Arch Dermatol Res (1999) 291 : 432–436
O R I G I N A L PA P E R
S. Tajima · H. Inoue · A. Kawada · A. Ishibashi ·
H. Takahara · N. Hiura
Alginate oligosaccharides modulate cell morphology,
cell proliferation and collagen expression
in human skin fibroblasts in vitro
Received: 16 October 1998 / Received after revision: 8 February 1999 / Accepted: 12 February 1999
Abstract Effects of alginate oligosaccharides on cell
proliferation and expression of collagen in cultured skin
fibroblasts were studied. The oligosaccharides were
found to suppress fibroblast proliferation to half the
level in control cultures at a dose of 10 mg/ml during a
period of 5 days. The inhibition was accompanied by a
change in cell shape. The inhibition of cell prolifera-
tion was reversible, since depletion of these oligosac-
charides led to a recovery of cell motility. Treatment of
confluent cells with 10 mg/ml oligosaccharides for 5 days
resulted in a reduction in collagen synthesis to one half
of that in control cultures and inhibition of steady
state levels of α1(I), α2(I), α1(III) and α1(VI) collagen
mRNAs. These results suggest that alginate oligosac-
charides are potential modulators of dermal fibro-
blasts and may provide a useful tool for the treatment
of disorders related to abnormal collagen metabolism.
Key words Alginate oligosaccharides · Collagen ·
Human skin fibroblasts
Introduction
Alginates are naturally occurring polysaccharides found
in brown seaweed and are composed of two building
blocks, mannuronic acid and guluronic acid. Alginates are
extracted from the seaweed and utilized commercially in
the food and pharmaceutical industry. For a long time al-
ginates have been known to have homeostatic properties
S. Tajima (쾷) · H. Inoue · A. Kawada · A. Ishibashi
Department of Dermatology, National Defense Medical College,
3-2 Namiki, Tokorozawa, Saitama 359, Japan
Tel. +81-429-95-1665; Fax +81-429-96-5209
H. Takahara
Maruha Corporation Central Research Institute,
School of Agriculture, Ibaraki University, Japan
N. Hiura
Department of Bioresource Science, School of Agriculture,
Ibaraki University, Japan
© Springer-Verlag 1999
and alginate dressing has been reported to result in accel-
erated healing of experimental burn wounds [1–3]. The
basis of these effects is thought to be the formation of a
hydrophilic gel at the wound surface through an ion ex-
change reaction between the calcium of the alginate fila-
ments and the sodium in the blood and exudate, resulting
in the production of soluble calcium-sodium alginate [4,
5]. Although they have been used as biological skin dress-
ings for delayed wound healing, their usefulness remains
to be determined.
Alginate polysaccharides are extremely insoluble and
it is therefore hard to evaluate their effects on skin metab-
olism. A homogeneous suspension of insoluble alginate
has been shown to stimulate the proliferation of normal
human skin fibroblasts [6]. Alginate oligosaccharides are
a soluble from derived from the digestion of the polysac-
charides by alginate lyase. Alginate oligosaccharides have
been shown to be a potential stimulator of human kera-
tinocyte proliferation. They stimulated keratinocyte pro-
liferation in the presence of epidermal growth factor
(EGF) more than in the absence of EGF, suggesting that
their stimulating effects may be mediated by interaction
with EGF [7] We studied the effects of alginate oligosac-
charides on the morphology and proliferation cultured
skin fibroblasts and on their production of collagens.
Materials and methods
Materials
[2,3-3H]Proline (2.1 TBq/mmol) and [32P]dCTP (110 TBq/mmol)
were purchased from Amersham. A Multiprime DNA labeling kit
was from Amersham, and Dulbecco’s modified Eagle’ medium
(DMEM) and fetal bovine serum (FBS) were from Gibco. Chro-
matographically purified bacterial collagenase was from Advance
Biofactures Company, pepsin from Cooper Biomedical, and nitro-
cellulose filters from Schleicher & Schuell. Sodium alginate was
purchased from Wako Pure Chemical Co. (Osaka, Japan).
Preparation of alginate oligosaccharides
Sodium alginate was digested at 40 °C for 6 h with alginate lyase
prepared from the broth of Alteromonas sp. in 0.05 M sodium
 433
phosphate buffer, pH 7.5. After the reaction mixture had been fil-
tered, sodium and phosphate ions were removed using an electro- Results
dialyzer (Micro Acilizer, Asahi Chemical Industries Co.). The
mixture was then lyophilized and used for the experiments [7]. Effect of alginate on cell shape

Fibroblast culture Treatment of fibroblasts with alginate oligosaccharides

Fibroblasts were explanted from skin biopsy samples. Cells grown
in 35-mm Petri dishes in DMEM containing 10 FBS were rou-
tinely subcultured every 8 days. Cells in passages 2 and 3 were
used.
Metabolic labeling, collagen synthesis and pepsin treatment
(10 mg/ml) for 4 days resulted in the cells changing shape
with the formation of elongated dendrites. This change
was not caused by the cell toxicity of alginate, since de-
pletion of alginate from the culture medium restored the
initial cell morphology (not shown).

Confluent cultures were incubated for 72 h with various concen- Effect of alginate on cell proliferation
trations of alginate oligosaccharides in DMEM supplemented with
1% FBS. During the last 24 h of incubation, cells were labeled The kinetics of fibroblast growth in the absence and the
with [2,3-3H]proline (20 µCi/ml) in the presence of ascorbic acid presence of alginate are shown in Fig. 1. Cells which had
(30 µg/ml). The medium was then collected and mixed with a pro-
tease inhibitor cocktail to yield 1 mM each on N-ethylmaleimide,
been treated with alginate on day 1 (closed arrow) ceased
EDTA (pH 7.0), and phenylmethylsulfonyl fluoride. The cells to grow during the next 4 days (5 days in culture) com-
were harvested with 0.05% trypsin, and their number was deter- pared with controls. Cells which had been treated with al-
mined using a Coulter counter. After centrifugation cells were re- ginate on day 4 (open arrow) showed significantly slower
combined with the medium and stored at –20 °C until analysis. The proliferative activity during the next 4 days (8 days in cul-
amount of radioactivity incorporated into collagen was determined
using purified bacterial collagenase [8] as previously described ture) than controls. When alginate was removed from the
[9]. Syntheses of collagen and total protein were expressed as col- culture medium on day 5 (arrowhead), the cells recovered
lagenase-sensitive and total protein counts per minute per cell, re-
spectively. The percentage collagen synthesis was calculated as
previously described [9]. For pepsin treatment, the culture medium
was harvested as described above, then the proteins were precipi-
tated with 30% saturated ammonium sulfate. Cells were removed
by scraping and extracted with 0.5 M acetic acid at 4 °C overnight
and cleared by centrifugation. Proteins of the medium and cells
were digested with pepsin (50 µg/ml) for 18 h at 4 °C. After neu-
tralization, the proteins were precipitated with 30% saturated am-
monium sulfate. Type I and III collagens were resolved by SDS-
polyacrylamide gel electrophoresis without reduction followed by
autoradiography. The percentage of type III collagen (α1(III)/α1(I)
+ α2(I) + α1(III)) was calculated with a densitometer [10].

Cell proliferation

Cultures were treated with alginate oligosaccharides for 48 h doses

of 0, 1, 10 mg/ml in DMEM supplemented with 10% FBS. At the
end of treatment, cells were counted with a Coulter counter.

Northern blot analysis

Fibroblasts at confluent density were treated with various doses of

alginate oligosaccharides for 48 h in DMEM supplemented with
1% FBS. Total RNA was isolated from the cells using a previously
described procedure [11]. RNA samples were denatured in deion-
ized 1 M glyoxal/10 mM phosphate buffer, pH 7.0, at 50 °C for 1 h,
then resolved by 1% agarose gel electrophoresis and blotted to ni-
trocellulose filters. RNA was immobilized on the filter by heating
at 80 °C for 2 h and hybridized to 32P-labeled probes in a solution
of 50% formamide, 5 × SSC and 5 × Denhardt’s solution contain-
ing 0.1% SDS and 250 µg/ml sonicated salmon sperm DNA at
42 °C for 18 h. The following cDNA probes radioactively labeled
to a specific activity of 108 cpm/µg DNA by random priming were
used: proα1(I) (Hf677) [12], proα2(I) (Hf32) [13], proα1(III)
(Hf934) [14] and proα1(VI) [15] collagens. The filters were
washed at a stringency of 0.1 × SSC at 42 °C for 1 h exposed to X-
ray film at –80 °C with an intensifying screen. The autoradiograms
were scanned with a densitometer. The abundance of mRNA was
expressed as a percentage of control cells.
Fig. 1 Effect of alginate oligosaccharides on cell proliferation.
Human skin fibroblasts were seeded at a density of 1 × 104/30-mm
diameter dish and cultured in Dullbecco’s modified Eagle’s me-
dium supplemented with 10% fetal bovine serum for 8 days. Some
cultures were treated with 10 mg/ml alginate on day 1 (closed ar-
row) or day 4 (open arrow) for 4 days. At the and of the treatment
with alginate which started on day 1 for 4 days, the medium was
replaced with fresh medium devoid of alginate (arrowhead), and
the culture continued for another 4 days (dotted line). At the end of
the treatment, cultures were harvested by trypsinization, and the
number of cells was determined. Values are mean ± SD calculated
from two sets of assays performed in triplicate
434

Fig. 2 Dose-response relationship for the inhibition of cell prolif-

eration by alginate. Human skin fibroblasts were seeded at a den-
sity of 1 × 104/30-mm diameter dish and cultured in Dulbecco’s
modified Eagle’s medium supplemented with 10% fetal bovine
serum. Cultures were treated with various concentrations of algi-
nate on day1 in culture for 2 day (closed circles) or 4 days (open
circles). At the end of the culture, cultures were harvested by
trypsinization, and the number of cells was determined. Values are
mean ± SD from two sets of assays performed in triplicate
Fig. 3 Northern blot assays of the expression of matrix-related
Table 1 Effect of alginate oligosaccharides on collagen synthesis
proteins. Fibroblasts at confluent density in culture were treated
in skin fibroblasts in culture. Cultures at confluent density were
with various doses of alginate oligosaccharides for 48 h. RNA was
treated with alginate oligosaccharides for 48 h at doses of 0, 0.1, 1
isolated from the cells and fractionated by agarose 1% gel elec-
and 10 mg/ml in DMEM containing 1% fetal bovine serum, then
trophoresis. RNA was blotted onto filters and hybridized with 32P-
labeled with [3H]proline for the last 18 h of the treatment. Collagen
labeled α1(I), α2(I), α1(III) and α1(VI) probes. The filters were
and noncollagen protein production was determined from 3H in-
subjected to autoradiography. Total RNA was stained with ethid-
corporation in collagenase-sensitive and -resistant proteins. The
ium bromide (lowest panel). Two major RNA species of 5.8 and
data are presented as percentage collagen synthesized taking into
4.8 kb in α1(I), 4.9 and 4.6 kb in α2(I) and 5.4 and 4.8 kb in α1(III)
consideration the frequency of occurrence of proline in collagen
mRNA are apparent (a). Autoradiograms were scanned with a den-
relative to that in noncollagen proteins [8]. Values are means ± de-
sitometer. Levels of mRNA were determined by estimating control
viation from two independent assays performed in duplicate.
values (without treatment) as 100%. Values are mean ± deviation
Alginate oligosaccharides Collagen from two sets of experiments (b)
(mg/ml) (%)
0 12.2 ± 0.7 in cell shape or cell number (not shown). Under cell growth-
0.1 12.1 ± 1.0 arrested conditions, alginate inhibited collagen synthesis
1 8.7 ± 0.7 at a dose of 10 mg/ml (Table 1). There was no significant
10 7.4 ± 0.8 difference between control and alginate-treated cells in
the ratio of type I collagen to type III collagen estimated
after pepsin treatment (type III/type I + type III 0.12, 0.11
their initial growth potential (Fig. 1). The inhibition of cell and 0.14 in control cells and cells treated with 1 mg/ml
proliferation was sharply dependent on alginate concen- and 10 mg/ml alginate, respectively). We measured mRNA
tration from 0.01 to 10 mg/ml (Fig. 2). levels of type I, III and VI collagens because they are ma-
jor matrix proteins in the dermal connective tissue. North-
Effect of alginate on collagen expression ern blot analysis revealed that alginate considerably re-
duced the mRNA levels α1(I), α2(I) and α1(II) at doses of
The cells at their confluent density were treated with algi- 1 and 10 mg/ml, and to a lesser extent reduced the mRNA
nate. Under these conditions, alginate did not cause changes level of α1(VI). Quantitative determination showed that

alginate reduced the mRNA levels of α1(I) and α1(III) to
one-tenth of the levels found in the control experiments at
a dose of 10 mg/ml (Fig. 3a, b).
Discussion
Various types of biological dressing are available which
provide a moist environment for wound healing. Alginate
polysaccharide sheets have been widely used as skin
dressings and have been shown to be nontoxic, nonaller-
genic and completely biodegradable without tissue reac-
tion [16, 17]. However, biochemical in vitro studies on the
mechanism of action of alginate dressings in the wound
healing process have not been carried out. The wound heal-
ing process is a complex process involving cellular migra-
tion, proliferation and differentiation with extracellular
matrix synthesis and remodeling. In the early phases of
wound healing fibroblasts migrate into the region of the
wound interface where they synthesize and secrete the
new granulation tissue extracellular matrix. In the late
phases of wound healing, the cells become stellate and
regress [18]. We demonstated that alginate oligosaccharides
caused inhibition of fibroblast growth within 5 days in a
manner dependent on the concentration from 0.01 to 10 mg/
ml, the latter resulting in almost complete cessation of cell
proliferation. This effect was not accompanied by cyto-
toxicity on the basis of the criteria of dye exclusion and
growth recovery after its depletion. The inhibiting effect
on cell growth was accompanied by a change ion cell
shape to stellate. Alginate also inhibited the expressions
of collagen in the cultured skin fibroblasts at doses of 1 and
10 mg/ml. These inhibiting effects on fibroblasts growth
and collagen expression as well as the change in cell
shape to stellate suggest that alginate oligosaccarides may
provide a useful tool for the suppression of hypergranula-
tion tissues which is caused by an abnormal response of
the fibroblasts during wound repair.
It has been demonstrated that alginate stimulates kera-
tinocyte proliferation by functioning as a cofactor for
EGF. This implies that these oligosaccharides posses he-
paran sulfate- or heparin-like activity (7), since heparan
sulfate and heparin are known to have high affinity for
EGF and basic fibroblast growth factor and to enhance
their physiological function [19, 20]. Since rapid epithe-
lialization is also essential for wound healing, the stimu-
lating effect of alginate on keratinocyte proliferation may
serve as a useful therapeutic tool for wound healing.
The mechanism by which alginate oligosaccharides
change the shape of cells is unclear. Since cultured cells
have been reported to undergo matrix-dependent changes
in shape in collagen-gel matrix, tissue equivalent matrix
and matrix-coated dishes [21, 22], the relatively high
doses of alginate oligosaccharides of 1 and 10 mg/ml in
the culture medium may have directly interacted with the
cell surface resulting in change in shape.
The effect of calcium alginate (insoluble alginate poly-
saccharide) on fibroblast proliferation has been examined
in vitro studies. The proliferation of mouse L929 fibro-
435
blasts is initially inhibited up to 25% by calcium alginate,
but recovers to control levels within 5 days of treatment
[23, 24]. In contrast, the proliferation of normal human
skin fibroblasts is stimulated beyond 5 days of treatment
by low concentrations of calcium alginate [6]. The results
of the latter study appear to be in disagreement with our
results. This difference may be due to the form of alginate
used. In the previous study calcium alginate was homoge-
nized in the culture medium, whereas in our experiments
liquid alginate oligosaccharides were used.
References
1. Blaine G (1947) Experimental observations on absorbable algi-
nate products in surgery. Ann Surg 125 : 102–114
2. Barnett SE, Varley SJ (1987) The effects of calcium alginate
on wound healing. Ann Roy Coll Surg 69 : 153–155
3. Attwood AI (1989) Calcium alginate dressing accelerates split
skin graft donor site healing. Br J Plast Surg 42 : 373–379
4. Oliver LC, Blaine G (1949) Homeostatis with absorbable algi-
nates in neurosurgical practice. Br J Surg 37 : 307–311
5. Barrett SE, Varkey SJ (1987) The effects of calcium alginate
on wound healing. Ann R Coll Surg Engl 69 : 153–155
6. Doyle JW, Roth TP, Smith RM, Li Y-Q, Dunn RM (1996) Ef-
fect of calcium alginate on cellular wound healing processes
modeled in vitro. J Biomed Mater Res 32 : 561–568
7. Kawada A, Hiura N, Shiraiwa M, Tajima S, Hiruma M, Hara
K, Ishibashi A, Takahara H (1997) Stimulation of human kera-
tinocyte growth by alginate oligosaccharides, a possible cofac-
tor for epidermal growth factor in cell culture. FEBS Lett 408 :
43–46
8. Phillip CL, Tajima S, Pinnell SR (1992) Ascorbic acid and
transforming growth factor-β1 increase collagen biosynthesis
via different mechanisms: coordinate regulation of proα1(I)
and proα1(III) collagens. Arch Biochem Biophys 295 : 397–403
9. Peterkofsky B, Diegelman R (1971) Use of a mixture of pro-
teinase-free collagenase for the specific assay of radioactive
collagen in the presence of other proteins. Biochemistry 10 :
988–994
10. Tajima S, Pinnell SR (1981) Collagen synthesis by human skin
fibroblasts in culture; studies of fibroblasts explanted from
papillary and reticular dermis. J Invet Dermatol 77 : 410–412
11. Chomczynski P, Sacchi N (1987) Single step method of RNA
isolation by acid guanidinium thiocyanate-phenol-chloroform
extraction. Anal Biochem 162 : 156–159
12. Bernard MP, Chu M-L, Myers JC, Ramirez F, Eikenberry EF,
Prockop DJ (1983) Nucleotide sequences of complementary
deoxyribonucleic acids for the proα1 chain of human type I
procollagen: statistical evaluation of structures that are con-
served during evolution. Biochemistry 22 : 5213–5223
13. Bernard MP, Myers JC, Chu M-L, Ramirez F, Eikenberry EF,
Prockop DJ (1983) Structure of a cDNA for the proα2 chain of
human type I procollagen. Comparison with chick cDNA for
proα2(I) identifies structurally conserved features of the protein
and the gene. Biochemistry 22 : 1139–1145
14. Chu M-L, Weil D, de Wet W, Bernard M, Sippola M, Ramirez
F (1985) Isolation of cDNA and genomic clones encoding hu-
man proα1 (III) collagen. J Biol Chem 260 : 4357–4363
15. Chu M-L, Conway D, Pan T, Baldwin C, Mann K, Deutzmann
R, Timpl R (1988) Amino acid sequence of the triple-helical
domain of human collagen type VI. J Biol Chem 263 : 18601–
18606
16. Attwood AI (1989) Calcium alginate dressing accelerates split
skin graft donor site healing. Br J Plast Surg 43 : 373–379
17. Barrett SE, Varley SJ (1987) The effects of calcium alginate on
wound healing. Ann R Coll Surg 69 : 153–155
436
18. Grinnell F, Nakagawa S (1991) Spatial regulation of fibroblast
proliferation. An explanation for cell regression at the end of
the wound repair. In: Barbul A, Caldwell MD, Eaglestein WH,
Hunt TK, Marshall D, Pines E, Skover G (eds) Clinical and ex-
perimental approaches to dermal and epidermal repair. Wiley-
Liss, New York, pp 155–166
19. Higashiyama S, Abraham JA, Miller J, Fiddes JC, Klagsbrun
M (1991) A heparin-binding growth factor secreted by macro-
phage-like cells that is related to EGF. Science 251 : 936–939
20. Kjellén L, Lindahl U (1991) Proteoglycans; structures and in-
teractions. Annu Rev Biochem 60 : 443–475
21. Nusgens B, Merrill C, Lapiere C, Bell E (1984) Collagen
biosynthesis by cells in a tissue equivalent matrix in vitro. Col-
lagen Relat Res 4 : 351–364
22. Hay ED, Svoboda KK (1989) Extracellular matrix interaction
with the cytoskeleton. In: Stein WD, Bronner F (eds) Cell
shape. Academic Press, London New York, pp 147–172
23. Schmitz RJ, Chung LY, Andrews AM, Spyratou O, Turner TD
(1993) Biocompatibility of wound management products:
a study of the effects of various polysaccharides on murine
L929 fibroblast proliferation and macrophage respiratory burst.
J Pharm Pharmacol 45 : 508–513
24. Rosdy M, Clauss L-C (1990) Cytotoxicity testing of wound
dressing using normal human keratinocytes in culture. J Bio-
med Mater Res 24 : 363–377