DISCUSSION
Our study compares two systems to generate recombinant baculovirus
vectors in the context of rAAV production. Indeed, baculovirus backbone
can be engineered by exploiting both homologous recombination (HR) or
Tn7 transposition, the latter being the most widely used system. First,
we show that the recombinant baculoviruses derived from the Tn7 system
tends to loose the exogenous expression cassette after few rounds of
viral amplification while HR system preserves baculoviral genome
stability over multiple passages. This is an advantage for manufacturing
at industrial scale, because multiple rounds of baculovirus
amplifications could be needed to infect large volumes of culture.
Secondly, we show that BEVs generated with the HR system preserve rAAV
yields and infectivity comparable to the standard Tn7 system.
Importantly, HR system is devoid of any bacterial DNA sequences in rAAV
biotherapeutic final product which is an additional safety advantage to
meet requirements from the regulatory agencies.
The genetic instability, and thus, substantial loss of the expression
cassette, observed during recombinant baculovirus amplification after
Tn7 transposition was specific to this system. The mechanism underlying
this genetic instability is not fully understood, although the presence
of residual bacterial elements could be detrimental for baculovirus
genome replication. Our data show that HR system resulted in increased
genetic stability of the inserted transgene, due to knock-out the BAC
sequence from parental bacmid, when homologous recombination occurred.
Nonetheless, some limitations of HR system have also been identified, i)
the low frequency of natural HR events that can occur in Sf9 insect cell
could lead to the generation of fewer number of clones of recombinant
baculoviruses; ii) the risk to generate baculovirus defective
interfering particles (DIs) due to cross-complementation of ORF1629
essential protein from viable recombinant baculovirus. Here we show that
the use of circular donor plasmid induced very low HR frequency and lead
to accumulation of DIs that are not carrying the cassette of interest.
However, we solved this issue by using a linearized donor plasmid that
maximizes the on-target cassette integration and finally leads to
similar yield of baculoviral clones than Tn7 system. This observation
suggests again the need for HR system to still proceed with one round of
viral plaque isolation to definitively eliminate DIs and generate a pure
baculoviral reagent before amplification for generating master viral
seed (MVS). Whereas the standard Tn7 system seemed to be limited to a
few numbers of passages, in this study we show that HR system may allow
extended number of rounds (up to 10 passages) without cassette
instability. This is a major advantage for industrial applications if
large batches of biotherapeutic products require the use of high doses
of baculovirus reagents.
Our study also shows that HR and Tn7-derived BEVs generated similar
amounts of rAAV vectors, despite the fact that total capsid titer seemed
to be slightly higher with HR. This observation was confirmed by
SDS-PAGE analyses when running AAV proteins normalized by the vector
genome copy number. Further investigations are necessary to determine if
the higher proportion of empty AAV capsids in the HR system is due to
the absence of chitinase and cathepsin in the BacMagic-2 system, as
suggested by another group (Chen, 2017), or a direct consequence of the
higher genetic stability of Cap expression along the rAAV production.
Importantly, we show that final rAAV products generated with HR-derived
BEV are exempted of encapsidated bacterial sequences including
antibiotic resistance genes, such as gentamycin and kanamycin. In
contrary to the Tn7 system, the HR system knock-out the BAC sequence
from parental bacmid. Gentamycin sequence is detectable in rAAV
generated with Tn7-derived BEVs, likely due to the proximal location
with the AAV ITRs. Nonetheless, none of the two methods can avoid this
reverse packaging phenomenon of baculoviral sequences from the AAV ITRs
(Penaud-Budloo et al., 2017).
In summary, we conducted a side-by-side comparison of different methods
to generate recombinant baculovirus vectors and concluded that all
methods are equally efficient to generate high yields of rAAV vectors.
HR-derived BEVs showed higher genetic stability over passages and lower
amounts of nucleic acid impurities (i.e. bacterial sequences) which are
significant advantages for large scale manufacturing of rAAV
biotherapeutics.