Moving frames - a blow on "virtual particles"? There is a polemic about whether virtual particles EXIST, or are only a tool for making our calculi easier. Here is a doubt about their existence, and on the other hand, an argument in support of their existence.
Against: consider a charge q regarded from an inertial frame in which it is in movement. Then, the movement of the charge q produces a magnetic field that can act on whatever other moving charge would appear. The magnetic field can be said to be transmitted through the vacuum by virtual particles.
But, in a frame of coordinates where q is at rest, it produces NO magnetic field. So, NO virtual particles. Then, virtual particles EXIST in one frame and DON'T EXIST in another?
In favor: there is NO action-at-a-distance. The nature works in one and same way when propagating some type of influence. E.g., inside a dielectric, the electric field between two charged plates is propagated from one atom-dipole to another. Why should the nature behave differently in vacuum, i.e. without some sort of dipoles?
 A: Your argument against is based on the fact that in one frame there is no magnetic field, but there is a magnetic field in a different frame. So there must be magnetic virtual particles in some frames but not in others. Hence magnetic virtual particles can't exist.
However there aren't separate magnetic and electric virtual particles. There are just virtual photons that are responsible for both magnetic and electric fields. Your argument simply shows the virtual photons look different when viewed from different inertial frames.
A: Virtual particles never been considered "real" in any serious interpretation of QFT for two reasons:
1) what we call virtual particles are just terms in a series at some order of a coupling constant,
2) there's no way to interact with them. 
An interaction between two physical objects is a black box in QFT. The way people usually use to "see" inside this black box is via perturbation theory. The concept of virtual particle is there a convenient way to make sense of such computation, because it allows you to interpret each terms via Feynman's diagrams, and that's all it is. In QFT, the "primitive" physical objects are not particles at all but fields. So if you take QFT seriously, you would have to forget about the particles interpretation of physics.  
