The complete crystallised structure of LAT is still not published because
it is still very difficult to crystallise membrane proteins. Therefore,
we can only predict possible secondary structures and functions of LAT.
The protein structure homology-modeling servers that we used to analyse
the amino acid sequence of LAT are Swiss-Model,
JPRED3 and
Kyte-Doolittle.
Figure. 1: Swiss-Model graph of linker
for activation of T-cells family member 1 isoform a [Homo sapiens].
H represents helix, C represents coiled and E represents extended β-strands.
In theory, the membrane-spanning region is frequently an α-helix comprising
of around 20 amino acid residues. When data on 3-D structure is absent
the presence of transmembrane α-helical segments in membrane protein
can be predicted by looking for amino acid sequences in regions that
are hydrophobic (high hydropathy values) [1].
Here, strong peaks are observed for this membrane protein. We anticipate
that they are constructed from helices arisen from 23 amino acids (red).
The finding is in agreement with the information obtained from JPRED3
and
Kyte-Doolittle as shown below. Extended β-strands are also predicted
to be located at 48-52, 132-135, 191-193 amino acids.
Figure 2: Predicted Hydrophathy plot of linker
for activation of T-cells family member 1 isoform a [Homo sapiens]
from Kyte-Doolittle.
A large positive hydropathic index is indicative of a hydrophobic region
of the polypeptide chain, whereas a large negative value is indicative
of a hydrophilic region [1]. The red square indicates a possible transmembrane
α-helical region.
Figure 3: Predicted secondary structures of linker for activation
of T-cells family member 1 isoform a [Homo sapiens] by JPRED3.
We can see a concrete agreement of the structure of the α-helices in
LAT.
As the average length of an α-helix is 0.15 nm per amino acid residue,
an α-helical sequence of 20 to 25 residues is long enough to span the
thickness (3 nm) of the lipid bilayer [2]. A polypeptide chain surrounded
by lipids, having no water molecules with which to hydrogen-bond, will
tend to form α-helices or β-sheets, in which intrachain hydrogen bonding
is maximized. If the side chains of all amino acids in a helix are nonpolar,
hydrophobic interactions with the surrounding lipids further stabilise
the helix.
Corresponding β-strands here that are also present in the Swiss-Model
graph are the amino acids at 48-52, 132-135, 162-164, 171-173 and 191-193.
Interestingly, there appears to be 2 distinct phosphorylation motifs
(YVNV) in LAT, which are involved with β-strands at the same time (171-173
and 191-193). They are likely to function as binding sites for recruiting
other proteins that have domains that can recognise phosphotyrosine
residues (i.e. SH2 domains)
Figure 4: Stereo, space-filling representation of an α-helical
segment of LAT determined by X-ray crystal structure analysis. Backbone
atoms are coloured according to type (N purple, O red, and H white)
and the side chain atoms are gold [3].
Figure 5: Ribbon diagram of a helical signal-anchor for type III membrane
protein [3].
The 3-D structure of the α-helices of LAT is provided by membranome
database.
The prediction of the transmembrane segment is based on the fact that
amino acids have preferred locations in transmembrane helices. Hydrophobic
amino acids (Ala, Val, Leu, Ile, and Phe) often reside in the hydrocarbon
interior, where charged and polar amino acids are almost never found.
Charged residues are commonly seen at the lipid-water interface, but
positively charged residues occur more often on the cytoplasmic face
of transmembrane proteins. Transmembrane protein sequences and structures
are adapted to the transition from water on one side of the membrane,
to the hydrocarbon core of the membrane, and then to water on the other
side of the membrane. The amino acids that constitute transmembrane
segments reflect these transitions [4].
The amino acids Lys and Arg frequently have novel behaviours at the
lipidwater interface. Both of these residues contain long aliphatic
side chains with positively charged groups at the end. In many membrane
proteins, there is an association between aliphatic chain of Lys or
Arg and the hydrophobic portion of the bilayer, with the positively
charged groups (amino or guanidinium) extending beyond to associate
with negatively charged phosphate groups. This behavior, with the side
chain pointing up out of the membrane core, has been termed snorkeling.
If a Phe residue resides near the lipidwater interface, there
will be a general arrangment of the residue with the aromatic ring oriented
toward the membrane core. This is termed anti-snorkeling [4].
In addition, if proline residues are present, a transmembrane α-helix
will be bent. Transmembrane α-helices often contain distortions and
kinksmore so than for water-soluble proteins. Helix
distortions may have evolved in membrane proteins because (1) helices,
even distorted ones, are highly stable in the membrane environment,
and (2) helix distortions may be one way to create structural diversity
from the simple helix building blocks of most membrane proteins. About
60% of known membrane helix distortions are kinks at proline residues.
Proline distorts the ideal α-helical geometry because of steric conflict
with the preceding residue and because of the loss of a backbone H bond.
Proline-induced kinks create weak points in the helix, which may facilitate
movements required for transmembrane transport channels [4].
References
[1]. M, L.A. et al (2012). Principles of Biochemistry. 5th ed. New York: Pearson Education.
[2]. L, D.L. & C., M.m. (2011). Lehninger Principles of Biochemistry. 4th ed. New York: W.H. Freeman & Sumanas.
[3]. http://www.membranome.org/protein.php?pdbid=LAT_HUMAN
[4]. Garrette, L.H. & Grisham, C.M. (2008). Biochemistry. 4th ed. Belmont: Brooks Cole.