, T., Belsham, G., Bødker, R., & Bøtner, A. (2012). Culicoids as Vectors of Schmallenberg Virus Emerging Infectious Diseases, 18 (7) DOI: 10.3201/eid1807.120385
Veronesi, E., Henstock, M., Gubbins, S., Batten, C., Manley, R., Barber, J., Hoffmann, B., Beer, M., Attoui, H., Mertens, P., & Carpenter, S. (2013). Implicating Culicoides Biting Midges as Vectors of Schmallenberg Virus Using Semi-Quantitative RT-PCR PLoS ONE, 8 (3) DOI: 10.1371/journal.pone.0057747