www.elsevier.com/locate/brainres

Available online at www.sciencedirect.com

Research Report

Morg1þ/� heterozygous mice are protected from

experimentally induced focal cerebral ischemia

A. Stahra, C. Frahma, A. Kretza, T. Bondevab, O.W. Wittea, G. Wolf b,n

aDepartment of Neurology, University of Jena, Jena, GermanybDepartment of Internal Medicine III, University of Jena, Jena, Germany

a r t i c l e i n f o

Article history:

Accepted 10 September 2012

Available online 13 September 2012

Keywords:

Cerebral ischemia

Stroke

Prolyl hydroxylases

Hypoxia-inducible transcription

factor (HIF)

Morg1

Middle cerebral artery occlusion

(MCAO)-model

a b s t r a c t

Focal cerebral ischemia (stroke) and reperfusion injury leads to acute and chronic brain

damage. The increase of the hypoxia-inducible transcription factor alpha (HIF-a), an

important transcription factor for several genes, may attenuate ischemic brain injury.

We recently identified a new WD-repeat protein designated Morg1 (MAPK organizer 1) that

interacts with prolyl hydroxylase 3 (PHD3), an important enzyme involved in the regulation

of HIF-1a and HIF-2a expression. While homozygous Morg1�/� mice are embryonically

lethal, heterozygous Morg1þ/� mice have a normal phenotype. Brain vasculature as well as

systolic blood pressure in Morg1þ/� mice were indistinguishable from wild-type (WT)

animals. We show here that Morg1þ/� mice were partially protected from cerebral

ischemia/reperfusion injury in comparison to WT (Morg1þ/þ) animals using the middle

cerebral artery occlusion model (MCAO). Morg1þ/� mice compared with WT animals

revealed a significantly reduced infarct volume as detected by Nissl and Map 2 staining

despite a similar restriction of blood flow in both mice genotypes as measured by laser

Doppler flowmetry. Immunohistochemistry revealed specific Morg1 expression in reactive

astrocytes in the ipsilateral (ischemic) hemisphere in Morg1þ/� and WT mice, especially in

the penumbral regions. In the contralateral hemisphere, Morg1 was not detectable.

Furthermore, Morg1 mRNA expression was significantly enhanced in the ischemic brain

of WT, but not in ischemic brain tissue obtained from Morg1þ/� animals. However, HIF-1a

was expressed with the same intensity in Morg1þ/� and WT mice with no difference

between the ipsilateral and contralateral hemispheres. No positive staining for HIF-2a was

found in ischemic (ipsilateral) and non-ischemic (contralateral) brain regions in Morg1þ/þ

and Morg1þ/� mice. Almost no PHD3 staining was found in the contralateral hemispheres

of either WT or heterozygous Morg1þ/� mice. Transcript expression for the HIF1a-

dependent genes erythropoietin (Epo) and vascular endothelial growth factor 164 (VEGF

164) were significantly reduced in the ischemic brain from Morg1þ/� mice. Positive staining

for PHD3 in the ipsilateral hemisphere of WT mice was suggested to occur in astrocytes. A

compensatory increase in Morg1 expression in astrocytes in the penumbra may negatively

influence infarct volume. It appears that these effects are independent of the PHD3-

HIF1a axis.

& 2012 Elsevier B.V. All rights reserved.

0006-8993/$ - see front matter & 2012 Elsevier B.V. All rights reserved.http://dx.doi.org/10.1016/j.brainres.2012.09.017

nCorresponding author. Fax: þ49 3641 9324302.E-mail address: Gunter.Wolf@med.uni-jena.de (G. Wolf).

b r a i n r e s e a r c h 1 4 8 2 ( 2 0 1 2 ) 2 2 – 3 1

1. Introduction

Acute ischemic stroke is common in humans and its compli-

cations are major causes of long-term morbidity and mortality

(Balami et al., 2011). Although arterial recanalization with

subsequent reperfusion may restore brain function (Molina,

2011), a better understanding of molecular mechanisms

underlying cerebral ischemia may help in the development

of novel therapeutic approaches to limit initial damage or even

enhance recovery after ischemic stroke (Ratan et al., 2007).

Cerebral cell populations including astrocytes respond to

reduced oxygen tension by the expression of many genes, most

of which are directly or indirectly induced by the hypoxia-

inducible transcription factor (HIF; Kim et al., 2009). One of the

two subunits of HIF is HIF-1a, which is continuously expressed

in neural tissue, but immediately degraded via the proteosomal

pathway after ubiquitination. In the presence of oxygen, prolyl

residues in HIF-1a become hydroxylated by the enzymes prolyl

hydroxylase domain (PHD)1–3. Recent work suggests that

upregulation of HIF-1a may protect the brain from acute

cerebral ischemia and may also foster neovascularization sub-

sequent to cerebral hypoxia (Scholzke and Schwaninger, 2007;

Marti et al., 2000).

We recently identified a b-transducin (WD) repeat protein

designated Morg1 (MAPK organizer 1) that interacts with PHD3

in vitro and in vivo (Hopfer et al., 2006). Binding to PHD3 occurs

at a conserved region predicted to the top surface of one

propeller blade. HIF-1a expression is reduced by Morg1 since

this protein activates PHD3, leading to an accelerated degrada-

tion of HIF-1a (Hopfer et al., 2006). We generated Morg1 knock-

out mice by homologous recombination (Hammerschmidt et al.,

2009). While homozygous Morg1�/� mice displayed embryonic

lethality due to severe neuronal development defects (anence-

phaly, neurulation defects; Wolf et al., unpublished), Morg1þ/�

animals show a normal phenotype and were protected from

renal hypoxia/reperfusion injury (Hammerschmidt et al., 2009).

Since we recently discovered a complex regulation of Morg1

expression in ischemic human brain (reduced expression in

neurons with ischemic damage, but upregulation of these

proteins in astrocytes surrounding the penumbra (Haase et al.,

2009)), the present study investigates the outcome of experi-

mentally induced cerebral ischemia using middle cerebral artery

occlusion (MCAO) in Morg1þ/� and wild-type (Morg1þ/þ) mice.

2. Results

2.1. Blood pressure values and cerebral blood vessels

Systolic blood pressure as measured with tail plethysmogra-

phy were not different between the genotypes (Morg1þ/þ:

9275 mmHg, Morg1þ/�: 9473 mmHg, n¼8). Furthermore,

immunohistological staining of cortical and subcortical

branches of the middle cerebral artery (MCA) with an anti-

body against von Willebrand factor (vWF) and quantitative

analysis of the vascular area revealed no significant differ-

ences in cortical and subcortical blood vessel architecture

between Morg1þ/� mice (n¼3) and Morg1þ/þ (n¼4), respec-

tively (Fig. 1A and B).

2.2. Infarct volume

During surgery, body temperature was maintained at physio-

logical level. Cerebral blood flow (CBF) was measured by laser

Doppler flowmetry (LDF).

CBF before, as well as following occlusion of the MCA, did

not differ between Morg1þ/� mice and their controls (Fig. 2).

However, significantly reduced infarct volumes were

observed in heterozygote Morg1þ/� mice compared to their

controls 48 h after MCAO (n¼6). Both staining methods (Nissl

and Map 2) lead to similar results (Fig. 3A and B).

2.3. Immunohistochemistry

Immunohistochemistry revealed a specific Morg1 expression

in reactive astrocytes in the ipsilateral hemisphere in Morg1þ/

� and WT mice, especially in penumbral regions. In the

contralateral hemisphere, Morg was undetectable in either

genotype (Fig. 4A). However, although we have previously

shown in humans that Morg1 expressing cells do also stain

positively for Glial fibrillary acidic protein (GFAP) (Haase et al.,

2009), we have not performed double labelings in the present

study. Therefore, we cannot totally exclude that these cells

may partly represent activated microglia. HIF-1a was

expressed with the same intensity in Morg1þ/� and WT mice

with no obvious difference between the contralateral and

ipsilateral hemispheres (Fig. 4B).

Almost no PHD3 staining was found in the contralateral

hemispheres of either Morg1þ/� or WT mice (Fig. 5A). How-

ever, astrocytes stained positively for PHD3 in the ipsilateral

hemisphere only in the WT, but not in the Morg1þ/� animals

(Fig. 5A).

No specific positive staining for HIF-2a was found in cells of

the ischemic and non-ischemic hemispheres of Morg1þ/� and

Morg1þ/þ mice (Fig. 5B).

2.4. qPCR analysis

We also investigated expressions of Morg1, PHD3, HIF1-a, HIF-

2a and the HIF-dependent genes Epo and VEGF164, employing

qPCR analysis. Our data unveil that Morg1 expression was

significantly up-regulated in WT mice after MCAO in the

ipsilateral hemisphere (po0.05, n¼6) compared with the

contralateral hemisphere (Fig. 6A). However, in the Morg1þ/�

mice Morg1 mRNA did not differ between the two hemi-

spheres (Fig. 6A), but it was significantly reduced in MCAO

contralaterally (n¼6) as well as in the ipsilateral hemispheres

(n¼6) relative to the WT animals (Fig. 6A). In addition,

expression of PHD3 was significantly decreased in Morg1þ/�

mice in both hemispheres compared to the WT mice (n¼6)

(Fig. 6B). The MCAO treatment only slightly, but not signifi-

cantly, augmented the expression of the PHD3 gene in the

ipsilateral hemispheres in Morg1þ/� as well as in WT mice

relative to the non-ischemic regions (Fig. 6B). We note here

that after MCAO there was no significant difference in mRNA

amounts of PHD3 between the non-ischemic hemisphere of

WT animals and the ischemic hemisphere of Morg1þ/� mice.

We next examined levels of HIF-1a and HIF-2a mRNA after

30 min MCAO. Both genes were significantly down-regulated

in the contralateral hemisphere of Morg1þ/� mice compared

b r a i n r e s e a r c h 1 4 8 2 ( 2 0 1 2 ) 2 2 – 3 1 23

with the WT animals (n¼6) (Fig. 6C and D). We detected an

induced expression of HIF-2a (n¼6) (Fig. 6D), but not of the

HIF-1a gene in the ipsilateral hemisphere of the WT mice. On

the other hand, in the Morg1þ/� mice the HIF-1a mRNA level

was reduced in the ipsilateral hemisphere relative to WT

(n¼6) (Fig. 6C), whereas an obvious difference between the

expression of the HIF-2a gene was not observed. Further-

more, we found that expression of the erythropoietin (Epo)

gene was significantly reduced in Morg1þ/� mice both in the

ischemic and in the non-ischemic hemisphere relative to WT

(n¼6) (Fig. 6E). Ischemic injury did not affect Epo mRNA levels

neither in Morg1þ/� nor in WT mice (Fig. 6E). Analysis of the

expression of the vascular endothelial growth factor 164

(VEGF 164) revealed that MCAO leads to a significant elevation

of VEGF 164 expression in ischemic hemispheres of both

Morg1þ/� (n¼6) and WT mice (n¼6) (Fig. 6F). Nevertheless,

the levels of VEGF 164 were found to be significantly lower in

Morg1þ/� mice compared with the WT animals (Fig. 6F).

3. Discussion

The major finding of this study is that Morg1þ/� mice are

partially protected from cerebral ischemia and reperfusion

subcort

ical

cort

ical

Morg1+/-

vWF DAPI

0.0

0.5

1.0

1.5

2.0

2.5

cortical subcortical

coro

nal are

a o

f vW

F+

MC

A b

ran

ch

es (

µm

2 x

10

3)

Morg1+/+, n = 4

Morg1+/-, n = 3

p = 0.40

p = 0.13

(ROI = 0.1mm2)

Morg1+/+

Fig. 1 – ((A) and (B)) Vascular architecture. (A) The vascular architecture of the MCA was immunohistochemically assessed

applying the endothelial marker vWF. Proximal and distal vessel branches displayed similar morphology in subcortical and

cortical brain areas of Morg1þ/� versus Morg1þ/þ (WT) mice, respectively. Nuclear counterstaining with DAPI facilitated

neuro-anatomic orientation. (B) Quantitative analysis of vWF positive vascularized areas on corresponding coronal brain

sections revealed similar results in Morg1þ/� and WT controls. Scale bar, 50 lm.

b r a i n r e s e a r c h 1 4 8 2 ( 2 0 1 2 ) 2 2 – 3 124

injury in comparison to the WT (Morg1þ/þ) animals. In

Morg1þ/� animals, MCAO induced less injury albeit laser

Doppler studies showed that the degree of CBF was similarly

restricted in both Morg1þ/� and WT mice. It seems unlikely

that in Morg1þ/� mice the effects on stroke size are caused

by differences in blood pressures (at least systolic blood

Fig. 2 – Cerebral blood flow. Changes in cerebral blood flow (CBF) during infarct induction were assessed in Morg1þ/� versus

WT mice. No significant differences were detected between the different genotypes.

* *

25

20

15

10

5

0

infa

rct siz

e / tota

l bra

in v

olu

me [%

]

Nissl staining Map2 staining

Morg1+/-WT

Fig. 3 – ((A) and (B)) Infarct volume. (A) The infarct volume was analyzed by Nissl staining and Map 2 immunohistochemistry

2 days after 60 min MCAO. Significantly smaller infarcts were detected in Morg1þ/� mice (grey bar; n¼12) compared to their

WT controls (black bar; n¼11; �pr0.05). (B) Coronal brain sections indicate that infarcts in Morg1þ/� mice were mainly

restricted to anterior striatal regions. Scale bar, 5 mm.

b r a i n r e s e a r c h 1 4 8 2 ( 2 0 1 2 ) 2 2 – 3 1 25

pressures), or a different vascular architecture between the

genotypes.

Brain ischemia leads to the induction of several genes

encoding a variety of proteins, which are mainly involved in

inflammatory and immune processes and in various cellular

metabolisms (Sieber et al., 2011; Lippoldt et al., 2005). Many of

these genes are transcriptionally stimulated by HIF-1 (Yi

et al., 2007). Hypoxic preconditioning protects the brain

against subsequent ischemia by inducing HIF-1a resulting in

an upregulation of HIF target genes including VEGF, Epo,

iNOS, glucose transporter-1, and many others (Ratan et al.,

2007). In healthy neurons during normoxia, little HIF-1a

protein is detected, because in the presence of oxygen and

2-oxoglutarate a group of enzymes called PHDs are activated

to hydroxylate 2 prolines on the HIF-1a protein resulting in

the initiation of a degradation pathway of this transcription

factor (Ratan et al., 2007). As the intracellular oxygen con-

centration falls below a critical threshold, PHDs fail to

hydroxylate HIF-1a leading to a stabilization and an increase

in its concentration. Emerging data indicate that inhibition of

all three PHD isoforms (PHD1–3) can promote cell survival in

the nervous system (Lomb et al., 2007; Ratan et al., 2007;

Siddiq et al., 2005). Although PHD2 appears to be the isoform

most important for tagging HIF-1a for degradation, PHD3 has

also been linked to neuronal survival (Lipscomb et al., 2001).

PHD3 was first identified by Freeman and coworkers as a

protein that is induced in neurons after growth factor

deprivation (Lee et al., 2005). Inhibition of PHD3 prevents

apoptosis of neurons by a HIF-1a dependent mechanism

(Lomb et al., 2007).

Recently, by using yeast two-hybrid assays we identified

Morg1 as a potential binding partner of PHD3 (Hopfer et al.,

2006). Morg1 has seven WD-repeat domains, and binding to

PHD3 occurs at the beginning of a b-sheet of strand b. PHD3

and Morg1 co-localize in the cytoplasm. Morg1 overexpres-

sion leads to an activation of PHD3 resulting in a decrease of

HIF-1a expression. In contrast, inhibition of Morg1 with

siRNA leads to a marked increase of HIF-1a activity (Hopfer

et al., 2006). However, more recently it was proposed that

PHD3 mainly regulates HIF-2a (Bishop et al., 2008). Although,

in accordance with these findings, we observed a reduction in

PHD3 expression in ischemic brain regions of Morg1þ/� mice

compared with WT animals, we did not find a compensatory

increase either in HIF-1a or in HIF-2a in the ischemic hemi-

spheres of Morg1þ/� mice. Previous studies have clearly

demonstrated that HIF-1a expression plays a role in stroke-

associated brain injury as well as recovery (Chen et al., 2010).

Morg1+/-

Morg1+/-

contralateral ipsilateral

contralateral ipsilateral

WT

(Morg1+/+)

WT

(Morg1+/+)

Fig. 4 – ((A) and (B)) Immunohistochemical staining for

Morg1 (A) and HIF-1a (B). (A) Morg1 positive astrocytes

were detected in the perilesional striatum of Morg1þ/� and

WT mice at the second day of reperfusion following 60 min

MCAO. (B). HIF-1a expression was revealed in both

hemispheres of Morg1þ/� and WT mice with the same

intensity. Scale bars, 50 lm.

WT

(Morg1+/+)

Morg1+/-

contralateral ipsilateral

Morg1+/-

contralateral ipsilateral

WT

(Morg1+/+)

Fig. 5 – ((A) and (B)) Immunohistochemical staining for PHD3

(A) and HIF-2a (B). (A) PHD3 expression was found

exclusively in the ipsilateral hemispheres of Morg1þ/� and

WT mice, with Morg1þ/� mice showing fewer positive cells

than WT animals. (B) No specific HIF-2a staining could be

detected in Morg1þ/� as well as in WT mice. Scale bars,

50 lm.

b r a i n r e s e a r c h 1 4 8 2 ( 2 0 1 2 ) 2 2 – 3 126

Likewise, it has been demonstrated that HIF-2a mediates the

transcriptional activation of Epo expression in astrocytes, and

this pathway promotes astrocytic paracrine-dependent neu-

ronal survival during ischemia (Chavez et al., 2006). The

differential expression of PHD1–3 has been incompletely

studied in the brain (Ran et al., 2005). Although we have

previously demonstrated that Morg1 interacts with PHD3

(Hopfer et al., 2006), we have not tested whether Morg1 may

also influence the activity of other PHD isoforms. One may

wonder why PHD2 does not compensate for the reduction of

PHD3 activity in Morg1þ/� mice in the ischemic brain. An

explanation may be the recent observation that both PHD2

and 3 mRNA expression is down-regulated after ischemia/

reperfusion (Schodeol et al., 2009). Thus, PHD2 may not

compensate for the reduced activity of PHD3 in this condi-

tion. In agreement with this hypothesis is our observation

that the non-ischemic brain area of Morg1þ/� animals did not

reveal an increase in HIF-1a and HIF-2a expression. It appears

from our studies that Morg1 expression negatively influences

infarct volume, apparently independently of the PHD3-HIF-1a

***

0

50

100

150

200

250

MCAO co

MCAO ip

si

MCAO co

MCAO ip

si

Mo

rg1

m R

NA

exp

ressio

n/

% o

f W

T M

CA

co

** ***

0

50

100

150

200

PH

D3

mR

NA

exp

ressio

n /

%W

T M

CA

co

*

*

**

MCAO co

MCAO ip

si

MCAO co

MCAO ip

si

MCAO co

MCAO ip

si

MCAO co

MCAO ip

si

Morg1+/-

Morg1+/+

*

MCAO co

MCAO ip

si

MCAO co

MCAO ip

si

200

HIF

-2α

mR

NA

exp

ressio

n/

% o

f W

T M

CA

co

0

50

100

150*

* **

0

50

100

150

200

*

HIF

-1α

m R

NA

exp

ressio

n/

% o

f W

T M

CA

co **

Morg1+/-

Morg1+/+

WT M

CAO co

WT M

CAO ip

si

MCAO co

MCAO ip

si0

50

100

150

200

250

VE

GF

16

4 m

RN

A e

xp

ressio

n/

% o

f W

T M

CA

co

*

**

*****

***

MCAO co

MCAO ip

si

MCAO co

MCAO ip

si0

50

100

150

200E

po

mR

NA

exp

ressio

n /

% o

f W

T M

CA

co *

*

*

Morg1+/-

Morg1+/+

Fig. 6 – ((A)–(F)) mRNA expressions of Morg1, PHD3, HIF1-a, HIF-2a Epo and VEGF 164 analyzed with qPCR (gene expression

was measured in mice at the second day after reperfusion following 60 min MCAO). (A) Expression of Morg1 in WT (black

columns) contralateral hemispheres (MCAO co), ipsilateral hemispheres (MCAO ipsi), Morg1þ/� mice (grey columns)

contralateral hemispheres (MCAO co), and Morg1þ/� mice ipsilateral hemispheres (MCAO ipsi). Morg1 was significantly

increased in Morg1þ/þ MCAO ipsi relative to Morg1þ/þ MCAO co (�po0.05, n¼6). Morg1 mRNA levels were found suppressed

in Morg1þ/� MCAO co versus Morg1þ/þ MCAO co (���po0.001), and Morg1þ/� MCAO ipsi versus Morg1þ/þ MCAO ipsi

(�po0.05, ���po0.001, n¼6). (B) Expression of PHD3. The level of PHD3 mRNA did not change after MCAO in Morg1þ/þ, or in

Morg1þ/� mice. Nevertheless, significantly lower amounts of PHD3 were detected in Morg1þ/� MCAO co versus Morg1þ/þ

MCAO co and Morg1þ/� MCAO ipsi versus Morg1þ/þ MCAO ipsi (�po0.05, ��po0.01, n¼6). (C) Expression of HIF-1a. HIF-1a

mRNA did not change after MCAO in Morg1þ/þ or in Morg1þ/� mice. A reduced expression of HIF-1awas detected in Morg1þ/�

MCAO co versus Morg1þ/þ MCAO co and Morg1þ/� MCA ipsi versus Morg1þ/þ MCA ipsi (�po0.05, ��po0.01, n¼6). (D)

Expression of HIF-2a mRNA level was significantly induced after MCAO in Morg1þ/þ, but not in Morg1þ/� mice (MCAO ipsi

versus Morg1þ/þ MCAO co). However, a significantly decreased expression of HIF-2a was detected in Morg1þ/� MCAO co

compared with Morg1þ/þ MCAO co, but not between Morg1þ/� MCAO ipsi and Morg1þ/þ MCAO ipsi (�po0.05, ��po0.01,

n¼6). (E) Erythropoietin (Epo) mRNA expression. Epo mRNA levels were not significantly affected after MCAO in Morg1þ/þ or

Morg1þ/� mice relative to non-ischemic contralateral hemispheres. However, a decreased Epo expression was found in

Morg1þ/� MCAO co versus Morg1þ/þ MCAO co and in Morg1þ/� MCAO ipsi versus Morg1þ/þ MCAO ipsi (�po0.05, n¼6). (F)

Expression of VEGF 164. mRNA levels significantly increased subsequently to MCAO in Morg1þ/þ (Morg1þ/þ MCAO ipsi versus

Morg1þ/þ MCAO co) as well as in Morg1þ/� mice. A significantly reduced expression of VEGF 164 was found in Morg1þ/�

MCAO co versus Morg1þ/þ MCAO as well as between Morg1þ/� MCAO ipsi and Morg1þ/þ MCAO ipsi (�po0.05, ��po0.01,

���po0.001, n¼6).

b r a i n r e s e a r c h 1 4 8 2 ( 2 0 1 2 ) 2 2 – 3 1 27

axis and the induction of HIF-1a-related genes such as Epo

and VEGF. In this regard, Morg1�/� embryos die at day 9.5

with severe neuronal development defects indicating a pivo-

tal role of Morg1 in brain development and remodeling (Wolf

et al., unpublished results).

The protective effects from reduced Morg1 expression are

not limited to the brain. We recently found that also renal

ischemia/reperfusion injury was reduced in Morg1þ/� mice

compared with the Morg1þ/þ animals (Hammerschmidt et al.,

2009).

Morg1 has also been identified as a binding partner of the

extracellular kinase (ERK) pathway scaffold protein MP1. It

interacts with MP1, Raf-1, MEK, and ERK, and stabilizes their

assembly into complexes (Vomastek et al., 2004).

Although we found no evidence that overexpression or

inhibition of Morg1 influences ERK activation (Hopfer et al.,

2006), it is possible that Morg1 serves as a link between HIF-1a

expression and ERK activation. Indeed, activation of the ERK

pathway has been found after hypoxia and reoxygenation in

astrocytes, and this mechanism contributes to neuronal

apoptosis in stroke (Szydlowska et al., 2010). A requirement

of the ERK-pathway for HIF-1a induction has been described

in neurons and a direct phosphorylation of HIF-1a by ERK 1,

and 2, leading to an increased half-life, has also been

reported (Li et al., 2005). However, the relationship of Morg1

to the ERK-signal cascade and its activation of PHD3 also

suggest an additional mechanism, for example that ERK-

activation leads to a change in PHD3-Morg1 assembly result-

ing in a decrease of PHD3 activity and a concomitant increase

in HIF-1a/HIF-2a. Further studies are necessary to test this

intriguing hypothesis.

We found an up-regulation of Morg1 expression in ipsilat-

eral astrocytes in the penumbra. Although this increase was

reduced in Morg1þ/� mice, it was not totally abolished. In

addition, similar findings were observed in humans in which

Morg1 is expressed in the naive brain, and we also observed

that Morg1 is down regulated following ischemic brain

damage. However, Morg1 expression was high in reactive

astrocytes adjacent to the ischemic brain areas (Haase et al.,

2009).

We have previously shown that angiotensin II down-

regulates PHD3 expression via AT2 receptors (Wolf et al.,

2002). Angiotensin II enhances HIF-1a expression by this

mechanism in neuronal PC12 cells (Wolf et al., 2004). Further-

more, it has been demonstrated that angiotensin II protects

against brain ischemia by interacting with AT2 receptors (Li

et al., 2005). Thus, one may propose a mechanism by which

suppression of PHD3 increases HIF-1a in the brain after

ischemia leading, via increased HIF-1a with the subsequent

induction of various growth factors, to enhanced neuronal

recovery. Certainly, the timing of HIF-1a expression may be

important (Yeh et al., 2011; Zhang et al., 2010). On the other

hand, brain ischemia also upregulates Morg1 expression

likely in astrocytes in the penumbra, thus increasing PHD3

stability and activity and thereby limiting availability of HIF-

1a. A major limitation of this first study is that we did not

investigate functional parameters such as motor or cognitive

function. Further studies are necessary to test whether the

reduced infarct volume in Morg1þ/� mice may transfer into

better motor or functional behavior of the animals.

4. Experimental procedures

4.1. Measurement of systolic blood pressure

Systolic blood pressure was measured using tail cuff plethys-

mography (TSE-Systems, Bad Homburg, Germany). Mice were

under slight anesthesia with ether. Measurements were

repeated three times (morning, noon and evening) with

n¼8 for Morg1þ/� and Morg1þ/þ. Mean values were calculated

from all measurements.

4.2. Induction of focal cerebral ischemia by MCAO

Morg1�/� mice with C57Bl/6 background were generated by

homologous recombination using standard techniques

(Hammerschmidt et al., 2009). Exons 1–5 of the mouse morg1

gene were disrupted with a cassette containing selection

markers and the gene for green fluorescence protein to detect

expression. Morg1þ/� mice revealed a normal phenotype

(Hammerschmidt et al., 2009).

MCAO was performed as previously described (Sieber et al.,

2010; Popp et al., 2009). In brief, 2 to 3 months old Morg1

heterozygous mice (Morg1þ/�) and control mice with the

same background (129/Bl6-K8tm1A7; designated as WT) were

anesthetized with 2.5% isoflurane in a N2O:O2 (3:1) mixture. A

total of 20 mice were used in each group. No mice died in the

sham-operated group, two mice died in the Morg1þ/� and one

animal in the Morg1þ/þ MCAO group. The right common

carotid artery (CCA), the external carotid artery (ECA) and

the internal carotid artery (ICA) were dissected from sur-

rounding tissue. A 7-0 nylon monofilament (70SPRe, Doccol

Corp, USA) was inserted into the ICA to occlude the MCA for

60 min. Cerebral blood flow was measured by using a laser

Doppler flowmetry (Peri Flux System 5000, Perimed, Sweden)

prior to and during MCAO. The body temperature of the mice

was maintained at 37 1C by a heating pad (rectal temperature

determination with a probe). Sham-operated animals under-

went the same procedure without occlusion of the MCA. The

investigators performing the experiments were unaware of

the genotype, and the phenotype of Morg1þ/þ and Morg1�/�

mice is indistinguishable. Mice were sacrificed 2 days after

MCAO. All animal procedures were approved by the local

government and complied with international and European

Union norms.

4.3. Sample preparation

For immunohistochemistry, anesthetized mice were fixed by

perfusion through the ascending aorta with 4% paraformal-

dehyde (PFA). Brains were removed and post fixed for

5 h in 4% PFA at 4 1C. After cryoprotection in 0.2 M

phosphate-buffered saline (PBS) containing 30% sucrose,

brains were frozen in methyl butane at �30 1C and stored at

�80 1C. Coronal sections were cut at 30 mm on a freezing

microtome (Microm International GmbH, Thermo Scientific,

Germany).

b r a i n r e s e a r c h 1 4 8 2 ( 2 0 1 2 ) 2 2 – 3 128

4.4. Analysis of cerebral vascularization

Brains obtained from nave WT and Morg1þ/� mice were

transcardially perfused as described above, and 16 mm cor-

onal sections from similar bregma levels were cut on a

cryotome and serially collected on glass slides. For immuno-

histochemical processing, slides were postfixed in 4% PFA (pH

7.4) for 20 min, washed in PBS and incubated with an

antiserum against von Willebrand factor (vWF; rabbit poly-

clonal, 1:500, Chemicon, Germany) containing 3% BSA/PBS

and 2% normal donkey serum (NDS) over night at 4 1C,

followed by incubation with appropriate Cy3 conjugated

secondary antisera for 1 h at room temperature. To avoid

unspecific antibody binding, slices were pre-incubated in 10%

NDS for 2 h. Tissue permeability was increased by supple-

mentation of 0.3% Triton X-100 to the solutions. For layer

orientation, specimens were counterstained with 40,6-diami-

dino-2-phenylindole dihydrochloride (DAPI).

In nave brains, the vascular area of proximal and distal MCA

branches was defined in corresponding perfusion regions on

adjacent slices from both hemispheres. On each slice, the area

of vWF positive MCA branches was calculated within a fixed

ROI of 0.15 mm2 using a semiquantitative imaging software

(Zeiss AutMess imaging program, Zeiss, Germany) supplemen-

ted by the Axiovision LE module (Zeiss). Vascularization was

separately assessed in cortical and subcortical regions of

Morg1þ/� and Morg1þ/þ groups. In each group, 3–4 animals

were analyzed applying a blinded protocol. For calculation,

vessels were pictured at 20� magnification.

4.5. Analysis of the infarct volume

Every 12th brain slice was processed for infarct volumetry.

Map 2 staining was performed as described below. For Nissl

staining slices were mounted on super frost slides (Menzel-

Glaser, Germany), air-dried and incubated in cresyl violet for

15 min (see above). Afterwards sections were dehydrated,

fixed and cover slipped with Neo-Mounts (Merck, Germany).

By using a CCD camera and the Scion Image software (NIH,

USA) the infarct area as well as the whole brain volume were

measured by an investigator unaware of the genotype of the

animals.

4.6. Immunohistochemistry

Free floating sections were treated with 0.8% H2O2 before

incubation with selected antibodies (Morg1, HIF-1a, HIF-2a,

PHD3 and Map 2, listed in Table 1) in TBS containing 3% NDS

and 0.2% Triton X-100 overnight at 4 1C. Sections were further

processed by the Vectastain Elite ABC Kit (Vector Labora-

tories, USA) using biotinylated secondary antibodies, listed in

Table 1. Finally, sections were incubated in 3,30-diaminoben-

zidine tetrahydrochloride (DAB, Sigma-Aldrich, Germany),

mounted on slides, air-dried, and cover slipped with Entellan

(Merck, Germany).

4.7. RNA isolation and quantitative PCR (qPCR)

MCAO was performed for 60 min, and the mice were sacri-

ficed 2 days after MCAO. Total RNA was isolated from

Table 1 – Antibodies for immunohistochemistry.

Primary antibodies IH Secondary antibodies IH

Morg1:

Abnova H00084292-M02,

Taiwan, 1:100

Donkey anti-mouse biotinylated

secondary antibody, Jackson

ImmunoResearch, USA, 1:600

HIF-1a:

R&D systems AF1935,

USA, 1:250

Donkey anti-goat biotinylated

secondary antibody,

Jackson ImmunoResearch, USA,

1:600

HIF2-a:

Abcam ab20654, USA,

1:100

Donkey anti-rabbit biotinylated

secondary antibody, Jackson

ImmunoResearch, USA, 1:600

PHD3:

Santa Cruz Biotechnology

sc-98792, USA, 1:100

Donkey anti-rabbit biotinylated

secondary antibody, Jackson

ImmunoResearch, USA, 1:500

Map2:

Clone AP-20: Sigma-

Aldrich M1406, USA,

1:1000

Donkey anti-mouse biotinylated

secondary antibody, Jackson

ImmunoResearch, USA, 1:500

Table 2 – Sequence and annealing temperature of the primers used for qPCR.

mGAPDH 50 0 TGTCAGCAATGCATCCTGCA 30 60 1C

50 ATGTCATCATACTTGGCAGGTT 30

mMorg1 50 CCTATCACCTGCACCTGCTT 30 58 1C

50 CACTTTCCCGTCTTCAGAGC 30

mPHD3 50 GCTATCCAGGAAATGGGACA 30 58 1C

50 GGCTGGACTTCATGTGGATT 30

mHIF-1a 50 TGAGCTTGCT CATCAGTTGC CAC 30 58 1C

50 TGTCCAGTTAGTTCAAACTGAGTTAACC 30

mHIF-2a 50 AAGCTCCTGTCCTCAGTCTG 30 58 1C

50 CATCCTCATGAAGAAGTCAC 30

Mepo 50 TGCCCGAACGTCCAACCCTGCTG 30 55 1C

50 TCACCTGTCCCCTCTCCTGCAG 30

mVEGF164 50 TTTACTGCTGTACCTCCACCATG 30 58 1C

50 TCACCGCCTTGGCTTGTCACAT 30

b r a i n r e s e a r c h 1 4 8 2 ( 2 0 1 2 ) 2 2 – 3 1 29

contralateral (non ischemic) and ipsilateral (ischemic) hemi-

sphere of the Morg1þ/� and WT mice using RNA-easy kit

(QIAGEN GmbH, Hilden, Germany). cDNAwas generated from

1 mg total RNA using M-MLV Reverse Transcriptase kit (Invi-

trogen, Karlsruhe, Germany) following the manufacture’s

protocols. The expressions of the genes of interest and

GAPDH mRNA were analyzed with monoplex analysis as

previously described (Bondeva et al., 2007; Livak and

Schmittgen, 2001). The expression levels were calculated

using the DDCT method (Livak and Schmittgen, 2001). The

mRNA expression level for each gene analyzed was normal-

ized to GAPDH and expressed in percent relatively to the WT

mRNA levels of the contralateral hemisphere. Primers used

for real-time PCR analysis of mRNA are presented in Table 2.

4.8. Data quantification and statistical analyses

The infarct sizes were measured and calculated as percen-

tage of the total brain volume. Statistical significance was

determined using Mann–Whitney U test (npr0.05) and sta-

tistical analysis was performed by SPSS (version 18, SPSS

GmbH Software, Munich, Germany). For comparison of more

than two groups (e.g., for real-time PCR data) the

Kruskall–Wallis test was used first followed by the

Mann–Whitney U test. Data are shown as mean7standard

error of the mean (SEM). A po0.05 was considered as

statistically significant.

r e f e r e n c e s

Balami, J.S., Chen, R.L., Grunwals, I.Q., Buchna, A.M., 2011.Neurological complications of acute ischemic stroke. LancetNeurol. 10, 357–371.

Bishop, T., Gallagher, D., Pascual, A., Lygate, C.A., de Bono, J.P.,Nicholls, L.G., Ortega-Saenz, P., Oster, H., Wijeyekoon, B.,Sutherland, A.I., Grosfeld, A., Aragones, J., Schneider, M., vanGeyte, K., Teixeira, D., Diez-Juan, A., Lopez-Barneo, J., Chan-non, K.M., Maxwell, P.H., Pugh, C.W., Davies, A.M., Carmeliet,P., Ratcliffe, P.J., 2008. Abnormal sympathoadrenal develop-ment and systemic hypotension in PHD3�/� mice. Mol. Cell.Biol. 28, 3386–3400.

Bondeva, T., Roger, T., Wolf, G., 2007. Differential regulation oftoll-like receptor 4 gene expression in renal cells by angio-tensin II: dependence of AP1 and PU1 transcriptional sites.Am. J. Nephrol. 27, 308–314.

Chavez, J.C., Baranova, O., Lin, J., Pichiule, P., 2006. The transcrip-tional activator hypoxia inducible factor 2 (HIF-2/EPAS-1)regulates the oxygen-dependent expression of erythropoietinin cortical astrocytes. J. Neurosci. 26, 9471–9481.

Chen, C., Ostrowski, R.P., Zhou, C., Tang, J., Zhang, J.H., 2010.Suppression of hypoxia-inducible factor-1a and its down-stream genes reduces acute hyperglycemia-enhanced hemor-rhagic transformation in a rat model of cerebral ischemia.J. Neurosci. Res. 88, 2046–2055.

Haase, D., Keiner, S., Mawrin, C., Wolf, G., 2009. Reduced Morg1expression in ischemic human brain. Neurosci. Lett. 455,46–50.

Hammerschmidt, E., Loeffler, I., Wolf, G., 2009. Morg1 heterozy-gous mice are protected from acute renal ischemia-reperfusion injury. Am. J. Physiol. Renal Physiol. 297,F1273–F1287.

Hopfer, U., Hopfer, H., Jablonski, K., Stahl, R.A.K., Wolf, G., 2006.The novel WD-repeat protein Morg1 acts as a molecular

scaffold for hypoxia-inducible factor prolyl hydroxylase 3(PHD3). J. Biol. Chem. 281, 8645–8655.

Kim, H.A., Mahato, R.I., Lee, M., 2009. Hypoxia-specific geneexpression for ischemic disease gene therapy. Adv. DrugDelivery Rev. 61, 614–622.

Lee, S., Nakamura, E., Yang, H., Wei, W., Linggi, M.S., Sajan, M.P.,Farese, R.V., Freeman, R.S., Carter, B.D., Kaelin, W.G., Sclisio, S.,2005. Neuronal apoptosis linked to EgIN3 prolyl hydroxylaseand familial pheochromocytoma genes: developmental cul-ling and cancer. Cancer Cell 8, 155–167.

Li, J., Culman, J., Hortnagl, H., Zhao, Y., Gerova, N., Timm, M.,Blume, A., Zimmermann, M., Seidel, K., Dirnagl, U., Unger, T.,2005. Angiotensin AT2 receptor protects against cerebralischemia-induced neuronal injury. FASEB J. 19, 617–619.

Lippoldt, A., Reichel, A., Moenning, U., 2005. Progress in theidentification of stroke-related genes: emerging new possibi-lities to develop concepts in stroke therapy. CNS Drugs 19,821–832.

Lipscomb, E.A., Sarmiere, P.D., Freeman, R.S., 2001. SM-20 is anovel mitochondrial protein that causes caspase-dependentcell death in nerve growth factor-dependent neurons. J. Biol.Chem. 276, 5085–5091.

Livak, K.J., Schmittgen, T.D., 2001. Analysis of relative geneexpression data using real-time quantitative pcr and the2(-delta delta c(t) method. Methods 25, 402–408.

Lomb, D.J., Straub, J.A., Freeman, R.S., 2007. Prolyl hydroxylaseinhibitors delay neuronal cell death caused by trophic factordeprivation. J. Neurochem. 103, 1897–1906.

Marti, H.J.H., Bernaudin, M., Bellail, A., Schoch, H., Euler, M., Petit,E., Risau, W., 2000. Hypoxia-induced vascular endothelialgrowth factor expression precedes neovascularization aftercerebral ischemia. Am. J. Pathol 156, 965–976.

Molina, C.A., 2011. Reperfusion therapies for acute ischemicstroke. Stroke 42 (Suppl. 1), S16–S19.

Popp, A., Jaenisch, N., Witte, O.W., Frahm, C., 2009. Identificationof ischemic regions in a rat model of stroke. PLoS One 4,e4764, http://dx.doi.org/10.1371/journal.pone.0004764.

Ran, R., Xu, H., Lu, A., Bernaudin, M., Sharp, F.H., 2005. Hypoxiapreconditioning in the brain. Dev. Neurosci. 27, 87–92.

Ratan, R.R., Siddiq, A., Smirnova, N., Krapisheva, K., Haskew-layton,R., McConoughey, S., Langley, B., Estevez, A., Sen, C.K., Gzaryan,I., Cho, S., Fink, M., Lamanna, J., 2007. Harnessing hypoxicadaptation to prevent, treat, and repair stroke. J. Mol. Med. 85,1331–1338.

Sieber, M.W., Guenther, M., Kohl, M., Witte, O.W., Claus, R.A.,Frahm, C., 2010. Inter-age variability of bona fide unvariedtranscripts normalization of quantitative PCR data inischemic stroke. Neurobiol. Aging 31, 654–664.

Sieber, M.W., Claus, R.A., Witte, O.W., Frahm, C., 2011. Attenuatedinflammatory response in aged mice brains following stroke.PLoS One 6, e26288.

Schodeol, J., Klanke, B., Weidemann, A., Buchholz, B., Berhardt,W., Bertog, M., Amann, K., Korbmacher, C., Wiesener, M.,Warnecke, C., Kurtz, A., Eckardt, K.U., Willam, C., 2009. HIF-prolyl hydroxylases in the rat kidney. Am. J. Pathol. 174,1663–1674.

Scholzke, M.N., Schwaninger, M., 2007. Transcriptional regulationof neurogenesis: potential mechanisms in cerebral ischemia.J. Mol. Med. 85, 577–588.

Siddiq, A., Ayoub, I.A., Chavez, J.C., Aminova, L., Shah, S.,LaManna, J.C., Patton, S.M., Connor, J.R., Cherny, R.A., Volitakis,I., Bush, A.I., Langsetmo, I., Seeley, T., Gunzler, V., Ratan, R.,2005. Hypoxia-inducible factor prolyl 4-hydroxlase inhibition. J.Biol. Chem. 280, 41732–41743.

Szydlowska, K., Gozdz, A., Dabrowski, M., Zawadzka, M.,Kaminska, B., 2010. Prolonged activation of ERK triggersglutamate-induced apoptosis of astrocytes: neuroprotectiveeffect of FK506. J. Neurochem. 113, 904–918.

b r a i n r e s e a r c h 1 4 8 2 ( 2 0 1 2 ) 2 2 – 3 130

Vomastek, T., Schaefferm, H.J., Tarcsafalvi, A., Smolkin, M.E.,Bissonette, E.A., Weber, M.J., 2004. Modular construction of asignaling scaffold: Morg1 interacts with components of theERK cascade and links ERK signaling to specific agonists. Proc.Nat. Acad. Sci. U.S.A. 101, 6981–6986.

Wolf, G., Schroeder, R., Stahl, R.A., 2004. Angiotensin II induceshypoxia-inducible factor-1 alpha in PC 12 cells through aposttranscriptional mechanism: role of AT2 receptors. Am. J.Nephrol. 24, 415–421.

Wolf, G., Harendza, S., Schroeder, R., Wenzel, U., Zahner, G.,Butzmann, U., Freeman, R.S., Stahl, R.A., 2002. AngiotensinII’s antiproliferative effects mediated through AT2-receptorsdepend on down-regulation of SM-20. Lab. Invest. 78, 59–71.

Yeh, S.H., Ou, L.C., Gean, P.W., Hung, J.J., Chnag, W.C., 2011.Selective inhibition of early-but not late-expressed Hif-1a isneuroprotective in rats after focal ischemic brain damage.Brain Pathol. 21, 249–262.

Yi, J.H., Park, S.W., Kapadia, R., Vemuganti, R., 2007. Role oftranscription factors in mediating post-ischemic cerebralinflammation and brain damage. Neurochem. Int. 50,1014–1027.

Zhang, X., Deguchi, K., Yamashita, T., Ohta, Y., Shnag, J., Tian, F.,Liu, N., Panin, V.L., Ikeda, Y., Matsuura, T., Abe, A., 2010.Temporal and spatial differences of multiple protein expres-sion in the ischemic penumbra after transient MCAO in rats.Brain Res. 1343, 143–152.

b r a i n r e s e a r c h 1 4 8 2 ( 2 0 1 2 ) 2 2 – 3 1 31