The Beauty of a Visualized Peroxo-diiron(III) Intermediate


Structure

Previews
The Beauty of a Visualized Peroxo-diiron(III)
Intermediate
Gerhard Grüber1,*
1School of Biological Sciences, Nanyang Technological University, 60 Nanyang Drive, Singapore 637551, Republic of Singapore
*Correspondence: grueber@ntu.edu.sg
http://dx.doi.org/10.1016/j.str.2015.04.010

Deoxyhypusine synthase (DHS) and deoxyhypusine hydroxylase (DOHH) are essential for hypusination of
eukaryotic translation initiation factor 5A (eIF-5A). In this issue, Han et al. use X-ray crystallograpy and UV/Vis
and Mössbauer spectroscopy to provide insights into the fundamental mechanism of the hypusination of
eIF-5A and a peroxo intermediate state.
The non-heme diiron enzymes catalyzing

the oxidation of hydrocarbons became

the focus of industrial applications, like

the soluble methane monooxygenase hy-

droxylase (sMMOH) from methanotrophic

bacteria, which oxidizes methane to

methanol (Tinberg and Lippard, 2011).

The first step of its reaction cycle

(Figure 1) is the interaction of the reduced

(Fe(II)-Fe(II)) form of the enzyme with

dioxygen to yield a peroxo-diiron(III) spe-

cies named P. This intermediate is subse-

quently converted into a species termed

Q, which contains an Fe(IV)-Fe(IV) core

(Wallar and Lipscomb, 1996), and is

believed to perform the oxidation of

methane.

The intermediates P and Q are only

short-lived. A few weeks ago, time-

resolved Resonance Raman spectros-

copy confirmed the Fe2(m-O)2 diamond

structure of Q (Banerjee et al., 2015).

The data showed that both m-oxygen

atoms of Q are derived from O2, and sug-
Figure 1. Catalytic Cycle of sMMOH
gested homolytic O-O bond cleav-

age via intermediates such as (I) on

the P/Q path (Figure 1), but not

the heterolytic mechanisms (II and

III) that require the protonation of

the peroxo ligand (Solomon et al.,

2000; Xue et al., 2008).

For the structure of the peroxo-

diiron(III) intermediate P, which pre-

cedes Q, various configurations

have been reviewed. These include

m-1,2 (h1,h1-1,2; either cis [I in

Figure 2A] or trans [II in Figure 2A]),

an h2,h2-1,2 (III in Figure 2A), and a

m-1,1 (h1,h1-1,1; IV in Figure 2A)

arrangement of the peroxo ligand

(Tinberg and Lippard, 2011). Like

Q, the P intermediate of sMMOH is

very short-lived (t1/2 of about 1.5 s
at 4�C) (Liu et al., 1995), making spectro-
scopic work difficult.

Importantly, there is a human non-

heme diiron oxygenase called deoxyhy-

pusine hydroxylase (hDOHH) in which

the peroxo-diiron(III) intermediate is sta-

ble. Previous Resonance Raman studies

of this intermediate suggested a m-1,2

interaction of the peroxide ion with the

diiron core (Vu et al., 2009). In this issue

of Structure, Han et al. (2015) now present

two crystal structures of hDOHH, one of

the peroxo-diiron(III) intermediates, and

the other of a complex with glycerol.

Through time-dependent UV/Vis spectra

of the intermediate, which is colored

blue (absorption maximum at �630 nm),
they show that the peroxo species, which

they have crystallized, does not corre-

spond to the one formed immediately af-

ter the reaction of the reduced form of

hDOHH with O2, but to a slightly different

one obtained over the time of crystalliza-

tion (48 hr). This is supported by Mössba-
Structure 23, May 5, 2015
uer spectra of the 57Fe-hDOHH in the

crystals, which differ slightly from those

obtained from freshly prepared hDOHH

in solution. In any case, Han et al.

describe here the crystal structure of a

native peroxo intermediate of a non-

heme diiron enzyme. The 1.7 Å struc-

ture reveals a cis-m-1,2 arrangement

(cis-h1,h1-1,2; scheme I of Figure 2A) of

the peroxo unit, with a gauche Fe-O-O-

Fe torsion angle. Remarkably, the O1-

atom of the peroxide is more centrally

placed between the two iron atoms (2.2

and 2.6 Å from Fe1 and Fe2, respectively),

whereas O2 is basically interacting only

with Fe2 (3.5 and 2.2 Å from Fe1 and

Fe2, respectively) (Figure 2B). While this

asymmetric arrangement is still cis-m-1,2

(h1,h1-1,2), it is not far from an h1,h2-1,2

arrangement (V of Figure 2A), a configura-

tion that has very recently been proposed

on the basis of spectroscopic data for the

peroxo intermediate of a bacterial aryl

oxygenase (Makris et al., 2015). The lack
ª201
of true isosbestic points in the

time-dependent UV/Vis spectra re-

ported by Han et al. (2015) suggests

that the peroxo intermediate may

adopt a continuum of configurations

and this may be reflected in the

slight differences observed between

the two molecules in the asymmetric

unit of the hDOHH crystal.

The crystallographic work of Han

et al. also reveals the reason for the

unusual stability of the peroxo inter-

mediate of hDOHH: in contrast to all

other well-characterized non-heme

diiron enzymes, hDOHH features a

histidine-rich coordination of the

diiron core lacking any bridging car-

boxylates. The histidines coordinate

iron through their Nε-atoms, not
5 Elsevier Ltd All rights reserved 805

mailto:grueber@ntu.edu.sg
http://dx.doi.org/10.1016/j.str.2015.04.010
http://crossmark.crossref.org/dialog/?doi=10.1016/j.str.2015.04.010&domain=pdf


Figure 2. Peroxo-diiron(III) Core
(A) Possible geometries for the peroxo-diiron(III) core of diiron-containing oxygenases; (B) Fo-Fc difference
density (contoured at 4.7s) for the non-protein ligands of the diiron core of hDOHH. The corresponding
difference map was calculated after building the protein ligands but before incorporating the peroxo,
and water ligands.

Structure

Previews
their Nd-atoms as in most other diiron

oxygenases. Furthermore, the secondary

ligation hemisphere surrounding the bind-

ing site for the peroxide ion is very hydro-

phobic, shielding the reactive intermedi-

ate by a methionine pair and a leucine

pair. Importantly, potential hydrogen-

bonding donors such as threonine, which

occur in many O2-activating monooxyge-

nases, where they seem to shuttle pro-

tons to the active site (Wallar and Lips-

comb, 1996), are completely absent

from the hDOHH active site; therefore,

the heterolytic breakdown of the peroxide

resulting from protonation is greatly

delayed.

A high-oxidation state intermediate

such as compound Q has so far not

been described for hDOHH, and perhaps

there is no requirement for it. In the

sMMOH system, Q is the active species

that breaks the unusually strong (DH =

105 kcal mol�1) C-H bond in methane to
insert oxygen. However, the peroxo-diiro-

n(III) intermediate P of sMMOH is by itself

able to oxidize electron-rich hydrocar-

bons such as diethyl ether or propylene
806 Structure 23, May 5, 2015 ª2015 Elsevie
(indicated by the direct pathway between

P and Hox in Figure 1) (Tinberg and Lip-

pard, 2010).

There is only one substrate known for

DOHH: deoxyhypusine eukaryotic trans-

lation initiation factor 5A (eIF-5A). The

various functions of this protein in transla-

tion initiation and elongation (in particular

of oligoproline-containing polypeptide

segments) and in nuclear export of retro-

viral RNA, CD83 mRNA, and iNOS

mRNA depend on the elongation of an

exposed lysine residue (K50 in the human

protein) by attachment of a 2-hydroxo-4-

amino butyl unit. This unique posttransla-

tional modification is performed in a

two-step reaction catalyzed by the en-

zymes deoxyhypusine synthase (DHS)

and DOHH (Park, 2006). In the first step,

DHS attaches a 4-aminobutyl group orig-

inating from spermidine to the Nz atom of

residue K50, to yield deoxyhypusine-eIF-

5A (Dhp-eIF-5A). In the second step,

DOHH hydroxylates Dhp-eIF-5A in the

2-position of the 4-aminobutyl group,

thereby affording hypusine-eIF-5A (Hpu-

eIF-5A). The structures presented by
r Ltd All rights reserved
Han et al. (2015) immediately suggest

how the substrate and the product would

bind to human DOHH. This in itself is a

major achievement, as hDOHH could

well act as a target for the development

of therapeutics for the treatment of HIV/

AIDS, certain cancers, and diabetes (Kai-

ser, 2012).
REFERENCES

Banerjee, R., Proshlyakov, Y., Lipscomb, J.D., and
Proshlyakov, D.A. (2015). Nature 518, 431–434.

Han, Z., Sakai, N., Böttger, L.H., Klinke, S., Hauber,
J., Trautwein, A.X., and Hilgenfeld, R. (2015).
Structure 23, this issue, 882–892.

Kaiser, A. (2012). Amino Acids 42, 679–684.

Liu, K.E., Valentine, A.M., Qiu, D., Edmondson,
D.E., Appelman, E.H., Spiro, T.G., and Lippard,
S.J. (1995). J. Am. Chem. Soc. 117, 4997–4998.

Makris, T.M., Vu, V.V., Meier, K.K., Komor, A.J., Ri-
vard, B.S., Münck, E., Que, L., Jr., and Lipscomb,
J.D. (2015). J. Am. Chem. Soc. 137, 1608–1617.

Park, M.H. (2006). J. Biochem. 139, 161–169.

Solomon, E.I., Brunold, T.C., Davis, M.I., Kemsley,
J.N., Lee, S.K., Lehnert, N., Neese, F., Skulan, A.J.,
Yang, Y.S., and Zhou, J. (2000). Chem. Rev. 100,
235–350.

Tinberg, C.E., and Lippard, S.J. (2010). Biochem-
istry 49, 7902–7912.

Tinberg, C.E., and Lippard, S.J. (2011). Acc. Chem.
Res. 44, 280–288.

Vu, V.V., Emerson, J.P., Martinho, M., Kim, Y.S.,
Münck, E., Park, M.H., and Que, L., Jr. (2009).
Proc. Natl. Acad. Sci. USA 106, 14814–14819.

Wallar, B.J., and Lipscomb, J.D. (1996). Chem.
Rev. 96, 2625–2658.

Xue, G., Fiedlera, A.T., Martinho, M., Münck, E.,
and Que, L., Jr. (2008). Proc. Natl. Acad. Sci.
USA 105, 20615–20620.


	The Beauty of a Visualized Peroxo-diiron(III) Intermediate
	References