BASIC RESEARCH www.jasn.

org

Minocycline Protects against Neurologic Complications

of Rapid Correction of Hyponatremia
Fabrice Gankam-Kengne,* Alain Soupart,* Roland Pochet, Jean Pierre Brion,
and Guy Decaux*
*Research Unit on Hydromineral Metabolism and Department of General Internal Medicine, Erasme University
Hospital, Brussels, Belgium; Laboratory of Histology, Histopathology, and Neuroanatomy, Faculty of Medicine,
Universite Libre de Bruxelles, Brussels, Belgium; Department of Internal Medicine, University of Texas
Southwestern Medical Center, Dallas, Texas; and Tubize Hospital, Tubize, Belgium

ABSTRACT
Osmotic demyelination syndrome is a devastating neurologic condition that occurs after rapid
correction of serum sodium in patients with hyponatremia. Pathologic features of this injury include
a well-demarcated region of myelin loss, a breakdown of the blood brain barrier, and infiltration of
microglia. The semisynthetic tetracycline minocycline is protective in some animal models of central
nervous system injury, including demyelination, suggesting that it may also protect against demy-
elination resulting from rapid correction of chronic hyponatremia. Using a rat model of osmotic
demyelination syndrome, we found that treatment with minocycline significantly decreases brain
demyelination, alleviates neurologic manifestations, and reduces mortality associated with rapid
correction of hyponatremia. Mechanistically, minocycline decreased the permeability of the blood
brain barrier, inhibited microglial activation, decreased both the expression of IL1 and protein
nitrosylation, and reduced the loss of GFAP immunoreactivity. In conclusion, minocycline modifies
the course of osmotic demyelination in rats, suggesting its possible therapeutic use in the setting of
inadvertent rapid correction of chronic hyponatremia in humans.

J Am Soc Nephrol 21: 2099 2108, 2010. doi: 10.1681/ASN.2010050467

Osmotic demyelination syndrome (ODS) is a se-
vere neurologic condition that is characterized
by severe demyelination in the central nervous
system (CNS) secondary to osmotic imbalance.
In a clinical setting, this syndrome often occurs
after too rapid correction of chronic hyponatr-
emia.15
In ODS, demyelination is widespread in the
brain, with predominance in hippocampus, basal
ganglia, and subcortical regions. The physiopathol-
ogy of this disorder is not yet fully understood, and
an experimental murine model has been developed
to better understand the mechanisms leading to
myelin damage after an osmotic injury.2,5,6 Previ-
ous experiments have suggested that ODS might
share key characteristics with other models of cen-
tral nervous system demyelination in which both
crogliamacrophage activation are involved in the
genesis of demyelinative changes.711
Minocycline is a second-generation tetracycline
that has been well studied in various models of brain
pathology including autoimmune or ischemic myelin
damage, and others have reported that administration
of minocycline in CNS injury was associated with a
striking reduction in BBB permeability, inhibition of
microgliamacrophage activation, and inflammatory
Received May 7, 2010. Accepted September 14, 2010.
Published online ahead of print. Publication date available at
www.jasn.org.
Correspondence: Dr. Guy Decaux or Dr. Gankam-Kengne, Service
de Medecine Interne Generale, Hopital Erasme, 808 Route de Len-
nik, 1070 Bruxelles, Belgium. Phone: 32-2-555-6286; Fax: 32-2-555-
6285; E-mail: guy.decaux@skynet.be or fgankamk@ulb.ac.be

opening of the blood brain barrier (BBB) and mi- Copyright 2010 by the American Society of Nephrology

J Am Soc Nephrol 21: 20992108, 2010 ISSN : 1046-6673/2112-2099 2099

 BASIC RESEARCH www.jasn.org

cytokines secretion, with subsequent benefits in mortality and treated with hypertonic saline alone and animals treated
functional outcome.1217 In this work, we aimed to investigate with minocycline early and 33 mEq/L for animals treated
whether minocycline has a protective effect in an animal model of with delayed minocycline; P not significant for SNa in
ODS. group 1 versus group 2 and group 3; P 0.001 for SNa day
We report that, in a rat model, minocycline decreases mortal- 4 versus SNa day 5).
ity and alleviates neurologic manifestations after rapid correction
of chronic hyponatremia in rats. The action of minocycline was Improved Functional Outcome after Minocycline
associated with restoration of normal BBB permeability, decrease Treatment in ODS
in microgliamacrophage activation, reduction in glial fibrillary After rapid correction of serum sodium, animals treated with hy-
acidic protein (GFAP) immunoreactivity loss, and decreased ex- pertonic saline alone exhibited severe neurologic manifestations
pression of proinflammatory cytokines, along with significant at- including seizures, paralysis, and coma. In contrast, animals
tenuation of demyelination lesions. treated with early minocycline only showed mild neurologic man-
ifestations. The incidence and severity of these neurologic mani-
festations was significantly lower after treatment with minocy-
RESULTS cline throughout the course of the disease (Figure 2A). Body
weight loss, which can be used as a marker of CNS pathology after
Serum Sodium Values before and after Correction of induction of ODS,5 was significantly greater in animals corrected
Hyponatremia with hypertonic saline alone compared with animals treated with
We successfully used a previously described model (Figure 1)5,6,18 to hypertonic saline and minocycline. (33 2% in group 1
induce chronic severe hyponatremia in all rats. In these ex- versus 17 7% in group 2; P 0.014; Figure 2B).
periments, only rats who had severe hyponatremia (SNa
120 mEq/L) were included in the analysis. Table 1 shows the Minocycline Reduces Death Associated with Rapid
SNa value in the different groups before and after the cor- Correction of Chronic Hyponatremia
rection for each set of experiments. SNa levels at day 4 were Table 1 shows mortality in animals treated with hypertonic
comparable in the different groups studied for all experi- saline only and with minocycline before and after hyper-
ments. Administration of hypertonic saline resulted in a tonic saline. Eighty-four percent (26/31) of the animals
significant and comparable rise of SNa in all groups (mean treated with hypertonic saline alone died 6 days after the
24-hour gradient of serum sodium of 31 mEq/L for animals correction, whereas after the same amount of time, only
48% (13/27) of animals treated with hy-
Day 0 pertonic saline and minocycline before
Day 4
Induction of hypoNa
Correction of hypoNa and during the correction of hyponatre-
Induction of ODS. mia died (P 0.0053). Survival analysis
showed that treatment with minocycline
Group 1 was associated with significant reduction
(n = 31) in mortality.
Vehicle
Delayed Administration of
Group 2 Minocycline Still Provides Protection
(n = 27) in ODS
Minocycline starting 12 hrs
In the clinical setting, rapid correction of hy-
before ODS induction ponatremia is generally inadvertent, the toxic
Group 3
gradient of serum sodium is achieved within
(n = 13) 12 to 24 hours, and the symptoms are usually
delayed for 48 to 72 hours after hyponatre-
Minocycline starting 18 hrs
after ODS induction mia is corrected. Therefore, we sought to
determine whether minocycline admin-
Figure 1. Schematic representation of the design used. Hyponatremia was induced by istered 18 hours after the toxic gradient
1-deamino-[8-D-arginine] vasopressin and liquid diet and maintained for 4 days. In of serum sodium was still beneficial. Six
group 2, 12 hours before the correction of hyponatremia, minocycline or saline was
of 13 animals who received minocycline
given intraperitoneally at a dose of 60 mg/kg of body weight. At time of the first
injection of hypertonic saline, the same dose was repeated. In group 3, the first dose
18 hours after the correction survived
of minocycline (60 mg/kg of body weight) was given 18 hours after the correction of versus 26 of 31 animals who received hy-
hyponatremia. Twelve hours after the initial dose, another dose of 60 mg/kg of body pertonic saline alone (P 0.056). In the
weight was given. Thereafter, in both groups, the injections were continued at the group of animals who received delayed
reduced dose of 45 mg every 12 hours until 3 days after the correction. Animals in minocycline 18 hours after the correction
group 1 received normal saline injection instead of minocycline. of hyponatremia, all seven rats who later

2100 Journal of the American Society of Nephrology J Am Soc Nephrol 21: 2099 2108, 2010
 www.jasn.org BASIC RESEARCH

(NaCl Minocycline)
died already exhibited early neurologic signs by the time

Experiment 1 was designed to address mortality on day 10 (6 days after correction of chronic hyponatremia). Experiment 2 was designed to address the weight loss on day 7, and experiment 3 addressed
minocycline was given, and in contrast, five of six animals

108 2a
137 1a
29 2a
Group 2
who survived were nonsymptomatic at the time of minocy-

(n 8)

NA
NA
cline administration.
Experiment 3

Early Minocycline Treatment Decreases the

Permeability of BBB 24 Hours after Rapid Correction
of Chronic Hyponatremia
Evans Blue Dye Analysis.
(NaCl Alone)

As shown in Figure 3A, rapid correction of chronic hyponatremia

103 1a
132 2a
29 1a
Group 1

(n 8)

NA
NA
resulted in a significant increase in the permeability of the BBB
(P 0.01 compared with uncorrected controls). Early treatment
with minocycline restored the permeability of the BBB to almost
normal levels (P 0.05 compared with group 1, P not signifi-
cant compared with uncorrected controls).
Early Minocycline)
Group 2 (NaCl

107 2a
132 3a
26 2a

17 6c
(n 12)

Immunostaining for IgG.

NA

We used immunolabeling for IgG to identify regions with an

Experiment 2

increased permeability of the BBB. Representative micropho-

tographs of the brains of animals in each group are shown in
Figure 3, B and C. In animals treated with hypertonic saline
alone, a massive immunoreactivity for IgG was found in corti-
(NaCl Alone)

cal and subcortical areas, the basal ganglia, and hippocampus.

106 2a
130 2a
24 2a

33 2c
Group 1

(n 9)

In contrast, animals treated with minocycline showed only a

NA

faint immunoreactivity for IgG in the same areas.

Table 1. Biological and clinical parameters of animals used for all three experiments

Minocycline Reduces Myelin Damage and

MicrogliaMacrophage Accumulation after Rapid
Delayed Minocycline)
Group 3 (NaCl

Correction of Chronic Hyponatremia

We next wanted to know whether the improvement in clin-
106 2a
138 1a
33 1a
(n 13)

7/13b

ical outcome achieved by minocycline was associated with

NA

decreased demyelination in the CNS. We investigated mye-

the BBB permeability 24 hours after the correction of hyponatremia. NA, not applicable.

lin integrity by myelin basic protein (MBP) immunoreac-

P 0.056 for group 1 versus group 3 and P 0.0026 for group 1 versus group 2.

tivity and we found that animals treated with hypertonic

saline alone developed large areas of demyelination in the
cortex, basal ganglia, and hippocampus (Figure 4, A and C);
Experiment 1

P not significant for group 1 versus group 2 in all three experiments.

Early Minocycline)
Group 2 (NaCl

in contrast, animals treated with hypertonic saline and mi-

nocycline had less demyelination lesions (Figure 4, B and
105 1a
135 1a
31 1a
(n 27)

13/27b

D). Using semiquantitative image analysis to evaluate the

NA

extent of myelin pathology in the two groups of animals, we

found that treatment with minocycline significantly re-
duced the surface of areas of demyelination (P 0.05 in
group 1 versus group 2; Figure 4E).
Because the effects of minocycline have been reported to be
(NaCl Alone)
Group 1

(n 31)
105 1
136 1
31 1

partly caused by inhibition of microglial activation, we ana-

P 0.014 for group 1 versus group 2.
26/31
NA

lyzed microgliamacrophage activation by immunolabeling

for CD68 (clone ED1). In animals treated with hypertonic sa-
line alone, there was a higher proportion of CD68-positive cells
than in animals treated with minocycline (P 0.05 in group 1
SNa day 4 (mEq/L)
SNa day 5 (mEq/L)

versus group 2). Activated microglia were seen at the border of

Percent of weight
Mortality day 10

and within demyelinative regions (Figure 5).

DSNa (mEq/L)

loss day 7

Treatment with Minocycline Decreases GFAP

Immunoreactivity Loss in ODS
GFAP immunoreactivity loss has been shown to be affected
b
a

J Am Soc Nephrol 21: 2099 2108, 2010 Minocycline Alleviates Osmotic Demyelination 2101
 BASIC RESEARCH www.jasn.org

A B
6 50
C
N aC l + Minocycline 12hrs before correction
100
p = 0.014
5 N aC l + m inocycline 18 hrs after the correction
NaC l + Minoc y c line 40

% of body weight loss

NaC l alone
Neurological Score

* 80 N aC l alone

Percent survival
4
* **
30
60
** **
3

20 40
2
p = 0,011

1 10 20

0 0 0
0 1 2 3 4 5 6 7 0 1 2 3 4 5 6 7
NaCl alone NaCl + Mino
Days after correction of hyponatremia Days after correction of hyponatremia

Figure 2. Minocycline reduces mortality and morbidity after rapid correction of chronic hyponatremia. (A) Treatment with minocycline
decreases neurologic manifestations in ODS. Less severe neurologic manifestations were observed in the group of animals treated with
minocycline (*P 0.05; **P 0.01). (B) Animals treated with minocycline also showed a decrease in weight loss 3 days after ODS
induction (P 0.014 for group 1 versus group 2). (C) Kaplan-Meyer survival curve showing a decreased mortality in animals treated with
minocycline (P 0.011 by log-rank test).

by minocycline in other neurologic conditions.19 We found limit the impact of our findings in the clinical setting. How-
that severe GFAP immunoreactivity loss occurs in ODS and ever, recently it was shown that local CNS delivery of minocy-
that animals treated with minocycline displayed smaller ar- cline combined with an intraperitoneally dose can reduce the
eas of loss of astrocytes compared with animals treated with amount of minocycline administered for neuroprotection and
hypertonic saline (P 0.05 in group 1 versus group 2; Figure 6). increase the protective effect in another model of CNS injury.24
It should also be pointed out that, in our study, the gradient of
Minocycline Decreases Expression of IL-1 and serum sodium achieved by animals in all groups is very high
Nitrosylation of Protein in ODS and rarely seen in the setting of correction of hyponatremia.
IL-1 is involved in the pathogenesis of demyelination, and nitric Another interesting conclusion of our work comes from the
oxide has been shown to be a potent inflammatory mediator in fact that delayed administration of minocycline still carries
demyelinative disorders.20,21 Previous reports suggested increased some protection. Of note, all rats that died after receiving de-
production of nitric oxide synthase22 by activated microglia in layed minocycline (18 hours after correction) were already
ODS.23 Therefore, we use immunohistochemistry to study the symptomatic by the time minocycline was given. Translated to
production of IL-1 and the effect of nitric oxide synthase (ni- clinical setting, it is rare to observe symptoms of ODS within
trosylation of protein) after minocycline administration. Treat- the first 24 hours of correction of hyponatremia, and the fact
ment with minocycline resulted in a decrease in IL-1 immuno- that all of six animals who survived after delayed administration of
reactivity and protein nytrosylation, as shown by a striking minocycline were pauci-symptomatic or asymptomatic, suggest
decrease in nitrotyrosine immunoreactivity in the CNS (Figure 7). that this approach might still be used in clinical practice.
Our results are in line with previously published observa-
tions on the effect of minocycline on BBB in other models of
DISCUSSION CNS demyelinative and nondemyelinative pathology.14,15,25
Regarding the possible mechanisms associated with the pro-
In this work, we studied the effect of minocycline, a second- tection of the BBB, previous work has suggested that minocy-
generation tetracycline, in an animal model of ODS. cline acts on matrix metalloproteases and inflammatory cyto-
Using a very large series of animals, we found that treatment kines secretion that are involved in the alteration of the BBB
with minocycline resulted in increased survival and decreased permeability.2527
severity of neurologic manifestations after rapid correction of Minocycline did not completely abrogate the histologic le-
chronic hyponatremia. We also found that treatment with mi- sions of myelinolysis, and some areas of myelin loss were found
nocycline reduces BBB permeability and reduced microglia in nonsymptomatic or pauci-symptomatic animals, suggesting
macrophage activation. dissociation between neurologic manifestations and myelin
The doses of minocycline used in this animal study, derived destruction. These findings reflect the vast spectrum of ODS
from the doses used in other models of experimental CNS pro- and clinically asymptomatic central pontine myelinolysis
tection, are very large and seldom used in humans and might found in humans. The fact that significant asymptomatic or

2102 Journal of the American Society of Nephrology J Am Soc Nephrol 21: 2099 2108, 2010
 www.jasn.org BASIC RESEARCH

p< 0.01 NaCl + Mino NaCl alone The same process could be applied to the
A
2.5 p< 0.05 B C role of microglia in ODS. In accordance with
previous studies,1315,29 we found that mino-
EB (g/gr brain tissue)

2.0 cycline reduced microgliamacrophage acti-

vation in CNS. Nevertheless, despite signifi-
1.5 cant microglial inhibition, demyelinating
p = NS lesions were still present in the brains of ani-
E D mals treated with minocycline. This further
1.0
questions the role of microglial activation in
0.5
ODS. It is possible that activated microglia
macrophages appear in the CNS after the
original demyelinative insult to remove mye-
0.0
NaCl alone NaCl + Mino Uncorrected Ctrl lin debris without contributing to the initial
demyelinative lesion. This was suggested by
Figure 3. Minocycline decreases the permeability of the BBB after rapid correction of
experiments showing that inhibition of mi-
chronic hyponatremia. (A) Evans blue dye extravasation 24 hours after the correction of
chronic hyponatremia showing increase permeability in animals treated with hyper-
crogliamacrophage activation with lova-
tonic saline alone in contrast to animals treated with hypertonic saline and minocycline satin was not associated with a net decrease
23

(n 8 for NaCl alone; n 7 for NaCl minocycline; n 6 for uncorrected Ctrl). P in mortality after inadvertent correction of
0.05 for group 1 versus group 2 and P 0.01 for group 1 versus group 3. Group 2 chronic hyponatremia. A very likely explana-
versus group 3, P not significant. (BE) Immunohistochemistry for IgG. Increase tion of these findings is that the secretion of
deposition of IgG in subcortical regions and basal Ganglia (C) in animals treated inflammatory cytokines by microglia exacer-
hypertonic saline alone compared with animals who also received minocycline (B). At bates neurologic deficits and demyelination
higher magnification, IgG are seen intraparenchymally in animals treated with NaCl but has marginal contribution to the main
alone (D), whereas mainly intravascular deposition is visible in animals treated with
demyelinative process.
minocycline (E). Scale bar: 200 m.
The decrease in inflammatory marker
pauci-symptomatic demyelination can occur despite mini- reactivity (IL-1 and nitro-tyrosine) ob-
mal opening of the BBB together with our recent study using served after minocycline administration further provides
dexamethasone in ODS6 challenge the assumption that strong evidence that minocycline attenuates inflammation
opening of the BB in ODS plays a crucial role in the devel- after ODS induction.
opment of demyelination.8,9,23,28 GFAP loss is an early phenomenon in diverse neurologic

A C E
NaCl alone

.
*
.

B D
NaCl + Mino

Figure 4. Minocycline reduces myelin loss after rapid correction of chronic hyponatremia. Immunohistochemistry for MBP showing
large area of demyelination in cortex, hippocampus, and basal ganglia (arrows) after correction of hyponatremia with NaCl alone
(A); less and smaller lesions are found after administration of minocycline (B). Image quantification of MBP lacking area (C) showing that
treatment with minocycline significantly attenuated the demyelination (P 0.05 compared with animals treated with NaCl alone) but
did not suppressed the myelin loss. (C and D) Higher magnification of hypothalamus in animals treated with hypertonic saline alone
(C) or hypertonic saline and minocycline (D) showing massive and well-delineated loss of myelin on axons bundles (*). (E) Represents
the quantification of myelin loss in the two groups compared with uncorrected control. Computerized quantification was applied
on MBP-stained images (as in A and B) (n 4 for NaCl alone; n 12 for NaCl Minocycline; n 5 for uncorrected controls).
Scale bar 200 m.

J Am Soc Nephrol 21: 2099 2108, 2010 Minocycline Alleviates Osmotic Demyelination 2103
 BASIC RESEARCH www.jasn.org

A associated with massive loss of GFAP reac-

tivity and that treatment with minocycline
C
NaCl + Mino

significantly reduced GFAP staining loss.

The full significance of GFAP loss with re-
gard to astrocyte viability and the role in
ODS is a yet to be determined.
After rapid correction of chronic hypo-
natremia, few therapeutic options are
B
available to prevent the development or
limits the consequences of ODS. Thus far,
NaCl alone

re-lowering serum sodium is the most ben-

eficial treatment when the initial osmotic
injury has already occurred.6,3335 Our re-
sults suggest that, in clinical practice, early
administration of minocycline might have
a role in preventing neurologic damage in
Figure 5. Minocycline decreases microglial activation in ODS. Strong immunostaining
hyponatremic patients who underwent
for CD68 was observed in the basal ganglia and cortex of animals corrected with NaCl
alone (B). In contrast, moderate immunoreactivity was seen after treatment with NaCl rapid correction of their sodium deficit.
and minocycline (A). Minocycline significantly reduced microglial activation in rat brain Nevertheless, re-lowering of serum sodium
after correction of chronic hyponatremia (C) (n 4 for NaCl alone; n 12 for NaCl should remain the treatment of choice
minocycline). Scale bar: 200 M. when a toxic gradient is achieved until fur-
ther work is done to fully explore the pos-
disorders. 30,31 Previous work reported a beneficial effect of mi- sibilities of minocycline in management of rapid correction of
nocycline after CNS injury in relation to decreased GFAP chronic hyponatremia in terms of the minimal effective dose
loss,19 and in vivo studies showed that minocycline prevents and possible combination with re-lowering of serum sodium.
astrocyte cell death.32 Here, we found that experimental ODS is In summary, this work addressed the effect of minocycline

D
NaCl alone NaCl + Mino
A B
*
*
* * *

C
C)

*
*
*

Figure 6. Minocycline reduces GFAP loss in osmotic demyelination. Immunohistochemistry for GFAP showing massive loss of GFAP
immunoreactivity (*) after correction of hyponatremia with NaCl alone (A). This occurs predominantly on subcortical regions, hippocam-
pus, and basal ganglia. Administration of minocycline decrease the loss of astrocytes (B). (C) Higher magnification of area of GFAP
reactivity loss in cortex of animals treated with NaCl alone. Scale bar: 200 m. (D) Quantitative evaluation of GFAP loss 6 days after the
correction showing significant decrease of astrocytes loss in animals treated with minocycline compared with animals treated with NaCl
alone. Computer-assisted quantification of GFAP-stained area was used (n 4 for NaCl alone; n 5 for uncorrected controls; n 12
for NaCl minocycline).

2104 Journal of the American Society of Nephrology J Am Soc Nephrol 21: 2099 2108, 2010
 www.jasn.org BASIC RESEARCH

NaCl alone NaCl + Mino minocycline; 12 hours before the correction of

A
A B
B hyponatremia, group 2 received intraperitoneal
injection of minocycline hydrochloride (Duch-
efa) at a dose of 60 mg/kg of body weight. At
3N T

time of correction (injection of hypertonic sa-

line), the same dose was repeated, and minocy-
cline injections were continued at a decreased
dose of 45 mg/kg of body weight every 12 hours
for 3 days (until day 7 of experiment). Group 1
C
C D
D served as controls animals and received only in-
traperitoneal injection of sterile saline instead of
minocycline.
To determine whether delayed administra-
Il1

tion of minocycline impacts survival, we used a

second administration scheme in another group
of 13 animals (group 3), where minocycline (60
mg/kg of body weight) was given 18 hours after
Figure 7. Minocycline decreases IL-1 and nitrotyrosine immunoreactivity in ODS. the correction of hyponatremia when the toxic
Representative images of nitrotyrosine (A and B) and IL-1 (C and D) immunoreactivity. gradient was already achieved. This dose was re-
Intense immunoreactivity for nitrotyrosine and IL-1 was found in brain of animals peated 12 hours later, and, as in group 2, a re-
submitted to ODS (A and C, respectively). In contrast, moderate immunoreactivity was duced dose of 45 mg/kg of body weight every 12
found after treatment with minocycline (B and D). hours was then given until day 7.
on ODS. We found that, in rats, minocycline decreased the The minocycline dosage and administration scheme were de-
neurologic manifestations and increased the survival after rived from previous reports studying the effects of that drug in
rapid correction of chronic hyponatremia. These effects were CNS injury.13,25
associated with a striking reduction in BBB permeability, a
decrease in microgliamacrophage activation and GFAP im- Correction of Hyponatremia
munoreactivity loss, and a decrease in inflammatory protein After 4 days of liquid feeding (day 4 of experiment), hyponatremia
expression. was rapidly corrected by intraperitoneal administration of 1 M NaCl
in a single dose.36,37 Pelleted chow was given to the animal on day 5 of
the experiment, after the blood sampling for the determination of
CONCISE METHODS final 24-hour SNa; to avoid incidental re-lowering occurring at the
day 5, water was given only at day 6 of the experiment.6,33
Animals
Male Wistar rats (250 to 300 g) were used for all experiments. They
Blood Measurements
were housed in individual cages under conditions of constant temper- Blood samples (0.3 ml) were collected via tail transection at days 4 and
ature (23C) and a 12-hour/12-hour light/dark cycle and used after a 5 after light halothane anesthesia for serum sodium analysis. In ani-
brief period of adaptation (3 to 4 days), during which ad libitum water mals of group 2, blood samples were taken after the first injection of
and standard pelleted chow was provided. All procedures were per- minocycline. Electrolytes measurements were performed using
formed in accordance with guidelines for animal care at Universite MODULAR p800 (Roche).
Libre de Bruxelles.
Evaluation of Neurologic Manifestations and Mortality
Induction of Hyponatremia A first set of experiments was designed to assess mortality and neuro-
As described previously, chronic hyponatremia was induced by
logic manifestations. Animals were corrected with 2 ml hypertonic
1-deamino-[8-D-arginine] vasopressin infusion and liquid diet.5,6,18 An
NaCl/100 g of body weight, and they were allowed to survive until day
osmotic minipump (Model 2001; Alzet, Palo Alto, CA) was filled with 4
10 (6 days after the correction) of the experiment.
g/ml 1-deamino-[8-D-arginine] vasopressin (Minirin, Ferring, Swe-
A total of 31 rats were included in the group of animals treated
den) and inserted in the back of the animal under light halothane anes-
with hypertonic saline alone; in a second group of 27 rats, minocycline
thesia at the beginning of the experiment (day 0). On the day of insertion,
was started 12 hours before the beginning of the correction, and in a
rats were switched to a liquid diet with low sodium content (AIN 76;
third group of 13 animals, minocycline was given 18 hours after cor-
Technilab BMI). The liquid diet was maintained until the morning of day
rection of hyponatremia.
4. Infusion was maintained until the end of the experiment.
Rats treated with early minocycline and hypertonic saline were
Drug Treatment observed daily for occurrence of neurologic manifestations. Using a
Three groups of animals were studied. Group 1 consisted of controls previously described scoring system,23 neurologic manifestation were
animals with no minocycline treatment, and group 2 received early scored on a daily basis as follows: 6 no neurologic manifestations;

J Am Soc Nephrol 21: 2099 2108, 2010 Minocycline Alleviates Osmotic Demyelination 2105
 BASIC RESEARCH www.jasn.org
5 slow or awkward gait; 4 limb weakness and/or paralysis; 3
seizures; 2 severe motor deficits; 1 complete inability to move;
and 0 death.
In a second set of experiments (experiment 2), weight loss was
evaluated in animals of group 1 and group 2 (early minocycline).
Because a correction gradient of 25 mEq/L per 24 hours has been
associated with increased mortality, we specifically aimed to reach
a lower gradient of correction to induce ODS with decreased mor-
tality. These animals were given 1 ml/100 g of body weight of
hypertonic saline, and weight was recorded before the correction
of hyponatremia and 3 days after the correction (day 7 of experi-
ment). For this experiment, we used an independent set of rats
divided into two groups: animals treated with hypertonic saline
alone (n 9) and animals treated with minocycline before the
correction (n 12).
Immunohistochemistry and Immunofluorescence
At day 10 of the experiment, surviving animals treated with NaCl
or early minocycline were killed by decapitation. The brain was
divided along the midline into its two hemispheres and was fixed
overnight in formalin 10% before rinsing with PBS. Immunohis-
tochemistry was performed on 7-m-thick paraffin section using
anti-myelin basic protein for myelin (anti-MBP 1/1000; Abcam,
Cambridge, UK), anti-CD68 clone ED1 for microglia (1/200;
AbDserotec, Dusseldorf, Germany), and anti-GFAP for astrocytes
(1/1000; Dako, Brussels, Belgium). Anti-IL-1 and anti-nitro-ty-
rosine were purchased from Pierce Endogen and Millipore (Brus-
sels, Belgium). Fluorescent conjugated secondary antibodies used
were goat anti-mouse Alexa 488 and goat anti-rabbit Alexa 594
(1:100), both from Invitrogen. Technical procedures have been
described elsewhere.6 IgG immunohistochemistry was performed
as an assessment of BBB reactivity as described previously, using
one-step detection with biotinylated anti-rat IgG.6,9
Myelin, MicrogliaMacrophage Activation, and
Astrocyte Loss Quantification
Image quantification of areas of demyelination and GFAP loss was
performed as described previously in animals treated with early
minocycline and hypertonic saline.38,39 Briefly, macro-images of
stained sagittal sections of brain were obtained with a Nikon cam-
era fitted with a Nikon macro-objective. Three sets of two adjacent
slides for each animal (six slides/animals) labeled for myelin (anti-
MBP) and astrocytes (anti-GFAP) were analyzed blindly using Im-
age J software (National Institutes of Health). Each set of slides
were separated by at least five slides. The three sets of two slides
were taken at a similar sagittal level in all of the groups studied. The
regions with loss of myelin and astrocytes were identified as MBP
and GFAP unstained areas with light microscope and were mea-
sured by the software after gray level thresholding of the corre-
sponding image. Results were reported over the total brain surface
area and expressed as percentage of demyelinated or astrocyte loss
area.
Microglia-macrophage activation was quantified on sagittal sec-
tions stained for CD 68. As described by others,13 area of CD68 im-
munoreactivity were was measured on non counterstained slides us-
2106 Journal of the American Society of Nephrology
ing image J software and the ratio of CD68 stained area over total
brain surface was calculated.
Evaluation of BBB by Evans Blue Dye Extravasation
In another set of experiments (experiment 3), Evans blue was used
to determine the BBB permeability 24 hours after the correction of
hyponatremia in animals treated with hypertonic saline and early
minocycline (n 8 for each group) and in a third control group of
uncorrected hyponatremic rats (n 6). In each animal, Evans blue
solution (Sigma Chemical, St. Louis, MO) (2% in saline; 3 ml/kg of
body weight) was injected intravenously in the tail vein 24 hours
after rapid correction of chronic hyponatremia and allowed to
circulate for 40 minutes. Under deep anesthesia, the chest was
dissected, and the left ventricle was perfused with saline for 20
minutes to remove any circulating dye. The skull was dissected to
extract the brain; brain tissue was weighed and allowed to incubate
in formamide for 72 hours at 50C. The formamide solution con-
taining extracted Evans blue was centrifuged at 20,000 g for 15
minutes, and the absorbance of the supernatant was analyzed spec-
trophotometrically at 610 nm. The content of Evans blue, ex-
pressed per gram of brain tissue, was calculated using a linear
standard curve built after absorbance of known amounts of dye in
formamide.
Statistical Analysis
Results were expressed as mean SEM. Paired and unpaired t tests
were used as a parametric test for normally distributed variables. Oth-
erwise, Mann-Whitney nonparametric tests were used. Fischer exact
test on proportions and Kaplan-Meir survival analysis were also used
to compared mortality and survival in both groups. One-way
ANOVA followed by Bonferonni least significant difference was used
for Evans blue dye contains analysis. Tests were run using statdirect
software. P 0.05 was considered significant.
ACKNOWLEDGMENTS
This study was supported by a grant from the Fonds National de la
Recherche Scientifique, conventions 3.4509.03, 3.4656.09, a grant
from the Belgian PAI/IAP P6/43, and a grant from Fondation Erasme.
Part of this study was presented as an abstract at Renal Week 2009
(J Am Soc Nephrol 20: 2009. THPO O69 p127A). We thank Michelle
Authelet and Charles Nicaise for their helpful technical assistance and
Will Renthal for his critical comments.
DISCLOSURES
None.
REFERENCES
1. Kleinschmidt-DeMasters BK, Norenberg MD: Rapid correction of hy-
ponatremia causes demyelination: Relation to central pontine myeli-
nolysis. Science 211: 1068 1070, 1981
J Am Soc Nephrol 21: 2099 2108, 2010

2. Sterns RH, Cappuccio JD, Silver SM, Cohen EP: Neurologic sequelae
after treatment of severe hyponatremia: A multicenter perspective.
J Am Soc Nephrol 4: 15221530, 1994
3. Sterns RH, Laureno R, Norenberg MD, Karp BI: Symptomatic hypona-
tremia in rats: Effect of treatment on mortality and brain lesions. Am J
Physiol 258: F1475F1477, 1990
4. Sterns RH, Riggs JE, Schochet SS Jr: Osmotic demyelination syn-
drome following correction of hyponatremia. N Engl J Med 314:
15351542, 1986
5. Verbalis JG, Martinez AJ: Neurological and neuropathological se-
quelae of correction of chronic hyponatremia. Kidney Int 39: 1274
1282, 1991
6. Gankam Kengne F, Soupart A, Pochet R, Brion JP, Decaux G:
Re-induction of hyponatremia after rapid overcorrection of hypo-
natremia reduces mortality in rats. Kidney Int 76: 614 621, 2009
7. Murase T, Sugimura Y, Takefuji S, Oiso Y, Murata Y: Mechanisms
and therapy of osmotic demyelination. Am J Med 119: S69 S73,
2006
8. Adler S, Verbalis JG, Williams D: Effect of rapid correction of hypo-
natremia on the blood-brain barrier of rats. Brain Res 679: 135143,
1995
9. Baker EA, Tian Y, Adler S, Verbalis JG: Blood-brain barrier disrup-
tion and complement activation in the brain following rapid cor-
rection of chronic hyponatremia. Exp Neurol 165: 221230, 2000
10. Yong VW, Giuliani F, Xue M, Bar-Or A, Metz LM: Experimental models
of neuroprotection relevant to multiple sclerosis. Neurology 68: S32
S37, 2007
11. Kleinschmidt-Demasters BK, Rojiani AM, Filley CM: Central and ex-
trapontine myelinolysis: Then and now. J Neuropathol Exp Neurol 65:
111, 2006
12. Arvin KL, Han BH, Du Y, Lin SZ, Paul SM, Holtzman DM: Minocycline
markedly protects the neonatal brain against hypoxic-ischemic injury.
Ann Neurol 52: 54 61, 2002
13. Popovic N, Schubart A, Goetz BD, Zhang SC, Linington C, Duncan ID:
Inhibition of autoimmune encephalomyelitis by a tetracycline. Ann
Neurol 51: 215223, 2002
14. Yenari MA, Xu L, Tang XN, Qiao Y, Giffard RG: Microglia potentiate
damage to blood-brain barrier constituents: Improvement by minocy-
cline in vivo and in vitro. Stroke 37: 10871093, 2006
15. Yrjanheikki J, Tikka T, Keinanen R, Goldsteins G, Chan PH, Koisti-
naho J: A tetracycline derivative, minocycline, reduces inflamma-
tion and protects against focal cerebral ischemia with a wide ther-
apeutic window. Proc Natl Acad Sci USA 96: 13496 13500,
1999
16. Chen M, Ona VO, Li M, Ferrante RJ, Fink KB, Zhu S, Bian J, Guo L,
Farrell LA, Hersch SM, Hobbs W, Vonsattel JP, Cha JH, Friedlander
RM: Minocycline inhibits caspase-1 and caspase-3 expression and
delays mortality in a transgenic mouse model of Huntington disease.
Nat Med 6: 797 801, 2000
17. Zhu S, Stavrovskaya IG, Drozda M, Kim BY, Ona V, Li M, Sarang S, Liu
AS, Hartley DM, Wu DC, Gullans S, Ferrante RJ, Przedborski S, Kristal
BS, Friedlander RM: Minocycline inhibits cytochrome c release and
delays progression of amyotrophic lateral sclerosis in mice. Nature
417: 74 78, 2002
18. Verbalis JG: An experimental model of syndrome of inappropriate
antidiuretic hormone secretion in the rat. Am J Physiol 247: E540
E553, 1984
19. Tomas-Camardiel M, Rite I, Herrera AJ, de Pablos RM, Cano J,
Machado A, Venero JL: Minocycline reduces the lipopolysaccha-
ride-induced inflammatory reaction, peroxynitrite-mediated nitra-
tion of proteins, disruption of the blood-brain barrier, and damage
in the nigral dopaminergic system. Neurobiol Dis 16: 190 201,
2004
20. Badovinac V, Mostarica-Stojkovic M, Dinarello CA, Stosic-Grujicic S:
Interleukin-1 receptor antagonist suppresses experimental autoim-
mune encephalomyelitis (EAE) in rats by influencing the activation and
J Am Soc Nephrol 21: 2099 2108, 2010
www.jasn.org BASIC RESEARCH
proliferation of encephalitogenic cells. J Neuroimmunol 85: 8795,
1998
21. Jacobs CA, Baker PE, Roux ER, Picha Ks, Toivola B, Waugh S, Kennedy
MK: Experimental autoimmune encephalomyelitis is exacerbated by
IL-1 alpha and suppressed by soluble IL-1 receptor. J Immunol 146:
29832989, 1991
22. Kikuchi K, Kawahara K, Biswas KK, Ito T, Tancharoen S, Morimoto Y,
Matsuda F, Oyama Y, Takenouchi K, Miura N, Arimura N, Nawa Y,
Meng X, Shrestha B, Arimura S, Iwata M, Mera K, Sameshima H,
Ohno Y, Maenosono R, Yoshida Y, Tajima Y, Uchikado H, Kuramoto
T, Nakayama K, Shigemori M, Hashiguchi T, Maruyama I: Minocy-
cline attenuates both OGD-induced HMGB1 release and HMGB1-
induced cell death in ischemic neuronal injury in PC12 cells. Bio-
chem Biophys Res Commun 385: 132136, 2009
23. Takefuji S, Murase T, Sugimura Y, Takagishi Y, Hayasaka S, Oiso Y,
Murata Y: Role of microglia in the pathogenesis of osmotic-induced
demyelination. Exp Neurol 204: 88 94, 2007
24. Xue M, Mikliaeva EL, Casha S, Zygun D, Demchuk A, Yong VW:
Improving outcomes of neuroprotection by minocycline: Guides from
cell culture and intracerebral hemorrhage in mice. Am J Pathol 176:
11931202, 2010
25. Wasserman JK, Schlichterf LC: Minocycline protects the blood-brain
barrier and reduces edema following intracerebral hemorrhage in the
rat. Exp Neurol 207: 227237, 2007
26. Paemen L, Martens E, Norga K, Masure S, Roets E, Hoogmartens J,
Opdenakker G: The gelatinase inhibitory activity of tetracyclines
and chemically modified tetracycline analogues as measured by a
novel microtiter assay for inhibitors. Biochem Pharmacol 52: 105
111, 1996
27. Yong VW, Power C, Forsyth P, Edwards DR: Metalloproteinases in
biology and pathology of the nervous system. Nat Rev Neurosci 2:
502511, 2001
28. Rojiani AM, Cho ES, Sharer L, Prineas JW: Electrolyte-induced demy-
elination in rats. 2. Ultrastructural evolution. Acta Neuropathol 88:
293299, 1994
29. Yrjanheikki J, Keinanen R, Pellikka M, Hokfelt T, Koistinaho J:
Tetracyclines inhibit microglial activation and are neuroprotective
in global brain ischemia. Proc Natl Acad Sci USA 95: 15769 15774,
1998
30. Itoyama Y, Sekizawa T, Openshaw H, Kogure K, Goto I: Early loss of
astrocytes in herpes simplex virus-induced central nervous system
demyelination. Ann Neurol 29: 285292, 1991
31. Liu D, Smith CL, Barone FC, Ellison JA, Lysko PG, Li K, Simpson IA:
Astrocytic demise precedes delayed neuronal death in focal ischemic
rat brain. Brain Res Mol Brain Res 68: 29 41, 1999
32. Kuang X, Scofield VL, Yan M, Stoica G, Liu N, Wong PK: Attenuation
of oxidative stress, inflammation and apoptosis by minocycline pre-
vents retrovirus-induced neurodegeneration in mice. Brain Res 1286:
174 184, 2009
33. Soupart A, Pennicnckx R, Crenier L, Stenuit A, Perier O, Decaux G:
Prevention of brain demyelination in rats after excessive correction of
chronic hyponatremia by serum sodium lowering. Kidney Int 45: 193
200, 1994
34. Soupart A, Penninckx R, Stenuit A, Perier O, Decaux, G. Reinduc-
tion of hyponatremia improves survival in rats with myelinolysis-
related neurologic symptoms. J Neuropathol Exp Neurol 55: 594
601, 1996
35. Perianayagam A, Sterns RH, Silver SM, Grieff M, Mayo R, Hix J,
Kouides R: DDAVP is effective in preventing and reversing inadvertent
overcorrection of hyponatremia. Clin J Am Soc Nephrol 3: 331336,
2008
36. Soupart A, Penninckx R, Stenuit A, Perier O, Decaux G: Treatment of
chronic hyponatremia in rats by intravenous saline: Comparison of rate
versus magnitude of correction. Kidney Int 41: 16621667, 1992
37. Soupart A, Stenuit A, Perier O, Decaux G: Limits of brain tolerance to
daily increments in serum sodium in chronically hyponatraemic rats
Minocycline Alleviates Osmotic Demyelination 2107
 BASIC RESEARCH www.jasn.org
treated with hypertonic saline or urea: Advantages of urea. Clin Sci
(Lond) 80: 77 84, 1991
38. Mi S, Hu B, Hahm K, Luo Y, Kam Hui ES, Yuan Q, Wong WM, Wang
L, Su H, Chu TH, Guo J, Zhang W, So KF, Pepinsky B, Shao Z, Graff
C, Garber E, Jung V, Wu EX, Wu W: LINGO-1 antagonist promotes
spinal cord remyelination and axonal integrity in MOG-induced
experimental autoimmune encephalomyelitis. Nat Med 13: 1228
1233, 2007
39. Bauer J, Elger CE, Hans VH, Schramm J, Urbach H, Lassmann H, Bien
2108 Journal of the American Society of Nephrology
CG: Astrocytes are a specific immunological target in Rasmussens
encephalitis. Ann Neurol 62: 67 80, 2007
See related editorial, Managing Overly Rapid Correction of Chronic Hypona-
tremia: An Ounce of Prevention or a Pound of Cure? on pages 20152016, and
related article, Minocycline Prevents Osmotic Demyelination Syndrome by
Inhibiting the Activation of Microglia, on pages 2090 2098.
J Am Soc Nephrol 21: 2099 2108, 2010