REDES DE ÁREA LOCAL INALÁMBRICA (WLAN)

Previous evidence suggested that the intracellular compartmentalisation of AGT is dependent on the presence or absence of TL-1 (i.e. the codon at nucleotide positions -66 to -64) (see figures 4 and 10). In its absence, all translation would be expected to start at the TL-2 (nucleotide positions +1 to +3). Although one or other of the ancestral transcription start sites (see figure 10) can also be lost, this has usually been considered

to be a neutral secondary event For example, previous studies have shown that the

presence of ATG at the TL-1 in the common marmoset and its mutation to AT A in

the human ancestral line is entirely responsible for the presence of both mitochondrial

and peroxisomal AGT in the former, yet only peroxisomal AGT in the latter

In addition to the absence of TL-1 in human'^^ and saki monkey^^, the present study has shown that it is also absent in the following catarrhines: chimpanzee, gorilla, and diana monkey (see figures 11 and 12). Of these, the human, chimpanzee, gorilla and saki monkey are known to have a peroxisomal distribution of AGT"^^ (and Danpure et al, unpublished observations). The subcellular distribution of AGT in the gibbon and diana monkey is not known. In addition to the presence of TL-1 in the marmoset"^^ and

baboon^^, it has been shown in this study that it is also present in the spider monkey, golden lion tamarin, goeldis monkey, and squirrel monkey (all platyrrhines), as well as

Chapter 3 - M olecular Adaptation o f AGT Targeting in Primates

the Celebes macaque (a catarrhine) (see figures 11 and 12). AGT is known to be both mitochondrial and peroxisomal in the marmoset and yet peroxisomal in the baboon. The distribution in the other species is unknown, but as all Callitrichidae studied to date have both mitochondrial and peroxisomal AGT"^^, it is likely that the golden lion

tamarin, goeldis monkey and common squirrel monkey do as well. In addition, it is likely that Celebes macaque has peroxisomal AGT as its close relative the Japanese macaque does^^. Unfortunately, there are no clues to the subcellular distribution of AGT in the spider monkey.

With the notable exception of two of the Cercopithecidae, the baboon and the Celebes macaque, the sequences at TL-1 in the primate AGT genes studied in this paper are compatible with the known or likely subcellular distributions of AGT. Thus, when the triplet ATG is present at this site, AGT is both mitochondrial and peroxisomal.

However, when any other sequence (e.g. ATA, ATC, GTG or CTG in this study) is present at this site, AGT is only peroxisomal. Thus in most of the primates studied, loss of mitochondrial AGT targeting is due to loss of TL-1 and hence the exclusion of the

region encoding the MTS from the ORF, as suggested previously Clearly, this

cannot be the case for the baboon and macaque which must have lost mitochondrial AGT targeting by a different mechanism.

There are at least two possible mechanisms by which the baboon and macaque could have lost the ability to target AGT to the mitochondria without loss of TL-1. They could have lost the TS-A (figure 10), or they could have accumulated non-synonymous mutations in the MTS that prevent it from functioning as such. Loss of TS-A, which is upstream of both translation start sites, would result in the exclusion of region 1

Chapter 3 - M olecular Adaptation o f AGT Targeting in Primates

guinea pig, which has also lost the ability to target AGT to the mitochondria, but this could be a secondary event to the putative earlier loss of TL-l"^^.

On the assumption that region 1 is contained within the ORF in the baboon and macaque, the deduced amino acid sequence shows a number of differences from those expected for an efficient MTS, which are usually positively-charged amphiphilic a-

helices For example, R(-2) and R(-8) have both been replaced by Q (figure 12). This

would not only decrease substantially the net positive charge of the sequence, but also

the loss of R(-2) would be predicted to interfere with presequence cleavage In

addition, the presence of three juxtaposed helix breakers, P (-ll), G(-IO) and P(-9), would call into question the ability of this region to fold into an a-helix.

Although both of the mechanisms proposed above for the loss of mitochondrial

targeting in the baboon and macaque are possible, the former is the more probable (i.e. the loss of the TS-A). This is because the SDS-PAGE (figure 14) shows that AGT in the baboon liver is similar in size to human AGT (i.e. it lacks the 22 amino acid

mitochondrial leader sequence). If the MTS was included within the ORF it would not be expected to cleave (due to the lack of a cleavage site) and therefore it would result in a larger protein. If these conclusions are correct then this is the first example of the loss of mitochondrial AGT targeting caused by loss of TS-A.

Chapter 3 - M olecular Adaptation o f AGT Targeting in Primates

3.3.2 Possible temporal relationship of the mutational events leading to the varied