T. Matsunami et al.128
occurred in pancreatic β-cells under HBO treat-
ment. Importantly, we used pressure levels and
duration of exposure to HBO typical of the clini-
cal setting, in order to evaluate the toxic effects of
HBO in diabetes.
MATERIALS AND METHODS
Animals and the usage policy
A total of 24 adult, 10-week-old, male Wistar rats
(weight range, 280 to 320 g), bred in our laboratory, were
used for the experiment. They were housed at 23 to
25°C with light from 7:00 AM to 7:00 PM and free
access to water at all times. All animals were fed a com-
mercial diet during the experiment. All study procedures
were implemented in accordance with the Institutional
Guidelines for Animal Experiments at the College of
Bioresource Sciences, Nihon University and with the
permission of the Committee of Experimental Animal in
The rats (n = 24) were allowed to acclimatize for 1
week, prior to treatment. They were randomly distribut-
ed into four groups, 6 rats per group, as follows: no dia-
betic induction and no HBO treatment (Control); HBO
treatment without diabetic induction (HBO group); dia-
betic induction without HBO treatment (DM: Diabetes
Mellitus group); and diabetic induction and HBO treat-
ment (DM + HBO group). HBO treatment was started 5
days after the diabetic induction. The mean body weight
of the animals in all groups was measured at the initial
and final visits during the study period.
Induction of diabetes
Wistar rats were injected intraperitoneally (i.p.) with
40mg/kg streptozotocin (STZ, Wako Pure Chemical
Industries, Ltd. Japan), dissolved in citrate buffer (pH
4.5), to induce diabetes as described previously (Like et
al. 1976). Control and HBO rats were treated with equal
volumes and concentrations of the STZ injection vehicle,
citrate buffer. The blood glucose levels in the auricular
veins were measured 5, 8 and 12 days after the STZ in-
jection, by an automated glucose measurement equip-
ment, G meter®
(Sanofi-Aventis K.K. Japan). Diabetes
was considered to have been induced when the glucose
level reached at least 280 mg/dl (Matkovics et al. 1997).
HBO exposure was set at a pressure of 2.8 ATA for
oxygen toxicity (Harabin et al. 1990; Chavko et
al. 1996; Speit et al. 2002). Reported side effects
have often been associated with exposure to
levels of HBO much higher than those generally
used for treating clinical conditions. The clinical-
ly approved maximum pressure and duration of
HBO exposure are 3 ATA and 120 min, respec-
tively (Feldmeier 2003), although the most com-
monly-used protocol for standard therapeutic
purposes is slightly lower (1.8 to 2.8 ATA for 60
to 90 min) (Benedetti et al. 2004). The oxidative
effects of HBO have been investigated in animals
and humans (Oter et al. 2005; Eken et al. 2005).
One study on the effects of HBO in rats revealed
that, following 2 h of HBO exposure at 3 ATA,
elevated levels of oxidative stress markers, thio-
barbituric acid reactive substances (TBARS) and
superoxide dismutase (SOD) were found in the
lung, brain and erythrocytes (Oter et al. 2005).
However, these authors used pressure conditions
that were higher than the standard clinical treat-
ment. Therefore, it appears that adverse side
effects of HBO exposure have occurred under
conditions of overdose, at least as far as the stan-
dard clinical levels of exposure are concerned.
Therefore, in order to evaluate the safety of HBO
exposure, it is necessary to use conditions (in
terms of pressure and duration) similar to those
used in the clinical setting.
Although HBO has been used to treat several
medical conditions, its potential side effects
should not be ignored. As mentioned above,
HBO therapy has been used to facilitate the repair
of diabetic lower limb wounds (Tibbles et al.
1996). However, diabetes is a condition that
induces oxidative stress (Oberley 1996; Brownlee
et al. 2001), and further exposure to HBO may
only serve to exacerbate this stress, and thereby
accelerate the progress of the illness. The side
effects of HBO therapy in diabetic patients have
been poorly investigated. Moreover, there is an
insufficient number of adequate models available
with which to evaluate the toxic effects of HBO
associated with this stress. In this study, using an
animal model of diabetes, we examined the physi-
ological parameters associated with glucose
homeostasis and the histological changes that
Enhancement of Glucose Toxicity by HBO 129
2 h once daily, for 7 days for all experiments; this proto-
col reflects that used in the clinical setting. A hyperbaric
chamber for small animals (Nakamura Tekko-sho K.K.,
Tokyo, Japan) was used for HBO exposure. The ventila-
tion rate was 4-5 L/min. All administrations were started
at the same hour in the morning (10 AM) to exclude any
confounding issues associated with biological rhythm
After the final exposure to HBO, the animals were
weighed and anesthetized with sodium pentobarbital 40
mg/kg i.p. Pancreatic tissues were harvested from the
sacrificed animals, and the fragments from tissues were
fixed in 10% neutral formalin solution, embedded in par-
affin and then stained with immunostaining for insulin,
according to the method of Coskun et al. (2005) for his-
The areas of insulin immunoreactive β-cells in each
group of rats were compared using images obtained
using the Image J software version 1.8 system (Wayne
Rasband National Institutes of Health, USA). Forty
Langerhans islets were chosen at random from each
group of individual rats. The percentage of insulin
immunoreactiveβ-cells was calculated for each group.
The collected and calculated data were expressed as
mean ± standard deviation (S.D.). Student’s t-test was
used to determine significant differences between the
control and hyperbaric groups. For the analysis of the
immunohistochemical data, a nonparametric test
(Kruskal-Wallis) was used. Differences were considered
statistically significant if p < 0.05.
The baseline body weight of the rats at
beginning of the study was similar and in all
groups. At the end of the treatment (after 12
days), there was no difference in body weight
between control and HBO rats (mean group
weights ranged from 307 to 313 g). However dia-
betic rats presented with weight loss, and the body
weight of DM + HBO rats (262 ± 10 g; mean ±
S.D.), in particular, decreased significantly (P <
0.01), when compared with the rats in the other 3
groups (control, 313 ± 5 g; HBO, 307 ± 4 g; DM,
280 ± 9 g).
The blood glucose levels of the DM + HBO
group were significantly higher (P < 0.05) 8 and
12 days after diabetes induction, compared with
the DM group (Fig. 1). No significant differences
were observed between any other groups or
The pancreatic islet cells were histologically
normal in the control and HBO groups. Strongly
stained, anti-insulin positive cells were observed
in the islets of the pancreatic tissues from the non-
diabetic rats in these control and HBO groups (Fig.
2). In contrast, in the DM and the DM + HBO
groups, insulin antigen-positive signals were
weak in the majority of the islets (Fig. 3). Insulin
Fig. 1. Glucose levels in non-diabetic Wistar (WR) and diabetic WR rats. Values are expressed as the
mean ± standard deviation (n = 6).
*P < 0.01, **P < 0.001 compared with control value.
T. Matsunami et al.130
immunoreactive β-cells were measured in the
islets of Langerhans. The percentage of insulin
immunoreactive β-cells is shown in Table 1. The
area of insulin immunoreactive β-cells in the DM
and DM + HBO groups decreased significantly
(P < 0.001), when compared with the HBO group.
Furthermore, HBO treatment was associated with
a significant decrease in insulin immunoreactive
β-cells (P < 0.001; DM + HBO vs. DM).
HBO therapy is a unique method used in the
Fig. 2. Non-diabetic rats in the non-HBO (control) and hyperbaric (HBO) groups, showing normal cells
in the islets of Langerhans and also showing β-cells in the islets of Langerhans that are strongly
stained with the anti-insulin antibody. Immunoperoxidase, haematoxylin counterstain × 320. Scale
bar = 50μm.
Fig. 3. Diabetic rats in the non-HBO (DM) and hyperbaric (DM + HBO) groups, showing β-cells in the
islets of Langerhans with weak insulin-immunoreactivity in the majority ofβ-cells. Immunoperoxi-
dase, haematoxylin counterstain × 320. Scale bar = 50μm.
TABLE 1. Comparison of the areas with the insulin immunoreactive β-cells in the islets of
Group Mean area of insulin immunoreactiveβ-cells in islets
Control 81.32 ± 6.66%
HBO 80.91 ± 8.65%
DM 18.33 ± 3.29%*
DM + HBO 8.01 ± 3.31%**
Non-diabetic, non-HBO rats (control) and hyperbaric (HBO) groups; diabetic,
non-HBO rats (DM) and hyperbaric (DM + HBO) groups.
Kruskal-Wallis test was used for statistical analysis. Values are expressed as mean ±
S.D., and n = 6 animals for all groups.
*P < 0.001; DM vs. HBO. **P < 0.001; DM + HBO vs. DM.
Note that HBO treatment was associated with a significant decrease in insulin
immunoreactiveβ-cells in DM rats.
Enhancement of Glucose Toxicity by HBO 131
treatment of a variety of illnesses and conditions,
such as carbon monoxide poisoning, decompres-
sion sickness, osteomyelitis, diabetic foot, and
impaired wound healing (Feldmeier 2003). The
Undersea and Hyperbaric Medical Society
(UHMS) has an approved list of clinical condi-
tions where HBO treatment is indicated, includ-
ing, amongst others, the above mentioned disor-
ders (Feldmeier 2003). Nevertheless, this should
not be considered as exhaustive, as there is an
abundance of reports which suggest many suc-
cessful cures obtained with HBO treatment for
conditions other than those on the UHMS list.
Although HBO treatment seems to be a ‘cure-all’
therapy, nevertheless, potential and known side
effects should be studied further to avoid the
excessive use of HBO in the clinical setting.
The causes of the observed side effects fall
into two categories: firstly, those due to the poten-
tial risk of oxygen toxicity itself during the treat-
ment period; secondly, those due to the use of a
pressure and duration of HBO administration that
exceeds the approved therapeutic limits (currently
set at, a maximum 3 ATA pressure and 2 h dura-
tion). Oxygen toxicity, manifested for example
by a slight increase in SOD and reduced glutathi-
one peroxidase (GPx) and catalase (CAT) activity,
has been reported in both the lung and brain of
rats and guinea pigs, in situations where although
the HBO exposure pressures were clinically
acceptable, the duration of the session was rather
longer than the recommended 2 hours (Harabin et
al. 1990). Direct toxic effects of HBO have been
observed under conditions with relatively higher
pressures and longer durations, such as 4 to 5 ATA
and more than 2 hours, respectively (Harabin et
al. 1990). In the present study, we used an HBO
exposure protocol which followed clinically used
pressures and duration, and investigated the levels
of blood glucose, and histological changes in the
pancreatic β-cells of STZ-induced diabetic rats, a
recognized model of type 1 diabetes mellitus
(Kakkar et al. 1995). Our results showed that
glucose levels were not improved in HBO-treated
diabetic rats, suggesting that HBO administration
might have the ability to enhance the functional
disorder of pancreatic islet cells. A recent study
has reported that HBO exposure at a pressure of
2.8 ATA for 2 h once daily for 7 days elevated
hemorheological data and the TBARS levels of
erythrocyte membranes in diabetic rats (Liu et al.
2003). These authors showed significant changes
of the hemorheological parameters in diabetic rats
associated with the administration of HBO, how-
ever, they did not study the duration of the chang-
es of blood glucose levels or perform any histo-
logical analysis. We have shown for the first time
in STZ-diabetic rats, that the rise in blood glucose
levels started after the beginning of HBO treat-
ment, administrated at pressure levels and for a
duration of time used in the clinical setting.
Furthermore, this present study also showed that
the insulin-producingβ-cells areas of diabetic rats
receiving HBO treatment were significantly
reduced compared with those of the diabetic con-
trol group (i.e., no HBO treatment). STZ, a
mono-functional nitrosourea derivative, is one of
the most commonly used substances to induce
diabetes in experimental animals (Evans et al.
1965; Szkudelski 2001). Previous evidence sug-
gests that STZ may damage pancreaticβ-cells and
could induce oxidative stress (Ohkuwa et al.
On the other hand, it has been shown that
blood glucose levels were decreased by
HBO treatment in GK rats, an animal model of
spontaneous type 2 diabetes, suggesting that HBO
could interfere with the activity of the natural
anti-oxidative defense system and thus promote
de novo generation of free radicals (Ihara et al.
1999; Maritim et al. 2003). Our results suggest
that the mode of action of HBO against the glu-
cose control system might be different in STZ-
and GK-induced diabetic rats. The possible side
effects of HBO seen in this study, namely the
elevation of glucose levels in the peripheral blood
and the damage to β-cells, might occur as a result
of the enhancement of STZ cytotoxicity by HBO
administration. One factor behind the enhance-
ment of glucose toxicity might be oxidative stress,
as described previously (Ihara et al. 1999). Our
experimental system, which uses STZ-induced
diabetes plus HBO treatment, could be used as a
model for evaluating oxidative stress, through the
T. Matsunami et al.132
monitoring of unusual increases in glucose levels.
Nevertheless, in cases where cytotoxicity occurs
during HBO administration, due to factors other
than STZ, it will also be necessary to investigate
the influence of HBO. For the safe employment
of HBO for therapeutic purpose, it will be impor-
tant to fully understand the nature of the toxic
effects of HBO with respect to oxidative stress.
This study was partially supported by the
Academic Frontier Project “Surveillance and control
for zoonoses” from Ministry of Education, Culture,
Sports, Science and Technology, Japan.
Benedetti, S., Lamorgese, A., Piersantelli, M., Pagliarani, S.,
Benvenuti, F. & Canestrari, F. (2004) Oxidative stress and
antioxidant status in patients undergoing prolonged expo-
sure to hyperbaric oxygen. Clin. Biochem., 37, 312-317.
Brownlee, M. (2001) Biochemistry and molecular cell biology
of diabetic complications. Nature, 414, 813-820.
Chavko, M. & Harabin, A.L. (1996) Regional lipid peroxida-
tion and protein oxidation in rat brain after hyperbaric oxy-
gen exposure. Free Radic. Biol. Med., 20, 973-978.
Coskun, O., Kanter, M., Korkmaz, A. & Oter, S. (2005) Quer-
cetin, a flavonoid antioxidant, prevents and protects strep-
tozotocin-induced oxidative stress and beta-cell damage in
rat pancreas. Pharmacol. Res., 51, 117-123.
Eken, A., Aydin, A., Sayal, A., Ustündağ, A., Duydu, Y. &
Dündar, K. (2005) The effects of hyperbaric oxygen treat-
ment on oxidative stress and SCE frequencies in humans.
Clin. Biochem., 38, 1133-1137.
Evans, J.S., Gerritsen, G.C., Mann, K.M. & Owen, S.P. (1965)
Antitumor and hyperglycemic activity of streptozotocin
(NSC-37917) and its cofactor, U-15,774. Cancer
Chemother. Rep., 48, 1-6.
Feldmeier, J.J. (2003) Hyperbaric oxygen: indications and
results; the hyperbaric oxygen therapy committee report.
Undersea. Hyperb. Med. Soc.
Harabin, A.L., Braisted, J.C. & Flynn, E.T. (1990) Response of
antioxidant enzymes to intermittent and continuous hyper-
baric oxygen. J. Appl. Physiol., 69, 328-335.
Ihara, Y., Toyokuni, S., Uchida, K., Odaka, H., Tanaka, T.,
Ikeda, H., Hiai, H., Seino, Y. & Yamada, Y. (1999) Hyper-
glycemia causes oxidative stress in pancreatic beta-cells of
GK rats, a model of type 2 diabetes. Diabetes, 48,
Kakkar, R., Kalra, J., Mantha, S.V. & Prasad, K. (1995) Lipid
peroxidation and activity of antioxidant enzymes in diabet-
ic rats. Mol. Cell. Biochem., 151, 113-119.
Like, A.A. & Rossini, A.A. (1976) Streptozotocin-induced pan-
creatic insulitis: new model of diabetes mellitus. Science.,
Lipsky, B.A., Berendt, A.R., Deery, H.G., Embil, J.M., Joseph,
W.S., Karchmer, A.W., LeFrock, J.L., Lew, D.P., Mader,
J.T., Norden, C. & Tan, J.S. (2004) Diagnosis and treat-
ment of diabetic foot infections. Clin. Infect. Dis., 39,
Liu, D.Z., Chien, S.C., Tseng, L.P. & Yang, C.B. (2003) The
influence of hyperbaric oxygen on hemorheological param-
eters in diabetic rats. Biorheol., 40, 605-612.
Maritim, A.C., Sanders, R.A. & Watkins, J.B. 3rd. (2003)
Effects of alpha-lipoic acid on biomarkers of oxidative
stress in streptozotocin-induced diabetic rats. J. Nutr.
Biochem., 14, 288-294.
Matkovics, B., Kotorman, M., Varga, I.S., Hai, D.Q. & Varga, C.
(1997) Oxidative stress in experimental diabetes induced
by streptozotocin. Acta Physiol. Hung., 85, 29-38.
Oberley, L.W. (1996) Free radicals and diabetes. Free Radic.
Biol. Med., 5, 113-124.
Ohkuwa, T., Sato, Y. & Naoi, M. (1995) Hydroxyl radical for-
mation in diabetic rats induced by streptozotocin. Life Sci.,
Oter, S., Korkmaz, A., Topal, T., Ozcan, O., Sadir, S., Ozler, M.,
Ogur, R. & Bilgic, H. (2005) Correlation between hyper-
baric oxygen exposure pressures and oxidative parameters
in rat lung, brain, and erythrocytes. Clin. Biochem., 38,
Speit, G., Dennog, C., Radermacher, P. & Rothfuss, A. (2002)
Genotoxicity of hyperbaric oxygen. Mutat. Res., 512,
Szkudelski, T. (2001) The mechanism of alloxan and strepto-
zotocin action in β cell of the rat pancreas. Physiol. Res.,
Tibbles, P.M. & Edelsberg, J.S. (1996) Hyperbaric-oxygen
therapy. N. Engl. J. Med., 334, 1642-1648.