Peptides 32 (2011) 895–899

Contents lists available at ScienceDirect

Peptides
journal homepage: www.elsevier.com/locate/peptides

Enhancement of the cancer targeting specificity of buforin IIb by fusion with an

anionic peptide via a matrix metalloproteinases-cleavable linker
Ju Hye Jang a , Min Young Kim a , Jin-Won Lee b , Sun Chang Kim c , Ju Hyun Cho a,∗
a
Department of Biology, Research Institute of Life Science, Gyeongsang National University, 900 Gajwa-dong, Jinju 660-701, Republic of Korea
b
Department of Life Science and Research Center for Natural Sciences, Hanyang University, Seoul 133-791, Republic of Korea
c
Department of Biological Sciences, Korea Advanced Institute of Science and Technology, Daejeon 305-701, Republic of Korea

a r t i c l e i n f o a b s t r a c t

Article history: Buforin IIb is a novel cell-penetrating anticancer peptide derived from histone H2A. In this study, we
Received 21 December 2010 enhanced the cancer targeting specificity of buforin IIb using a tumor-associated enzyme-controlled
Received in revised form 12 February 2011 activation strategy. Buforin IIb was fused with an anionic peptide (modified magainin intervening
Accepted 12 February 2011
sequence, MMIS), which neutralizes the positive charge of buforin IIb and thus renders it inactive, via a
Available online 18 February 2011
matrix metalloproteinases (MMPs)-cleavable linker. The resulting MMIS:buforin IIb fusion peptide was
completely inactive against MMPs-nonproducing cells. However, when the fusion peptide was admin-
Keywords:
istrated to MMPs-producing cancer cells, it regained the killing activity by releasing free buforin IIb
Buforin IIb
Anticancer peptide
through MMPs-mediated cleavage. Moreover, the activity of the fusion peptide toward MMPs-producing
Antimicrobial peptide cancer cells was significantly decreased when the cells were pretreated with a MMP inhibitor. Taken
Fusion peptide together, these data indicate that the cancer targeting specificity of MMIS:buforin IIb is enhanced com-
Matrix metalloproteinase pared to the parent peptide by reactivation at the specialized areas where MMPs are pathologically
produced.
© 2011 Elsevier Inc. All rights reserved.

1. Introduction likelihood of resistance development and additive effects in com-

bination therapy [15,25].
Although numerous chemotherapeutic drugs, including alky- Most anticancer peptides selectively kill cancer cells by dis-
lating agents, antimetabolites and hormone agonists/antagonists, ruption of the cancer cell membrane or permeation and swelling
have been developed and successfully used for the treatment of of mitochondria, where the electrostatic attraction between the
metastatic cancers, the efficacy of cancer chemotherapy is lim- negatively charged membrane components of cancer cells and
ited by severe side-effects and dose limitations [30]. Current the cationic anticancer peptides is believed to play a crucial role
chemotherapeutic drugs cannot distinguish between cancer cells [29]. Among the anticancer peptides, buforin IIb-a synthetic analog
and proliferating normal cells, and kill both. Moreover, cancer cells of buforin II that contains a proline hinge between the two ␣-
develop resistance to these drugs that is mediated by the overex- helices and a model ␣-helical sequence at the C-terminus (3×RLLR)
pression of multidrug-resistance proteins that pump the drugs out [26]-selectively targets cancer cells through interaction with the
of cells and thus render the drugs ineffective [28]. To overcome cell-surface gangliosides. Buforin IIb then traverses cancer cell
the limits of current chemotherapeutic drugs, many researchers membranes without damaging them and induces mitochondria-
have labored to identify new anticancer molecules. Recently, dependent apoptosis [17]. Buforin IIb also displays powerful
anticancer peptides, cationic antimicrobial peptides (AMPs) with cytotoxic activity when injected into solid tumors in p53-deficient
cancer-selective toxicity, have received attention as alternative mice [6]. These results suggest that buforin IIb may be developed
chemotherapeutic agents that overcome the limits of current drugs. into a novel therapeutic agent for the treatment of cancers. How-
These peptides have several advantages over currently used anti- ever, preliminary study showed that buforin IIb also killed normal
cancer therapeutics, such as low intrinsic cytotoxicity, decreased cells at higher concentrations, though not efficient compared to
cancer cells. Moreover, neurons, of which gangliosides are abun-
dant in the plasma membranes [31], may be vulnerable to buforin
IIb. Therefore, the cancer-targeting specificity of buforin IIb should
Abbreviations: AMP, antimicrobial peptide; MMIS, modified magainin interven- be enhanced to fully exploit its therapeutic potential.
ing sequence; MMP, matrix metalloproteinase; MTT, 3-(4,5-dimethylthiazol-2-yl)-
In this study, we used a tumor-associated enzyme-controlled
2,5-diphenyl tetrazolium bromide.
∗ Corresponding author. Tel.: +82 55 751 5950; fax: +82 55 754 0086. activation strategy to enhance the cancer targeting specificity of
E-mail address: juhyun.cho@gnu.kr (J.H. Cho). buforin IIb. Matrix metalloproteinases (MMPs) are a family of over

0196-9781/$ – see front matter © 2011 Elsevier Inc. All rights reserved.
doi:10.1016/j.peptides.2011.02.010
896 J.H. Jang et al. / Peptides 32 (2011) 895–899

Table 1 in the serum-free medium to inhibit endogenous MMPs expressed

Peptide sequences.
by the cells. After 48 h of incubation, cells were treated with pep-
Peptide name Peptide sequence tides (buforin IIb or MMIS:buforin IIb, 0–8 ␮M) and incubated
Buforin IIb RAGLQFPVG[RLLR]3 for another 24 h. Cell viability was measured with the 3-(4,5-
MMIS:buforin IIb DAEAVGPEAADEEKDEDGPLG/IAGQRAGLQFPVG[RLLR]3 dimethylthiazol-2-yl)-2,5-diphenyl tetrazolium bromide (MTT)
assay using the CellTiter 96-cell proliferation assay kit (Promega,
Note: the MMP-2 cleavage site is underlined. The cleavage position is indicated by
“/”. Madison, WI, USA). The percentage of cell viability was determined
as follows:

25 secreted and membrane-bound zinc endopeptidases that par- As − A0

Viability (%) = × 100%
ticipate in numerous normal and pathologic extracellular matrix Ac − A0
(ECM) remodeling events, including tumor progression, metasta- where As is the absorbance value of the sample, Ac is the absorbance
sis and angiogenesis [5,10,22,32]. They have also been implicated of control and A0 is the background absorbance. Each experiment
as key mediators of the cellular responses to inflammation and was repeated at least three times independently.
degeneration [16,27,34]. Among the MMPs, MMP-2 (gelatinase A)
and MMP-9 (gelatinase B) especially play a critical role in ECM 2.4. Hemolysis assay
breakdown, and many type of cancer have shown increased lev-
els of MMP-2 and MMP-9 [7–9,12,20,23]. MMP-2 and MMP-9 Hemolytic activity was assayed as described by Aboudy et al.
substrate sequences have successfully been used in prodrug strate- [1] with a slight modification. Three milliliters of freshly pre-
gies attempting to release doxorubicin, melittin and methotrexate pared human red blood cells (RBCs) were washed with isotonic
as the cytotoxic agents to tumor tissues [3,14,21]. Therefore, we phosphate-buffered saline (PBS), pH 7.4, until the color of the super-
fused buforin IIb with an anionic peptide which can neutralize the natant turned clear. The washed RBCs were then diluted to a final
positive charge of buforin IIb and thus render it inactive via a MMP- volume of 20 ml with the same buffer. Peptide samples (10 ␮l), seri-
2/MMP-9-cleavable linker, and analyzed the anticancer activity of ally diluted in PBS, were added to 190 ␮l of the cell suspension
the fusion peptide using cell lines of different MMPs expression in microfuge tubes. Following gentle mixing, the tubes were incu-
levels. Here we have shown that the cancer targeting specificity of bated at 37 ◦ C for 30 min and then centrifuged at 4000 × g for 5 min.
the fusion peptide was enhanced compared to the parent peptide One hundred microliters of supernatant were taken, diluted to 1 ml
by reactivation through MMPs-mediated cleavage. with PBS, and absorbance at 567 nm was measured to monitor the
release of hemoglobin that indicated RBC membrane damage. Zero
2. Materials and methods hemolysis and 100% hemolysis consisted of RBC suspended in PBS
and 0.2% Triton X-100, respectively. The percentage of hemolysis
2.1. Peptides was determined as follows:
As − A0
Buforin IIb and MMIS:buforin IIb (Table 1) were synthesized on Hemolysis (%) = × 100%
A100 − A0
a MilliGen 9050 peptide synthesizer (Peptron, Korea). Synthetic
peptides were purified by reversed-phase high performance liq- where As is the absorbance of the sample, A100 is the absorbance of
uid chromatography and characterized by mass spectroscopy and completely lysed RBC in 0.2% Triton X-100, and A0 is the absorbance
amino acid analysis. To test whether MMIS:buforin IIb is digested of zero hemolysis.
by MMPs, 10 ␮M of MMIS:buforin IIb was incubated with 10 nM
of recombinant human active MMP-2 or MMP-9 (Calbiochem, San 2.5. Reverse transcription (RT)-PCR analysis
Diego, CA, USA) in the digestion buffer (50 mM Tris–HCl, pH 7.5,
200 mM NaCl, 5 mM CaCl2 , 5 ␮M ZnCl2 ) at 37 ◦ C. At the designated Total RNAs were extracted from cultured cells using a Trizol
time points, the digestion mixture was sampled and analyzed by reagent (Invitrogen, Carlsbad, CA, USA), and 1 ␮g of each RNA
15% SDS–PAGE. was used for semi-quantitative RT-PCR. Three sets of gene-
specific primers were used to specifically amplify MMP-2 and
2.2. Cells MMP-9 fragments [MMP-2F (5 -CAATACCTGAACACTTTCTATGG-
3 )/MMP-2R (5 -CTGTATGTGATCTGGTTCTTG-3 ) for human and
FSF (normal human foreskin fibroblast) was obtained from the mouse MMP-2 genes; hMMP-9F (5 -CCTGGAGACCTGAGAACCAA-
National Institute of General Medical Sciences Human Genetic 3 )/hMMP-9R (5 -GGACCACAACTCGTCATCG-3 ) for human
Mutant Cell Repository (NIGMS). HeLa (human cervix adenocarci- MMP-9 gene; mMMP-9F (5 -CCCAAAGACCTGAAAACCTCCAA-
noma), B16-F0 (mouse melanoma), HT1080 (human fibrosarcoma) 3 )/mMMP-9R (5 -CGACCACAACTCGTCGTCG-3 ) for mouse
and U87MG (human glioblastoma) were purchased from the MMP-9 gene]. ␤-Actin gene expression, which was amplified
American Tissue Cell Culture (ATCC). Cells were cultured in by ␤-actin-F (5 -GCATCACACCTTCTACAATGAGC-3 )/␤-actin-R
a complete medium [DMEM (FSF, HeLa, B16-F0 and U87MG) (5 -GCTCATAGCTCTTCTCCAGGG-3 ), was used as an internal con-
or RPMI-1640 (HT1080) supplemented with 10% FBS and 0.1% trol. The RT-PCR was performed using the OneStep RT-PCR Kit
penicillin–streptomycin] in a humidified atmosphere of 5% CO2 at (QIAGEN, Hilden, Germany) with a temperature profile of 50 ◦ C
37 ◦ C. Trypsin–EDTA (0.05%) was used to detach cells in subcultur- for 30 min, 94 ◦ C for 2 min, followed by 35 cycles of 94 ◦ C for 30 s,
ing. All the cell culture media and reagents were purchased from T ◦ C (depending on the target gene) for 30 s, 72 ◦ C for 60 s, and
Lonza (Basel, Switzerland). final extension at 72 ◦ C for 10 min. The annealing temperature
(T) was optimized for each pair of primers: MMP-2F/MMP-2R,
2.3. In vitro cytotoxicity assay 55 ◦ C; hMMP-9F/hMMP-9R and mMMP-9F/mMMP-9R, 66 ◦ C;
␤-actin-F/␤-actin-R, 60 ◦ C. After amplification, the PCR products
Cells were seeded onto 96-well plates at a density of 5000 were analyzed by 1% agarose gel electrophoresis, and bands
cells/well (HeLa, B16-F0 and U87MG) or 10,000 cells/well (FSF and were visualized by ethidium bromide staining. The predicted PCR
HT1080) in 0.1 ml of serum-free medium. In some experiments, product sizes for MMP-2, MMP-9 and ␤-actin are 227 bp, 539 bp
MMP-2/MMP-9 inhibitor III (0–80 ␮M, Calbiochem) was included and 467 bp, respectively.
 J.H. Jang et al. / Peptides 32 (2011) 895–899 897

Fig. 1. Cell viability of the MMPs-nonproducing cancer cells treated with buforin IIb and MMIS:buforin IIb. (A) In vitro cytotoxicity. Buforin IIb (䊉) or MMIS:buforin IIb ()
was added to FSF and HeLa cells. After a 24-h incubation with the peptides, cell viability was measured by the MTT assay. (B) Hemolytic activity. Fresh RBC suspension was
incubated with peptides. The release of hemoglobin into supernatant was monitored to indicate the membrane damage of RBC. Data in (A) and (B) represent the mean ± SD
of 3 independent experiments. Error bars in (B) are smaller than the size of the symbols and are not shown in the figure.

2.6. Gelatin zymography tested against HeLa cells, the free buforin IIb released from the
fusion peptide (IAGQ-buforin IIb) had the same cytotoxic activ-
Cells were grown to about 70% confluence before the growth ity as the parent peptide, which indicated that the four amino
media was exchanged to the serum-free media. After 72 h of incu- acids (IAGQ) added to the N-terminal of buforin IIb as a result of
bation, the conditioned media was collected by centrifugation MMPs-mediated cleavage had no adverse effects on the cytotoxic
and concentrated with Microcon YM-10 filters (Millipore, Biller- activity of buforin IIb (data not shown). These results showed that
ica, MA, USA). 10 ␮g of the total protein was electrophoresed on a free buforin IIb was released from MMIS:buforin IIb by MMP-2 and
10% SDS–polyacrylamide gel containing 1 mg/ml gelatin (Sigma, St. MMP-9, and regained the cytotoxic activity.
Louis, MO, USA). The gel was washed twice for 20 min in 2.5% (v/v)
Triton X-100, and then incubated for 20 h at 37 ◦ C in the develop- 3.3. Cytotoxicity of MMIS:buforin IIb against MMPs-producing
ment buffer (50 mM Tris–HCl, pH 7.5, 200 mM NaCl, 5 mM CaCl2 , cancer cells
and 5 ␮M ZnCl2 ). The gel was stained with Coomassie Brilliant Blue
and destained. The white bands in the dark backgrounds indicate Based on the in vitro cleavage data, we assumed that the free
the presence of the gelatinases. buforin IIb can be released from MMIS:buforin IIb and regain the
cytotoxic activity if cancer cells secrete MMP-2 or MMP-9. To test
3. Results this hypothesis, we first checked the production of MMP-2 and
MMP-9 in four cancer cell lines (HeLa; B16-F0; HT1080; U87MG) by
3.1. Design of the MMIS:buforin IIb fusion peptide semi-quantitative RT-PCR. As shown in Fig. 3A, HeLa did not express
either MMP-2 or MMP-9, while B16-F0 and HT1080 expressed both
Though buforin IIb displayed selective cytotoxicity against can- MMP-2 and MMP-9. In U87MG, MMP-2 was strongly expressed, but
cer cells, it also affected the viability of normal cells such as human MMP-9 was weakly expressed. The production of MMPs in these
fibroblasts and erythrocytes at higher concentrations. As shown cell lines was further confirmed by gelatin zymography of the cell-
in Fig. 1, buforin IIb killed 35.13 ± 4.66% of FSF at 8 ␮M (A), and conditioned media. To examine whether MMP-2 and MMP-9 were
lysed 14.75 ± 0.14% of RBC at 50 ␮M (B). Therefore, to increase secreted to the media, gelatinolytic activities of MMP-2 and MMP-
the specificity of buforin IIb toward cancer cells, we designed a 9 were analyzed using the conditioned media from the four cell
fusion peptide to generate a pro-form of buforin IIb that is inac- lines (Fig. 3B). No clear band was shown in the HeLa conditioned
tive by itself and becomes active when buforin IIb part is released
from the fusion peptide. We fused an anionic peptide (modified
magainin intervening sequence, MMIS) which can neutralize the
positive charge of buforin IIb and thus render it inactive [18], at
the N-terminal of buforin IIb. A MMP-2/MMP-9-cleavable linker
(GPLGIAGQ) was introduced at the fusion site of buforin IIb and
MMIS (Table 1). Unlike buforin IIb, the MMIS:buforin IIb fusion
peptide was completely inactive against human fibroblasts and ery-
throcytes at tested concentrations (Fig. 1). Moreover, it was also
inactive against HeLa cells, which display very little MMP-2 and
MMP-9 activity (Figs. 1A and 3). These results showed that the
fusion partner MMIS efficiently shielded the cytotoxicity of buforin
IIb, thus rendering the MMIS:buforin IIb fusion peptide inactive
against MMPs-nonproducing cells.

3.2. Cleavage of MMIS:buforin IIb by MMP-2 and MMP-9

To regain the cytotoxic activity, free buforin IIb should be

released from the MMIS:buforin IIb fusion peptide by MMPs-
mediated cleavage. To test if MMPs can cleave the fusion peptide in Fig. 2. Cleavage of MMIS:buforin IIb by MMPs. 10 ␮M of MMIS:buforin IIb was incu-
bated with 10 nM of recombinant human active MMP-2 (A) or MMP-9 (B) in the
vitro, MMIS:buforin IIb was incubated with active MMP-2 or MMP-
digestion buffer at 37 ◦ C. At the designated time points, the digestion mixture was
9. As shown in Fig. 2, both MMP-2 and MMP-9 cleaved the fusion sampled and analyzed by 15% SDS–PAGE. Molecular markers are shown in the left.
peptide and generate a band corresponding to buforin IIb. When Buforin IIb is shown in the right.
898 J.H. Jang et al. / Peptides 32 (2011) 895–899

Fig. 5. The influence of MMP-2/MMP-9 inhibitor III pretreatment on the cell viabil-
ity. After treatment with MMP-2/MMP-9 inhibitor III, the viability of B16-F0, HT1080
and U87MG cells exposed to MMIS:buforin IIb was determined by the MTT assay.
Data represent the mean ± SD of 3 independent experiments.

growth inhibition compared to an untreated control) of the fusion

Fig. 3. Expressions of MMP-2 and MMP-9 in various cancer cells. (A) Semi-
quantitative RT-PCR analysis. 1 ␮g of each RNA extracted from the cultured cells peptide against B16-F0, HT1080 and U87MG were 4.9 ␮M, 8.5 ␮M
was subjected to RT-PCR using the gene-specific primers for MMP-2 and MMP-9. and 5.9 ␮M, respectively. Moreover, the cytotoxic activity of the
␤-Actin gene expression was used as internal control. The size of each PCR product fusion peptide was significantly decreased when these cells were
matches the predicted size (see Section 2). (B) Gelatin zymography. The conditioned pretreated with MMP-2/MMP-9 inhibitor III. The MMP-2/MMP-9
media collected from each cell culture were separated on a 10% SDS–polyacrylamide
gel containing gelatin as a substrate. After a 20-h incubation at 37 ◦ C, the gel was
inhibitor III-treated cells showed 80–90% viability as compared
stained with Coomassie Brilliant Blue and destained. The white bands in the dark to 30–40% viability for the untreated cells (Fig. 5). Overall, these
backgrounds indicate the presence of gelatinases (MMP-2 and MMP-9). results demonstrate that when the MMIS:buforin IIb fusion pep-
tide was administrated to MMPs-producing cancer cells, the fusion
peptide was cleaved at the MMPs cleavage site and the liberated
media. In the conditioned media from B16-F0 and HT1080 cells, buforin IIb killed cancer cells.
the bands at 72 kDa and 62 kDa indicated the presence of pro- and
active MMP-2, and the bands at 100 kDa and 86 kDa indicated the
presence of mouse- and human MMP-9, respectively. In the U87MG 4. Discussion
conditioned media, the prominent bands corresponding to pro- and
active MMP-2 were present, but only a faint band at 86 kDa was Over the last two decades many anticancer peptides, cationic
present showing little amount of MMP-9. AMPs with cancer-selective toxicity, have appeared [15,25,29].
We then tested if the MMIS:buforin IIb fusion peptide regains However, most of them do not have enough specificity-low cyto-
the cytotoxic activity when administrated to the MMPs-producing toxic activity to normal mammalian cells-to be developed into a
cancer cells cultured in the serum free media. The absence of serum novel cancer therapeutics. Among the anticancer peptides, buforin
protein in the culture media was intended to mimic the condi- IIb has received attention as a potential candidate for a novel anti-
tions in the tumor vicinity in vivo. Because of the large sizes of cancer drug since it displayed selective cytotoxicity against 62
the serum protease inhibitors such as ␣2-macroglobulin, they are cancer cell lines [17]. The selectivity of buforin IIb for cancer cells
unable to cross the capillary endothelium and therefore are absent results largely from the overall positive charge (+7) of the peptide.
in the interstitial fluid. In the tumor tissues, the activities of MMP- Positively charged buforin IIb efficiently targets cancer cells whose
2 and MMP-9 are not inhibited by serum proteins although they outer surfaces of the plasma membranes contain high concentra-
are regulated more specifically by tissue inhibitors of metallopro- tions of negatively charged gangliosides [2,11,13]. In contrast, the
teinases (TIMPs) [4,33]. As shown in Fig. 4, the fusion peptide killed surface of normal mammalian cell membranes is composed mainly
B16-F0, HT1080 and U87MG cells as potent as the parent buforin of neutral zwitterionic phospholipids and sterols [19], and thus not
IIb. The IC50 values (the concentration of peptide that induced 50% a good target for buforin IIb.

Fig. 4. Cell viability of the MMPs-producing cancer cells treated with buforin IIb and MMIS:buforin IIb. After a 24-h incubation with buforin IIb (䊉) or MMIS:buforin IIb (),
the viability of B16-F0, HT1080 and U87MG cells was measured by the MTT assay. Data represent the mean ± SD of 3 independent experiments.
 J.H. Jang et al. / Peptides 32 (2011) 895–899 899

Though buforin IIb showed greater specificity toward cancer by matrix metalloproteinase II and matrix metalloproteinase IX. Bioconjug
cells than other anticancer peptides, it also affected the viabil- Chem 2004;15:931–41.
[5] Chiarug V, Ruggiero M, Magnelli L. Angiogenesis and the unique nature of tumor
ity of normal cells such as human fibroblasts and erythrocytes at matrix. Mol Biotechnol 2002;21:85–90.
higher concentrations (Fig. 1). Therefore, to increase the specificity [6] Cho JH, Sung BH, Kim SC. Buforins:histone H2A-derived antimicrobial peptides
of buforin IIb toward cancer cells, tumor-associated enzyme- from toad stomach. Biochim Biophys Acta 2009;1788:1564–9.
[7] Davies B, Miles DW, Happerfield LC, Naylor MS, Bobrow LG, Rubens RD, et al.
controlled activation strategy was explored in this study. The Activity of type IV collagenases in benign and malignant breast disease. Br J
MMIS:buforin IIb fusion peptide was constructed by joining an Cancer 1993;67:1126–31.
anionic peptide and buforin IIb with a linker cleavable by the [8] Davies B, Waxman J, Wasan H, Abel P, Williams G, Krausz T, et al. Levels of matrix
metalloproteases in bladder cancer correlate with tumor grade and invasion.
tumor-associated enzymes MMP-2 and MMP-9 (Table 1). Because Cancer Res 1993;53:5365–9.
the positive charge is critical to the cytotoxicity of anticancer pep- [9] D’Errico A, Garbisa S, Liotta LA, Castronovo V, Stetler-Stevenson WG, Grigioni
tides such as buforin IIb [29], we reasoned that fusion with an WF. Augmentation of type IV collagenase, laminin receptor, and Ki67 prolif-
eration antigen associated with human colon, gastric, and breast carcinoma
anionic peptide would suppress the cytotoxic activity of buforin
progression. Mod Pathol 1991;4:239–46.
IIb by counterbalancing the overall positive charge. For an anionic [10] Egeblad M, Werb Z. New functions for the matrix metalloproteinases in cancer
fusion partner, a modified magainin intervening sequence (MMIS), progression. Nat Rev Cancer 2002;2:161–74.
which has been proven to efficiently shield the toxicity of expressed [11] Guo Z, Wang Q. Recent development in carbohydrate-based cancer vaccines.
Curr Opin Chem Biol 2009;13:608–17.
buforin IIb on the host Escherichia coli cells, was selected [18]. MMIS [12] Hamdy FC, Fadlon EJ, Cottam D, Lawry J, Thurrell W, Silcocks PB, et al. Matrix
and buforin IIb was joined by the octapeptide linker (GPLGIAGQ), metalloproteinase 9 expression in primary human prostatic adenocarcinoma
which is cleaved efficiently at the G–I bond by MMP-2 and MMP-9 and benign prostatic hyperplasia. Br J Cancer 1994;69:177–82.
[13] Hanai N, Nakamura K, Shitara K. Recombinant antibodies against ganglioside
[21,24]. The embedded linker provides the targeting function since expressed on tumor cells. Cancer Chemother Pharmacol 2000;46 Suppl.:S13–7.
MMP-2 and MMP-9 are specifically overexpressed and secreted at [14] Holle L, Song W, Holle E, Wei Y, Wagner T, Yu X. A matrix metalloproteinase 2
tumor tissues. On contact with MMP-2 or MMP-9 producing cells, cleavable melittin/avidin conjugate specifically targets tumor cells in vitro and
in vivo. Int J Oncol 2003;22:93–8.
the embedded linker should be cleaved, restoring the cytotoxic [15] Hoskin DW, Ramamoorthy A. Studies on anticancer activities of antimicrobial
activity of the fusion peptide by releasing free buforin IIb. peptides. Biochim Biophys Acta 2008;1778:357–75.
In agreement with our hypothesis, the MMIS:buforin IIb fusion [16] Kawasaki Y, Xu ZZ, Wang X, Park JY, Zhuang ZY, Tan PH, et al. Distinct roles of
matrix metalloproteases in the early- and late-phase development of neuro-
peptide exhibited MMPs-related selectivity. Cancer cells that pathic pain. Nat Med 2008;14:331–6.
expressed high amounts of MMPs (B16-F0, HT1080 and U87MG) [17] Lee HS, Park CB, Kim JM, Jang SA, Park IY, Kim MS, et al. Mechanism of
were sensitive to the fusion peptide (Fig. 4), whereas the MMPs- anticancer activity of buforin IIb, a histone H2A-derived peptide. Cancer Lett
2008;271:47–55.
deficient HeLa cells were resistant (Fig. 1A), suggesting that the
[18] Lee JH, Minn I, Park CB, Kim SC. Acidic peptide-mediated expression of the
anticancer activity of the fusion peptide was enzyme dependent. antimicrobial peptide buforin II as tandem repeats in Escherichia coli. Protein
This was supported by the facts that MMP-2/MMP-9 inhibitor III Expr Purif 1998;12:53–60.
was able to protect the MMPs-producing cancer cells from the [19] Lehrer RI, Lichtenstein AK, Ganz T. Defensins: antimicrobial and cytotoxic pep-
tides of mammalian cells. Annu Rev Immunol 1993;11:105–28.
cytotoxicity of the fusion peptide (Fig. 5) and MMP-2/MMP-9 pre- [20] Levy AT, Cioce V, Sobel ME, Garbisa S, Grigioni WF, Liotta LA, et al. Increased
digestion could restore the activity of the fusion peptide against expression of the Mr 72,000 type IV collagenase in human colonic adenocarci-
HeLa (data not shown). Taken together, these data indicate that noma. Cancer Res 1991;51:439–44.
[21] Mansour AM, Drevs J, Esser N, Hamada FM, Badary OA, Unger C, et al. A new
the cancer targeting specificity of MMIS:buforin IIb was enhanced approach for the treatment of malignant melanoma: enhanced antitumor effi-
compared to the parent peptide by reactivation at the specialized cacy of an albumin-binding doxorubicin prodrug that is cleaved by matrix
areas where MMPs are pathologically produced. Ongoing studies metalloproteinase 2. Cancer Res 2003;63:4062–6.
[22] Mott JD, Werb Z. Regulation of matrix biology by matrix metalloproteinases.
are aimed at evaluating the activity of MMIS:buforin IIb fusion pep- Curr Opin Cell Biol 2004;16:558–64.
tide and its potential as a therapeutic agent in vivo using xenograft [23] Naylor MS, Stamp GW, Davies BD, Balkwill FR. Expression and activity of MMPs
mouse models. Furthermore, the design strategy used in the con- and their regulators in ovarian cancer. Int J Cancer 1994;58:50–6.
[24] Netzel-Arnett S, Sang QX, Moore WG, Navre M, Birkedal-Hansen H, Van Wart
struction of MMIS:buforin IIb, which joins an anionic peptide and an HE. Comparative sequence specificities of human 72- and 92-kDa gelati-
anticancer peptide with a linker cleavable by the tumor-associated nases (type IV collagenases) and PUMP (matrilysin). Biochemistry 1993;32:
enzymes MMP-2 and MMP-9, could be applied to increase the can- 6427–32.
[25] Papo N, Shai Y. Host defense peptides as new weapons in cancer treatment.
cer targeting specificity of other cationic anticancer peptides.
Cell Mol Life Sci 2005;62:784–90.
[26] Park CB, Yi KS, Matsuzaki K, Kim MS, Kim SC. Structure–activity analysis of
Acknowledgements buforin II, a histone H2A-derived antimicrobial peptide: the proline hinge is
responsible for the cell-penetrating ability of buforin II. Proc Natl Acad Sci USA
2000;97:8245–50.
This work was supported by a grant of the Korea Healthcare [27] Parks WC, Wilson CL, Lopez-Boado YS. Matrix metalloproteinases as mod-
technology R&D Project, Ministry for Health, Welfare & Family ulators of inflammation and innate immunity. Nat Rev Immunol 2004;4:
Affairs, Republic of Korea (A084115). 617–29.
[28] Perez-Tomas R. Multidrug resistance: retrospect and prospects in anti-cancer
drug treatment. Curr Med Chem 2006;13:1859–76.
References [29] Schweizer F. Cationic amphiphilic peptides with cancer-selective toxicity. Eur
J Pharmacol 2009;625:190–4.
[1] Aboudy Y, Mendelson E, Shalit I, Bessalle R, Fridkin M. Activity of two synthetic [30] Smith LL, Brown K, Carthew P, Lim CK, Martin EA, Styles J, et al. Chemopreven-
amphiphilic peptides and magainin-2 against herpes simplex virus types 1 and tion of breast cancer by tamoxifen: risks and opportunities. Crit Rev Toxicol
2. Int J Pept Protein Res 1994;43:573–82. 2000;30:571–94.
[2] Bitton RJ, Guthmann MD, Gabri MR, Carnero AJ, Alonso DF, Fainboim L, et al. [31] Sonnino S, Mauri L, Chigorno V, Prinetti A. Gangliosides as components of lipid
Cancer vaccines: an update with special focus on ganglioside antigens. Oncol membrane domains. Glycobiology 2007;17:1R–3R.
Rep 2002;9:267–76. [32] Vartak DG, Gemeinhart RA. Matrix metalloproteases: underutilized targets for
[3] Chau Y, Padera RF, Dang NM, Langer R. Antitumor efficacy of a novel drug delivery. J Drug Target 2007;15:1–20.
polymer–peptide–drug conjugate in human tumor xenograft models. Int J Can- [33] Woessner JF, Nagase H. Matrix metalloproteinases and TIMPs. New York:
cer 2006;118:1519–26. Oxford University Press; 2000.
[4] Chau Y, Tan FE, Langer R. Synthesis and characterization of [34] Yong VW. Metalloproteinases: mediators of pathology and regeneration in the
dextran–peptide–methotrexate conjugates for tumor targeting via mediation CNS. Nat Rev Neurosci 2005;6:931–44.