Subject: Re: GpO in Drosophila melanogaster
It is clear now that the gene referred to in FlyBase as Dmel\Gpo
(aka aGPO, = FBgn0001129) is really the gene Dmel\Iap2
(= FBgn0015247 = CG8293).  The Ross et al. (1994) paper mis-concluded
that the gene described in GenBank accession M96581 was Gpo
(glycerophosphate oxidase).  It is now clear from the work of Bruce
Hay and colleagues that the gene called Gpo is in reality Iap2 and
has nothing to do with the glycerophosphate oxidase enzyme. As a
side note, FlyBase correctly assigns the M96581 accession to Iap2. 
Thus, the gene called Gpo in FlyBase should be suitably retired.
Let me try to clarify the aGPO picture.  Since the IAP2 paper from
Bruce Hay came out, we did clone and sequence the flight muscle GPO
using a rat cDNA as a probe.  We never published, since the genome
sequence was coming out and GPO would be suitably "exposed".
The situation is this - there are actually three GPO paralogs in D.
melanogaster, CG2137  at 43D, CG7311 at 34C and the most abundantly
expressed gene CG8256 at 52C7 in the flight muscle.  The latter is
the gene we have worked on for many years, i.e. with regard to making
flightless mutants and analyzing its role in the aglycerophosphate
cycle.  So, we know from RT-PCR using primers to each of the three
GPO genes, that CG8256 is the only GPO expressed in the flight muscle
and must account for >90% of the enzyme activity in the adult. I 
propose to name the 52C7 gene Gpo-1 as a paralog of the other Gpo's.
CG8256 is called l(2)k05713 in Flybase.  It is not lethal, however,
since the insertion is viable over Df(2R)WMG268.12.  Obviously a 
'hit and run' lethal must have been generated somewhere on 
chromosome 2 during the generation of the insertion in 52C7.