Neurochemical Research

, Volume 32, Issue 11, pp 1927–1931

Morphological Alteration of Golgi Apparatus and Subcellular Compartmentalization of TGF-β1 in Golgi Apparatus in Gerbils Following Transient Forebrain Ischemia


  • Zhiping Hu
    • Department of Neurology, Second Xiangya HospitalCentral South University
    • Department of Neurology, Second Xiangya HospitalCentral South University
  • Lesi Xie
    • Department of Anatomy, Xiangya School of MedicineCentral South University
  • Wei Lu
    • Department of Neurology, Second Xiangya HospitalCentral South University
  • Jie Zhang
    • Department of Neurology, Second Xiangya HospitalCentral South University
  • Ting Li
    • Department of Neurology, Second Xiangya HospitalCentral South University
  • Xiang Wang
    • Department of Neurology, Second Xiangya HospitalCentral South University
Original Paper

DOI: 10.1007/s11064-007-9382-1

Cite this article as:
Hu, Z., Zeng, L., Xie, L. et al. Neurochem Res (2007) 32: 1927. doi:10.1007/s11064-007-9382-1


Golgi apparatus (GA) is a very important organelle involved in the metabolism of numerous proteins. TGF-β1 plays an important role in supporting neuronal survival after ischemic insults. Little is known, however, about the morphological alteration of GA and subcellular compartmentalization of TGF-β1 in brain after ischemia. Therefore, our present study was designed to check for GA morphological alterations and TGF-β1 subcellular localization. GA immunoreactivities were examined in the somatosensory cortex of gerbils after 10 min transient forebrain ischemia. Confocal Immunofluorographs of TGF-β1 and TGN38 were also taken. Results indicated that no fragmentation of GA was found in gerbils of norm, shams and 6, 24 and 72 h postocclusion, but some of the cortical cells showed fragmentation of GA in gerbils 7 days postocclusion. TGF-β1 was colocalized with TGN38, a marker molecule for the GA. We conclude that there was morphological alterations of GA and TGF-β1 was present in GA in the somatosensory cortex after 10 min ischemia.


TGF-β1Golgi apparatus (GA)Gerbils


In neurons, Golgi apparatus (GA) is a very important organelle that is involved in the packaging and axoplasmic flow of numerous endogenous proteins and of exogenous macromolecules transported by the orthograde, retrograde and transsynaptic routes. Previous studies using GA-specific antibodies indicated that fragmentation of the organelle can be detected at a high frequency in both upper and lower motor neurons of patients with amyotrophic lateral sclerosis (ALS) and in mice models of ALS [1, 2], in nigral neurons with alpha-synuclein-positive inclusions in patients with Parkinson’s disease [3, 4], in the ballooned neurons in patients with corticobasal degeneration and Creutzfeldt-Jakob diseasem [5] and in patients with Alzheimer’s disease [6]. These findings led to the proposal that the GA may play an important role in the course of neurological diseases. The fragmentation could have detrimental effects on the secretory activity of the GA and might contribute to the alteration of neuronal activity. But till now, there has been no such study in brain after ischemia yet.

TGF-β1 is one of the most well known neuroprotective growth factors, which can be found in vitro and in vivo, predominantly as a precursor molecule in which the proregion is still associated with the mature molecule by noncovalent binding in a so-called latent, biologically inactive complex (LTGF-β). Its subcellular compartmentalization and release from neurons, however, are largely unknown. Few studies have addressed the subcellular localization of TGF-β. TGF-β1 immunoreactivity is found to be in hepatocyte and heart muscle cell mitochondria [7], lysosomes of hepatocytes [8], the endoplasmic reticulum (ER) and GA of fibroblasts [9] and HEL erythroleukemia cells [10], the mitochondria of normal T cells [11]. A substantial portion of TGF-β2 was also found to be secreted by the regulated secretory pathway in PC12 cells and hippocampal neurons [12]. However, the subcellular localization of TGF-β1 in brain are largely enigmatic.

In this study, we investigated morphology alteration of GA in gerbils models subjected to 10-min bilateral carotid artery occlusion (BCAO) by immunohistological method using antiserum against components of the trans-Golgi network, TGN38 [13]. We also studied intracellular compartmentalization of TGF-β1 by immunofluorescence colocalization of TGF-β1 and TGN38.

Materials and methods

Animals and surgery

All animal experiments were conducted in accordance with National Institutes of Health and institutional guidelines. Male Mongolian gerbils (Hanzhou, Zhejiang, China) weighing 80–90 g were housed in standard temperature (22 ± 1°C) and light-controlled (light on 07:00–20:00) environment with ad libitum access to food and water. The animals were randomly divided into nine groups. They were sham and BCAO and there were four time points (6, 24, 72 h and 7 days after surgery) for either of the treatment. Additionally, there were a group of gerbils without any treatment. There were eight animals in each group. Immediately before surgery, each subject was anesthetized with 5% halothane (70% N2O/30% O2). During surgery, halothane was lowered to 1.5–2%. A midline incision was made in the neck and surgical silk was loosely placed around isolated carotid arteries. Anesthesia was disconnected and atraumatic miniature aneurysm clips were attached to occlude both carotid arteries for 10 min. At the end of the occlusion period, blood flow was reestablished. In sham-operated animals, the arteries were exposed but not occluded. The wound was then sutured, and the animals were allowed to recover. The body temperature was maintained at 36–37°C during the surgery with a heating pad. After surgery, the animals maintained the environmental temperature at 28°C.

Immunohistochemical methods

At the time points, the animals were perfused transcardially with 30 ml of 0.9% saline followed by 100 ml of 4% paraformaldehyde solution. The brains were removed and placed in 10% Formalin for less than 24 h, processed, and embedded in paraffin wax. Coronal sections (5 μm) of somatosensory cortex were taken using a microtome. Immunohistochemical staining was performed on paraffin-embedded coronal sections (5 μm). Sections were deparaffinized in xylene and rehydrated in ethanol and distilled water. Sections were incubated in 0.3% hydrogen peroxide (H2O2) for 30 min to block endogenous peroxidase activity, rinsed, incubated in pepsin/HCl for 30 min, rinsed in water, and then rinsed in PBS. Normal goat serum (1.5% in PBS) was used as a preblock for 1 h, after which the sections were incubated with primary antibody TGN38 (Santa Cruz Biotechnology, Santa Cruz, CA) 1:200 overnight at 4°C in a humidified immunostaining chamber. After the incubation with primary antibodies, sections were rinsed in PBS (3 × 10 min) and incubated in biotinylated secondary antibody for 20 min at room temperature, after which they were again washed in PBS (3 × 10 min). Subsequently, sections were incubated in avidin-conjugated horseradish peroxidase for 20 min at room temperature and washed in PBS (3 × 10 min). Finally, sections were visualized by using 0.05% diaminobenzidine and 0.02% H2O2 until color developed and then washed in tap water, dehydrated, coverslipped using DPX mountant, and examined under a light microscope. A detailed examination of the immunostaining product observed in the cortex was made with a CDD cooling camera (Motic 5000, Germany) monitored with Motic Image Advanced 3.2 software (Motic, Germany).

Immunofluorescence confocal microscopy

Immunofluorescence double labeling was also performed on paraffin-embedded coronal sections (5 μm). Sections were deparaffinized , rehydrated and so on like above, but not incubated in 0.3% hydrogen peroxide (H2O2). Sections were incubated with the primary antibody TGN38 (1:200) overnight at 4°C, rinsed and then incubated with secondary antibody conjugated to fluorescein isothiocyanate (FITC) for 2 h. TGF-β1(Santa Cruz Biotechnology, Santa Cruz, CA) (1:200) antibody was used followed by and was revealed by a TRITC conjugated secondary antibody. Confocal images were obtained with a laser scanning confocal microscope (Zeiss LSM 510 META,Germany) through a 100 oil lens.


Morphological alteration of GA in gerbils models postocclusion

The layers III–VI of the somatosensory cortex were adequately and specifically immunostained with the anti-TGN38 antiserum, for cortical cells in layers III and VI of somatosensory cortex were exceptionally sensitive to ischemia [14]. We found that the GA of somatosensory cortex in gerbils models of norm, shams and 6, 24 and 72 h postocclusion showed normal profiles, that were in the form of larger, angular, or elongated profiles which filled the cell body and extended into the proximal segments of dendrites (Fig. 1). For cortical cells from gerbils survived 7 days postocclusion, most of the GA were normal, but some of the cortical cells showed fragmentation of GA, for which the organelle lost the normal network-like configuration and which were replaced by numerous disconnected smaller round elements (Fig. 2).
Fig. 1

The GA of somatosensory cortex in gerbils models of norm, shams and 6, 24 and 72h postocclusion showed normal profiles. Picture above showed a representative photograph of cortical cells with a normal network of the Golgi apparatus, which were immunostained with antiserum against TGN38 (×40)
Fig. 2

Anti-TGN38 immunostaining of somatosensory cortex in gerbils models 7 days postocclusion. Most cortical cells showed a normal network of the Golgi apparatus (small arrows), however some of the cells showed fragmentation and reduction in number of the Golgi apparatus (large arrows). The fragmented GA were not strongly stained compared to normal GA (left cells) (×40)

Confocal immunofluorographs of TGF-β1 and TGN38 in gerbils models postocclusion

The question whether TGF-β1 is colocalized with the TGN was addressed by a double-labeling approach. As a marker molecule for the TGN, a 38-kDa integral membrane protein (TGN38) was chosen. In all cortical cells, intracellular staining using an antibody for TGF-β1 and TGN38 revealed perinuclear staining and confocal images revealed an extensive overlap of TGF-β1 immunoreactivity (red) and TGN38 immunoreactivity (green) (Fig. 3).
Fig. 3

Representative photograph of immunofluorescence colocalization of TGF-β1 (red) and a marker molecule for the trans-Golgi network, TGN38 (green) in cortical cells. Confocal microscopy revealed areas of overlapping signals. Bars = 10 μm


The annual incidence of symptomatic stroke in the United States is estimated to be above 700,000 [15]. Cerebral vessel occlusions account for approximately 85% of all strokes, while the remaining 15% arise from intracerebral bleeding [16]. Though significant effort is under way to develop therapies for storke, there is still no promising form of therapy to maximize functional recovery and no neuroprotective agent has been proven to be valuable in the clinical setting. Recently, Golgi apparatus was found to play an important role in neurological disease. Fragmentation of the GA have been observed in many neurodegenerative diseases, such as ALS, Parkinson’s disease and Alzheimer’s disease [16]. Here, for the first time, we demonstrated that in gerbils models following 10 min transient forebrain ischemia, no fragmentation of GA were observed in gerbils models norm, shams and 6, 24 and 72 h postocclusion, but some of the cortical cells showed fragmentation of GA in gerbils survived 7 days postocclusion. It was possible that fragmentation of the GA was chronic and needed quit a long time to occur after acute ischemic insults. If gerbils survived longer, more cortical cells with fragmented GA may be found. We will further study the morphological alteration of GA in gerbils, which are ischemia longer or survive longer postocclusion. Further studies measuring the GA protein or gene expression will also be taken to further clarify the alteration of GA after ischemic insults.

The mechanisms underlying fragmentation of the GA remain to be resolved. Cerebrospinal fluid from amyotrophic lateral sclerosis patients caused fragmentation of the Golgi apparatus in the neonatal rat spinal cord and it was suggested that the putative toxin(s) present in ALS-CSF may cause impairment in the GA [17]. Some researches indicated that Golgi complex disassembly involved MAPK and PKA [18] and fragmentation of the Golgi apparatus was an early apoptotic event independent of the cytoskeleton [19]. Fragmentation of GA was also found in gerbils survived 7 days postocclusion in our research. It was supposed that, firstly, once neurons experienced energy failure after ischemic insults, calcium would accumulate in the intracellular space as a result of disturbed ion homeostasis. This, in turn, activated many cellular processes, which culminated in GA fragmentation. Secondly, many proteases, kinases and lipases were activated directly or indirectly by the ischemic insult, which would further result in the fragmentation of GA. Then, it was found that the Golgi apparatus were early targets for extensive tyrosine nitration in a model of hypoxia-asphyxia, leading to the damage to the Golgi apparatus [20]. Finally, it was reported that transient cerebral ischemia caused severe protein aggregation in neurons and the aggregates were associated with the GA. The accumulation of protein aggregates would cause fragmentation of GA after ischemic insults [21]. Additionally, some investigations also implied that lipid peroxidation and the vasogenic and cytotoxic brain edema induced by ischemia may be causally related to the disturbance in Golgi structure.

Many organelles were found to play important roles in the pathology and pathogenesis of neurological disease as investigations furthered into subcellular levels. In neurodegenerative diseases, GA was supposed to be an early target of the pathological process which initiated neuronal degeneration [1]. In our experiments, no fragmentation of the GA has been found in the first 3 days postocclusion as suggested above, that could have important functional effects helping gerbils surviving from acute ischemic insults. For in neurons, all newly synthesized proteins destined for fast axoplasmic transport pass through the GA. But when gerbils survived 7 days postocclusion, some of the cortical cells showed fragmentation of GA. It was suggested that fragmentation of a structure with such crucial roles in the cellular handling of proteins was likely to be associated with significant impairment of function and may contribute to the development of irreversible injury by interfering with the normal maintenance of plasma membranes and axonal transport. Thus, the structural abnormalities in the GA may participate in functional changes critical to irreversible neuronal injury following cerebral ischemia and novel approaches targeting at GA in management of cerebral ischemia may be of great importance. Of course, extensive research efforts are needed to clarify the contribution of GA dysfunction to cerebral ischemia and its pathophysiological basis, which will certainly significantly impact our ability to develop more effective therapies for ischemic stroke.

TGF-β1 is a multifunctional cytokine triggering different physiological situations including: cell cycle control, haematopoiesis control, cell differentiation, angiogenesis, induction of apoptosis, and formation of extracellular matrix [22]. It also plays an important role in supporting neuronal survival after ischemia [23]. Several studies have investigated the subcellular localization of TGF-β1 in nonneural cells, such as hepatocyte, heart muscle cell, fibroblasts, HEL erythroleukemia cells and normal T cells [711]. Its subcellular compartmentalization and release in the cortical cells, however, are largely unknown. Here we demonstrated for the first time that TGF-β1 colocalized with the trans-Golgi network marker TGN38 in the cortical cells, which indicated that TGF-β1 was presented in the Golgi apparatus.

It is well known that the GA plays a important role in the transport, processing and targeting of numerous proteins destined for secretion, plasma membrane and lysosomes. We clearly demonstrated the presence of TGF-β1 in the GA, suggesting that both continuous protein synthesis and transportation via the Golgi complex were required for TGF-β1 secretion. The findings indicated that TGF-β1 passed through the GA after mRNA transcription in the nucleus and translation in the endoplasmic reticulum, before secretion in the cortical cells. After ischemic insults, TGF-β1 levels were enhanced and such an upregulation was beneficial [23]. Fortunately, just as we studied above, there was no obvious fragmentation of GA observed in the first few days after ischemia, which could help gerbils surviving greatly. It was found that treatment of neurons with high potassium can liberate biologically active TGF-β by GA [12]. Thus, approaches targeting at GA such as high potassium may can be adopted to improve the function of GA and make the struggle to survive easier for gerbils after ischemic insults.

The mechanism of TGF-β1-mediated neuroprotection remains to be resolved, but there was evidence suggested that TGF-β1 regulated the expression and ratio of apoptotic (Bad) and antiapoptotic proteins (Bcl-2, Bcl-x1), creating an environment favorable for cell survival [24]. It was found that hepatocytes from TGF-β1 knock out mice showed increased numbers of hypertrophied or enlarged Golgi complexes [25], suggesting that TGF-β1 may be required for regulation of subcellular organelles in hepatocytes and may be essential for physiological functions involving GA. Here, we found the presence of TGF-β1 in the GA in cortical cells, which indicated that TGF-β1 may have significant regulatory functions within the GA after ischemic insults, distinct from its many actions mediated by its external cell membrane receptor system. Understood the potential intracellular functions of TGF-β1 would further clarify the mechanism of the neuroprotection of TGF-β1 and may significantly improve our therapy in stroke.

In summary, our research have for the first time demonstrated the morphological alteration of GA in the gerbils models postocclusion. Our data further provided evidence that TGF-β1, a prominent TGF-β isoform, was present in GA. The role of GA in stroke and potential intracellular functions of TGF-β1 after ischemic insults need further evaluation.

Copyright information

© Springer Science+Business Media, LLC 2007