Experimental approaches to membrane topology

residues are indicative of potential transmembrane helices.

The positive-inside rule is based on the observation that positive charged amino acids are manifold more abundant in cytoplasmic location, as compared to periplasmic location. This is consistent with positively charged loops serving as cytoplasmic anchors for membrane-spanning regions (von Heijne 1989, 1992).

There were also some other hydropathy systems used to predict transmembrane domains of membrane proteins, such as the TopPred (Von Heijne, 1992; Sipos and Von Heijne, 1993) and TMpred (Hofmann and Stoffel, 1993). All these methods are following the similar principles as described above but more complex. There are examples of hydropathy plot shown in Fig. 2.

1.3.2 Experimental approaches to membrane topology

A variety of biochemical techniques have been employed to elucidate topological structure, including chemical labeling, in situ proteolysis, immunological methods, and molecular genetic approaches (Jennings, 1989). Recently, a method of using the in vitro

transcription/translation technique was also developed to detect the topological structure of certain membrane proteins (Holland and Drickamer, 1986).

The chemical labeling is a well-established and simple method in its principle. It requires a labeling agent that only accesses to one side of the membrane. Labeling agents can be covalently reacting radioactive,

fluorescent, or spin-labeled small molecules, or they can be antibodies or proteolytic enzymes (Jennings, 1989). The most frequently used chemical

Hydropathy plot of cadA (average of 11 aa)

-7 -6 -5 -4 -3 -2 -1 0 1 2 3

1 51 101 151 201 251 301 351 401 451 501 551 601 651 701

Amino acid sequence

Hydropathy index

Fig. 2 Hydropathy plots. (a.) Staphylococcal CadA by Kyte and Dollittle method

a.

b.

c.

15

is the site-directed chemical labeling of cysteine residues using a

sulfhydryl reagent (Loo and Clarke, 1995). From generating a collection of mutants, each with a single unique cysteine residue, these mutated proteins can be probed with sulfhydryl reagents in oriented membranes to determine the surface accessibility at different residues (Long et al., 1998).

In situ proteolysis method is using the proteolytic enzymes, which have been used extensively to identify sites that are exposed to the surfaces of membrane proteins. The most common use of this method is to generate and isolate relatively large hydrophobic fragments (Steck et al., 1976; Jennings et al., 1986). End group determination using the proteolytic method makes it possible to localize the cleavage site in the primary structure. Proteolytic enzymes are usually used on the outer surface of sealed membranes, or for bilateral cleavage of an unsealed membrane. However, sealing proteolytic enzymes may also work inside RBC ghosts to establish intracellular cleavage sites (Lepke and passow, 1976; Jennings et al., 1986).

Immunological tools have been used increasingly in the study of membrane protein topography recently. In order to be useful for such purposes, the antibody is prepared directly against a specific portion of the primary structure of the protein or an extra epitope inserted to the protein using molecular techniques. The antibodies can be used in topographical work by determining the accessibility of a particular antibody to its target portion of a membrane protein at which side of a sealed membrane. The work of Eckhardt et al. (1999) on mammalian CMP-sialic acid transporter has exemplified this approach in membrane topology studies.

In 1986, Manoil and Beckwith reported a new method for assessing membrane protein topography based on the sidedness of fusion proteins formed from portions of the membrane protein of interest attached to reporter enzyme alkaline phosphatase, which is active only when exported to the periplasm (Hoffman and Wright, 1985). The fusion protein is constructed with the membrane polypeptide upstream from alkaline phosphatase, and the eventual location of the phosphatase will reflect the sidedness of the C-terminus of the membrane protein. Years later, Froshauer et al. (1988) developed a complementary method that employed fusions with β-galactosidase, which is active only in the

cytoplasmic location. On the other hand, fusions to periplasmic C-termini of a membrane polypeptide give low β-galactosidase activity. The

combined uses of periplasmic- and cytoplasmic- fusions provide positive activity signals for the domain determinations on both sides of the

membrane (Manoil et al., 1990; Manoil, 1991, Fig. 3). Hirata et al. (1998) did another example for the combined uses of reporter enzymes in

staphylococcal tetracycline efflux protein, Tet (K), using periplasmic β-lactamase and cytoplasmic chloramphenicol acetyl transferase.

A novel method described for the membrane protein topography was derived from using in vitro transcription/translation of DNA

constructs encoding fusion proteins containing potential transmembrane segments in the presence of microsomes (Holland and Drickamer, 1986).

When the fused target peptide is transmembraneous, the

C-terminus-fused signal protein will be transferred into the microsome and glycosylated after in vitro transcription/translation. On the other hand, when the fused peptide is not behind a transmembrane segment, the

signal protein will be stayed out of the microsome, and

17

Fig. 3. Use of fusions to analyze membrane protein topology. Fusion of alkaline

1

2

1

AP+

AP - 2

β-gal - (toxic)

β-gal +

Periplasmic Cytoplasmic N

N

N

Periplasmic Cytoplasmic

without glycosylation. Using the protein glycosylation or not, the

transmembrane segments could then be defined (Holland and Drickamer, 1986). The topographies of a mammalian P-type ATPase, H⁺,K⁺-ATPase (Bamberg and Sachs, 1994), and two CPx-type ATPases from

Helicobacter pylori (Melchers et al., 1996; Bayle et al., 1998) were established using the approach.

1.3.3 Previous membrane topology studies in CPx-type ATPase