The transport of ammonium across biological membranes is an important physiological process in all domains of life (Kleiner 1993; Knepper et al 1989). Genes encoding high affinity ammonium transporters were first described in Saccharomyces cerevisiae and Arabidopsis thaliana (Marini et al. 1994; Ninnemann et al. 1994) since when more than 50 homologs have been identified. The genes encode highly hydrophobic proteins generally of between 400 and 450 amino acids that represent a unique family (Amt) of transporters found in archaea, bacteria, fungi, plants and animals (Saier Jr. et al 1999). The Rhesus (Rh) family of blood antigens also show significant similarity to the Amt proteins (Marini et al 1997b) and by complementation of a S. cerevisiae Amepl,2,3 mutant it has recently been demonstrated that the Rhesus-associated glycoprotein (RhAG) and the non-erythroid Rhesus-related glycoprotein (RhCG) can indeed transport ammonium (Marini et al 2000b).
With the recent explosion in bacterial genome sequences it has become apparent that nearly all bacteria and archaea encode members of the Amt family. In many cases organisms encode multiple Amt proteins, with as many as three copies being found in Synechocystis sp. PCC6803 and Archaeoglobus fulgidus. There are also three copies in Saccharomyces cerevisiae and six copies in Arabidopsis thaliana. The functions of the different paralogs are not understood but they often have different Km values and in plants they exhibit different tissue specificities and expression profiles. In both the bacteria and the archaea the amt genes are almost invariably linked to a gene, designated glnK, encoding a small signal transduction protein that is a member of the Pn protein family. Conservation of gene linkage in distantly related bacteria often reflects a functional relationship between the gene products, and the linkage of amtB and glnK led us to propose that the GlnK and AmtB proteins may interact physically (Thomas et al 2000a).
While there is a considerable body of data to support the proposal that the Amt proteins facilitate the entry of ammonium into the cell, the mode of action of these proteins is still a matter of debate. In its uncharged form ammonia (NH3) is highly membrane-permeable, but energy-dependent uptake systems for ammonium have been reported for many organisms (Kleiner 1993) and the prevailing view has been that these systems function in active transport of the charged species, NH/ (Kleiner 1993; Ninnemann et al 1994; Marini et al. 1997a). However an alternative model for ammonium acquisition has been proposed by Soupene et al (Soupene et al 1998) who investigated the physiological role of the E. coli AmtB protein. The authors concluded that the AmtB protein recognizes the ammonia molecule as its substrate and that it uses a facilitated diffusion mechanism, similar to the glycerol facilitator (Heller et al 1980), to catalyze equilibration of ammonia across the cytoplasmic membrane.
Secondary structural predictions for members of the Amt family suggest that they encode proteins with 10-12 trans-membrane (TM) helices with a C-terminal cytoplasmic extension (Ninnemann et al. 1994; Marini et al. 1994; Marini et al 2000a; Tate et al 1998). In a detailed in silico and in vivo empirical topological analysis of the E. coli AmtB protein we concluded that this protein has 12 TM helices with both the N-terminus and C-terminus in the cytoplasm (Thomas et al 2000b). A detailed understanding of both the mode of action and the structure of Amt proteins can only be obtained through purification and biochemical analysis for which we believe the E. coli AmtB protein provides an excellent model system. In our recent work we have used this system to address a number of these questions.
Was this article helpful?