How do adenoviruses interact with muscle tissue that it makes them acceptable as vector viruses?

Question

From what I understand certain attenuated adenoviruses are popular as a vector virus, meaning a virus that is used as the active agent in a vaccine to infect cells and trigger a helpful immune response.

Of course, this use assumes that the vector virus can indeed infect cells and cause them to replicate the target protein in the first place. From what I can tell, how adenoviruses attach to cells and their receptors is imperfectly understood and furthermore live adenoviruses seem to often prefer infecting cells in organ tissue, not muscle tissue, so what makes researchers confident that adenoviruses are effective as vector viruses when injected into a muscle?

Answer 1

There's research going back three decades that adenoviruses can infect muscle fibers, e.g. Quantin et al. 1992. This one in particular was conducted for the purpose of evaluating potential to treat muscle disorders, although the gene incorporated in the virus was just a reporter gene, i.e. "does nothing" besides allowing easy identification of the protein expressed.

We constructed a recombinant adenovirus containing the beta-galactosidase reporter gene under the control of muscle-specific regulatory sequences. This recombinant virus was able to direct expression of the beta-galactosidase in myotubes in vitro. We report its in vivo expression in mouse muscles up to 75 days after infection. [...]

To test whether the adenovirus construct is able to direct beta-galactosidase expression in muscle, we injected 10⁸ plaqueforming units into the thigh of either newborn, 1-week-old, or 4-week-old mice. Very strong histochemical staining (>30 positive fibers) was detected in the muscle around the injection site up to 2 weeks after infection, even in mice that were infected at 4 weeks [...]. In the latter case, staining was present over a distance of at least 8-9 mm, and most of the positive fibers appeared stained over a length of 1.5-2 mm. [...] These data were obtained by staining the muscles as a whole. The internal fibers may not have been accessible to the reagents because of the tight structure of the muscle, thus leading to an underestimation of the number of stained fibers. This was confirmed by incising some infected muscles before staining: in addition to external fibers, some internal fibers appeared positive.

FIG. 3. Expression of beta-galactosidase in muscle of infected mice. (A and B) Muscle of 4-week-old mice analyzed 14 days after injection. (C and D) Muscle of newborn mice analyzed 75 days after injection. (Bars = 2 mm.)

The vector used was just "run of the mill" Ad5 with E1 and E3 deleted, i.e. non-replicating.

Of the newer "stars" of vaccinology, ChAdOx1 has definitely been tested in this regard, in fact also compared with the Ad5 "gold standard":

To further investigate the difference in luciferase expression between vectors, BALB/c mice were vaccinated with HAdV-5 or ChAdOx1 vectors expressing luciferase and muscle tissue samples from the site of injection and draining popliteal lymph nodes were harvested 24 h post vaccination. Luciferase expression was detected ex vivo in both muscle and lymph node tissue of vaccinated animals (Fig. 2A,B). Since the extraction procedure differed between muscle and lymph node samples, absolute luminescence units cannot be compared between tissues. However, in lymph node samples, mean luminescence activity after HAdV-5 vaccination was approximately 20-fold higher than after ChAdOx1 vaccination (Fig. 2A). By contrast, mean luminescence activity within muscle samples was less than 3-fold greater after HAdV-5 vaccination (Fig. 2B). These data suggest that the greatest differences in transduction efficiency between HAdV-5 and ChAdOx1 in vivo occur within the draining lymph nodes after intramuscular vaccination.

Given this, I strongly suspect Ad26 was tested in this regard as well, but I can't be bothered to search for that paper.