Báo cáo khoa học: The ubiquitin-like protein monoclonal nonspecific suppressor factor b conjugates to endophilin II and regulates phagocytosis potx

The ubiquitin-like protein monoclonal nonspecific
suppressor factor b conjugates to endophilin II and
regulates phagocytosis
Morihiko Nakamura and Shunsuke Shimosaki
Department of Cooperative Medical Research, Collaboration Center, Shimane University, Izumo, Japan
Introduction
Monoclonal nonspecific suppressor factor b (MNSFb)
is a 14.5 kDa fusion protein consisting of a protein
with 36% identity with ubiquitin and ribosomal pro-
tein S30. The ubiquitin-like segment is cleaved from
ribosomal protein S30 in the cytosol [1]. MNSFb is
covalently attached to certain lysines of specific target
proteins, including Bcl-G, a proapoptotic protein [2].
MNSFb conjugates to Bcl-G and regulates the
Keywords
endophilin; macrophage; phagocytosis;
ubiquitin-like protein
Correspondence
M. Nakamura, Department of Cooperative
Medical Research, Collaboration Center,
Shimane University, Izumo 693-8501, Japan
Fax: +81 853 20 2913
Tel: +81 853 20 2916
E-mail:
(Received 13 July 2009, revised 31 July
2009, accepted 3 September 2009)
doi:10.1111/j.1742-4658.2009.07348.x
Monoclonal nonspecific suppressor factor b (MNSFb) is a ubiquitously
expressed member of the ubiquitin-like family that has been implicated in
various biological functions. Previous studies have demonstrated that
MNSFb covalently binds to the intracellular proapoptotic protein Bcl-G in

cells of the macrophage cell line Raw264.7, suggesting involvement of this
ubiquitin-like protein in apoptosis. In this study, we purified a 62 kDa
MNSFb adduct from murine liver lysates by sequential chromatography
on DEAE and anti-MNSFb IgG-conjugated Sepharose. MALDI-TOF MS
fingerprinting revealed that this MNSFb adduct consists of an 8.5 kDa
MNSFb and endophilin II, a member of the endophilin A family. MNSFb
may conjugate to endophilin II with a linkage between the C-terminal
Gly74 and Lys294. We confirmed this result by immunoprecipita-
tion ⁄ western blot studies. Endophilin II was ubiquitously expressed in vari-
ous tissues, although a truncated form was observed in liver. The 62 kDa
MNSFb–endophilin II was specifically expressed in liver and macrophages.
Small interfering RNA-mediated knockdown of endophilin II and ⁄ or
MNSFb promoted phagocytosis of zymosan in Raw264.7 cells. Conversely,
cotransfection of endophilin II and MNSFb cDNAs inhibited the phagocy-
tosis of zymosan. Such inhibition was not observed in cells expressing a
mutant of endophilin II in which Lys294 was replaced by arginine. These
results suggest that the post-translational modification of endophilin II by
MNSFb might be implicated in phagocytosis by macrophages.
Structured digital abstract
l
MINT-7261558, MINT-7261537, MINT-7261546: MNSF beta (uniprotkb:P35545) physically
interacts (
MI:0915) with Endophilin-2 (uniprotkb:Q62419)byanti bait coimmunoprecipitation
(
MI:0006)
Abbreviations
EGF, epidermal growth factor; ERK, extracellular signal-regulated kinase; MAP kinase, mitogen-activated protein kinase; MNSF, monoclonal
nonspecific suppressor factor; siRNA, small interfering RNA; tEnd-II, 30 kDa truncated form of endophilin II.
FEBS Journal 276 (2009) 6355–6363 ª 2009 The Authors Journal compilation ª 2009 FEBS 6355
mitogen-activated protein kinase (MAP kinase) path-

way by inhibiting the activation of extracellular signal-
regulated kinase (ERK) [3]. Several ubiquitin-like
proteins are implicated in interferon signaling.
MNSFb, ISG15, FAT10 and NUB1L are induced by
interferon-c [4–7]. Among them, MNSFb and FAT10
are closely related to apoptosis [2,8].
The endophilins are a family of proteins identified in
a search for SH3 domain-containing proteins. The
most extensively studied isoform of the mammalian
endophilins, brain-specific endophilin I [9,10], is
necessary for synaptic vesicle endocytosis [11–13].
Endophilin II, which is ubiquitously expressed [14], is
a member of the endocytic endophilin family, but its
function in regulating endocytosis remains unclear. In
this study, we observed that the authentic 45 kDa
endophilin II is expressed in various mouse tissues,
including brain, testis, and spleen. Interestingly, heter-
ogeneous species with molecular masses of about 62,
45 and 25 kDa were observed in liver lysates. We dem-
onstrated that MNSFb covalently conjugates to endo-
philin II in liver and cells of the macrophage cell line
Raw264.7, and that MNSFb–endophilin II complex
formation might be implicated in phagocytosis in
macrophages.
Results
Purification of MNSFb adducts from murine liver
In previous studies, we have shown that MNSFb cova-
lently conjugates to various target proteins [2,4]. Dur-
ing the studies, we observed that several MNSFb
adducts were found in liver extracts of mice. To fur-

ther study the mechanism of action of MNSFb,we
tried to isolate the MNSFb conjugates. When an
extract from fresh murine liver was chromatographed
through a DEAE–Sepharose column, and various frac-
tions were subjected to western blotting by using
antibody against MNSFb, several MNSFb antibody-
reactive conjugates were eluted from the column by
100 mm NaCl (Fig. 1, lane 5). We further purified
these conjugates by MNSFb antibody affinity chroma-
tography. The final preparation gave two silver-stained
bands on SDS ⁄ PAGE with mobilities corresponding to
62 and 95 kDa under reducing conditions (Fig. 1, lane 3).
We observed that antibody against MNSFb recognized
these two bands (Fig. 1, lane 6). These MNSFb anti-
body-reactive proteins were first subjected to N-termi-
nal sequence analysis after electroblotting. Despite
repeated attempts, ambiguous signals were obtained
from each protein (about 100 pmol). For internal
sequencing, the protein bands were digested in-gel with
trypsin. Selected peptides from the 62 kDa band were
subjected to sequence analysis, and the results showed
the amino acid sequences of MNSFb and an irrelevant
protein (data not shown). No clear results were
obtained from the 95 kDa band.
MALDI-TOF MS analysis of MNSFb adducts
To identify the target molecule of MNSFb, MALDI-
TOF MS was performed by separating the 62 kDa
MNSFb adduct by SDS ⁄ PAGE under reducing condi-
tions. Bands corresponding to the 62 kDa MNSFb
adduct were excised and subjected to in-gel digestion

with trypsin as described. Then, the resulting mixtures
of peptides were analyzed by MALDI-TOF MS.
Table 1 shows the peptide masses observed by
MALDI-TOF MS of the 62 kDa MNSFb adduct puri-
fied from murine liver. The resulting sets of peptide
masses were then used to search the NCBI database
for potential matches. Seven of the peptides in the
MALDI spectrum matched endophilin II protein, pro-
viding a sequence coverage of 32%. This result
indicates that the 62 kDa MNSF b adduct is an
MNSFb–endophilin II protein complex. Endophilin II
is a member of the endophilin A family, and is ubiqui-
tously expressed in various tissues. Endophilin II
possesses the SH3 domain, which is critical for associa-
tion with synaptojanin-1 and dynamin. A signal was
-95 kDa MNSF adduct
-62 kDa MNSF adduc
t
1 2 3 4 5 6
97 kDa-
65 kDa-
45 kDa-
31 kDa-
CBB Silver WB: anti-MNSF
Fig. 1. Purified fractions of MNSFb adducts analyzed by
SDS ⁄ PAGE (12% polyacrylamide gel) and immunostained for pro-
tein. Lanes 1 and 4 contain an aliquot (100 lg of total protein) from
liver lysates prepared from Balb ⁄ c mice; lanes 2 and 5 contain an
aliquot from the DEAE chromatography purification step; lanes 3
and 6 contain an aliquot from the anti-MNSFb affinity chromatogra-

phy purification step; lanes 1 and 2, CBB (Coomassie brilliant blue)-
stained; lane 3, silver-stained; lanes 4–6, immunostained with
antibodies against MNSFb. Mobilities of the 95 and 62 kDa MNSFb
adducts and the molecular mass standards (kDa) are indicated to
the right and left of the figure, respectively.
MNSFb conjugates to endophilin II M. Nakamura and S. Shimosaki
6356 FEBS Journal 276 (2009) 6355–6363 ª 2009 The Authors Journal compilation ª 2009 FEBS
detected at 2096.3 Da that corresponds to amino acids
7–25 of MNSFb. Importantly, a pair of signals was
detected at 1161.4 and 1605.0 Da (Table 2). These sig-
nals correspond to digestion fragments in which amino
acids 291–297 of endophilin II are covalently linked by
an isopeptide bond to amino acids 71–74 of MNSFb,
and amino acids 291–301 are linked by an isopeptide
bond to amino acids 71–74. Collectively, MNSFb may
conjugate to endophilin II with a linkage between the
C-terminal Gly74 and Lys294. Lys294 is specific for
endophilin II among all members of the endophilin
family (Fig. 2).
Immunoblot and immunoprecipitation analyses
To confirm that MNSFb covalently conjugates to
endophilin II, we performed immunoblotting of
samples purified by MNSFb antibody affinity chroma-
tography. The results of the blotting revealed that anti-
body against endophilin II specifically recognized the
62 kDa, but not the 95 kDa, MNSFb adduct
(Fig. 3A), indicating that the 62 kDa band is a com-
plex of MNSFb with endophilin II.
To further study these interactions, we performed
immunoprecipitation experiments using antibodies

against endophilin. Cell lysates were prepared from
mouse liver and immunoprecipitated with antibodies
directed against MNSFb, and associated proteins were
analyzed by western blot analysis by using antibodies
against endophilin. As shown in Fig. 3B, MNSFb
associated with endophilin II. Immunoprecipitation of
cell lysates with normal IgG followed by western blot
analysis revealed no detectable association of endophi-
lin II, indicating the specificity of MNSFb for endo-
philin II. In addition, converse immunoprecipitation
with antibody against endophilin II and immunoblot
analysis with antibody against MNSFb confirmed the
association between endophilin II and the MNSFb
adduct. Brain extracts were also examined by immuno-
precipitation ⁄ western blot. As depicted in Fig. 3C,
neither antibody against endophilin I nor antibody
against endophilin II recognized the MNSFb adduct.
Thus, we concluded that the 62 kDa MNSFb adduct
is an MNSFb–endophilin II complex and is specifically
expressed in liver. Collectively, the results of immuno-
blotting, together with the internal peptide sequences
in Table 1 and MALDI-TOF MS analysis, show that
MNSFb covalently binds to endophilin II via an
isopeptide bond.
We next performed immunoblotting analysis to mea-
sure endophilin II levels in various organs of mice. As
shown in Fig. 4, the 45 kDa endophilin II is expressed
Table 1. Assignment for peptide fragments from a trypsin digest of the 62 kDa MNSFb adduct. The 62 kDa MNSFb adduct was digested
by trypsin and subjected to MALDI-TOF MS analysis. The data in the second column are the mass values obtained experimentally, whereas
the results in the third column are those calculated from the trypsin fragmentation of the gene products of endophilin II and MNSFb. The

fourth column indicates the numbers of the first and last amino acid of the indicated endophilin II and MNSFb peptides, whereas the fifth
shows the corresponding amino acid sequences.
Mass (MH
+
)
Protein Observed Calculated Residues Sequence
Endo-II
919.1 919.0 129–136 DSLDIEVK
1007.0 1007.2 68–76 LTMLNTVSK
1260.5 1260.5 302–312 SMPPLDQPSCK
1389.4 1389.6 54–65 TIEYLQPNPASR
1605.9 1605.8 137–149 QNFIDPLQNLCDK
2039.9 2041.2 103–123 ELGGESNFGDALLDAGESMK
2385.6 2385.7 261–282 EPFELGELEQPNGGFPCAPAPK
MNSFb
2096.3 2096.4 7–25 AQELHTLEVTGQETVAQIK
Table 2. Isopeptide bonds between the C-terminal glycine of
MNSFb and the lysine of endophilin II. The 62 kDa MNSFb adduct
was digested by trypsin and subjected to MALDI-TOF MS analysis.
The data in the first column are the mass values obtained experi-
mentally, whereas the results in the second column are those
calculated from the trypsin-fragmented peptide complexes. The
third column shows the corresponding amino acid sequences of
MNSFb and endophilin II (in bold).
Mass (MH
+
)
Observed Calculated Sequence
1161.3 1161.4 SSDKPIR(291–297)
MLGG(71–74)

1605.2 1605.0 SSDKPIRMPSK(291–301)
MLGG(71–74)
M. Nakamura and S. Shimosaki MNSFb conjugates to endophilin II
FEBS Journal 276 (2009) 6355–6363 ª 2009 The Authors Journal compilation ª 2009 FEBS 6357
in different mouse tissues, including brain (cerebrum
and cerebellum) and testis. It should be pointed out
that the 30 kDa truncated form of endophilin II
(tEnd-II) was consistently observed in liver (Fig. 4A,
lane 4). The N-terminal region of endophilin II might
be cleaved, because we employed antibodies against
the middle (in this study) and C-terminal (not shown)
regions of endophilin II. We examined the expression
of endophilin II in the murine Raw264.7 macrophage-
like cell line and peritoneal macrophages, because
endophilin II is a member of the endocytic endophilin
family. We observed two bands (62 kDa MNSFb–
endophilin II and 45 kDa endophilin II) in the western
blot of the macrophages (Fig. 4B, lanes 1 and 2).
Interestingly, heterogeneous bands, including the
62 kDa band (MNSFb–endophilin II), were reproduc-
ibly observed in liver lysates but not in brain lysates
(Fig. 4B, lanes 3 and 4: dialyzed against NaCl ⁄ P
i
in
the presence of protease inhibitors). This observation
is consistent with the results for the formation of the
62 kDa complex of MNSFb and endophilin II in liver
(Fig. 3).
Regulation of phagocytosis by endophilin II
Endophilin II is one of three members of the subgroup

endophilin A, but its function in regulating endocytosis
remains unclear. It has been reported that endophilin I
and endophilin III are implicated in receptor-mediated
endocytosis [15]. Thus, we assessed the possible roles
of endophilin II in the phagocytosis of zymosan in
macrophages. We employed Raw264.7 cells, which
have been studied in our laboratory [3]. The 62 kDa
MNSFb adduct and endophilin II are expressed in
untreated Raw264.7 cells (Figs 4B and 5A). Endophi-
lin II small interfering RNA (siRNA), but not control
scramble siRNA, specifically reduced the expression of
endophilin II but not of a-tubulin (Fig. 5A). It should
be noted that the expression of 62 kDa MNSFb–endo-
philin II was also decreased. As can be seen in Fig. 5B,
Raw264.7 cells phagocytized zymosan particles
(30.3% ± 3.6%). Cells capturing more than two ops-
onized zymosan particles were judged as positive. We
examined the effect of endophilin II siRNA on the
phagocytosis of opsonized zymosan in Raw264.7 cells.
The treatment of Raw264.7 cells with endophilin II
siRNA significantly enhanced phagocytosis (1.9-fold)
(Fig. 5C). Interestingly, MNSFb siRNA, which has
been used in the previous studies [3], showed similar
effects on phagocytosis, although the effect was less
potent (1.4-fold) (Fig. 5C). To determine whether
MNSFb–endophilin II formation in Raw264.7 cells is
involved in the phagocytosis of zymosan, immunoblot-
ting was performed with the use of antibodies against
endophilin II. We did not observe 62 kDa MNSFb–
endophilin II in the cells treated with MNSFb siRNA

(Fig. 5A), suggesting that MNSFb siRNA was effec-
tive in reducing MNSFb protein levels. Importantly,
Raw264.7 cells transfected with endophilin II and
MNSFb siRNA (double knockdown) enhanced phago-
cytosis to a similar extent as observed for cells treated
with endophilin II siRNA alone. Thus, it is unlikely
that free MNSFb plays a central role in the phagocy-
tosis. The facilitatory effect of MNSFb siRNA was
lower than that of endophilin II siRNA. In addition,
the expression of 62 kDa MNSFb–endophilin II was
much lower than that of endophilin II. Together, these
results strongly indicate that MNSFb–endophilin II
has a much higher inhibitory activity than endophilin
II alone. Transient DNA transfection experiments were
performed to confirm this idea. As can be seen in
Fig. 5D, transfection with pcDNA3.1–endophilin II
resulted in significant inhibition of the phagocytosis
of zymosan (71.1%) in Raw264.7 cells, consistent
with the data from siRNA knockdown experiments
(Fig. 5C). No such inhibition was observed in cells
Fig. 2. Primary structure of mouse endophi-
lins. End-I, endophilin I; End-II, endophilin II;
End-III, endophilin III. The SH3 domain is
boxed. In End-II, the Lys294 residue respon-
sible for isopeptide formation is in bold. The
lengths and numbers of the endophilins are
indicated.
MNSFb conjugates to endophilin II M. Nakamura and S. Shimosaki
6358 FEBS Journal 276 (2009) 6355–6363 ª 2009 The Authors Journal compilation ª 2009 FEBS
expressing a mutant endophilin II that has an arginine

in place of the lysine at position 294 that is responsible
for MNSFb conjugation. Cotransfection of the expres-
sion vectors encoding endophilin II and MNSFb
caused stronger inhibition of the phagocytosis of
zymosan (75.1%). We observed weak but significant
inhibition of zymosan phagocytosis in cells transfected
with pcDNA3.1–MNSFb. Immunoblotting analysis
showed that the expression of MNSFb–endophilin II
was increased in cotransfected cells (Fig. 5E). Collec-
tively, these observations demonstrate that Lys294 of
endophilin II is the only site of MNSFb conjugation
A
B
C
65 kDa-
A
B
45 kDa-
31 kDa-
65 kDa-
45 kDa-
31 kDa-
8
-End-II
-tEnd-II
1234567
123
4
-End-II
- tEnd-II

-MNSF/End-II
Fig. 4. Tissue distribution of endophilin II. (A) Tissue homogenates
(50 lg of protein each) obtained from the indicated organs were
subjected to immunoblotting analysis using antibody against endo-
philin II. Lane 1: cerebrum. Lane 2: cerebellum. Lane 3: brain stem.
Lane 4: liver. Lane 5: testis. Lane 6: epididymis. Lane 7: prostate
gland. Lane 8: seminal vesicle. (B) Lane 1: peritoneal macrophage
lysates. Lane 2: Raw264.7 cells. Lane 3: liver lysates dialyzed
against NaCl ⁄ P
i
in the presence of protease inhibitors. Lane 4: brain
lysates dialyzed as liver in lane 3.
Fig. 3. Western blot (WB) and immunoprecipitation (IP) analysis of
MNSFb adducts. (A) Samples purified by anti-MNSFb affinity chro-
matography from mouse liver cell lysates were analyzed by wes-
tern blot. Lane 1: control rabbit IgG. Lane 2: antibodies against
MNSFb. Lane 3: antibodies against endophilin II. (B) Liver cell
extracts (300 lg) were immunoprecipitated with antibodies against
MNSFb (lane 2) or normal IgG (lane 1). Immunoprecipitates were
analyzed by western blot with antibodies against endophilin II
(lanes 1 and 2). Conversely, the extracts (300 lg) were immunopre-
cipitated with antibodies against endophilin II (lane 4) or normal IgG
(lane 3). Immunoprecipitates were analyzed by western blot with
antibodies against MNSFb (lanes 3 and 4). (C) Brain cell extracts
(300 lg) were immunoprecipitated with antibodies against MNSFb
(lanes 1 and 2). Immunoprecipitates were analyzed by western blot
with antibodies against endophilin II (lane 1) or endophilin I (lane 2).
The cell extracts (5 lg) were analyzed by western blot with anti-
bodies against endophilin I (lane 3). In (A) and (B), mobilities of the
62 kDa MNSFb adduct (arrowhead) and the molecular mass stan-

dards (kDa) are indicated to the right and left of the figure, respec-
tively. In (C), the mobility of 40 kDa endophilin I (End-I) is indicated
to the right of the figure.
M. Nakamura and S. Shimosaki MNSFb conjugates to endophilin II
FEBS Journal 276 (2009) 6355–6363 ª 2009 The Authors Journal compilation ª 2009 FEBS 6359
and that this post-translational modification may nega-
tively regulate phagocytosis in macrophages.
Discussion
We have previously demonstrated that MNSFb forms
an isopeptide bond with Bcl-G, a proapoptotic protein.
MNSFb becomes isopeptide-linked via its C-terminal
diglycine motif, by analogy with ubiquitin and other
ubiquitin-like modifiers, such as SUMO. MNSFb–Bcl-G
directly binds to ERKs, and inhibits ERK activation by
MAP kinase kinase 1 [3]. Similarly, in this study, we
showed that MNSFb covalently binds to endophilin II in
liver and Raw264.7 cells. We isolated a 62 kDa MNSFb
adduct from mouse liver lysates (Fig. 1). MALDI-TOF
MS analysis of the 62 kDa MNSFb adduct showed that
MNSFb conjugates to endophilin II with a linkage
between the C-terminal Gly74 and Lys294. The molecu-
lar mass of 62 kDa was somewhat larger than that of
endophilin II conjugated by a single molecule of MNSFb
(calculated molecular mass, 49.242 Da). One may specu-
late that MNSFb might conjugate to at least one lysine
in addition to Lys294 in endophilin II. However, we did
not observe any candidate fragments with theoretical
masses by MALDI-TOF MS. Additionally, experiments
with mutants indicate that Lys294 of endophilin II is
the only site of MNSFb conjugation (Fig. 5E).

Endophilins are SH3 domain-containing proteins.
Three isoforms of endophilin have been identified:
endophilin I, which functions during synaptic vesicle
recycling; endophilin II, which is ubiquitously distrib-
uted throughout many tissue types; and endophilin III,
which is expressed in brain and testis [16]. Although
many studies have been focused on endophilin I, the
underlying mechanism of action of endophilin II
remains unclear. Lua and Low have reported [17] that
overexpression of endophilin II enhances epidermal
growth factor (EGF)-stimulated receptor endocytosis
1
A
C
D E
B
2 3 4 5
MNSF/End-II-
End-II-
-tubulin-
Untreated cells
Opsonized zymosan
Control End-II siRNA End-II cDNA
Phagocytosis index
MNSF
––++–+
End-II
mEnd-II
–+–+
– –––

1
0
–
+
–
+
**
**
*
3
0
**
**
*
Phagocytosis index
2
1
End-II siRNA
MNSF siRNA
–+ –+
– – + +
MNSF/End-II-
MNSF
– – + + – +
End-II
mEnd-II
–+ –+
–– ––
–
+

–
+
End-II-
-tubulin-
Fig. 5. Effect of endophilin II on formation of phagosomes in Raw264.7 cells. (A) Expression of endophilin II and MNSFb–endophilin II were
analyzed by western blot after treatment with siRNAs for 72 h. Lanes 1, 3 and 4: transfected with scramble (control) siRNA. Lane 2: endo-
philin II siRNA. Lane 5: MNSFb siRNA. Lanes 1, 2, 4 and 5: immunostained with antibody against endophilin II. Lane 3: control rabbit IgG.
Mobilities of 62 kDa MNSFb ⁄ endophilin II (open arrowhead) and 45 kDa endophilin II (closed arrowhead) are indicated to the left and right of
the figure. (B) Raw264.7 cells were untreated or incubated with IgG-opsonized zymosan particles (Alexa488-labeled) for 30 min. Fluo-
rescence was viewed with confocal microscopy. Cells were transfected with siRNA or cDNA for endophilin II as described in Experimental
procedures. (C) Raw264.7 cells were treated with siRNAs for 72 h before addition of zymosan particles. The phagocytosis assay was
performed as described in Experimental procedures. Cells capturing more than two zymosan particles were regarded as positive cells.
Values are shown as the mean ± standard deviation (n = 5). *P < 0.05 versus untreated; **P < 0.01, one-way ANOVA test. (D) Raw264.7
cells were transiently cotransfected with expression vectors encoding either the endophilin II or endophilin II mutant, along with expression
vector for MNSFb. The phagocytosis assay was performed after 72 h of transfection as described in Experimental procedures. Values are
shown as the mean ± standard deviation (n = 3). *P < 0.05 versus empty vector; **P < 0.01, one-way ANOVA test. (E) Western blot
analysis of extracts of cells transfected with the wild-type or mutant endophilin II cDNA. The results of immunoblotting using antibodies
against endophilin II and a-tubulin are shown.
MNSFb conjugates to endophilin II M. Nakamura and S. Shimosaki
6360 FEBS Journal 276 (2009) 6355–6363 ª 2009 The Authors Journal compilation ª 2009 FEBS
and ERK1 ⁄ 2 phosphorylation. Angers et al. [18] have
reported that the E3 ubiquitin ligase Itch ubiquitinates
endophilin I and localizes to the endosomal system fol-
lowing EGF stimulation. They also mentioned that
endophilin I binds to germinal center kinase-like
kinase, suggesting a role for endophilin I in the c-Jun
N-terminal kinase signaling pathway [19]. Together
with our previous studies on the regulation of ERK
activity by MNSFb, this indicates that endophilins,
ubiquitin and ubiquitin-like protein(s) may be closely

involved in MAP kinase pathways. Experiments are
underway to isolate E3 ubiquitin ligase-like enzyme(s)
involved in MNSFb conjugations.
Sugiura et al. [15] mentioned that the variable region
of endophilin III is important in regulating transferrin
endocytosis. In this study, we showed that MNSFb
covalently modifies the Lys294 in the variable region
of endophilin II and that this modification is involved
in the phagocytosis of zymosan in Raw264.7 cells.
Thus, as in the case of endophilin III, the variable
region of endophilin II may be important in regulating
endocytosis in macrophages. Interestingly, both endo-
philin II and endophilin III negatively regulate recep-
tor-mediated internalization [15]. It has been reported
that endophilin family members bind to synaptojanin
and dynamin via a Grb2-like Src homology 3 domain
(constant region) [9,20]. In this context, MNSFb may
not affect the binding of endophilin II to these part-
ners. It should be noted that the residue responsible
for MNSFb conjugation (Lys294) is only found in
endophilin II (Fig. 2). Indeed, we did not observe
MNSFb–endophilin I in brain extracts (Fig. 3C).
In preliminary experiments, we observed that dectin-
1(b-glucan receptor) involves the mechanism of action
of endophilin II on phagocytosis in macrophages. Dec-
tin-1-mediated intracellular signaling pathways regulat-
ing phagocytosis in macrophages remain largely
unknown. Although Syk involves most of the functions
of dectin-1, this kinase is not required for particle
uptake in macrophages [21]. Investigations are under-

way to clarify whether MNSFb–endophilin II is a
mediator in dectin-1-mediated signaling.
It is interesting that MNSFb–endophilin II was
detected in liver cells in terms of endocytosis. The
truncated 30 kDa form may be a consequence of the
MNSFb adduct. It is possible that the N-terminal
region may be deleted from endophilin II in these tis-
sues, because specific antibodies against C-terminal
(Fig. 4) and central (not shown) regions detected the
truncated form. Investigations are underway to clarify
the mechanism of action of endophilin II in liver cells.
Monoubiquitination is thought to regulate receptor
internalization and endosomal sorting [22]. It is evident
that monoubiquitination is involved in endocytosis of
EGF receptor [23]. In this study, we presented data
showing that endophilin II may act in concert with
MNSFb to inhibit the phagocytosis of zymosan (Fig. 5).
Thus, ubiquitin and ubiquitin-like protein(s) may regu-
late endocytosis by modification of target proteins.
Taken as a whole, the present study demonstrates
that endophilin II may negatively regulate phagocyto-
sis, and that MNSFb–endophilin II formation might
be important for potent regulation of phagocytosis in
macrophages.
Experimental procedures
Materials
Rabbit polyclonal antibodies against MNSF b were prepared
as previously described [4]. Rabbit polyclonal antibodies
against endophilin I and endophilin II were purchased
from Santa Cruz Biotechnology (Santa Cruz, CA, USA).

Alexa488–zymosan A was purchased from Molecular Probes.
The zymosan particles were opsonized by using Opsonizing
Reagents (Molecular Probes, Eugene, OR, USA) derived
from purified rabbit polyclonal IgG antibodies that are
specific for zymosan. Rabbit TrueBlot was obtained from
eBioscience (San Diego, CA, USA). DEAE cellulose was
purchased from Sigma-Aldrich (Tokyo, Japan).
Purification of the MNSFb adduct
Mouse liver was obtained from BALB ⁄ c mice. All purifica-
tion steps were performed at 4 °C. A portion of a mouse liver
(8 g) was homogenized in buffer A (20 mm Tris ⁄ HCl, pH 7.5,
50 mm NaCl, 1 mm dithiothreitol) containing a protease
inhibitor mixture [1 mm 4-(2-aminoethyl)-benzenesulfonyl
fluoride, 10 mgÆL
)1
leupeptin, 5 mgÆL
)1
aprotinin, and
1mgÆL
)1
pepstatin A], using a tissue blender (3 · 30 s). The
homogenate was centrifuged at 13 000 g for 20 min, filtered
through gauze, and centrifuged at 13 000 g for another
20 min. The resulting supernatant was further centrifuged at
120 000 g for 60 min to obtain a cytosolic extract, which was
incubated overnight with DEAE (Sigma, Tokyo, Japan) pre-
equilibrated with buffer A. Following extensive washing with
buffer A, bound proteins were eluted in 20 mm Tris ⁄ HCl
(pH 7.5) and 100 mm NaCl, and the protease inhibitor mix-
ture. Additional purification was achieved by anti-MNSFb

affinity chromatography, as previously described [1].
Immunoblotting
Cell extracts in SDS sample buffer were subjected to 12%
SDS ⁄ PAGE, and blotted onto poly(vinylidene difluoride)
membranes. The membrane was incubated overnight at
4 °C in a Tris-buffered saline solution with 5% milk to
M. Nakamura and S. Shimosaki MNSFb conjugates to endophilin II
FEBS Journal 276 (2009) 6355–6363 ª 2009 The Authors Journal compilation ª 2009 FEBS 6361
block nonspecific binding sites. Subsequently, the mem-
branes were incubated with anti-MNSFb rabbit IgG in
the blocking buffer, after which they were incubated with
peroxidase-conjugated anti-rabbit IgG. Detection was
performed according to the enhanced chemiluminescence
detection system (Amersham Biosciences, Chalfont St Giles,
UK). We have previously demonstrated that antibody
against MNSFb does not cross-react with ubiquitin [4].
In-gel digestion and MALDI-TOF MS
All experimental procedures were described in the previous
study [2]. Briefly, silver-stained spots were cut out of the
gels for in-gel digestion and digested with trypsin (Sigma).
The peptides were extracted from the gel matrix by vortexing
for 30 min, and then concentrated using Zip Tips (Millipore
Corp., Bedford, MA, USA). Peptide mass fingerprinting
was performed using a PerkinElmer ⁄ PerSeptive Biosystems
(Framingham, MA, USA) Voyager-DE-RP MALDI-TOF
mass spectrometer. The peptide samples were cocrystallized
with matrix on a gold-coated sample plate using 0.5 lLof
matrix (a-cyano-4-hydroxytranscinnamic acid) and 0.5 lLof
sample. Cysteines were treated with iodoacetamide to form
carboxyamidomethyl cysteine, and methionine was consid-

ered to be oxidized.
Immunoprecipitation
Immunoprecipitation was performed with a horseradish
peroxidase-conjugated antibody that recognizes native
rabbit IgG (TrueBlot), according to the manufacturer’s
instructions. RIPA buffer (50 mm Tris, 1% Nonidet P-40,
0.25% deoxycholate, 150 mm NaCl, 1 mm EDTA, 1 mm
phenylmethylsulfonyl fluoride, pH 7.4, containing
1mgÆmL
)1
each of the protease inhibitors aprotinin, leu-
peptin, and pepstatin) extracts of Raw264.7 cells were pre-
cleared with 50 mL of anti-rabbit IgG beads for 1 h on ice.
Subsequently, 5 mg of primary antibody against MNSFb
or endophilin II was added to precleared lysates and incu-
bated on ice for an additional 1 h. Samples were then incu-
bated overnight at 4 °C with 50 lL of anti-rabbit IgG
beads. The beads were washed five times with RIPA buffer,
and immunoprecipitates were released from the beads by
10 min of boiling in NuPAGE LDS sample buffer (Invitro-
gen). Immunoblotting was performed with antibody against
MNSFb or antibody against endophilin II. A rabbit IgG
TrueBlot was employed as a second antibody.
Cell culture, the siRNAs, and transfection of cells
Cells of the Raw264.7 macrophage-like cell line (ATCC
TIB-71) was cultured routinely in DMEM with 10% fetal
bovine serum and penicillin ⁄ streptomycin at 37 °C and 5%
CO
2
. Small interfering RNA duplexes were synthesized and

purified by Qiagen, Inc. (Chatsworth, CA, USA). The tar-
get sequences were as follows: MNSFb siRNA-437,
5¢-CCACCCTGCCATGCTAATAAA-3¢ [3]; and endophi-
lin II siRNA, 5¢-AAGGTGCTCTAGAAACACTAA-3¢.
Scramble siRNA directed against 5¢-GGACTCGACGC
AATGGCGTCA-3¢ was the negative control. Cells were
treated with siRNA according to the instructions provided
with the RNAiFect transfection reagent (Qiagen, Inc.).
Raw264.7 cells (1.2 · 10
5
) were treated with 3 lg of siRNA
in RPMI-1640 medium supplemented with 10% fetal boo-
vine serum in the presence of the RNAiFect transfection
reagent. After a 48 h incubation at 37 °C, the medium con-
taining the mixture of RNAiFect and siRNA was replaced
by DMEM containing 10% fetal bovine serum, and cells
were incubated for a further 24 h. Mutant endophilin II
(K294R) was generated by replacing the codon for Lys294
with the codon for arginine by utilizing the LA PCR
in vitro Mutagenesis Kit (Takara Bio, Ohtsu, Japan). In
some experiments, Raw264.7 cells were transiently cotrans-
fected with expression vectors encoding either the endophi-
lin II or endophilin II mutant along with expression vector
for MNSFb, as previously described [3].
Collection of peritoneal macrophages
Peritoneal macrophages were obtained using mice injected
4 days previously with 2 mL of a sterile 3% brewer thio-
glycolate broth (Difco, Detroit, MI, USA). The cells were
collected by centrifugation (at 400 g for 5 min), washed,
and resuspended in DMEM containing 10% fetal bovine

serum. Lysates of the collected macrophages were used for
immunoblotting.
Phagocytosis assays
Raw264.7 cells were treated with siRNAs as described above.
Cells were seeded at 5 · 10
4
cells per well in four-well cham-
ber plates (Nunc, Roskilde, Denmark). Cells were washed,
and Alexa488–zymosan A (Molecular Probes) was added
(10 particles per cell) and allowed to bind for 1 h at 4 °C.
Following this incubation, unbound zymosan was removed
by washing, and the cells were incubated at 37 °C for 30 min
to allow particle uptake. To determine the internalization of
zymosan, cells capturing zymosan were treated with 0.05%
Trypan blue in saline for a few minutes before observations
by fluorescence microscopy. To determine the phagocytosis
index, we identified > 200 Alexa-positive cells in randomly
chosen fields of view, and the percentage of cells capturing
more than two zymosan particles was determined.
Acknowledgement
This study was supported in part by Grants-in-Aid for
Scientific Research (19570131 to M. Nakamura).
MNSFb conjugates to endophilin II M. Nakamura and S. Shimosaki
6362 FEBS Journal 276 (2009) 6355–6363 ª 2009 The Authors Journal compilation ª 2009 FEBS
References
1 Nakamura M, Xavier RM, Tsunematsu T & Tanigawa
Y (1995) Molecular cloning and characterization of a
cDNA encoding monoclonal nonspecific suppressor
factor. Proc Natl Acad Sci USA 92, 3463–3467.
2 Nakamura M & Tanigawa Y (2003) Characterization of

ubiquitin-like polypeptide acceptor protein, a novel pro-
apoptotic member of the Bcl2 family. Eur J Biochem
270, 4052–4058.
3 Nakamura M & Yamaguchi S (2006) The ubiquitin-like
protein MNSFb regulates ERK–MAPK cascade. J Biol
Chem 281, 16861–16869.
4 Nakamura M & Tanigawa Y (1998) Ubiquitin-like
polypeptide conjugates to acceptor proteins in conca-
navalin A- and interferon gamma-stimulated T-cells.
Biochem J 330, 683–688.
5 Haas AL, Ahrens P, Bright PM & Ankel H (1987)
Interferon induces a 15-kilodalton protein exhibiting
marked homology to ubiquitin. J Biol Chem 262,
11315–11323.
6 Raasi S, Schmidtke G, Giuli RD & Groettrup M (1999)
A ubiquitin-like protein which is synergistically induc-
ible by interferon-c and tumor necrosis factor-a. Eur J
Immunol 29, 4030–4036.
7 Kito K, Yeh ETH & Kamitani T (2001) NUB1, a
NEDD8-interacting protein, is induced by interferon
and down-regulates the NEDD8 expression. J Biol
Chem 276, 20603–20609.
8 Raasi S, Schmidtke G & Groettrup M (2001) The
ubiquitin-like protein FAT10 forms covalent conjugates
and induces apoptosis. J Biol Chem 276, 35334–35343.
9 Ringstad N, Nemoto Y & De Camilli P (1997) The
SH3p4 ⁄ Sh3p8 ⁄ SH3p13 protein family: binding partners
for synaptojanin and dynamin via a Grb2-like Src
homology 3 domain. Proc Natl Acad Sci USA 94,
8569–8574.

10 Huttner WB & Schmidt A (2000) Lipids, lipid modifica-
tion and lipid–protein interaction in membrane budding
and fission – insights from the roles of endophilin A1
and synaptophysin in synaptic vesicle endocytosis. Curr
Opin Neurobiol 10, 543–551.
11 Anggono V, Smillie KJ, Graham ME, Valova VA,
Cousin MA & Robinson PJ (2006) Syndapin I is the
phosphorylation-regulated dynamin I partner in
synaptic vesicle endocytosis. Nat Neurosci 9, 752–760.
12 Ringstad N, Gad H, Lo
¨
w P, Di Paolo G, Brodin L,
Shupliakov O & De Camilli P (1999) Endophi-
lin ⁄ SH3p4 is required for the transition from early to
late stages in clathrin-mediated synaptic vesicle endo-
cytosis. Neuron 24, 143–154.
13 Schmidt A, Wolde M, Thiele C, Fest W, Kratzin H,
Podtelejnikov AV, Witke W, Huttner WB & So
¨
ling HD
(1999) Endophilin I mediates synaptic vesicle formation
by transfer of arachidonate to lysophosphatidic acid.
Nature 401, 133–141.
14 Hirayama S, Bajari TM, Nimpf J & Schneider WJ
(2003) Receptor-mediated chicken oocyte growth: differ-
ential expression of endophilin isoforms in developing
follicles. Biol Reprod 68, 1850–1860.
15 Sugiura H, Iwata K, Matsuoka M, Hayashi H, Take-
miya T, Yasuda S, Ichikawa M, Yamauchi T, Mehlen
P, Haga T et al. (2004) Inhibitory role of endophilin 3

in receptor-mediated endocytosis. J Biol Chem 279,
23343–23348.
16 Araki Y, Ogawa K, Toh Y, Akimutsu N, Hmamoto H,
Sekimizu K, Matsusue K, Kono A, Iguchi H &
Takiguchi S (2005) Direct interaction between
metastasis-associated protein 1 and endophilin 3. FEBS
Lett 579, 3731–3736.
17 Lua BL & Low BC (2005) Activation of EGF receptor
endocytosis and ERK1 ⁄ 2 signaling by BPGAP1 requires
direct interaction with EEN ⁄ endophilin II and a func-
tional RhoGAP domain. J Cell Sci 118 , 2707–2721.
18 Angers A, Ramjaun AR & McPherson PS (2004) The
HECT domain ligase itch ubiquitinates endophilin and
localizes to the trans-Golgi network and endosomal
system. J Biol Chem 279, 11471–11479.
19 Ramjaun AR, Angers A, Legendre-Guillemin V, Tong
XK & McPherson PS (2001) Endophilin regulates JNK
activation through its interaction with the germinal cen-
ter kinase-like kinase. J Biol Chem 276, 28913–28919.
20 Gout I, Dhand R, Hiles ID, Fry MJ, Panayotou G,
Das P, Truong O, Totty NF, Hsuan J, Booker GW
et al. (1993) The GTPase dynamin binds to and is acti-
vated by a subset of SH3 domains. Cell 75, 25–36.
21 Underhill DM, Rossnagle E, Lowell CA & Simmons
RM (2005) Dectin-1 activates Syk tyrosine kinase in a
dynamic subset of macrophages for reactive oxygen
production. Blood 106, 2543–2550.
22 Haglund K, Shimokawa N, Szymkiewicz I & Dikic I
(2002) Cbl-directed monoubiquitination of CIN85 is
involved in regulation of ligand-induced degradation of

EGF receptors. Proc Natl Acad Sci USA 99, 12191–
12196.
23 Galan JM & Haguenauer-Tsapis R (1997) Ubiquitin
lys63 is involved in ubiquitination of a yeast plasma
membrane protein. EMBO J 19, 5847–5854.
M. Nakamura and S. Shimosaki MNSFb conjugates to endophilin II
FEBS Journal 276 (2009) 6355–6363 ª 2009 The Authors Journal compilation ª 2009 FEBS 6363