Báo cáo khoa học: The potent inhibitory activity of histone H1.2 C-terminal fragments on furin doc

The potent inhibitory activity of histone H1.2 C-terminal
fragments on furin
Jinbo Han
1
, Ling Zhang
2
, Xiaoxia Shao
2
, Jiahao Shi
1
, and Chengwu Chi
1,2
1 Institute of Biochemistry and Cell Biology, Chinese Academy of Sciences, Graduate School of the Chinese Academy of Sciences,
Shanghai, China
2 Institute of Protein Research, Tong-ji University, Shanghai, China
Following protein biosynthesis, the post-translational
modifications ultimately lead to the maturation of
bioactive molecules. Within the secretory pathway,
these modifications include cleavage at specific sites by
endo- or exo-peptidase, amidation, glycosylation and
sulfonation, etc. [1]. Among these modifications, the
limited proteolysis of proproteins is a mechanism
widely used to regulate the activation of peptides and
proteins that play important roles in various biological
events from homeostasis to diseases. Many inactive
precursors are cleaved at paired or multiple basic
amino acids by a family of proteolytic enzymes called
proprotein convertases (PCs). PCs are calcium-depend-
ent serine proteases whose catalytic domain shares
some homology with that of the bacterial subtilisin. To
date, seven distinct PCs (furin, PC2, PC1 ⁄ PC3, PC4,

PACE4, PC5 ⁄ PC6 and LPC ⁄ PC7 ⁄ PC8 ⁄ SPC7) have
been identified in mammalian species [2].
Furin was the first identified mammalian PC and
the most extensively studied member of the known
seven PCs. It is responsible for the activation of var-
ious substrates ranging from the blood clotting fac-
tors, serum proteins, growth factors, and hormone
receptors to matrix metalloproteinases [3]. Recently,
some ion channels such as the epithelial sodium
channel and the yeast chloride channel were also
found to be processed by furin-like enzymes [4,5]. In
addition to endogenous proteins, many pathogens
such as viral envelope glycoproteins and bacterial
exotoxins are also activated by furin [6]. Thus, furin
is an attractive target for therapeutic agents. Many
peptide- or protein-based inhibitors were designed,
including the peptidyl inhibitor decanoyl-Arg-Val-
Lys-Arg-CH
2
Cl, the bioengineered variant of a1-anti-
trypsin Portland (a1-PDX) [7], polyarginines [8],
Drosophila Serpin 4 [9,10], and the serpin-derived
Keywords
furin; gene expression; histone H1;
inhibitory activity; limited proteolysis;
peptide synthesis
Correspondence
C. Chi, Shanghai Institute of Biochemistry
and Cell Biology, Chinese Academy of
Sciences, 320 Yue Yang Road,

Shanghai 200031, China
Fax: +86 21 54921011
Tel: +86 21 54921165
E-mail:
(Received 28 April 2006, revised 27 July
2006, accepted 4 August 2006)
doi:10.1111/j.1742-4658.2006.05451.x
Many physiologically important proproteins, pathogenic bacterial exo-
toxins and viral envelope glycoproteins are activated by the proprotein con-
vertase furin, which makes furin inhibitor a hot target for basic research
and drug design. Although synthetic and bioengineered inhibitors of furin
have been well characterized, its endogenous inhibitor has not been directly
purified from mammalian tissues to date. In this study, three inhibitors
were purified from the porcine liver by using a combination of chromato-
graphic techniques, and identified to be the C-terminal truncated fragments
with different sizes of histone H1.2. The gene of porcine histone H1.2 was
cloned and sequenced, further confirming the determined sequences. These
three C-terminal fragments inhibited furin with K
i
values around
2 · 10
)7
m while the full-length histone H1.2 inhibited it with a lesser activ-
ity, suggesting that the inhibitory activity relies on the C-terminal lysine-
rich domain. Though the inhibition was temporary, these inhibitors were
specific, and the reactive site of one C-terminal fragment was identified.
A 36 amino acid peptide around the reactive site was synthesized, which
could still inhibit furin with a K
i
of 5.2 · 10

)7
m.
Abbreviation
PCs, proprotein convertases.
FEBS Journal 273 (2006) 4459–4469 ª 2006 The Authors Journal compilation ª 2006 FEBS 4459
peptides, as well as the barley serine proteinase
inhibitor 2-derived cyclic peptides [11]. Some of these
inhibitors are used to prevent the activation of bac-
terial toxin, the processing of envelope glycoprotein
in viral replication and the metastasis of cancer [12–
14]. The propeptide of furin itself [15,16], the inter-
alpha-inhibitor protein [17] and human proteinase
inhibitor 8 [18] have been found to be potent furin
inhibitors, and our earlier work identified a high
positively charged protein, namely, nonhistone chro-
mosomal protein HMG-17, from porcine kidney as a
suicide substrate inhibitor of a furin like enzyme
kexin [19]. However, no other protein that possesses
furin inhibitory activity has been directly purified
from mammalian tissue.
In this study, in contrast to constructing artificial
furin inhibitors, we purified three fractions from the
porcine liver using a combination of chromatographic
techniques. They all possessed high inhibitory activity
against furin and have been identified to be the C-ter-
minally truncated fragments generated from histone
H1.2 with 126, 120 and 103 amino acid residues,
respectively. The activity assay showed that the full-
length histone H1.2 could also inhibit furin with a K
i

value of 4.6 · 10
)7
m. The identification of lysine-rich
histone H1.2 and its C-terminal fragments as inhibitors
of furin will undoubtedly pave the way for the devel-
opment of therapeutically useful furin inhibitors and
for the mechanistic studies of the regulation of furin
activity in vivo.
Results
Purification and identification of the endogenous
furin inhibitors from the porcine liver
Porcine liver was used as the raw material in the
search for the endogenous furin inhibitor for two
reasons: firstly, furin is expressed more abundantly
in the liver than in other tissues or organs; and sec-
ond, a variety of furin substrates are precisely proc-
essed in the liver, compelling the existence of furin
inhibitor to modulate the enzyme activity. The purifi-
cation procedure is described in Fig. 1. To avoid
possible proteolytic degradation, the fresh porcine
livers were immediately treated as an acetone powder
and extracted with 2.5% trichloroacetic acid. After
centrifugation, the supernatant was precipitated with
two-step ammonium sulfate fractionation. The active
portion was subjected to a cation exchange chroma-
tography, and the inhibitory activity was found in
the fraction eluted by 0.4 m NaCl (data not shown).
The active fraction was then pooled and loaded onto
a phenyl Sepharose CL-4B column, the highest inhibi-
tory activity was found in the unbound fraction (data

not shown). The unbound fraction was then further
purified on a Superdex-75 column (Fig. 2A) and then
on a Hamilton PRP-3 column (Fig. 2B). Six fractions
were finally obtained and assayed for their inhibitory
activity. Among them, the major fractions P2, P3 and
P4 have a strong inhibitory activity against furin. The
homogeneity of the three fractions was detected by
SDS ⁄ PAGE, and their apparent sizes were 21, 24 and
25 kDa, respectively (Fig. 2C).
The three purified proteins were sequenced by
Edman degradation. Unexpectedly, their N-terminal
partial sequences were found to be overlapping, indi-
cating that they are derived from the same protein
(Fig. 3A). A database search revealed that the N-ter-
minal sequences of these three proteins matched the
C-terminal sequence of human histone H1.2, except
for a few sites which were not conserved between
human and porcine histone H1.2. In order to eluci-
date the whole protein sequence, we cloned the
porcine histone H1.2 gene (Genebank Accession
#DQ060698) from the porcine genomic DNA, as
described in the experimental procedures. The predic-
ted protein sequence of the porcine histone H1.2 was
aligned with that of human histone H1.2, because,
as shown in Fig. 3(B), they share 92% identity. The
Porcine liver TCA extraction
Ammonium sulfate precipitation
CM-52 (cation) column
Phenyl-sepharose CL-4B
Superdex 75 column

HPLC
Fig. 1. Diagram showing the purification procedure. Fresh porcine
livers were immediately treated as an acetone powder and
extracted with 2.5% trichloroacetic acid. The extraction was precipi-
tated with a two-step ammonium sulfate fractionation and further
separated on a CM-52 cation exchange chromatography, a phenyl
Sepharose CL-4B column, a Superdex-75 column and a Hamilton
PRP-3 column. Finally, three fractions that possessed high inhibi-
tory activity against furin were purified.
Endogenous furin inhibitors from porcine liver J. Han et al.
4460 FEBS Journal 273 (2006) 4459–4469 ª 2006 The Authors Journal compilation ª 2006 FEBS
three fragments P4, P3 and P2 start from 88th, 94th
and 111th residue of histone H1.2 with 126-, 120-
and 103- amino acid residues, respectively. It is
worth pointing out that both fragments P4 and P3
were generated by a proteolytic cleavage between the
Leu–Val peptide bond. Obviously this bond was
cleaved by the same protease, while the fragment P2
was most probably generated by furin itself or by a
furin-like enzyme, as there are paired basic residues
prior to the cleaved bond.
A
B
C
Absorbance (214 nm)
Absorbance (214 nm)
0.80
0.70
0.60
0.50

0.40
0.30
0.20
0.10
0.00
Retention time (min)
Retention time (min)
P3
P4
P2
P1
P5
P6
0.00
0.90
0.80
0.70
0.60
0.50
0.40
0.30
0.20
0.10
0.00
0.00 5.00 10.00 15.00 20.00 25.00 30.00 35.00 40.00 45.00 50.00
5.00
10.00 15.00 20.00 25.00 30.00
Fig. 2. Purification of three fractions with
furin inhibitory activity from porcine liver.
(A) The active fraction, separated from the

phenyl Sepharose CL-4B column, was
loaded onto a Superdex 75 column. The
column was equilibrated and eluted with
20 m
M sodium acetate ⁄ acetic acid buffer,
pH 5.4, at a flow rate of 0.5 mLÆmin
)1
. The
fractions with inhibitory activity marked by a
bar were pooled. (B) The pooled fraction
from the Superdex 75 column was further
separated on a HPLC Hamilton PRP-3
column equilibrated with 0.1% (v ⁄ v)
trifluoroacetic acid, and the bound proteins
were eluted with a linear gradient of 0–20%
(v ⁄ v) acetonitrile in 0.1% (v ⁄ v) trifluoroacetic
acid in 0–20 min and 20–100% (v ⁄ v) aceto-
nitrile in 0.1% (v ⁄ v) trifluoroacetic acid in
20–50 min at a flow rate of 1 mLÆmin
)1
. The
peaks marked (P1-P6) were collected and
the peaks marked by P2, P3 and P4 exhib-
ited a high inhibitory activity against furin.
(C) Aliquots of the P2, P3 and P4 fractions
were subjected to electrophoresis on 15%
SDS ⁄ PAGE and visualized by silver staining.
Protein markers are indicated on the left.
J. Han et al. Endogenous furin inhibitors from porcine liver
FEBS Journal 273 (2006) 4459–4469 ª 2006 The Authors Journal compilation ª 2006 FEBS 4461

Temporary inhibition and identification of the
reactive site of the inhibitor
The inhibitory activity of the three fragments of his-
tone H1.2 against furin was assayed and analyzed
using Dixon’s plot to determine their inhibition con-
stants K
i
. Table 1 showed that the K
i
values of three
H1.2 fragment inhibitors were around 2 · 10
)7
m. Pro-
longed incubation over half an hour caused a gradual
loss of inhibitory activity, suggesting that the inhibi-
tion was temporary.
To identify the reactive site of the inhibitor, the P4
fragment was incubated with furin and the reaction
was stopped at the indicated times and analyzed on
SDS ⁄ PAGE (Fig. 4). The P4 fragment was gradually
degraded, yielding two smaller fragments within the
first 2 h (Fig. 4A). Obviously, this cleaved site should
be the reactive site of the P4 fragment inhibitor. The
two degraded peptides were then separated on
SDS ⁄ PAGE, trans-blotted to the polyvinylidene difluo-
ride membrane and sequenced separately (Fig. 4B).
According to the sequence of the P4 fragment inhib-
itor, its reactive site was deciphered to be K91-A92
(Fig. 4C). As known, in most cases, the preferential P1
residue for furin is arginine [20]; it is understood that

the cleavage of the P4 fragment by furin is very slow,
and was detected only after half an hour under our
experimental conditions. We believe that this cleavage
does not affect the K
i
determination since all reactions
for inhibitory activity analysis were finished in less
than 10 min.
Expression of full-length porcine histone H1.2
and its N- and C-terminal fragments and their
inhibitory activity assay toward furin
In order to check whether the full-length histone H1.2
or its N-terminal fragment also exhibit an inhibitory
activity against furin, and to obtain enough C-terminal
fragment P4 (C-H1) for further study, they were
expressed as His-tag fusion proteins. The recombinant
proteins were purified by metal affinity column and
RP-HPLC, and analyzed by SDS ⁄ PAGE and mass
spectrometry (Fig. 5). The molecular masses of the
recombinant full-length histone H1.2 and its N- and
C-terminal fragments determined by mass spectrometry
Fig. 3. Identification of the P2, P3 and P4 fractions by N-terminal sequencing. (A) The partial N-terminal sequences of the P2, P3 and P4 frac-
tions determined by Edman degradation. (B) Alignment of the protein sequences of histone H1.2 from Homo sapiens (human) and Sus
scrofa (pig). Arrows indicated the starting residue of P2, P3, and P4, respectively. The different residues between porcine and human his-
tone H1.2 are marked with gray or black shadow for the less conserved or unconserved residues, respectively. The underlined amino acids
indicated the putative furin recognition site (Fig. 5).
Table 1. Inhibition constants of the three purified C-terminal frag-
ments P2, P3 and P4 of histone H1.2 against furin. The rate of
hydrolysis of pyrArg-Thr-Lys-Arg-7-amino-4-methylcoumarin (MCA)
by furin was determined in the presence of various concentrations

of the different proteins, as described in the Experimental proce-
dures. The results obtained were used to compute the K
i
values.
Each value represents the mean ± SD determined from three inde-
pendent experiments.
Inhibitor Inhibition constant K
i
(M)
P2 2.4 ± 0.08 · 10
)7
P3 2.3 ± 0.17 · 10
)7
P4 2.1 ± 0.18 · 10
)7
Endogenous furin inhibitors from porcine liver J. Han et al.
4462 FEBS Journal 273 (2006) 4459–4469 ª 2006 The Authors Journal compilation ª 2006 FEBS
matched their calculated ones very well (Fig. 5), but
the apparent molecular weights of the proteins on
SDS ⁄ PAGE were about 10 kDa larger. A similar result
has been reported by Konishi et al. [21]. The aberrant
behavior of recombinant and native histone H1 frag-
ments on SDS ⁄ PAGE (Figs 2,4 and 5) are most likely
caused by their net basic charges, as basic proteins
migrate slower and acidic proteins migrate faster than
neutral proteins with the same molecular weight on
SDS ⁄ PAGE [22].
The inhibitory assay of these recombinant proteins
showed that the full-length histone H1.2 also inhibited
furin with a K

i
value of 4.6 · 10
)7
m, comparable
to the C-terminal fragment P4 (Table 2). But the basic
N-terminal 87- amino acid fragment of histone H1.2
(N-H1) and another highly basic protein cytochrome c
hardly inhibited furin, with the K
i
values being several
orders of magnitude higher. This huge difference
strongly indicates that the inhibitory activity of full-
length histone H1.2 and its C-terminal fragment is spe-
cific, and not a general property of positively charged
proteins. This is in accordance with the existence of a
specific active site in P4 (Fig. 4).
Synthesis of a smaller peptide inhibitor
As the three naturally occurring fragments P2, P3 and
P4 have similar K
i
values against furin, some N- and
C-terminal residues may not be necessary for the
inhibitory activity. In order to know whether a smaller
fragment around the observed cleavage site is still
active, an appropriately sized peptide (PAAATVTK
KVAKSP
KKAKAAKPKKAAKSAAKAVKPK) with
36- amino acid residues was synthesized. The deter-
mined molecular weight of the purified peptide was
consistent with the theoretical one (Fig. 5B). More-

over, this synthetic 36- amino acid peptide was still a
potent furin inhibitor with a K
i
value only about two-
fold higher than that of P2, P3 and P4 (Table 2).
Based on these results, it was possible to further design
a smaller potent furin inhibitor.
The secondary structure determination of histone
H1 and its fragments
The secondary structures of histone H1 and its frag-
ments were also examined by CD spectroscopy
(Fig. 6). As previously reported [23], histone H1.2 and
its fragments do not have a clear secondary structure.
Their secondary structures were found to be domin-
ated by a random coil (negative peak at 196–193 nm),
and the contents of the a-helix (estimated from the
ellipticity value at 222 nm [24]) are only 1.6, 1 and
0.5% for full-length H1, P4 and the 36- amino acid
peptide, respectively.
Discussion
Due to the physiological importance of furin substrates,
furin is a hot target for functional and mechanistic
A
C
B
kDa
97
66
43
31

20
14
0
VSKGTLVQTK
PKKATGSATP
SAAKAVKPKA
AKPKVAKPKK AAPKKK
KKAAKKTPKK
AKKPAAAAVT
KKVAKSPKKA KAAKPKKAAK
GTGASGSFKL NKKAATGETK PKAKKSGAAK
PKKSAGAAKK
15
30
60
120
kDa
97
66
43
31
20
30 60 (min)
50
100
90
80
Reactive site
70 60
110 120

40
30 20 10
P4
14
VSKGTLVQ
AAKPKKAA
(min)
Fig. 4. Determination of the reactive site of the P4 fragment inhibitor. (A) The degradation of the P4 fragment incubated with furin at differ-
ent time at 37 °C in 100 m
M Hepes buffer, pH 7.5, 1 mm CaCl
2
, 0.5% Triton X-100 and 1 mm 2-mercaptoethanol. At the indicated time, the
reaction was immediately terminated by boiling the sample at 100 °C for 5 min, then separated on 15% SDS ⁄ PAGE and stained with Coo-
massie brilliant blue. Protein markers are indicated on the left. (B) The cleaved fragments were separated and transferred to the polyvinylid-
ene difluoride membrane. The membrane was stained with ponceaus, and the bands were cut out for sequencing. Arrows showed the
partial N-terminal sequences of two cleaved fragments, respectively. (C) The reactive site within the P4 fragment is shown by the arrow.
J. Han et al. Endogenous furin inhibitors from porcine liver
FEBS Journal 273 (2006) 4459–4469 ª 2006 The Authors Journal compilation ª 2006 FEBS 4463
studies from basic-research and clinical-application
viewpoints. Many efforts have been made to develop
peptidyl, nonpeptidyl and protein-derived furin inhibi-
tors. The two most widely used inhibitors are the
peptide inhibitor decanoyl-Arg-Val-Lys-Arg-CH
2
Cl and
the bioengineered variant of a1-antitrypsin Portland
(a1-PDX), the latter of which is highly selective for furin
in vitro (K
i
¼ 0.6 nm) and has been used to prevent the

replication of pathogenic viruses, and the activation of
bacterial exotoxin and cancer metastasis [6]. Some
protein-derived inhibitors were obtained by engineering
other protease inhibitors, such as eglin C mutants
[25–27], turkey ovomucoid third domain [28] and
a
2
-macroglobulin [29], based on the consensus substrate
recognition sequence of furin. Recently, polyarginine
or polyarginine-containing peptides [30] were found
to be able to inhibit furin in vitro and, as a result, they
were also able to inhibit the maturation of the glyco-
protein of HIV type 1 gp160, and to prevent the
Pseudomonas aeruginosa exotoxin A activation in vivo
[13,14].
Based on the principle that wherever a protease
exists, its counterpart inhibitor can also be found, we
embarked on the search for an endogenous furin inhib-
A
B
Mass reconstruction of + EMS: 1.030 to 3.088 min from Sample 30 (kkak_p1) of 040812wiff (Turbo
Max. 2.9e8cps.
3570.0
2.9e8
2.8e8
2.6e8
2.4e8
2.2e8
2.0e8
1.8e8

1.6e8
1.4e8
1.2e8
1.0e8
8.0e7
6.0e7
4.0e7
2.0e7
3000 3100 3200 3300 3400 3500
Mass, amu
3600 3700 3800 3900 4000
3976.0
3922.0
3836.0
3739.0
3667.0
3640.0
3552.0
3498.0
3441.0
3409.0
3313.0 3242.0
3126.0
3046.0
3028.0
Fig. 5. SDS ⁄ PAGE and mass spectrometry
analysis of the recombinant full-length his-
tone H1.2 and its N- and C-terminal frag-
ments. (A) The recombinant full-length
histone H1.2 and its N- and C-terminal frag-

ments were expressed in E. coli and puri-
fied by TALON superflow metal affinity
column and RP-HPLC (see Experimental
procedures). The purified recombinant pro-
teins were examined on SDS ⁄ PAGE (left
panel). The determined molecular masses
and the calculated ones of the recombinant
proteins were aligned in the table. (B) The
sequence of the synthetic 36- amino acid
peptide around the reactive site (underlined)
and the mass spectrometry of its molecular
mass.
Table 2. Inhibition constants of various recombinant proteins and
the synthetic peptide against furin.
Inhibitor Inhibition constant K
i
(M)
C-H1 3.1 ± 0.12 · 10
)7
Full length histone H1 4.6 ± 0.02 · 10
)7
N-H1 7.8 ± 0.06 · 10
)5
Cytochrome c 3.3 ± 0.21 · 10
)5
36- amino acid peptide 5.2 ± 0.16 · 10
)7
Endogenous furin inhibitors from porcine liver J. Han et al.
4464 FEBS Journal 273 (2006) 4459–4469 ª 2006 The Authors Journal compilation ª 2006 FEBS
itor from the porcine liver where furin is relatively

abundant. Unexpectedly, three C-terminal fragments
of histone H1.2 with different sizes were found to be
potent inhibitors of furin with K
i
values around
2 · 10
)7
m, comparable with that of polyarginines
such as L6R (hexa-l-arginine) (K
i
¼ 1.14 · 10
)7
m) [8].
The structure of the catalytic domain of mouse furin
revealed that the active site of furin resides in an
extended substrate-binding groove that is lined with
many negatively charged residues [31]. Previous works
of peptide inhibitor indicated that positively charged
residues are preferred for being a furin inhibitor
[30,32]. In our purified fragments P2, P3 and P4, the
multiple positively charged Lys should contribute to
the potency of inhibition. However, compared with the
peptide inhibitor nona-l-arginine (K
i
¼ 4.2 nm ), which
produced hexa- and heptamers when cleaved by furin
[8], the inhibition of the histone H1 P4 fragment is
specific to the cleavage site being K178-A179 (the
sequence number of histone H1 is shown in Fig. 4).
The inhibition by histone H1 fragments is tempor-

ary, as the incubation with furin over half an hour
resulted in digestion at the specific active site (Fig. 4).
However, the temporary inhibition of histone H1
against furin is understandable, as furin is involved in
many subtly regulated physiological events, for which
the permanent inhibition is not desirable. A similar
case has been reported on 7B2, an endogenous PC2
inhibitor [33]. The neuroendocrine protein 7B2
contains two domains, a 21-kDa chaperon domain
required for the maturation of prohormone convertase
2 (PC2) and a C-terminal peptide capable of inhibiting
PC2 at nanomolar concentration [34]. When the 7B2
C-terminal peptide was incubated with PC2, a smaller
peptide (CT peptide 1–18 containing Lys-Lys at the
C-terminus) was generated, and its inhibitory activity
was lost when incubated with carboxypeptidase E to
remove the last two Lys residues [35].
As histone H1 forms little secondary structure in
solution (Fig. 6), the specific conformation may not be
necessary for the inhibitory activity. These indicate
that furin recognizes specific sequence around
KKAKflA in the histone fragments, which explains
why the three fragments, P2, P3 and P4, have similar
inhibitory activity, as well as the full-length histone H1
and the synthetic 36- amino acid peptide around the
cleavage site. It would be interesting to further identify
the minimum sequence around the cleavage site
required for inhibition.
Furin is predominantly located at the trans-Golgi
network and cell surface in vivo [36], whereas histone

H1 normally binds to the linker DNA of chromosome
in nucleus. However, some studies have shown that
nuclear proteins could be located on the surface of var-
ious cells, including intestinal microvilli, monocytes
and lymphocytes [37,38]. Histone fragments released
from epithelia were shown to have strong antimicrobial
activities [39–41]. During the cell apoptosis induced by
virus or bacteria, histones are released and bind to the
negatively charged surfaces of neighboring viable cells
[42]. In addition, it is interesting that the N-terminals
of both fragments P3 and P4 were Val, generated from
the cleavage of Leu-Val bond by an unknown protease.
It has been indicated that an endopeptidase in the
DNase I-containing extract from the bovine pancreas
was able to cleave human H1 into two fragments of
$8 and 14 kDa [43]. We speculate that there might be
a specific protease in the porcine liver to cleave histone
H1 at Leu-Val site into smaller fragments, thus facilita-
ting its transport to the cell surface or to other subcel-
lular compartments. These suggest that the inhibition
of furin by histone H1 fragments may be physiologi-
cally relevant, which remains to be clarified.
It is well known that, besides the housekeeping
role of chromosomal condensation, histone H1 has
some other biological functions, such as the regulation
of gene expression and the stimulation of myoblast
proliferation [44]. Moreover, histone H1.2 was found
to be a signal molecule that triggers the release of
cytochrome c from mitochondria in the DNA damage-
induced apoptosis [21], to selectively inhibit the activa-

tion of calmodulin-dependent enzymes [45], and to be
20000
10000
–10000
–20000
–30000
190 200 210
220
230 240 250
Wavelength (nm)
Proteins or peptide
H1
C-H1
36 aa peptide
1.6
1
0.5
Percentages of
alpha-helix %
Molar ellipticity
0
Fig. 6. The conformation of full-length H1,
C-H1 and the 36-amino acid peptide meas-
ured by far-UV CD spectra in 10 m
M phos-
phate buffer, pH 7.0 at 20 °C.——,
and ÆÆÆÆÆÆ indicate the full-length H1, C-H1 and
the 36- amino acid peptide, respectively.
J. Han et al. Endogenous furin inhibitors from porcine liver
FEBS Journal 273 (2006) 4459–4469 ª 2006 The Authors Journal compilation ª 2006 FEBS 4465

the intestinal protein receptor for 987P fimbriae of
enterotoxigenic Escherichia coli [46]. The C-terminal
domain of histone H1 was reported to be capable of
binding to an apoptotic nuclease (a DNA fragmenta-
tion factor, DFF40 ⁄ CAD) and of stimulating the
DNA cleavage [47], and this current work indicates
another potential function of the multifunctional pro-
tein histone H1.
In summary, this study has shown for the first time
that poly-lysine protein histone H1 and its C-terminal
fragments are potent furin inhibitors. In contrast to
other synthetic furin inhibitors, histone H1 and its
C-terminal fragments are endogenous proteins and
should exhibit little toxicity if used clinically. Our results
give a new indications for understanding the regulation
of furin activity in vivo, as well as for developing novel
tools to inhibit furin-mediated pathogenic processes.
Experimental procedures
Materials
Phenyl-Sepharose 4B and Superdex 75 column were from
Amersham Pharmacia (Uppsala, Sweden), Hamilton PRP-3
column from Hamilton Co (Reno, NV, USA). TALON
superflow metal affinity resin was from Clontech (Mountain
View, CA, USA). The fluorogenic substrate pyrArg-
Thr-Lys-Arg-7-amino-4-methylcoumarin (MCA) was from
Bachem Bioscience (San Diego, CA, USA), cytochrome c
from Sigma (St Louis, MO, USA). The purified recombin-
ant mouse furin was a generous gift from I. Lindberg (Lou-
isiana State University, New Orleans, LA, USA).
Purification of furin inhibitors

The porcine liver was excised and homogenized with five
volumes of cold acetone previously kept in )20 °C freezer.
About 100 g of acetone powder was obtained per kilogram
of liver. The 300 g of acetone powder was extracted with
10 volumes of 2.5% trichloroacetic acid overnight. After
centrifugation, the supernatant was subjected to stepwise
precipition with 0.5 and 0.9 saturated ammonium sulfate.
The pellet of the 0.9 saturated ammonium sulfate portion
was dissolved in a small volume of distilled water and was
dialyzed with 20 mm sodium acetate ⁄ acetic acid buffer
(pH 4.5). The dialyzed sample was loaded onto a CM-52
column pre-equilibrated with 20 mm sodium acetate ⁄ acetic
acid buffer, pH 4.5 (buffer A), washed with three column
volumes of buffer A and eluted stepwise. The fraction
eluted with buffer A containing 0.4 m NaCl was found to
have furin inhibitory activity. This fraction was adjusted
to 2 m ammonium sulfate and applied onto a phenyl
Sepharose CL-4B column pre-equilibrated with 2 m
ammonium sulfate in buffer A. The unbound fraction from
the column was collected, dialyzed with water and lyophi-
lized. The lyophilized fraction was then dissolved in 250 lL
buffer A, loaded onto a Superdex 75 column equilibrated
with buffer A. The fraction with a furin inhibitor activity
from the gel filtration column was further loaded onto a
Hamilton PRP-3 (150 · 4.1 mm) column equilibrated with
0.1% trifluoroacetic acid on a Waters 510 HPLC pump
and 2487 dual absorbance detector (Milford, MA, USA).
The bound proteins were eluted with a linear gradient of
0–20% acetonitrile in 0.1% (v ⁄ v) trifluoroacetic acid at a
flow rate of 1 mLÆmin

)1
in 0–20 min, and of 20–100%
acetonitrile in 0.1% (v ⁄ v) trifluoroacetic acid at a flow rate
of 1 mLÆmin
)1
in 20–50 min. The elute was monitored at
214 nm, collected and assayed for furin inhibitory activity.
Enzyme activity assay
The fluorogenic MCA substrate, pyrArg-Thr-Lys-Arg-
MCA, was used for the furin activity assay as previously
described [25]. To determine the inhibitory activity, differ-
ent amounts of the sample were preincubated with a fixed
amount of enzyme (2 · 10
)3
units) at 37 °C in 100 mm
Hepes buffer, pH 7.5, containing 1 mm CaCl
2
for 5 min,
the residual enzyme activity was then measured. The final
substrate concentration for all assays was 1 lm. The fluor-
escence of the released MCA was measured on-line with a
Hitachi F-2500 spectrofluorimeter using an excitation and
an emission wavelength of 370 nm (slit width, 10 nm) and
460 nm (slit width, 10 nm), respectively.
N-terminal sequencing
Amino acid sequencing was performed on a Perkin-Elmer
Applied Biosystems 494 pulsed-liquid phase protein
sequencer [Procise, PE Applied Biosystems (Foster City,
CA, USA)] with an on-line 785 A PTH-amino acid analyzer.
Gene cloning of the porcine histone H1.2

As there is no intron in the genes of histones, the
genomic DNA from porcine liver was used as a template
to clone the gene of histone H1.2. The human and
murine histone H1 cDNA sequences from the gene
database were referred to design a pair of PCR primers
as follows: 5¢-ATGTCCGAGAC(C ⁄ T)GCTCC(T ⁄ C)GC-3¢
and 5¢-GGTGGCTCTGAAAAGAGCC(G ⁄ T)TTTG-3¢.
The PCR products were ligated into the pGEM-T Easy
vector (Promega, Madison, WI, USA) and sequenced.
Expression and purification of full-length histone
H1.2 and its N- and C-terminal fragments
The genes of the full- length histone H1.2 (F-H1), its N-ter-
minal fragment with 87- amino acid residues (N-H1) and
Endogenous furin inhibitors from porcine liver J. Han et al.
4466 FEBS Journal 273 (2006) 4459–4469 ª 2006 The Authors Journal compilation ª 2006 FEBS
C-terminal fragment of 126-amino acid residues (C-H1)
were cloned through the flanking NcoI and XhoI restriction
sites into the expression vector pET28a. The sequences of
the constructions were verified by DNA sequencing. The
primer pairs for the cloning were as follows: F-h1:
5¢-ccatgggcatgtccgagactgctcctgc-3¢,5¢-ctcgagcttctttttgggtgca
gcctt-3¢; n-h1 : 5¢-ccatgggcatgtccgagactgctcctgc-3¢,5¢-ctcgagc
aggctcttgagacccagct-3¢; c-h1 : 5¢-ccatgggcgtgagcaagggcacctt
g-3¢,5¢-ctcgagcttctttttgggtgcagcctt-3¢.
The expression vectors were transformed into E. coli strain
BL21. Cells grown in LB medium containing 10 lgÆmL
)1
kanamycin were induced with isopropylthiogalactoside when
OD
600

reached 0.8. The harvested cells were lysed by soni-
cating. The recombinant proteins with His-tag were purified
by TALON superflow metal affinity column (BD Clontech)
according to the manufacturer’s instructions. The eluted frac-
tion was further purified by RP-HPLC on Hamilton PRP-3
(150 · 4.1 mm) column with gradient elution from 100%
buffer A (0.1% trifluoroacetic acid) to 100% buffer B (70%
acetonitrile with 0.1% trifluoroacetic acid) in 50 min. The
purified recombinant proteins were lyophilized for inhibitory
activity assay.
Measurement of the kinetic parameter K
i
The K
i
values of inhibitors against furin were determined
by Dixon’s plot (1 ⁄ V against I) using two different concen-
trations of substrate pyrArg-Thr-Lys-Arg-MCA (1.5 lm,
and 3.0 lm). Data from three measurements were averaged
and graphically analyzed with equation to obtain the equi-
librium inhibition constant, K
i
, as previously described [25].
Peptide synthesis
The 36- amino acid peptide (PAAATVTKKVAKSP
KKAKAAKPKKAAKSAAKAVKPK) derived from his-
tone H1.2 around the identified cleavage site was synthes-
ized using ABI 433 peptide synthesizer starting from
Fmoc-Lys
Boc
-Wang resin. The protected amino acids are:

Fmoc-Thr (tBu), Fmoc-Ser (tBu), Fmoc-Lys (Boc). The
resin was cleaved by trifluoroacetic acid containing 8%
p-cresol and 0.2% H
2
O for 1 h at room temperature. The
product was extracted by 0.1% trifluoroacetic acid contain-
ing 20% acetonitrile. The extract was then lyophilized and
purified on a Sephadex G-15 column, equilibrated with
0.1% trifluoroacetic acid. The eluted fraction was lyo-
philized and further purified on a RP-HPLC Zorbax C
8
column (9.4 · 250 mm) equilibrated with buffer A
(0.1% trifluoroacetic acid) at a flow rate of 2 mLÆmin
)1
.
The peptide was eluted by a two-step gradient system:
0–12% buffer B (70% acetonitrile, 0.08% trifluoroacetic
acid) in 10 min and 12–45% buffer B in 10–45 min. The
purified peptide was checked by the ABI API2000 Q-trap
mass spectroscope.
CD spectroscopy
Samples for CD spectroscopy were at a final concentration
of 200 l g ÆmL
)1
in 10 mm phosphate buffer, pH 7.0. Spectra
were obtained on a Jasco J-715 spectrometer in 1 mm of
cells at 20 °C. The results were analyzed with standard ana-
lysis software (jasco) and expressed as mean residue molar
ellipticity (h). The helical content was estimated from the
ellipticity value at 222 nm (h

222
), according to the empirical
equation of Chen et al. [24]:
% helical content ¼ 100½h
222
=À39 500 Âð1 À 2:57=nÞ
where n is the number of residues per helix.
Acknowledgements
We would like to thank Dr I. Lindberg (Louisiana
State University) for the purified recombinant mouse
furin. We also would like to appreciate Dr C. Wang
for discussion.
References
1 Han KK & Martinage A (1992) Post-translational
chemical modification(s) of proteins. Int J Biochem 24,
19–28.
2 Nakayama K (1997) Furin: a mammalian subtil-
isin ⁄ Kex2p-like endoprotease involved in processing of
a wide variety of precursor proteins. Biochem J 327,
625–635.
3 Rockwell NC, Krysan DJ, Komiyama T & Fuller RS
(2002) Precursor processing by kex2 ⁄ furin proteases.
Chem Rev 102, 4525–4548.
4 Wachter A & Schwappach B (2005) The yeast CLC
chloride channel is proteolytically processed by the
furin-like protease Kex2p in the first extracellular loop.
FEBS Lett 579, 1149–1153.
5 Hughey RP, Bruns JB, Kinlough CL, Harkleroad KL,
Tong Q, Carattino MD, Johnson JP, Stockand JD &
Kleyman TR (2004) Epithelial sodium channels are acti-

vated by furin-dependent proteolysis. J Biol Chem 279,
18111–18114.
6 Thomas G (2002) Furin at the cutting edge: from pro-
tein traffic to embryogenesis and disease. Nat Rev Mol
Cell Biol 3, 753–766.
7 Jean F, Stella K, Thomas L, Liu G, Xiang Y, Reason
AJ & Thomas G (1998) alpha1-Antitrypsin Portland, a
bioengineered serpin highly selective for furin: applica-
tion as an antipathogenic agent. Proc Natl Acad Sci 95,
7293–7298.
8 Cameron A, Appel J, Houghten RA & Lindberg I
(2000) Polyarginines are potent furin inhibitors. J Biol
Chem 275, 36741–36749.
J. Han et al. Endogenous furin inhibitors from porcine liver
FEBS Journal 273 (2006) 4459–4469 ª 2006 The Authors Journal compilation ª 2006 FEBS 4467
9 Oley M, Letzel MC & Ragg H (2004) Inhibition of
furin by serpin Spn4A from Drosophila melanogaster.
FEBS Lett 577, 165–169.
10 Osterwalder T, Kuhnen A, Leiserson WM, Kim YS &
Keshishian H (2004) Drosophila serpin 4 functions as a
neuroserpin-like inhibitor of subtilisin-like proprotein
convertases. J Neurosci 24, 5482–5491.
11 Villemure M, Fournier A, Gauthier D, Rabah N, Wil-
kes BC & Lazure C (2003) Barley serine proteinase inhi-
bitor 2-derived cyclic peptides as potent and selective
inhibitors of convertases PC1 ⁄ 3 and furin. Biochemistry
42, 9659–9668.
12 Bassi DE, Lopez De Cicco R, Mahloogi H, Zucker S,
Thomas G & Klein-Szanto AJ (2001) Furin inhibition
results in absent or decreased invasiveness and tumori-

genicity of human cancer cells. Proc Natl Acad Sci USA
98, 10326–10331.
13 Kibler KV, Miyazato A, Yedavalli VS, Dayton AI,
Jacobs BL, Dapolito G, Kim SJ & Jeang KT (2004)
Polyarginine inhibits gp160 processing by furin and
suppresses productive human immunodeficiency virus
type 1 infection. J Biol Chem 279, 49055–49063.
14 Sarac MS, Cameron A & Lindberg I (2002) The furin
inhibitor hexa-d-arginine blocks the activation of Pseu-
domonas aeruginosa exotoxin A in vivo. Infect Immun
70, 7136–7139.
15 Basak A & Lazure C (2003) Synthetic peptides derived
from the prosegments of proprotein convertase 1 ⁄ 3 and
furin are potent inhibitors of both enzymes. Biochem J
373, 231–239.
16 Bissonnette L, Charest G, Longpre JM, Lavigne P &
Leduc R (2004) Identification of furin pro-region deter-
minants involved in folding and activation. Biochem J
379, 757–763.
17 Opal SM, Artenstein AW, Cristofaro PA, Jhung JW,
Palardy JE, Parejo NA & Lim YP (2005) Inter-alpha-
inhibitor proteins are endogenous furin inhibitors and
provide protection against experimental anthrax intoxi-
cation. Infect Immun 73, 5101–5105.
18 Dahlen JR, Jean F, Thomas G, Foster DC & Kisiel W
(1998) Inhibition of soluble recombinant furin by
human proteinase inhibitor 8. J Biol Chem 273, 1851–
1854.
19 Fei H, Li Y, Wang LX, Luo MJ, Ling MH & Chi CW
(2000) Nonhistone protein purified from porcine kidney

acts as a suicide substrate inhibitor on furin-like
enzyme. Acta Pharmacol Sin 21, 265–270.
20 Duckert P, Brunak S & Blom N (2004) Prediction of
proprotein convertase cleavage sites. Protein Eng Des
Sel 17, 107–112.
21 Konishi A, Shimizu S, Hirota J, Takao T, Fan Y,
Matsuoka Y, Zhang L, Yoneda Y, Fujii Y, Skoultchi
AI et al. (2003) Involvement of histone H1.2 in apop-
tosis induced by DNA double-strand breaks. Cell 114,
673–688.
22 Weber K, Pringle JR & Osborn M (1972) Measurement
of molecular weights by electrophoresis on SDS-acryla-
mide gel. Methods Enzymol 26, 3–27.
23 Roque A, Iloro I, Ponte I, Arrondo JR & Suau P
(2005) DNA-induced secondary structure of the
carboxyl-terminal domain of histone H1. J Biol Chem
280, 32141–32147.
24 Chen YH, Yang JT & Chau KH (1974) Determination
of the helix and b
form of proteins in aqueous solution
by circular dichroism. Biochemistry 13, 3350–3359.
25 Liu ZX, Fei H & Chi CW (2004) Two engineered eglin
c mutants potently and selectively inhibiting kexin or
furin. FEBS Lett 556, 116–120.
26 Komiyama T, VanderLugt B, Fugere M, Day R,
Kaufman RJ & Fuller RS (2003) Optimization of
protease–inhibitor interactions by randomizing advent-
itious contacts. Proc Natl Acad Sci USA 100, 8205–
8210.
27 Komiyama T & Fuller RS (2000) Engineered eglin c

variants inhibit yeast and human proprotein processing
proteases, Kex2 and furin. Biochemistry 39, 15156–
15165.
28 Lu W, Zhang W, Molloy SS, Thomas G, Ryan K,
Chiang Y, Anderson S & Laskowski M Jr (1993)
Arg15-Lys17-Arg18 turkey ovomucoid third domain
inhibits human furin. J Biol Chem 268, 14583–14585.
29 Van Rompaey L, Ayoubi T, Van De Ven W &
Marynen P (1997) Inhibition of intracellular proteolytic
processing of soluble proproteins by an engineered
alpha 2-macroglobulin containing a furin recognition
sequence in the bait region. Biochem J 326, 507–514.
30 Kacprzak MM, Peinado JR, Than ME, Appel J, Henrich
S, Lipkind G, Houghten RA, Bode W & Lindberg I
(2004) Inhibition of furin by polyarginine-containing
peptides: nanomolar inhibition by nona-d-arginine. J Biol
Chem 279, 36788–36794.
31 Henrich S, Cameron A, Bourenkov GP, Kiefersauer R,
Huber R, Lindberg I, Bode W & Than ME (2003) The
crystal structure of the proprotein processing proteinase
furin explains its stringent specificity. Nat Struct Biol
10, 520–526.
32 Apletalina E, Appel J, Lamango NS, Houghten RA &
Lindberg I (1998) Identification of inhibitors of prohor-
mone convertases 1 and 2 using a peptide combinatorial
library. J Biol Chem 273, 26589–26595.
33 Martens GJ, Braks JA, Eib DW, Zhou Y & Lindberg I
(1994) The neuroendocrine polypeptide 7B2 is an endo-
genous inhibitor of prohormone convertase PC2. Proc
Natl Acad Sci USA 91, 5784–5787.

34 van Horssen AM, van den Hurk WH, Bailyes EM,
Hutton JC, Martens GJ & Lindberg I (1995) Identi-
fication of the region within the neuroendocrine
polypeptide 7B2 responsible for the inhibition of
prohormone convertase PC2. J Biol Chem 270, 14292–
14296.
Endogenous furin inhibitors from porcine liver J. Han et al.
4468 FEBS Journal 273 (2006) 4459–4469 ª 2006 The Authors Journal compilation ª 2006 FEBS
35 Zhu X, Rouille Y, Lamango NS, Steiner DF &
Lindberg I (1996) Internal cleavage of the inhibitory
7B2 carboxyl-terminal peptide by PC2: a potential
mechanism for its inactivation. Proc Natl Acad Sci USA
93, 4919–4924.
36 Molly SS, Thomas L, Vanslyke JK, Stenberg PE &
Thomas G (1994) Intracellular trafficking and activation
of the furin proprotein convertase: localization to the
TGN and recycling from the cell surface. EMBO J 13,
18–33.
37 Mecheri S, Dannecker G, Dennig D, Poncet P &
Hoffmann MK (1993) Anti-histone autoantibodies react
specifically with the B cell surface. Mol Immunol 30,
549–557.
38 Watson K, Edwards RJ, Shaunak S, Parmelee DC,
Sarraf C, Gooderham NJ & Davies DS (1995) Extra-
nuclear location of histones in activated human peri-
pheral blood lymphocytes and cultured T-cells. Biochem
Pharmacol 50, 299–309.
39 Fernandes JM, Molle G, Kemp GD & Smith VJ (2004)
Isolation and characterisation of oncorhyncin II, a his-
tone H1-derived antimicrobial peptide from skin secre-

tions of rainbow trout, Oncorhynchus mykiss. Dev Comp
Immunol 28, 127–138.
40 Richards RC, O’Neil DB, Thibault P & Ewart KV
(2001) Histone H1: an antimicrobial protein of Atlantic
salmon (Salmo salar). Biochem Biophys Res Commun
284, 549–555.
41 Rose FR, Bailey K, Keyte JW, Chan WC, Greenwood
D & Mahida YR (1998) Potential role of epithelial cell-
derived histone H1 proteins in innate antimicrobial
defense in the human gastrointestinal tract. Infect
Immun 66, 3255–3263.
42 Watson K, Gooderham NJ, Davies DS & Edwards RJ
(1999) Nucleosomes bind to cell surface proteoglycans.
J Biol Chem 274, 21707–21713.
43 Kanai Y (2003) The role of non-chromosomal histones in
the host defense system. Microbiol Immunol 47, 553–556.
44 Henriquez JP, Casar JC, Fuentealba L, Carey DJ &
Brandan E (2002) Extracellular matrix histone H1 binds
to perlecan, is present in regenerating skeletal muscle
and stimulates myoblast proliferation. J Cell Sci 115,
2041–2051.
45 Rasmussen C & Garen C (1993) Activation of calmodu-
lin-dependent enzymes can be selectively inhibited by
histone H1. J Biol Chem 268 , 23788–23791.
46 Zhu G, Chen H, Choi BK, Del Piero F & Schifferli
DM (2005) Histone H1 proteins act as receptors for the
987P fimbriae of enterotoxigenic Escherichia coli. J Biol
Chem 280, 23057–23065.
47 Widlak P, Kalinowska M, Parseghian MH, Lu X,
Hansen JC & Garrard WT (2005) The histone H1

C-terminal domain binds to the apoptotic nuclease,
DNA fragmentation factor (DFF40 ⁄ CAD) and
stimulates DNA cleavage. Biochemistry 44, 7871–7878.
Supplementary material
The following supplementary material is available
online:
Fig. S1. Dixon’s plots (1 ⁄ V against [I]) of the three
purified C-terminal fragments P2, P3 and P4 of histone
H1.2, the recombinant full-length histone H1.2, its
N- and C-terminal fragments, cytochrome c and the
36-amino acid peptide around the reactive site.
This material is available as part of the online article
from
J. Han et al. Endogenous furin inhibitors from porcine liver
FEBS Journal 273 (2006) 4459–4469 ª 2006 The Authors Journal compilation ª 2006 FEBS 4469