ORIGINAL ARTICLE

NF-jB activation and cell death after intracerebral hemorrhagein patients

Zeli Zhang • Yuguang Liu • Qibing Huang •

Yuxing Su • Yuan Zhang • Guanghui Wang •

Feng Li

Received: 2 December 2013 / Accepted: 23 January 2014

� Springer-Verlag Italia 2014

Abstract Nuclear factor-jB (NF-jB) plays an important

role in secondary damage after intracerebral hemorrhage

(ICH). We explored NF-jB activation and the relationship

between NF-jB and cell death in the perihematomal brain

tissue of patients after ICH. According to the interval

between onset of hemorrhage and specimen collection, 53

cases of patients with basal ganglia hemorrhage were

divided into six experimental groups: 0–6, 7–12, 13–24,

25–48, 49–96, and [96 h group. Brain tissues of the

experimental groups and control group were collected. IL-

1b, TNF-a, and NF-jB p65 expressions at the protein level

were detected by immunohistochemistry. Cell death was

detected by terminal deoxynucleotidyl transferase-medi-

ated dUTP-biotin nick end labeling (TUNEL) assay. All of

the detection items of immunohistochemistry and TUNEL

showed significant differences between the experimental

groups and control group. At the protein level, nuclear NF-

jB p65, IL-1b, and TNF-a achieved maximum values at

13–48, 0–24, and 13–48 h, respectively. Maximum cell

death was reached at 13–48 h. NF-jB activation increased

dramatically in perihematomal brain tissue after ICH. NF-

jB activation was closely related with cell death and had

an important function in secondary brain damage after ICH

in patients.

Keywords Intracerebral hemorrhage � NF-jB � IL-1b �TNF-a � Cell death

Introduction

Intracerebral hemorrhage (ICH) is one of the most common

diseases that threaten the health of the elderly [1]. It is the

least treatable form of stroke and is associated with high

mortality rates [2]. To date, no effective method has been

found to treat the neurological deficit caused by ICH [3, 4].

A series of pathophysiological changes in brain tissue

arise after ICH. Previous studies revealed that a large

number of inflammatory cells surround the hematoma in

the rat model of ICH and the inflammatory response is an

important mechanism of secondary brain damage after ICH

[5, 6].

Nuclear factor-jB (NF-jB) has been recognized as a

critical regulator of inflammatory responses since its dis-

covery [7]. In unstimulated cells, inactive NF-jB is

sequestered in the cytoplasm by inhibitory protein IjB.

NF-jB can be activated by a wide array of factors such as

TNF-a, IL-1, oxidative stress, and growth factors [8]. After

activation, the free NF-jB rapidly migrates into the

nucleus, binds to DNA, and promotes the transcription of

genes for the release of proinflammatory substances. So

NF-jB plays a major role in the inflammatory response [9].

Previous studies on ICH animal models indicate that

NF-jB activation increases after ICH and is closely related

to perihematomal cell death [10–14]. However, no sys-

tematic study has been conducted on NF-jB activation in

the perihematomal brain tissue of patients after ICH and on

the relationship between NF-jB activation and cell death.

In this study, we observed 53 ICH patients, studied the

clinical data and perihematomal brain tissue, and analyzed

Z. Zhang � Q. Huang � Y. Su � Y. Zhang � G. Wang

Department of Emergency Surgery, Qilu Hospital of Shandong

University, Jinan 250012, Shandong, People’s Republic of China

Y. Liu (&) � F. Li (&)

Department of Neurosurgery, Qilu Hospital of Shandong

University, Jinan 250012, Shandong, People’s Republic of China

e-mail: ns3000@126.com

F. Li

e-mail: feng.li@hotmail.com

123

Neurol Sci

DOI 10.1007/s10072-014-1657-0

NF-jB activation, TNF-a and IL-1b expressions, and the

relationship between NF-jB activation and cell death.

Materials and methods

Clinical data and grouping

A total of 53 patients hospitalized with basal ganglia

hemorrhage from November 2010 to November 2012 were

selected as the experimental cases. For all 53 cases, the

hematoma volume was 40 to 80 mL, and hematoma

evacuation operation was conducted along the non-func-

tional cortex. The patients in this study included 33 males

(62.26 %) and 20 females (37.74 %) aged 30–70 years,

with an average age of 56.7 years. According to the time

interval between the onset of hemorrhage and specimen

collection, the 53 cases were divided into six experimental

groups (Table 1). A total of eight patients with benign

disease and requiring neuroendoscope operations were

selected as the control group (Ctrl), including two cases of

intraventricular cyst, five cases of obstructive hydroceph-

alus, and one case of intraventricular meningioma. The Ctrl

group included five males and three females aged

30–69 years, with an average age of 54.3 years. Exclusion

criteria were established to eliminate factors that may

affect the study, including bleeding, inflammation, trauma,

surgery, use of drugs that affect the immune system (such

as ibuprofen, hormones, etc.), and the presence of under-

lying diseases within the previous month.

Ethics approval

The study protocol was approved by the ethics committee

of the hospital. All patients’ families received a

comprehensive description of the study and gave a written

informed consent for their relatives’ participation.

Specimen collection

Brain tissue located 1 cm away from the hematoma was

collected from the experimental groups, and tissue at the

junction of the gray and white matter was collected during

surgery from the Ctrl group. 0.5 cm3 of brain tissue was

collected per patient. The brain tissue was quickly fixed

with 10 % formalin and embedded in wax for immuno-

histochemistry and detection of cell death.

Immunohistochemistry

Immunohistochemical analysis was conducted using the

following primary antibodies: anti-NF-jB p65 antibody

(B7162 rabbit polyclonal, ANBO, USA), anti-IL-1b anti-

body (AP8531c rabbit polyclonal, ABGENT, USA), and

anti-TNF-a antibody (ab9579 mouse monoclonal, Abcam,

USA). Firstly, the tissue sections (4 lm thick) were de-

waxed, rehydrated, rinsed with distilled water and PBS,

repaired with EDTA, quenched with 3 % H2O2, exposed to

primary antibodies, and incubated at 4 �C overnight. Sec-

tions were then washed with PBS, incubated in polymer

helper for 25 min at room temperature, washed again, and

incubated with non-biotin rabbit/mice hypersensitivity two-

step secondary antibody (PV-9001/9002, GBI, USA) for

25 min at room temperature. Finally, the sections were

stained with diaminobenzidine–H2O2 solution, washed,

dehydrated in graded ethanol, immersed in xylene, and

covered with a coverslip.

The sections were observed under the light microscope

by five professors of pathology, and the positive cells and

cell types were identified. A total of five no-repeat fields

Table 1 Grouping of all cases and numbers of positive cells in immunohistochemistry and TUNEL detection

Group Time intervala Cases Immunohistochemistry TUNEL

Nucleus NF-jB p65 IL-1b TNF-a

A 0–6 h 8 35.4 ± 11.7� 50.7 ± 10.8b 27.7 ± 6.7� 28.5 ± 11.7�

B 7–12 h 14 54.4 ± 14.1# 48.2 ± 9.9 34.4 ± 9.9# 51.3 ± 13.3�

C 13–24 h 12 71.2 ± 16.7b 45.5 ± 12.7 43.5 ± 10.0 69.2 ± 13.3

D 25–48 h 8 64.4 ± 13.9 39.4 ± 7.5# 46.1 ± 12.9b 72.2 ± 15.0b

E 49–96 h 6 38.3 ± 13.8� 32.0 ± 9.2� 27.0 ± 8.9� 55.3 ± 13.2#

F [96 h 5 23.4 ± 10.0� 22.2 ± 7.1� 18.2 ± 7.3� 36.6 ± 11.8�

Ctrl 8 10.3 ± 3.2� 10.0 ± 3.7� 8.0 ± 3.4� 8.2 ± 3.7�

# P \ 0.05 and � P \ 0.01 indicate significant difference compared with the highest expression groupa The time interval between onset and specimen collectionb The highest expression group

Neurol Sci

123

(4009 high magnification) were randomly selected.

Nucleus NF-jB p65-positive cells and IL-1b and TNF-a-

positive cells were calculated, and the numbers of positive

cells in the five fields were recorded.

Terminal deoxynucleotidyl transferase-mediated

dUTP-biotin nick end labeling (TUNEL) assay

DNA fragment was detected with the TUNEL assay. Tis-

sue sections were deparaffinized and hydrated, and then

TUNEL assay was performed according to the instructions

provided by the manufacturer (In Situ Cell Death Detection

Kit, POD, Roche, Germany). After being incubated in the

TUNEL reaction mixture and rinsed with PBS, samples

were analyzed in a drop of PBS under a fluorescence

microscope (Olympus BX51, Olympus, Japan) by using an

excitation wavelength from 450 to 500 nm and detection

wavelength from 515 to 565 nm (green). After being

incubated in the DAB substrate solution, rinsed, and

mounted under glass cover slip, the slide was analyzed

under light microscope. Five no-repeat fields (4009 high

magnification) were randomly selected, and the numbers of

positive cells in the five fields were counted and recorded.

Statistical analysis

SPSS 13.0 statistical analysis software and Excel 2003

were used in the statistical analysis and charting. The

variables were shown as mean ± standard deviation. Stu-

dent’s t test and one-way ANOVA were used to evaluate

the results, and Pearson’s correlation coefficient to evaluate

the linear relationship between two variables. Statistical

significance was set at P \ 0.05.

Results

Immunohistochemistry detection of nucleus NF-jB

p65-positive cells, IL-1b, and TNF-a-positive cells

Immunohistochemical detection of the experimental groups

showed that NF-jB p65 was expressed in the nucleus of

Fig. 1 The microscopic images and column chart of NF-jB p65, IL-

1b, and TNF-a expression detected with immunohistochemistry.

Microscopic images (9400) of the experimental groups showed that

NF-jB p65 was expressed in nucleus of neurons and glial cells, while

IL-1b and TNF-a were expressed in the cytoplasm of neurons and

glial cells. The column chart showed the statistical results of the

nucleus NF-jB p65-positive cells, IL-1b-positive cells, and TNF-a-

positive cells in the experimental groups and the control group. Note

that the numbers of nucleus NF-jB p65-positive cells, IL-1b-positive

cells, and TNF-a-positive cells in each experimental group were all

significantly different from the control group, and the expression peak

hours were 13–48, 0–24 and 13–48 h separately. 1 ? indicates

positive-expression neurons, and 2 ? indicates positive-expression

glial cells. *P \ 0.001 indicates very significant difference compared

with the control group; #P \ 0.05 and �P \ 0.01 indicate significant

difference compared with the highest expression group (group C for

nucleus NF-jB p65-positive cells, group A for IL-1b-positive cells,

and group D for TNF-a-positive cells)

Neurol Sci

123

neurons and glial cells (Fig. 1), thus suggesting that NF-jB

was activated and migrated into the nucleus. Significant

differences were observed between each experimental group

and the Ctrl group (P \ 0.001). Nucleus NF-jB p65-positive

cells gradually increased over time after ICH and reached the

highest level in the 13–24 h group, compared with which the

25–48 h group did not show a significant difference

(P [ 0.05) and all the other experimental groups showed

significant differences (P \ 0.05). Thus, the expression peak

hours were 13–48 h (Fig. 1; Table 1).

Both IL-1b and TNF-a were expressed in the cytoplasm

of neurons and glial cells (Fig. 1), and significant differ-

ences were observed between each experimental group and

the Ctrl group (P \ 0.001). The number of IL-1b-positive

cells reached the highest level in the 0–6 h group, compared

with which the 7–12 and 13–24 h groups did not show

significant differences (P [ 0.05) and all the other experi-

mental groups showed significant differences (P \ 0.05).

Thus, the expression peak hours were 0–24 h. TNF-a-

positive cells gradually increased over time after ICH and

reached the highest level in the 25–48 h group, compared

with which the 13–24 h group did not show a significant

difference (P [ 0.05) and all the other experimental groups

showed significant differences (P \ 0.05), suggesting that

the expression peak hours were 13–48 h (Fig. 1; Table 1).

Time-dependent changes of cell death

Cell death was detected by the TUNEL assay. The sections

were observed under a fluorescence microscope (Olympus

BX51) and under an ordinary light microscope after DAB

color development. A large number of TUNEL-positive

cells were observed for the experimental group (Fig. 2a, b).

Significant differences were observed between the experi-

mental groups and Ctrl group (P \ 0.001). The number of

TUNEL-positive cells gradually increased over time after

ICH and reached the highest level in the 25–48 h group,

compared with which the 13–24 h group did not show a

significant difference (P [ 0.05) and all the other experi-

mental groups showed significant differences (P \ 0.05),

suggesting that the peak hours of cell death were 13–48 h

(Fig. 2c; Table 1).

Relationship between cell death and NF-jB activation

The activation peak hours of NF-jB and the expression

peak hours of IL-1b and TNF-a at the protein level were

13–48, 0–24, and 13–48 h respectively. The peak hours of

cell death were 13–48 h, thus suggesting a consistent var-

iation between cell death and NF-jB activation (Fig. 3).

Pearson’s correlation analysis revealed a strong linear

correlation between cell death and NF-jB activation

(P \ 0.0001) (Fig. 4).

Discussion

The pathophysiological changes of the perihematomal

brain tissue are complex after ICH and mainly include

Fig. 2 The microscopic images and column chart of cell death

detected with TUNEL assay. Images of the experimental groups

observed under fluorescence microscope (a 9400) and under ordinary

microscope after DAB color development (b 9400) all showed

TUNEL-positive neurons 1 and glial cells 2. c Column chart of

TUNEL-positive cells in the experimental groups and the control

group. Note that the numbers of TUNEL-positive cells in each

experimental group were all significantly different from the control

group, and the peak hours of cell death were 13–48 h. *P \ 0.001

indicates very significant difference compared with the control group;#P \ 0.05 and �P \ 0.01 indicate significant difference compared

with group D

Fig. 3 Dynamic trends of IL-1b, TNF-a, nucleus NF-jB p65-

positive cells, and TUNEL-positive cells. The results suggested that

the peak hours of the four were 13–48, 13–48, 0–24, and 13–48 h,

respectively. Solid symbols represent the peak hours of expression

Neurol Sci

123

primary and secondary damages. Primary damage is caused

by local brain tissue damage and compression on the sur-

rounding brain tissue, microvasculature, and distant struc-

tures. Reduction of primary damage after ICH is possible

by preventing rebleeding, dehydration, and hematoma

evacuation [15]. Secondary damage after ICH is mainly

caused by the reaction of the perihematomal brain tissue to

the products of hematoma, thus resulting in inflammation,

ischemia, and edema [3, 16, 17]. Secondary damage is

closely related to the outcome of ICH, so finding methods

to reduce the secondary damage around the hematoma is a

popular topic in recent research.

Ubiquitous transcription factor NF-jB is a critical reg-

ulator of inflammatory responses [7]. It has neuropatho-

logical and neuroprotective effects [18]. In ICH rat model

studies [9, 10, 19, 20], NF-jB activation can be detected

from 15 min to 2 h after ICH. Thereafter, activation will

gradually increase, reach its peak at 48 h, and then

decrease. The expression peak times of IL-1b and TNF-awere 3 and 48 h after ICH, respectively.

Previous studies were mostly limited to animal model

experiments. In this study, NF-jB activation and IL-1b and

TNF-a expressions after ICH were detected systematically.

The results showed that the nucleus NF-jB p65-positive

cells in each experimental group were all significantly

different from the Ctrl group, thus indicating that NF-jB

was activated, migrated into the nucleus, and was ready to

promote gene transcription for its downstream cytokines.

The activation peak hours of NF-jB were 13–48 h after

ICH. The expression peak hours of IL-1b and TNF-a at the

protein level were 0–24 and 13–48 h, respectively. All the

results were consistent with the results obtained in exper-

iments on ICH animal models.

Cell death is considered the direct cause of neurological

dysfunction and deterioration after ICH. In ICH rat models,

TUNEL-positive cells were observed in the perihematomal

brain tissue 8 h after ICH and a large number of TUNEL-

positive cells were observed 16 h later [11]. In studies on

patients with ICH, TUNEL-positive cells were found in the

perihematomal brain tissue 1, 2, and 5 days after ICH, and

mainly visible at the region 1.5 cm or farther from the edge

of the hematoma [21].

In this present study, the variation of cell death after

ICH in patients was detected with TUNEL assay. Signifi-

cant differences were found between the experimental

groups and the Ctrl group. TUNEL-positive cells were

found in the perihematomal brain tissue 6 h after ICH. The

TUNEL-positive gradually increased and achieved maxi-

mum values at 13–48 h. Thereafter, the TUNEL-positive

cells gradually decreased in number, but were still visible

96 h after ICH. A consistent variation was observed

between cell death and NF-jB activation. Pearson’s cor-

relation analysis also revealed a linear correlation between

cell death and NF-jB activation, thus indicating the close

relationship between them. Hickenbottom [11] found that

some cells were only positive for NF-jB activation, but

nearly all TUNEL-positive cells were also positive for

activated NF-jB. Previous studies also indicated the rela-

tionship between cell death and the expression of the

downstream cytokines of NF-jB, such as IL-1b and TNF-

a. In the NF-jB activated cells, downstream cytokine IL-

1b can result in brain edema and cell death [22–24] and

TNF-a can destroy the blood–brain barrier, increase brain

edema, and induce cell death [25].

The close relationship between NF-jB activation and

cell death after ICH was obvious. However, it is still

questionable whether NF-jB activation promotes cell

death or NF-jB is a compensatory mechanism to promote

cell survival. Some previous studies proved that NF-jB

activation promoted cell death [8, 12], whereas other

studies suggested that NF-jB activation inhibited cell

death [13, 14]. Clemens [26] indicated that sustained NF-

jB activation could induce nerve cell death, whereas

transient activation might be neuroprotective. To clarify

these issues further, studies are currently being conducted

in our research center.

Acknowledgments This study was supported by the National Nat-

ural Science Foundation of China (81301127) and the Natural Science

Foundation of Shandong Province, P.R. China (ZR2011HQ001).

References

1. Woo D, Broderick JP (2002) Spontaneous intracerebral hemor-

rhage: epidemiology and clinical presentation. Neurosurg Clin N

Am 13:265–279

2. Ciccone A, Pozzi M, Motto C, Tiraboschi P, Sterzi R (2008)

Epidemiological, clinical, and therapeutic aspects of primary

intracerebral hemorrhage. Neurol Sci 29(Suppl 2):S256–S257

Fig. 4 Correlation between TUNEL-positive cells and nucleus NF-

jB p65-positive cells. The scatter plot and the result of Pearson’s

correlation coefficient analysis suggested a positive linear relationship

between the two variables

Neurol Sci

123

3. Hwang BY, Appelboom G, Ayer A et al (2011) Advances in

neuroprotective strategies: potential therapies for intracerebral

hemorrhage. Cerebrovasc Dis 31(3):211–222

4. Sterzi R, Vidale S (2004) Treatment of intracerebral hemorrhage:

the clinical evidences. Neurol Sci 25(Suppl 1):S12

5. Xi G, Keep RF, Hoff JT (2006) Mechanisms of brain injury after

intracerebral haemorrhage. Lancet Neurol 5:53–63

6. Gong C, Hoff JT, Keep RF (2000) Acute inflammatory reaction

following experimental intracerebral hemorrhage in rat. Brain

Res 871:57–65

7. Barnes PJ, Karin M (1997) Nuclear factor-kappaB: a pivotal

transcription factor in chronic inflammatory diseases. N Engl J

Med 336:1066–1071

8. Ridder DA, Schwaninger M (2009) NF-jB signaling in cerebral

ischemia. Neuroscience 158:995–1006

9. Fang H, Wang PF, Zhou Y et al (2013) Toll-like receptor 4

signaling in intracerebral hemorrhage-induced inflammation and

injury. J Neuroinflammation 10:27

10. Zhang X, Li H, Hu S et al (2006) Brain edema after intracerebral

hemorrhage in rats: the role of inflammation. Neurol India

54(4):402–407

11. Hickenbottom SL, Grotta JC, Strong R et al (1999) Nuclear

factor-jB and cell death after experimental intracerebral hemor-

rhage in rats. Stroke 30:2472–2478

12. Hu YY, Huang M, Dong XQ et al (2011) Ginkgolide B reduces

neuronal cell apoptosis in the hemorrhagic rat brain: possible

involvement of Toll-like receptor 4/nuclear factor-kappa B

pathway. J Ethnopharmacol 137:1462–1468

13. Li WL, Yu SP, Chen D et al (2013) The regulatory role of NF-jB

in autophagy-like cell death after focal cerebral ischemia in mice.

Neuroscience 244:16–30

14. Song YS, Lee YS, Narasimhan P et al (2007) Reduced oxidative

stress promotes NF-jB-mediated neuroprotective gene expres-

sion after transient focal cerebral ischemia: lymphocytotrophic

cytokines and antiapoptotic factors. J Cereb Blood Flow Metab

27:764–775

15. Y.Nievas MC, Toktamis S, Haas E et al (2005) Benefits of

adapting minimal invasive techniques to selected patients with

spontaneous supratentorial intracerebral hematomas. Neurol Res

27(7):755–761

16. Yang GY, Betz AL, Chenevert TL et al (1994) Experimental

intracerebral hemorrhage: relationship between brain edema,

blood flow, and blood brain barrier permeability in rats. J Neu-

rosurg 81:93–101

17. Wagner KR, Xi G, Hua Y et al (1996) Lobar intracerebral

hemorrhage model in pigs: rapid edema development in perihe-

matomal white matter. Stroke 27:490–497

18. Wagner KR (2007) Modeling intracerebral hemorrhage: gluta-

mate nuclear factor-jB signaling and cytokines. Stroke

38:753–758

19. Zhao X, Zhang Y, Strong R et al (2007) Distinct patterns of

intracerebral hemorrhage-induced alterations in NF-jB subunit,

iNOS, and COX-2 expression. J Neurochem 101:652–663

20. Aronowski J, Hall CE (2005) New horizons for primary intra-

cerebral hemorrhage treatment: experience from preclinical

studies. Neurol Res 27:268–279

21. Qureshi AI, Suri MF, Ostrow PT et al (2003) Apoptosis as a form

of cell death in intracerebral hemorrhage. Neurosurgery

52(5):1041–1048

22. Holmin S, Mathiesen T (2000) Intracerebral administration of

interleukin-1beta and induction of inflammation, apoptosis and

vasogenic edema. J Neurosurg 92:108–120

23. Rothwell Nancy (2003) Interleukin-1 and neuronal injury:

mechanisms, modification, and therapeutic potential. Brain

Behav Immun 17:152–157

24. Denes A, Pinteau XE, Rothwell NJ et al (2011) Interleukin-1 and

stroke: biomarker, harbinger of damage, and therapeutic target.

Cerebrovasc Dis 32(6):517–527

25. Liu T, Clark RK, McDonnell PC et al (1994) Tumor necrosis

factor-a expression in ischemic neurons. Stroke 25(7):1481–1488

26. Clemens JA, Stephenson DT, Yin T et al (1998) Drug-induced

neuroprotection from global ischemia is associated with preven-

tion of persistent but not transient activation of nuclear factor-jB

in rats. Stroke 29:677–682

Neurol Sci

123