-Mannosidases and EDEM Homolog Proteins: Their Roles in Glycoprotein ERAD |
|||||||||||||||||
Glycoprotein ERAD The roles of N-linked oligosaccharides on the ER quality control (ERQC) of newly synthesized glycoproteins are discussed in this series of GlycoWord, and in this session, I would like to introduce recently published data focusing on ER-associated degradation (ERAD) and mannose trimming from the N-linked glycans in the ER. Class I 1,2-mannosidase family and EDEM The mannose trimming from the N-linked oligosaccharides in the ER is shown in Fig. 1 as a very simplified schema. ER -mannosidase I (ER ManI) in yeast and mammals trims the mannose from the middle branch of the Man9 oligosaccharide (Man9GlcNAc2) creating the Man8B glycan (Man8GlcNAc2 isomer B). In mammalian ER, the existence of another -mannosidase named ER mannosidase II (ER ManII) has been reported, which creates Man8C oligosaccharide (Man8GlcNAc2 isomer C), but this molecule has not yet been cloned. |
|||||||||||||||||
|
|||||||||||||||||
Mammalian EDEM (ER degradation enhancing -mannosidase-like protein) and yeast homolog Htm1p/Mnl1p are reported as proteins with an -mannosidase-like domain, but they lack enzyme activity and accelerate glycoprotein ERAD. Two novel EDEM homolog proteins were recently reported in mammals by searching the database (1, 2, 3). The domain organization of ER ManI and three EDEM proteins are schematically shown in Fig. 2. They have Class I 1,2-mannosidase domains (glycosylhydrolase family 47 motifs). The amino acid identities within this domain are as high as 45% among the three EDEM homolog proteins, whereas they are approximately 33% between ER ManI and each of the EDEM proteins. There is another subfamily consisting of Golgi a-mannnosidase I (Golgi ManI) in the Class I 1,2-mannosidase family, as shown in Fig. 3. The amino acid identity in this conserved domain is approx. 38% between the Golgi ManI and ER ManI, and about 30% between the Golgi ManI and EDEM, respectively. There is a protease-associated domain within the relatively long C-terminal region of EDEM3, although the function of this domain is currently unknown.
|
|||||||||||||||||
|
|||||||||||||||||
The roles of ER ManI and EDEMs in glycoprotein ERAD
It was recently reported that each mammalian EDEMs (EDEM1, 2, 3) enhanced glycoprotein ERAD (1, 2, 3). ER ManI also accelerates the degradation of misfolded glycoproteins when they are transfected into mammalian cells. ER ManI is an enzyme which trims the mannose from the N-linked oligosaccharides in the ER, and this trimming becomes the signal for newly synthesized glycoproteins entering the degradation pathway from the productive folding cycle (refer to QS-A00, QS-A01). Although the schema of glycoprotein ERQC looks more valid by the addition of the molecules such as ER ManI and EDEMs, new questions arise and it is not so simple as the model shows. Functional analysis of EDEM3 gives rise to a new question as to why EDEM 1, 2, and 3 form a group distinct from ER ManI We have recently shown that EDEM3 enhances the ERAD of misfolded glycoproteins and mannose trimming from the N-linked oligosaccharides when it is transfected to mammalian cells. When we introduce a mutation to one of the conserved acidic amino acid residues of EDEM3, which is known to be important for the enzyme activity, mannose trimming from the glycoprotein is inhibited (3). We are now trying to detect the enzyme activity of EDEM3 in vitro. Because the mannosidase activity is not detected in yeast Htm1p/Mnl1p and in mammalian EDEM1, 2, these proteins are expected to work as lectins which recognize N-linked glycans, although the lectin activity has not yet been demonstrated in vitro. The next question then is why EDEMs including proteins with or without mannosidase activity form a group distinct from ER ManI, which apparently has enzyme activity. Our working hypothesis is that the -mannosidase (glycosylhydrolase family 47) domain of EDEMs can act both as an enzyme and as a lectin. One sequence would act as an enzyme, while the other sequence lacks enzyme activity, in spite of their amino acid similarity. Another interpretation is that it would act as a lectin in vivo when the enzyme activity is very weak. Very recently, Dr. Parodis group published an interesting hypothesis (4), which is very suggestive to the understanding of the function of EDEMs. In fission yeast Schizosaccharomyces pombe, there is a gene encoding an ER -mannosidase (ER Man), but its enzymatic activity was not demonstrated, and only a faint amount of glycoproteins with Man8B oligosaccharides were detected in this yeast. In the present study, they have disrupted the gene encoding ER Man and found that glycoprotein ERAD was prevented. Therefore, they propose that ER Man of S. pombe acts as a lectin rather than as an enzyme, similar to Htm1p/Mnl1p/EDEM. We are now trying to analyze the biochemical properties of EDEMs to further analyze the function of EDEM homolog proteins and the molecular mechanism of ERAD. |
|||||||||||||||||
Nobuko Hosokawa (Institute for Frontier Medical Sciences, Kyoto University) | |||||||||||||||||
|
|||||||||||||||||
Jun.5, 2006 | |||||||||||||||||
|
|||||||||||||||||
|
|||||||||||||||||