Comparative analysis of MACROD1, MACROD2 and TARG1 expression, localisation and interactome

Abstract

The posttranslational modification ADP-ribosylation is involved in many cellular processes, with distinct roles for poly- and mono(ADP-ribosyl)ation (PAR- and MARylation, respectively). Reversibility of intracellular MARylation was demonstrated with the discovery of MACROD1, MACROD2 and TARG1, three macrodomain-containing enzymes capable of reversing MARylation of proteins and RNA. While the three enzymes have identical activities in vitro, their roles in cells are unclear and published data are partially contradictory, possibly due to a lack of validated reagents. We developed monoclonal antibodies to study these proteins and analysed their tissue distribution and intracellular localisation. MACROD1 is most prevalent in mitochondria of skeletal muscle, MACROD2 localises to nucleo- and cytoplasm and is found so far only in neuroblastoma cells, whereas the more ubiquitously expressed TARG1 is present in nucleoplasm, nucleolus and stress granules. Loss of MACROD1 or loss of TARG1 leads to disruption of mitochondrial or nucleolar morphology, respectively, hinting at their importance for these organelles. To start elucidating the underlying mechanisms, we have mapped their interactomes using BioID. The cellular localisation of interactors supports the mitochondrial, nucleolar and stress granule localisation of MACROD1 and TARG1, respectively. Gene ontology analysis suggests an involvement of MACROD1 and TARG1 in RNA metabolism in their respective compartments. The detailed description of the hydrolases’ expression, localisation and interactome presented here provides a solid basis for future work addressing their physiological function in more detail.

Behavioural Characterisation of Macrod1 and Macrod2 Knockout Mice

“…Previous studies have demonstrated that reduced latency to fall from the rotarod can be a result of muscle disease [ 76 ], mitochondrial disease [ 77 ] as well as mitochondrial dysfunction induced by ischemic injury [ 78 ]. Therefore, one plausible explanation for Macrod1 KO’s reduced latency to fall from the rotarod, based on the mitochondrial subcellular location of Macrod1 and enrichment in skeletal muscles [ 29 , 33 ] could be, loss of aerobic (mitochondrial) fitness. Indeed, Loss of MACROD1 in RD cells has been shown to cause mitochondrial fragmentation [ 33 ] which might become a problem under continued muscle use and increasing energy demands [ 79 ], such as the rotarod test.…”

“…Therefore, one plausible explanation for Macrod1 KO’s reduced latency to fall from the rotarod, based on the mitochondrial subcellular location of Macrod1 and enrichment in skeletal muscles [ 29 , 33 ] could be, loss of aerobic (mitochondrial) fitness. Indeed, Loss of MACROD1 in RD cells has been shown to cause mitochondrial fragmentation [ 33 ] which might become a problem under continued muscle use and increasing energy demands [ 79 ], such as the rotarod test. Certainly, altered mitochondrial structure has previously been linked to muscle fatigue [ 80 ].…”

“…Localization of MACROD1 seems largely in mitochondria [ 29 , 31 ] and thus it follows that MACROD1 is highly concentrated in skeletal muscles [ 29 ]. Since ADPr- mono-hydrolase activity can be highly specific [ 32 ] and perhaps dependent on certain cell types and/or stress, Žaja et al studied the effect of MACROD1 loss in a skeletal muscle cell line (RD cells—rhabdomyosarcoma) [ 33 ]. Genetic knockout of Macrod1 in RD cells did not result in any major growth defects but did subtly alter mitochondrial structure [ 33 ], which is in-line with our previous work that demonstrated Macrod1 KO mice are healthy and viable [ 29 ].…”

“…Since ADPr- mono-hydrolase activity can be highly specific [ 32 ] and perhaps dependent on certain cell types and/or stress, Žaja et al studied the effect of MACROD1 loss in a skeletal muscle cell line (RD cells—rhabdomyosarcoma) [ 33 ]. Genetic knockout of Macrod1 in RD cells did not result in any major growth defects but did subtly alter mitochondrial structure [ 33 ], which is in-line with our previous work that demonstrated Macrod1 KO mice are healthy and viable [ 29 ]. Previous in vitro studies of MACROD1, using different cell types, have reported MACROD1 as an essential cofactor for androgen receptor [ 34 ], estrogen receptor [ 35 , 36 ] and NFκB (nuclear factor kappa-light-chain-enhancer of activated B cells) [ 37 , 38 ] transcriptional activity.…”

“…On the other hand, MACROD2 resides largely in the cytoplasm [ 31 , 33 ] and is highly expressed in the brain [ 44 ]. Indeed, several genome–wide association studies (GWAS) have reported links between MACROD2 and various neurological and psychiatric conditions such as autism [ 45 ], schizophrenia [ 46 ], ADHD [ 47 ] as well as congenital heart defects [ 48 ] (28% of which are linked to neurodegeneration [ 49 ]).…”