Protein Structure Theory Group

INSTITUTE OF BIOCHEMISTRY - CHARITÉ UNIVERSITÄTSMEDIZIN BERLIN

---

 

Analysis and Prediction of Helical Membrane Proteins

peter.hildebrand@charite.de

Motivation

Because of the difficulties in membrane protein expression and crystallisation, theoretical approaches are highly relevant in order to elucidate the mechanisms of their stability and function. Sorting out structural patterns that are specific for helical membrane proteins is therefore a crucial base for the understanding of their folding and a valuable source for the modelling of their tertiary structure.

Approach

Protein helices spanning biological membranes generally have five portions: two terminal parts outside the membrane, tow boundary regions flanked by lipid head groups and the core part surrounded by the hydrophobic tails of the fatty acids. Because lipids are only rarely recorded in membrane protein structures, the position of the lipid head groups was outlined by water molecules or solvent-exposed electron donors and acceptors that lack a hydrogen-bonding partner. The aromatic belt was alternatively used to define the expansions of the lipid bilayer. In order to assess the basic structural principles of membrane proteins with different functions, we subdivided the data set of membrane protein structures into two classes: Membrane-gates comprising ion channels and solute transporters and membrane-coils embracing metabolic driven proton pumps, receptors and photosystems.

Geometrical features

The measuring of main chain and side chain torsion angles and of intrahelical hydrogen bonds provide appropriate methods to describe the geometry of secondary structures (e.g.: DSSP).

a) the core part: In order to quantify the influence of the different milieu, solely those parts of the transmembrane helices have been selected that are located within the hydrophobic part of the membrane. Transmembrane helices of membrane gates are stabilized by a higher content of bifurcated hydrogen bonds and therefore appear more rigid than helices of membrane coils and globular proteins. These distinct architecture probably accounts for the specific function of membrane channels and transporters, that is based on the movement of rigid transmembrane helices relative to each other.

The investigation of the Phi-, Psi- und Chi1–angles yielded remarkable peculiarities of a-helical membrane proteins that correlate strongly in particular for some polar (asn, asp) and aromatic (trp, tyr) amino acids. Accordingly, transmembrane helices of membrane channels and transporters have to be modelled using different torsion angles, while we propose that the rotamer libraries derived from helices of globular proteins are also valid for transmembrane helices.

b) the helix termini: At the C-termini of transmembrane helices structural motifs equivalent to the Gly-caps of helices in globular proteins have been found, with two third of the transmembrane Gly-caps taking up a primary structure that is typically not found at helix termini exposed to a polar solvent. Hence helix caps of transmembrane domains are substantially specific structural patterns that must be distinguished from the classical caps, known from a-helices of globular proteins. The Gly-caps found at the C-termini of transmembrane helices are therefore very typical for membrane proteins and indicate their relative position to the polar lipid head groups.

Our future aim is to apply these findings in order to improve the tertiary structure prediction of helical membrane proteins.

Fig. 1: Arrangement of transmembrane helices within the lipid bilayer. The lengths of the helical sections that span the hydrophobic interior of the membrane (gray) vary with their tilt angles (not imaged). Helices terminated by Gly are indicated according to the polarity of amino acid types in the n-2 and n-3 position. Trp and Tyr that label the border to the aqueous milieu are additionally denoted as are the N-termini of the protein subunits.

Molecular packing and packing defects

With the help of the Voronoi cell procedure we discovered differences in helix-helix packing densities in membrane proteins of different functions:

a) molecular packing: Membrane channels and transporters are packed less efficiently than other membrane proteins where molecular rearrangements are supposed to occur only on a small scale. The analysis of an updated set of 20 helical membrane proteins lead us to a point of view that is contrary to the proposed opinion that helical membrane proteins are generally packed more densely than other helical proteins.

b) packing defects (cavities): The loose packing of membrane channels and transporters is in turn mainly caused by the frequent placement of polar side chains close to voids or pockets that are lined along the pores which pervade these proteins (figure 2). It is inferred by focal packing defects, i.e. cavities, rather than by steadily increased distances between midpoints of atoms. These cavities allow for the structural flexibility that is required for the proper functioning of membrane channels and transporters.

Fig. 2: Polar (red) cavities are predominantly positioned in helix cap regions that are exposed to the polar milieu or within the gated pore of the glycerol-3-phosphate transporter (pdb-code: 1pw4). Non?polar (blue) cavities are placed in the proposed hinge regions that facilitate the rocker-switch-type movement of the helices that occur upon substrate binding. The cavities are depicted as balls that are sized according to the number of atomic neighbours. The centres of the cavities were calculated from the atom coordinates of the cavities neighbour atoms.

Nevertheless, if the proper functioning of membrane channels and transporters is based on a relatively loose packing of helical interfaces, how are these proteins stabilized compared to other membrane proteins? The comparatively close packing of the transmembrane backbones indicates that main chain interactions probably compensate for the loose packing of side chains. This close packing is predominantly realized between helix pairs, crossing at right-handed angles. Because two thirds of the helix crossings in membrane-gates are right-handed, whereas only one-third of the transmembrane helices in membrane coils cross right-handed, this seems to be another characteristic structural motif of membrane channels and transporters.

Tertiary structure prediction

The use of a sequence based matrix prediction method highlights that the information for residual helix-helix contacts is much more specific in membrane channels and transporters, as in other membrane proteins. These contacts are again mainly accomplished by small amino acids (Gly, Ser) that create contacts at every 4.0 residue, typical for right-handed helix crossings. In contrast, in other membrane proteins, aromatic (Phe, His) and polar amino acids (Asp, Glu) create characteristic contacts at every 3.5 residues, which is a signature for left-handed helix crossings. These patterns are conserved in many membrane protein families.

Publications

Hildebrand PW, Preissner R, Frömmel C. 2004. Structural features of transmembrane helices. FEBS Lett 559(1-3):145-51.

Hildebrand PW, Rother K, Goede A, Preissner R, Frömmel C. 2005. Molecular packing and packing defects in helical membrane proteins. Biophys J 88(3):1970-1977

Material

pdb-code	chain start		chain end		length
1c3w	A	14	A	29	16
1c3w	A	46	A	59	14
1c3w	A	81	A	94	14
1c3w	A	111	A	125	15
1c3w	A	136	A	150	15
1c3w	A	176	A	189	14
1c3w	A	206	A	221	16
mean					14.8
1f88	A	43	A	60	18
1f88	A	74	A	92	19
1f88	A	115	A	130	16
1f88	A	155	A	170	16
1f88	A	206	A	222	17
1f88	A	256	A	270	15
1f88	A	292	A	306	15
mean					16.6
1aig	P	13	P	29	17
1aig	N	34	N	51	18
1aig	N	86	N	100	15
1aig	N	119	N	136	18
1aig	N	170	N	187	18
1aig	N	234	N	250	17
1aig	O	57	O	74	18
1aig	O	115	O	130	16
1aig	O	147	O	164	18
1aig	O	199	O	217	19
1aig	O	270	O	285	16
mean					17.4
1jb0	A	75	A	94	20
1jb0	A	159	A	176	18
1jb0	A	196	A	215	20
1jb0	A	392	A	410	19
1jb0	A	299	A	315	17
1jb0	A	356	A	375	20
1jb0	A	442	A	463	22
1jb0	A	536	A	554	19
1jb0	A	594	A	613	20
1jb0	A	674	A	689	16
1jb0	A	729	A	747	19
1jb0	B	47	B	66	20
1jb0	B	134	B	153	20
1jb0	B	173	B	193	21
1jb0	B	272	B	289	18
1jb0	B	337	B	356	20
1jb0	B	373	B	391	19
1jb0	B	424	B	446	23
1jb0	B	523	B	541	19
1jb0	B	581	B	600	20
1jb0	B	654	B	673	20
1jb0	B	710	B	731	22
1jb0	F	64	F	83	20
1jb0	I	10	I	30	21
1jb0	J	11	J	30	20
1jb0	L	45	L	63	19
1jb0	L	77	L	95	19
1jb0	L	121	L	139	19
1jb0	M	9	M	26	18
1jb0	X	13	X	29	17
mean					19.0
1lgh	A	20	A	38	19
1lgh	B	21	B	39	19
mean					19.0
2occ	A	18	A	38	21
2occ	A	57	A	77	21
2occ	A	100	A	117	18
2occ	A	143	A	163	21
2occ	A	187	A	208	22
2occ	A	234	A	254	21
2occ	A	272	A	282	11
2occ	A	306	A	325	20
2occ	A	338	A	357	20
2occ	A	373	A	392	20
2occ	A	410	A	426	17
2occ	A	454	A	471	18
2occ	B	27	B	46	20
2occ	B	61	B	75	15
2occ	C	17	C	35	19
2occ	C	42	C	54	13
2occ	C	83	C	98	16
2occ	C	129	C	145	17
2occ	C	163	C	180	18
2occ	C	198	C	216	19
2occ	C	238	C	255	18
2occ	D	78	D	99	22
2occ	G	17	G	35	19
2occ	I	17	I	34	18
2occ	J	34	J	52	19
2occ	K	14	K	32	19
2occ	L	22	L	42	21
2occ	M	16	M	34	19
mean					19.6
1ezv	C	27	C	43	17
1ezv	C	88	C	107	20
1ezv	C	110	C	124	15
1ezv	C	187	C	202	16
1ezv	C	226	C	242	17
1ezv	C	296	C	310	15
1ezv	C	321	C	336	16
1ezv	C	354	C	372	19
1ezv	D	272	D	288	17
1ezv	E	58	E	76	19
1ezv	G	55	G	70	16
1ezv	I	16	I	36	21
mean					17.3
1qla	C	32	C	47	16
1qla	C	80	C	96	17
1qla	C	128	C	145	18
1qla	C	171	C	186	16
1qla	C	209	C	228	20
mean					18.0
1jvm	A	29	A	44	16
1jvm	A	91	A	110	20
mean					18.0
1msl	A	21	A	38	18
1msl	A	72	A	86	15
mean					16.5
1kpl	C	36	C	59	24
1kpl	C	81	C	97	17
1kpl	C	127	C	140	14
1kpl	C	150	C	159	10
1kpl	C	177	C	189	13
1kpl	C	194	C	201	8
1kpl	C	217	C	232	16
1kpl	C	255	C	276	22
1kpl	C	291	C	308	18
1kpl	C	332	C	350	19
1kpl	C	361	C	378	18
1kpl	C	387	C	397	11
1kpl	C	405	C	413	9
1kpl	C	422	C	434	13
mean					15.7
1j4n	A	13	A	30	18
1j4n	A	54	A	69	16
1j4n	A	78	A	86	9
1j4n	A	96	A	113	18
1j4n	A	142	A	156	15
1j4n	A	171	A	185	15
1j4n	A	195	A	200	6
1j4n	A	213	A	228	16
mean					13.6
1eul	A	51	A	68	18
1eul	A	91	A	107	17
1eul	A	257	A	271	15
1eul	A	298	A	305	8
1eul	A	761	A	777	17
1eul	A	790	A	805	16
1eul	A	833	A	850	18
1eul	A	903	A	914	12
1eul	A	933	A	946	14
1eul	A	968	A	985	18
mean					14.7
1l7v	A	15	A	27	13
1l7v	A	61	A	75	15
1l7v	A	93	A	103	11
1l7v	A	117	A	133	17
1l7v	A	145	A	158	14
1l7v	A	193	A	206	14
1l7v	A	233	A	249	17
1l7v	A	257	A	264	8
1l7v	A	275	A	293	19
1l7v	A	305	A	321	17
mean					14.5
1iwg	A	10	A	28	19
1iwg	A	339	A	356	18
1iwg	A	366	A	386	21
1iwg	A	394	A	412	19
1iwg	A	442	A	458	17
1iwg	A	468	A	492	25
1iwg	A	541	A	557	17
1iwg	A	873	A	891	19
1iwg	A	898	A	919	22
1iwg	A	926	A	943	18
1iwg	A	975	A	993	19
1iwg	A	1000	A	1021	22
mean					20.0
all mean					17.4

Methods

The Student Test can be used for the comparison of equally distributed samples of different size (n' and n''), as the average hydrogen bonding length and the mean main chain angles. For a value of t = 2.576 two samples are judged to be different with the reliability of 99%.

The generally accepted test for comparing differences between two binned distributions as the populations of the rotamer states of a certain amino acid in different data sets is the Chi-square Test. Ri is the relative number (%) of angles in a single rotamer for transmembrane helices and Si the relative number (%) of angles in a single rotamer for helices of globular proteins (with the absolute numbers of all rotameric states >90).

The collected data are relative distribution and the sum of Si is equal to the sum of the Ri. Therefore the number of degrees of freedom is equal to one less the number of bins (NB=1). With q=0.95, values for X^2>=6 exhibited a significant difference between the two data sets.

Rotameric states in amino acids of transmembrane helices and helices from globular proteins

rotamer	%	%
Ile	memb	glob	x^2
g+	89	84	0.1
g-	6	7	0.1
t	6	9	0.8
total	100	100	1.1	k=2
				p=0.95
				6
Leu	memb	glob	x^2
g+	61	63	0.0
g-	0	0	0.3
t	39	36	0.1
total	100	100	0.5	k=2
				p=0.95
				6
Met	memb	glob	x^2
g+	62	70	0.4
g-	0	4	3.6
t	38	27	2.0
total	100	100	6.0	k=2
				p=0.95
				6
Phe	memb	glob	x^2
g+	43	42	0.0
g-	1	2	0.2
t	56	56	0.0
total	100	100	0.2	k=2
				p=0.95
				6
Ser	memb	glob	x^2
g+	58	41	2.8
g-	29	41	2.3
t	13	17	0.5
total	100	100	5.6	k=2
				p=0.95
				6
Thr	memb	glob	x^2
g+	79	67	0.9
g-	20	30	2.2
t	1	2	0.3
total	100	100	3.3	k=2
				p=0.95
				6
Trp	memb	glob	x^2
g+	43	44	0.0
g-	3	14	6.8
t	54	42	1.6
total	100	100	8.4	k=2
				p=0.95
				6
Val	memb	glob	x^2
g+	10	11	0.0
g-	6	7	0.0
t	83	82	0.0
total	100	100	0.1	k=2
				p=0.95
				6
ASP, ASN	memb	glob	x^2
g+	73	79	0.3
g-	0	4	4.5
t	28	17	2.6
total	100	100	7.3	k=2
				p=0.95
				6

 

---