Download research/1999 - Paxton Provitera

Abstract
HIV (human immunodeficiency virus) and EIAV (equine infectious anemia
virus) are closely related lentiviruses that both infect immune cells but whose
pathogenesis differs. Membrane binding of the (matrix) MA protein of HIV
appears to be primarily driven by a cluster of basic residues in the MA domain
and possibly assisted by an N-myristoylation signal. Interestingly, the MA
protein of EIAV does not contain either of these signals. To understand what
factors may promote EIAV assembly we characterized the membrane binding
properties of its MA proteins using fluorescence methods and compared them
to our previous HIV-MA results. We find that like HIV-MA, EIAV-MA exists
as a multimer in solution whose protein-protein interactions are destabilized by
membrane binding. Unlike HIV-MA, EIAV-MA binds strongly to electrically
neutral membranes (POPC) as well as negatively charged (POPS) ones and our
results indicate a different exposure of the EIAV-MA Trp residues when bound
to the two types of membranes. Based on these data and the known structures
of closely related matrix proteins, we constructed a structural model of EIAVMA. This model predicts that EIAV-MA binds to POPS similar to HIV-MA,
but EIAV-MA has an additional membrane binding region that allows for
hydrophobic membrane interactions.
Figure 1: Gene map of Gag and depiction of mature virus assembly
Matrix protein is instrumental in membrane binding of Gag
-Decrease in fluorescence EIAV homotransfer as seen by an increase in
anisotropy as the pH is lowered
-Decrease in homotransfer indicates subunit dissociation
-This was verified by SDS-PAGE electrophoresis (data not shown)
0.04
Anisotropy Intensity
0.035
0.03
0.025
0.02
0.015
0.01
0.005
0
2.5
3.5
4.5
5.5
6.5
pH
Figure 2: pH dependence of oligomerization.
7.5
A - Lipid Compositions
1.1
0.9
0.9
0.7
0.7
0.5
2.05uM kd=0.77 POPC
0.3
(pH=7)
1.1
Intensity
Intensity
B - Salt
(pH=7)
0.5
2.05uM kd=2.66 POPS at 1 M NaCl
0.3
2.05uM kd=-0.39 POPS at 0.1M NaCl
2.05uM kd=2.66 POPS
0.1
0.1
-0.1
-0.1
-10
40
90
140
190
240
-10
40
90
[Lipid] uM
190
240
C - pH
Figure 3:
Membrane Binding of EIAV
(1M NaCl)
1.3
1.1
0.9
Intensity
-Binding to LUVs was followed by
the decrease in intrinsic
fluorescence of EIAV-MA
-Lipid membrane was added to a
solution containing EIAV-MA.
This may promote protein-protein
interactions so an alternate assay
was done (Fig. 4)
140
[POPS] uM
0.7
2.05uM pH=7
2.05uM pH=4
2.05uM pH=3
0.5
0.3
0.1
-0.1
-10
40
90
140
[POPS]
190
240
Normalized Change in Emission Energy
Binding of EIAV-MA to POPC and POPS
Bilayers (50 M) Labeled with Laurdan
1.2
1.0
0.8
0.6
0.4
0.2
POPS
POPC
0.0
0.0
0.2
0.4
0.6
0.8
1.0
1.2
1.4
1.6
1.8
2.0
[EIAV-MA] M
Figure 4: Membrane Binding of EIAV-MA to POPS and POPC
- Membranes were labeled with the environmentally-sensitive probe Laurdan. The
shift in Laurdan fluorescence as the protein displaces water from the surface was
followed and is shown.
Figure 5: Quenching of Trp by KI differs when EIAV is
bound to POPS and POPC
138 nM EIAV-MA in 20 mM Hepes 0.16 mM KCl
3
2.5
80 uM POPS slope 0.0056
Io/I
2
1.5
80 uM POPC slope 0.0039
1
0.5
-50
0
50
100
150
200
[KI] mM
250
300
350
400
Results
• Oligomerization of EIAV-MA is pH dependent as dissociation occurs at a
pH of ~ 4.5 (Fig. 2)
• EIAV-MA shows strong binding to lipid membranes regardless of
membrane composition (A), salt concentration (B), or pH (C). (Fig. 3 and
Fig. 4)
• Trp quenching by KI is an indication of where the hydrophobic residues are
oriented after membrane binding. The observed difference between POPC
and POPS indicates that EIAV-MA is binding differently to the two
membranes. (Fig. 5)
Figure 6: Molecular model for the EIAV-MA structure.
The amino acid sequences of the EIAV and HIV-1 matrix domains were aligned
with the multiple sequence alignment program ClustalW (Higgins et al., 1991).
The sequence identity of the alignment is 19% (23 identities over 120 residues).
Homology models of the EIAV matrix protein in monomeric and trimeric forms
were constructed with the model routine of the homology modeling program
Modeller (Sali and Blundell, 1993). The ClustalW alignment was used to match
the EIAV matrix sequence to the HIV-1 matrix sequence; the NMR structure of
HIV-1 matrix protein (PDB code, 1tam) was used as the structural template for the
monomer model and the x-ray structure of the HIV-1 matrix protein trimer (PDB
code, 1hiw) was used as the template for the trimer model. Residues present in the
HIV-1 MA trimer interfaces are only partly conserved in EIAV MA:
HIV-1 MA trimer interface residues:
ERFAVNQQQTGS-EE
Corresponding residues in EIAV MA:
DLFHDTDLQTLSGEE
Models of the EIAV monomer based on other alignments share common properties
with the model described above: 1) electrostatic polarity: the “front” surface is
basic, while the “back” surface is slightly acidic with exposed hydrophobic
residues that may penetrate the membrane interface; 2) surface Trp residues are
more prominent on the hydrophobic face than the basic face; 3) the sole Cys
residue is partially buried in the hydrophobic core. Fluorescence studies measuring
the accessibility of this Cys support this model.
Figure 7: Theoretical model of Membrane Binding of EIAV-MA
to POPS and POPC
From the binding data presented here we theorize that EIAV-MA binds to
POPS through electrostatic interactions. The binding then promotes subunit
dissociation through the use of the anionic phosphates of the membrane
surface. We believe that EIAV-MA binds to POPC and the subunits dissociate
through the same mechanisms as POPS. However, the dissociation exposes
hydrophobic residues found within the slightly acidic region on the “back” of
the protein, and “rolling” of the protein for the hydrophobic residues to
penetrate the membrane occurs on POPC.
Conclusions
• Similar to HIV-MA:
– The EIAV-MA solution structure is an oligomer that dissociates upon
membrane binding.
– The penetration into the membrane surface is negligible.
• Unlike HIV-MA:
– EIAV-MA has neither a myristoylation signal nor an apparent cluster of
basic residues.
– The oligomerization of EIAV-MA is pH-dependent and the protein must
use the anionic phosphates of the membrane surface to induce
dissociation.
– EIAV-MA binds to electrically neutral membranes, however this binding
may occur by alternate interaction sites that change the accessibility of
its Trp residues. Energy transfer from EIAV-MA Trp to membraneincorporated anthroyl stearic acid acceptors support this idea.
Biochemical studies are now underway to test this model.
Reference List
1. Higgins, D.G., Bleasby, A.J., and Fuchs, R. (1991). ClustalW: improved
software for multiple sequence alignment. CABIOS 8, 189-191.
2. Sali, A. and Blundell, T.J. (1993). Comparative protein modeling by
satisfaction of spatial restraints. J.Molec.Biol. 234, 779-815.
3. Scarlata, S., Ehrlich, L.S., and Carter, C.A. (1998). Membrane-induced
alterations in HIV-1 Gag and matrix protein-protein interactions.
J.Mol.Biol. 277, 161-169.
4. Ehrlich, L.S., Fong, S., Scarlata, S., Zybarth, G., and Carter, C. (1996).
Partitioning of HIV-1 Gag and Gag-related proteins to membranes.
Biochemistry. 35, 3933-3943.
Acknowledgements
N. Tijandra, I. Jayatilaka
Supported by NIH 05827101