Rat Neuronal Nitric Oxide Synthase (nNOS) Reductase Domain
Michael Castle, '06 and Adam Sutter, '06
Contents:
I. Introduction
Nitric oxide (NO) is a small, gaseous, uncharged molecule
produced by the nitric oxide synthase (NOS) family of enzymes, which convert
l-arginine and O2 to l-citrulline and NO. NO possesses a relatively
long half life and diffuses readily through cellular membranes, allowing it
to serve a variety of unique roles as a neurotransmitter, primarily by stimulating
the enzyme soluble guanylyl cyclase (sGC) to produce cyclic GMP. NO synthesis
is coupled to Ca2+ influx by the Ca2+-activated signaling molecule
Calmodulin (CaM), which binds NOS directly. In mammals, NO has been implicated
in the regulation of systemic vasomotor tone and blood pressure, bronchial and
gastrointestinal smooth muscle relaxation, penile erection, inflammatory responses,
and leukocyte-mediated microbe destruction, as well as neuromodulatory effects
on behavior, memory, and nociception1.
Three distinct isoforms of the NOS protein are found
in mammals: endothelial NOS (eNOS), inducible NOS (iNOS), and neuronal NOS (nNOS)
[Diagram]
[Reset]. All NOS isoforms are dimers
of two identical subunits, with the continuous polypeptide chain of a single
nNOS subunit being 1429 amino acids in length and possessing a mass of approximately
160 kDa. Each NOS subunit contains three conserved domains, the reductase domain,
the calmodulin binding domain, and the oxygenase domain2 [Diagram]
[Reset]. The CaM binding domain (residues
720-743) binds Ca2+-activated CaM, allowing NOS to respond to changes
in intracellular calcium levels. The reductase domain (residues 743-1429) binds
three ligands, nicotinamide adenine dinucleotide phosphate (NADPH), flavin adenine
dinucleotide (FAD), and flavin mononucleotide (FMN), and serves to transfer
electrons from NADPH to the oxygenase heme-binding domain. Using NADPH-derived
electrons, the oxygenase domain (residues 1-720) converts O2 and
l-arginine to NO and l-citrulline. The protein at left is the reductase domain
of nNOS (nNOSred) isolated from rat cortical tissue.
II. General Structure
nNOSred assembles as a homodimer of two identical reductase subunits
,
each composed of 686 amino acid residues. It belongs to a family of reductase
proteins that share a conserved organization of
FMN-binding
,
FAD-binding, and
NADPH-binding
domains.
Electron transfer proceeds from NADPH to FAD to FMN
to
the heme region of the trans oxygenase domain, where production of NO is catalyzed.
Transfer to heme is rate-limiting, and is induced by conformational changes following
Ca
2+/CaM binding to the NOS enzyme
3. The
FAD-
and
NADPH-binding domains form a continuous
unit , and
an
α-helical
connecting domain (CD) links the
FAD
and
FMN domains in way that aligns
the bound flavins in a position ideal for electron transfer
. This conformation is enforced by direct interaction between the
FAD
and
FMN domains. In addition,
the
FMN domain possesses considerable
rotational flexibility, and is theorized to undergo large scale movements
that facilitate the transfer of electrons to the oxygenase domain. The
FMN
domain also contains an autoinhibitory insert that is essential for
the regulation of electron transfer, and the C-terminal tail (CT) of NOSred
plays a role in regulating its activity as well.
III. NADPH Binding Domain
Amino acids 1232-1396 constitute the NADPH
binding domain
. NADPH binds at one end of a five-stranded parallel β-sheet
. Gln-1324 hydrogen bonds to
the adenosine moiety of NADPH, which also undergoes hydrophobic stacking
interactions with Tyr-1322
. A network of ionic interactions anchors the NADPH in the binding site:
the 2' phosphate interacts with Ser-1313,
Arg-1314,
and Arg-1400
, located on the CTN. Arg-1010,
from the FAD-binding domain, and Arg-1284
contributing additional ionic interactions
.
IV. FAD Binding Domain
Electrons from NADPH are transferred to the bound
FAD molecule. Amino acids 990-1038 and 1171-1231 comprise the FAD
binding domain
. FAD binding occurs in an elongated conformation at one end of the flattened
anti-parallel β-barrel fold of the FAD
binding domain
The N1 atom of FAD binds to a conserved water molecule, the N5 atom makes
a long Hydrogen-bond (3.6 Å) to Ser-1176,
and the Tyr-1197
ring stacks with the terminal adenine ring
.
V. FMN Binding Domain
The FMN
binding domain consists of amino acids 750-942, with amino acids
943-989 comprising the connecting
domain (CD) and 1059-1078 comprising a beta-finger
. These domains are necessary for rotational flexibility of the FMN-binding
domain (see below). The FMN-binding
domain consists of a five-stranded parallel β-sheet flanked by
five α-helices
. When FMN binds, Gly-810 accepts
a Hydrogen-bond from the FMN N5 atom, and the ring plane of FMN is stacked
between the rings of Phe-809
and Tyr-889 residues
.
VI. The FMN/FAD Domain Interface
The flavin-containing face of the FMN-binding
domain inserts into the cup-shaped depression formed by the continuous
unit of the FAD, NADPH,
and connecting
domains (the FNR/CD
unit), burying a total surface of 2,500 Å
. A double salt bridge between residues Glu-816
and Arg-1229
directly links the FMN and FAD domains
, along with an extensive network of hydrophobic contacts, H-bonds, and
salt bridges3
. The CD
aligns the alloxazine rings of the bound flavins to facilitate electron
transfer from FAD to FMN: in this conformation, the xylene ring methyl groups
of the flavins are only 4.8 Å apart
.
In this FMN/FAD interface, the bound FMN is completely
buried and cannot transfer electrons to the oxygenase domain—rotational
flexibility of the FMN-binding domain
is necessary for this transfer to occur. It is theorized that the flexible
hinge region
at the C-terminus of the FMN-binding
domain serves as a pivot point, allowing the domain to rotate so
that the bound FMN is approximately 15 Å from the heme region of the
trans oxygenase domain across the dimer, close enough for electron transfer
to proceed. This requirement for rotation of the FMN-binding
domain explains why transfer of electrons to the heme is the rate
limiting step in NO synthesis. The gap in the hinge structure results from
weakly defined electron density, which reflects the flexible nature of this
region.
VII. Regulatory Elements
An autoinhibitory
insert composed of 45 amino acids (residues 827-872) is located within
the FMN-binding domain
. This insert contains an α-helix
anchored within a hydrophobic pocket between the FMN-
and NADPH-binding domains. The
autoinhibitory helix (AH) also
contains a conserved motif of three hydrophobic
residues, Tyr-841, Phe-845, and Val-848, preceded by two basic
residues, Arg-838 and Lys-839
, which has been termed the “AH motif”3. This pattern
resembles canonical Ca2+/CaM-binding motifs.
AH most
likely regulates NO synthesis by inhibiting the production of NO at low Ca2+
concentrations in one of two theoretical ways. First, AH
could competitively bind Ca2+/CaM, preventing activated CaM from
contacting the CaM-binding domain when it is present only at low concentrations.
Second, the interactions of AH with
the FMN- and NADPH-binding
domains could contribute to a locked position in which the FMN
domain is unable to rotate. In this case, high concentrations of Ca2+
would induce CaM binding and subsequent release of the FMN domain to participate
in electron transfer. Whichever of these theoretical mechanisms is correct,
it is expected that the ultimate function of AH
is to inhibit NO production under low Ca2+ concentrations.
The C-terminal tail (CT) of nNOSred, attached to the
terminus of the NADPH-binding domain,
also serves to repress NOS activity when CaM activation is low. CT can be divided
into two regions. The C-terminal segment (CTC, residues 1414-1429), which is
variable in length and sequence among NOS isozymes, is not included in this
protein structure. The N-terminal segment (CTN,
residues 1397-1413)
, on
the other hand, is highly conserved. CTN
contains an α-helical segment that sits across a negatively charged groove
at the FAD/FMN
interface
. Residues
1397-1400, at the base of the helix, are attached via hydrogen bonds to Tyr-889
and Asp-1351
of the FMN- and NADPH-binding
domains, respectively
. Arg-1400
on the CTN
also binds to the 2’ phosphate of NADPH, as noted above. These interactions
lock the FMN-binding domain in the electron-accepting position, inhibiting NO
synthesis. Activated CaM binding to the Ca2+/CaM domain disrupts
these interactions, allowing the FMN domain to rotate freely and NO synthesis
to occur.
A regulatory phosphorylation site, Ser-1412
, is
also found on CTN
at the end of the a-helix. The addition of a negatively charged phosphate causes
the helix to be repelled from the negative groove in which it lies, displacing
the CTN
from its normal position and releasing the FMN-binding
domain. Phosphorylation of Ser-1412 therefore serves to relieve the repression
of NOS activity.
These two regulatory mechanisms in the reductase
subunit ensure that NO synthesis by NOS is tightly coupled to Ca2+
influx, an essential control for the overall biological activity of the
NOS protein.
References
1Rawlingson A (2003). Nitric
oxide, inflammation and acture burn injury. Burns. 29: 631-640.
2Li H & Poulos TL (2005). Structure-function
studies on nitric oxide synthases. J. Inorg. Biochem. 99: 293-305.
3Garcin ED, Bruns CM, Lloyd SJ,
Hosfield DJ, Tiso M, Gachhui R, et al. (2004). Structural Basis for Isozyme-specific
Regulation of Electron Transfer in Nitric Oxide Synthase. J. Biol. Chem.
279: 37918-37927.