Liposome-mediated NGF gene transfection following ... - Nature

Gene Therapy (1999) 6, 994–1005  1999 Stockton Press All rights reserved 0969-7128/99 $12.00 http://www.stockton-press.co.uk/gt

Liposome-mediated NGF gene transfection following neuronal injury: potential therapeutic applications LL Zou1,2, L Huang3, RL Hayes4, C Black4, YH Qiu4, JR Perez-Polo5, W Le6, GL Clifton4 and K Yang1,2 1

Department of Neurosurgery, 2Center for Cell and Gene Therapy, 6Department of Neurology, Baylor College of Medicine, Houston, TX; 3Department of Pharmacology, University of Pittsburgh, Pittsburgh, PA; 4Department of Neurosurgery, University of Texas Health Science Center at Houston, Houston, TX; and 5Department of Human Biochemistry and Genetics, University of Texas Medical Branch at Galveston, Galveston, TX, USA

We have systematically investigated the therapeutic potential of cationic liposome-mediated neurotrophic gene transfer for treatment of CNS injury. Following determination of optimal transfection conditions, we examined the effects of dimethylaminoethane-carbamoyl-cholesterol (DC-Chol) liposome-mediated NGF cDNA transfection in injured and uninjured primary septo-hippocampal cell cultures and rat brains. In in vitro studies, we detected an increase of NGF mRNA in cultures 1 day after transfection. Subsequent ELISA and PC12 cell biological assays confirmed that cultured cells secreted soluble active NGF into the media from day 2 after gene transfection. Further experiments showed that such NGF gene transfection reduced the loss of chol-

ine acetyltransferase (ChAT) activity in cultures following calcium-dependent depolarization injury. In in vivo studies, following intraventricular injections of NGF cDNA complexed with DC-Chol liposomes, ELISA detected nine- to 12-fold increases of NGF in rat CSF. Further studies showed that liposome/NGF cDNA complexes could attenuate the loss of cholinergic neuronal immunostaining in the rat septum after traumatic brain injury (TBI). Since deficits in cholinergic neurotransmission are a major consequence of TBI, our studies demonstrate for the first time that DCChol liposome-mediated NGF gene transfection may have therapeutic potential for treatment of brain injury.

Keywords: liposomes; NGF; gene therapy; neuronal injury; ChAT; CNS

Introduction Investigators studying neuronal injury have long known that administration of neurotrophins could have important therapeutic potential for intervention in the pathological responses to injury and enhancement of functional recovery.1–10 Recent studies have shown the benefits of administering of exogenous neurotrophic factor to treat traumatic brain injury (TBI).11,12 Although the administration of exogenous neurotrophic proteins has therapeutic potential, the limitations imposed by protein degradation and the blood–brain barrier could restrict the clinical utility of this approach. Gene transfer is an alternative means of introducing neurotrophins into the central nervous system (CNS). During the past few years, significant progress has been made in the development of techniques for transfecting genes into the CNS and exploring their potential to treat CNS disorders.13–22 Cationic liposomes, which can condense DNA and increase transfection efficiency both in vitro and in vivo,23–29 have several attractive features as vectors for gene transfer. First, cationic liposomes are non-immunogenic and nontoxic at therapeutic doses. Second, cationic liposomes as DNA carriers can transfect postmitotic, non-dividing cells Correspondence: K Yang, Department of Neurosurgery and Center for Cell and Gene Therapy, Baylor College of Medicine, One Baylor Plaza, Houston, TX 77030, USA Received 5 November 1998; accepted 5 February 1999

including neurons. Third, cationic liposomes can be used to deliver multiple genes of any type (linear or supercoiled) nucleic acids. Finally, cationic liposomes are relatively simple to prepare and can be administered into the body by several different routes. Recent advances in the development of more efficient liposome formulations have raised the possibility that liposome-mediated neurotrophic gene transfection may prove useful for treating CNS injury.30 Dimethylaminoethane-carbamoyl-cholesterol (DCChol) liposomes have a relatively high transfection efficiency and low toxicity.28 Recently, DC-Chol liposomes were approved by the FDA for a clinical trial for cancer immunotherapy. However, no studies have systematically examined whether DC-Chol liposomes could carry neurotrophic genes into brain cells. Neurotrophins, nerve growth factor (NGF) for example, are critical for brain recovery structurally and functionally after injury. In studies regarding traumatically injured brain, the septo-hippocampal cholinergic system is important since it remains a target of damage and dysfunction.31 Exogenous administration with NGF has been shown to spare septal cholinergic neurons from the injury-induced death and degeneration.1–3 As we know, the CNS is less accessible to systematic administration of protein and multiple direct injections are inconvenient. It is therefore of special interest to develop an alternative method by which sustained and sufficient administration of neurotrophin can be achieved. Our laboratory previously has found that

Liposome-mediated NGF gene transfer for neuronal injury LL Zou et al

lipofectin-mediated transfection of NGF cDNA caused NGF expression and could enhance choline acetyltransferase (ChAT) activity in primary septo-hippocampal cell cultures.32 We have also transfected a reporter gene into the rat spinal cord employing DC-Chol liposomes.33 The goal in the present study was to transfer the gene for NGF into rat brain cells using DC-Chol liposomes and to study the effects on neuronal recovery from injury. Our results suggest that DC-Chol liposomemediated NGF gene transfection have therapeutic potential to treat traumatic brain injury.

Results In vitro studies DC-Chol liposome-mediated ␤-galactosidase gene transfection in primary septo-hippocampal cell cultures: To test the efficiency and potency of the expression vector, we used X-gal staining in septo-hippocampal primary cell cultures which were transfected using DC-Chol liposomes complexed with pCMV/␤-gal plasmid DNA. Two days after liposome-mediated pCMV/␤-gal transfection, X-gal staining was detected in a number of cells (Figure 1a). As we observed with lipofectin,34 the ratio of nucleic acid to DC-Chol liposomes during transfection is critical for optimizing transfection efficiency in cultured cells. Employing different ratios of pCMV/␤-gal cDNA to DCChol liposomes, ␤-galactosidase transfection efficiency was calculated from X-gal staining in septo-hippocampal cultures 2 days after transfection. In vitro studies suggested that highest transfection efficiency was seen in employing a 1:3 (␮g DNA/DC-Chol liposome ␮l) transfection ratio and less efficient transfection was observed with a 1:1 transfection ratio (Figure 1b). The differential efficiencies associated with varying ratios of cDNA to liposomes are consistent with the view that the higher the net positive charge of DNA–liposome complexes, the better the interaction with a negatively charged cell membrane.34 However, an excess of DC-Chol liposomes will also decrease the transfection efficiency (Figure 1b). To study the temporal profile of transgene expression levels, we employed ␤-galactosidase activity assays to evaluate the ␤-gal activity following transfection in septohippocampal cultures with pCMV/␤-gal cDNA and DCChol liposomes. We found that expression of ␤-galactosidase activity was highest at 3 days after transfection and gradually decreased for up to 1 month (Figure 1c). DC-Chol liposome-mediated NGF gene transfection in primary septo-hippocampal cell cultures: Using the optimal ratio determined above, we transfected the septo-hippocampal cultures with 1 ␮g NGF cDNA complexed with 3 ␮l DC-Chol liposomes. To determine mRNA expression after gene transfection, we conducted RT-PCR analyses and confirmed increases of mRNA for NGF 1 day after transfection (Figure 2a). To measure the induced NGF protein levels, ELISA analyses were used to detect increases in NGF protein in culture media 2, 4 and 8 days after transfection (Figure 2b). Since we added fresh media every time conditioned media were collected, the results suggest that transfected cells continued to secrete NGF for at least 8 days after transfection. Our results also suggest that incubation time could significantly affect trans-

gene expression levels. The 12-h incubation period had significantly higher NGF expression levels than the 6-h period (Figure 2b). Rat pheochromocytoma (PC12) cells were used to assay the specific biological activity of the NGF in the medium of transfected cell cultures. We observed a prominent neurite outgrowth of PC12 cells following treatment for 36 h with media conditioned by septo-hippocampal cultures with NGF gene transfection. NGF (20 ng/ml) was added to sister wells as a positive control. The media from NGF transfected cells produced biological effects similar to the NGF isolated from mouse submaxillary glands. As expected, the media from control cell cultures, incubated only with liposomes, had no neurotrophic effects (Figure 3). The neurotrophic effect of the conditioned media could be blocked by treatment with antibodies specific to NGF (data not shown).

DC-Chol liposome-mediated NGF gene transfer can enhance recovery of ChAT mRNA and activity following calcium-dependent depolarization injury: To determine whether the transfection of neurotrophin cDNA/DCChol liposome complexes into CNS cells can enhance cholinergic neuronal transmission, we used septo-hippocampal primary cell cultures for our in vitro studies. We have previously observed significant increases in ChAT activity without significant changes in the numbers of ChAT-positive cells in septo-hippocampal cultures following liposome-mediated NGF gene transfection.32 This result suggests that increased ChAT activity is most likely due to the induction of de novo ChAT expression rather than the increase in number of cholinergic septal neurons. In this study, we further confirmed that NGF gene transfection can enhance the de novo ChAT expression by detecting increases in ChAT mRNA levels. Two days after DC-Chol liposome-mediated NGF cDNA transfection, transfected cultures showed significant increases in ChAT mRNA in transfected cultures (data not shown). To examine potential therapeutic effects, we applied the same transfection paradigm to traumatized septo-hippocampal cell cultures.35 The trauma, which is induced by calcium-dependent potassium-mediated depolarization injury, caused a significant decrease in ChAT activity without causing significant cell death (Figure 4). However, in cells which were transfected after injury, ChAT activity levels were significantly higher than the non-transfected, injured group (⬎89%; Figure 4). Similar protection could also be produced by adding purified NGF in the culture medium (⬎64%; Figure 4), while adding liposome per se was without any effect (data not shown). These results suggest that liposome-mediated NGF gene transfection can attenuate injury-induced neurotransmission deficits in cholinergic neurons. In vivo studies DC-Chol liposome-mediated gene transfer can be achieved in rat brain by direct injection: To extend the above in vitro studies to our in vivo model, we initially applied stereotactic injection techniques combined with cDNA for alkaline phosphatase (AK), a reporter gene. Thirty-six ␮l liposomes or liposomes complexed with 12 ␮g AK cDNA was injected into rat cerebral lateral ventricle. At 3 and 7 days after injection, rats were killed and their brains sectioned for histochemical staining. We did not observe

995


996

A

Figure 1 ␤-Galactosidase transfection in vitro. (A) Primary cultures of hippocampal and septal neurons were prepared (see Materials and methods) and transfected with pCMV/␤-gal cDNA employing a ratio of 1 ␮g DNA to 3 ␮l of DC-Chol liposomes. X-gal staining was detected 2 days after transfection. Blue staining is evident in the transfected cells under the light microscope (a: 110×; b: 220×). (B) The transfection efficiency was calculated from Xgal staining at six different transfection ratios of liposome to DNA (1:1 to 1:6/DNA ␮g: DC-Chol ␮l). Cell counts were conducted 2 days after transfection. The values represent the average percentage of X-gal stained cells in each well (± s.e.m.). The data indicate that 1 ␮g DNA to 3 ␮l DCChol liposomes transfection efficiency was higher than observed using other DNA to liposome ratios. (C) ␤-Galactosidase activity in transfected septohippocampal cell cultures. The ␤-galactosidase activity was analyzed by a spectrophotometer. The results indicated that ␤-galactosidase activity was highest at 3 days after transfection and was elevated for up to 1 month.

any AK activity at 3 days after injection. However, we found strong AK expression in ependymal membrane tissue as reflected by intensely stained patches at 7 days after transfection (Figure 5A). Furthermore, the intraventricular injections did not cause any pathological cellular morphology changes as observed using light microscope. With the same paradigm, we injected the DC-Chol liposome/NGF cDNA complex into the lateral cerebral ventricle. At 3, 7 and 14 days after transfection, ELISA demonstrated that NGF in CSF levels was significantly (P ⬍ 0.01) increased at 7 and 14 days, with the levels of NGF nine- to 12-fold higher than the control, respectively (Figure 5b).

DC-Chol liposome-mediated NGF gene transfection can attenuate the cholinergic neuronal injury after traumatic

brain injury: To determine the effect of NGF gene transfection on cholinergic neurons in the traumatically injured brain, we transferred the NGF cDNA into the injured rat brain. One day after unilateral cortical impact injury, DCChol liposomes complexed with NGF cDNA were injected into the lateral cerebral ventricle. Following a 3-, 7- or 14-day survival period, rats were perfused and immunohistochemical procedures were performed to detect ChAT immunoreactivity in the medial septum. The unilateral traumatic injury significantly reduced (P ⬍ 0.05) the number of ChAT-positive neurons at all time-points in the side ipsilateral to injury and at 7 and 14 days contralaterally (Figure 6A). However, NGF gene transfection significantly reversed (P ⬍ 0.05) the decrease in the number of ChAT-positive neurons in the septal region compared with sham-injured controls at 7 and 14


997

Figure 1 Continued.

days. Injured rats displayed 52% to 55% ChAT-positive neuronal loss ipsilateral to injury and 30% to 41% loss contralaterally, while NGF cDNA transfection significantly (P ⬍ 0.05) increased ChAT-positive neuronal cell immunostaining by 76% to 105% ipsilaterally and 38.8% to 55.2% contralaterally (Figures 6A and B).

Discussion Recently, significant efforts have been made to develop gene transfer methods for treating CNS injury and disorders.13,14,16,17,21,22,36–38 Cationic liposome-mediated gene transfer is non-immunogenic and non-toxic at therapeutic doses.33,39 This paper is the first systematic report using DC-Chol liposome-mediated neurotrophic gene transfer to treat neuronal injury in in vitro and in vivo models of CNS injury. We examined the effects of DC-Chol liposome-mediated NGF gene transfection in primary septohippocampal cell cultures and in traumatically injured rat brains. In in vitro studies, we detected an increase of NGF mRNA and protein in post-transfection cultures. PC12 cell biological assays demonstrated this transgene product had the same neurotrophic properties as endogenous NGF. We also found that NGF gene transfection could attenuate injury-induced loss of ChAT activity in cultures. In in vivo studies, we observed that transfection occurred in rat ventricular ependymal tissue following

Figure 2 NGF gene transfectio in vitro. (a) Increases in NGF mRNA in septo-hippocampal cell cultures 1 day after DC-Chol liposome-mediated transfection of NGF cDNA detected by RT-PCR (see Materials and methods). Agarose gel was stained with ethidium bromide. The upper band (460 bp PCR product) is the NGF target band and the 210 bp PCR product band is the actin band. Under RT-PCR heading are the results from RT-PCR. (I) injured group; (C) control group; (N) cultures transfected with NGF; (L) cultures treated with liposomes alone. In non-NGF transfection cultures (L) showed basal levels of NGF mRNA from RT-PCR. In cultures treated with DC-Chol liposomes complexed with pCMV/NGF cDNA (N), the NGF band was dramatically increased 1 day after transfection. We also see no difference in DC-Chol liposome-mediated NGF transfection between injured (I) and uninjured (C) cultures. Under the PCR (RNA) heading, the gel shows no band in NGF PCR product, which suggests no DNA contamination in RNA preparation. Marker: 100 bp DNA ladder (Gibco BRL, Grand Island, NY, USA). (b) ELISA analysis of NGF protein in media from septo-hippocampal cell cultures after transfection with pCMV/NGF cDNA complexed with DC-Chol liposomes. Two days after NGF gene transfection, NGF protein was dramatically increased. There was a statistically significant difference between the transfected cultures and controls in all the time-points we studied (from 2 to 8 days; P ⬍ 0.01; n = 4). Increased NGF secretion following gene transfection persisted for at least 8 days. Also, 12 h liposome–DNA complex incubation time resulted in higher levels of NGF production than 6 h incubation (P ⬍ 0.01; n = 4). The values are mean ± s.e.m. from four different wells.

intraventricular injections of AK reporter genes. Moreover, using the same paradigm, we transfected neurotrophic genes into the rat lateral ventricles. ELISA detected nine- to 12-fold increases of NGF in rat CSF following intraventricular injections. Further experiments showed that intraventricular injections of DC-Chol liposome/NGF cDNA complexes could attenuate the loss of cholinergic neuronal immunostaining in the rat septum after traumatic brain injury. Our results further suggest that optimizing the cationic liposome/DNA ratio for neurotrophic gene transfer is


998

Figure 4 NGF gene transfection increases ChAT activity in injured cell cultures. ChAT activity assays showed DC-Chol liposome-mediated NGF DNA transfection significantly increased ChAT activity 8 days after transfection in uninjured (C) and injured (I) septo-hippocampal cell cultures. In uninjured control cultures, NGF DNA transfection (C + ND) increased ChAT activity compared with controls (C) (* P ⬍ 0.05; n = 4). Calcium-dependent depolarization injury induced significant loss of ChAT activity 8 days after injury in septo-hippocampal cell cultures (I) compared with uninjured control cultures (C) (** P ⬍ 0.05; n = 4). However, DCChol liposome-mediated NGF cDNA transfection after injury (I + ND) significantly attenuated loss of ChAT activity compared with injured cultures (I) (*** P ⬍ 0.05). Administration of exogenous NGF protein (100 ng/ml) in injured cultures (I + NP) had a similar effect (*** P ⬍ 0.05; n = 4). The values are mean ± s.e.m. from four different wells. Figure 3 Assays of bioactivity of NGF. Representative photograph of PC12 cells cultured for 36 h in conditioned media from primary septohippocampal mixed cell cultures transfected with NGF cDNA. Rat PC12 cells were dislodged from maintenance culture flasks by vigorous shaking, then collected in a 15 ml tube. Cells were spun at 60 g for 10 min and the pellet was washed twice in PBS. Cells were counted in a hemacytometer microscope slide chamber, then plated on a 24-well plate in RPM-1640 culture medium at a density of 2 × 104 cell/0.5 ml. These pictures show representative changes in PC12 cell morphology after incubation with media from NGF cDNA transfected septo-hippocampal cultures. (A) Medium from non-transfected cultures; (B) medium from cultures treated with DC-Chol liposomes alone; (C) medium from cultures 3 days after transfection with cDNA for NGF complexed with DC-Chol liposomes; (D) non-conditioned medium in which 20 ng/ml NGF was added as positive control. Note prominent neurite outgrowth and associated growth cones produced by the medium from NGF cDNA transfected cultures (C) similar to those seen following exogenous administration of NGF (D). The arrow indicates the neurite growth.

critical. The transfection efficiency and toxicity associated with liposome-mediated gene transfer are determined by the ratio of cDNA to liposomes. Different cell lines may show varying optimal conditions for transfection. Therefore, optimizing the transfection in individual systems is important. We found that ratios of 1 ␮g DNA:3 ␮l liposomes produce highest transfection efficiency in septohippocampal cultures. However, lower DC-Chol component (DNA:1 ␮g/liposome:1 ␮l) or higher DC-Chol component (DNA:1 ␮g/liposome:6 ␮l) will significantly decrease transfection efficiency. The decreased transfection with higher DC-Chol concentrations may be attributed to the toxic effect of high doses of liposomes to the cultures. The 1:3 ratio of DNA to DC-Chol liposomes corresponds to a 1:2 (DNA:liposomes) electrical charge ratio. This is different from the reported optimal ratio in epithelial cell cultures (eg 1:1 to 6:1/DNA:lipids ratios).32 This discrepancy suggests the optimal transfection conditions will vary in different systems, although the reason for this is still not clear. It may be related to differences

in cell surface composition that affect complex binding and differences in the mechanism of complex uptake by different cell types.40 We have also found the incubation time can affect transgene product levels. Twelve hours of liposome–DNA complex incubation in cultures resulted in higher levels of NGF production than 6 h of incubation. However, we did not observe further increases in transgene expression by increasing the incubation time to 24 h (data from our laboratory observations). Since the time course of non-viral vector-mediated transgene expression varies in different systems,33 knowledge of the temporal profile of DC-Chol-mediated gene transfection in CNS cells is important. Our results from septo-hippocampal cell cultures suggest that the transgene is expressed at least 1 month after transfection. Due to the significant cell death seen in primary cell cultures after 1 month, studies regarding the duration of transgene expression for longer periods are not practical in this model system. The low transfection efficiency of liposome-mediated gene transfection is a major concern for potential clinical applications. However, levels of transfection efficiency sufficient to produce some therapeutic effects will vary in different systems and applications, and therapeutic effect can be evaluated only by studies employing carefully designed therapeutic endpoints. NGF is such a potent molecule that even a few cells producing it can cause biological effects. Our results suggest that a 3 or 4% transfection efficiency in septo-hippocampal cell cultures can produce a dramatic increase in NGF protein in surrounding media and protect against cholinergic neuronal damage. In previous studies, we found that less than 4% transfection efficiency with brain-derived neurotrophic factor (BDNF) cDNA could prevent neurofilament loss caused by neuronal injury.41 Thus, 3–4% transfection levels of neurotrophins in our in vitro systems


999

A

Figure 5 Intraventricular gene transfection. (A) Expression of alkaline phosphatase (AK) activity in the rat brain ventricular ependymal cells after intraventricular injection of liposomes (DC-Chol) complexed with pCMV/AK plasmid DNA. Twelve micrograms of pCMV/AK plasmid DNA were complexed with 36 ␮l of liposomes (DC-Chol) and injected into the rat cerebroventricle. Seven days after injection, rats were killed and the brains sectioned. Histochemical staining demonstrated that AK activity was expressed in the ventricular ependymal cells in liposome/DNA transfected animals (b, 250×), not liposome only animals (a, 100×). The arrows in panel b indicate intensely staining ependymal cell patches. (B) ELISA analysis of NGF protein in cerebrospinal fluid (CSF) from rats after intraventricular injection of pCMV/NGF cDNA complexed with DC-Chol liposomes. NGF protein levels in CSF were expressed as percentage of corresponding control values from rats injected with liposome only. Note that there were significant elevations (P ⬍ 0.05) of NGF levels at 7 and 14 days after transfection, associated with 12.3 ± 3.16 (n = 6) and 9.9 ± 2.85 (n = 4) fold increases at 7 and 14 days, respectively. Three days after transfection, NGF displayed a slight but not statistically significant elevation (P ⬎ 0.05; 2.4 ± 0.89, n = 5). All data were expressed as mean ± s.e.m. from four to six rats. *P ⬍ 0.05 as compared with control. L, liposome alone; L + N, liposome-mediated NGF transfection.

may be enough to produce beneficial effects against neuronal injury. While these results further suggest that liposome-mediated neurotrophin gene transfection may be clinically applicable even with a limited transfection efficiency, additional in vivo studies, such as those discussed below, are critical for accurate preclinical assessments. Using different methods, we obtained results suggesting that DC-Chol-mediated gene transfection could be achieved by direct injection into the cerebral ventricles. This injection did not cause any overt tissue damage detected by light microscopic histological examination. AK histochemical staining showed transgene expression in the ventricular ependymal tissue. The transgene expression in this group of cells has the potential to provide significant dispersion of therapeutic proteins due to CSF circulation. Our NGF ELISA results support this possibility. The nine- to 12-fold increases of NGF in the CSF suggest the NGF produced by gene transfer can be widely circulated in the CSF and circulation of this transgene product in CSF may have significant therapeutic effects. The in vitro NGF bioassay results further suggest

that the products of transgene expression can produce the same biological effects as natural products without any alteration in physiological function. The restorative effect of NGF on CNS cholinergic deficits and memory after injury is well documented.6,7,11,19,42–46 Exogenous administration with NGF has been shown to spare septal cholinergic neurons from death and degeneration following injury.1–3,47 NGF increases ChAT activity both in vitro32,48and in intact animals.8,49 Also, intraparenchymal grafts of cells genetically modified to produce NGF can prevent cholinergic neuronal degeneration caused by fornix transection50,51 or by immunolesioning procedures.52 Recent studies have shown that administration of exogenous neurotrophic factors can reduce spatial memory deficits in rats following TBI.11,12,44,45 Although administration of exogenous proteins has proven therapeutically beneficial in animal models, the limitations imposed by protein degradation and the blood– brain barrier restrict the clinical utility of these approaches. There is a need to develop long-term delivery of therapeutic levels of NGF into CNS as a therapy for neuronal injury. Employing non-viral vector gene


1000

Ba

b

c

Figure 6 NGF gene transfer attenuates cholinergic neuronal injury. (A) Liposome-mediated NGF cDNA transfection significantly reduced the ChATpositive neuronal loss in rat medial septal regions. The data are means ± s.e.m. from three to seven rats, ChAT-positive neuronal cells were averaged from eight sections per rat. ChAT-positive neuronal cells were counted by two independent observers who were blinded to the treatment. The results suggest that injury can cause significant ChAT-positive neuronal cell loss (P ⬍ 0.05) ipsilaterally and contralaterally 7 and 14 days following cortical impact injury. However, in the liposome-mediated NGF cDNA transfected group, ChAT-positive neuronal cell number significantly increased (P ⬍ 0.05) up to normal levels on both sides compared with the animals that received liposome alone, indicative of an NGF-induced ChAT cell recovery. Three days after injury, ChAT-positive cells were significantly reduced (P ⬍ 0.05) only on the ipsilateral side and NGF transfection had no effect on ChATpositive cell loss. L: left side of medial septum (contralateral to the injury side); R, right side of medial septum (ipsilateral to the injury). (B) Liposomemediated NGF cDNA transfection prevented ChAT-positive neuronal loss in the rat medial septal regions as shown by ChAT immunohistochemical staining in rat septal regions. (a) Rat brain septal region in coronal section from an uninjured animal; (b) septum 7 days following cortical impact injury and liposome vehicle intraventricular injection (1 day post-injury injection); (c) Septum 7 days after cortical impact injury and liposome/NGF cDNA complex intraventricular injection. The arrows point to the medial septal ChAT neuronal staining. ‘ip’ notes the side ipsilateral to injury.

transfer techniques to achieve therapeutically useful levels of expression of neurotrophins in the CNS could provide a new strategy for intervention following CNS injury. The results from our study suggest that intraventricular delivery of the NGF gene complexed with cationic liposomes increases NGF levels in CSF, and that increased levels of NGF are associated with beneficial effects on cholinergic neurons in the brain parenchyma. As we previously discussed, although the relatively low

transfection efficiency of liposome-mediated gene transfer may limit its application for certain diseases, cationic liposome-mediated neurotrophic gene transfer may provide useful treatment for certain forms of CNS injury. The present study demonstrates that small groups of transfected cells can produce transgene products (neurotrophic peptides) which can diffuse in the brain parenchyma and ameliorate damage to cholinergic neurons following injury. Similar protection has been


reported when exogenous NGF is infused into the ventricles following traumatic brain injury.11 Recent studies have suggested that cationic liposomemediated gene transfer of therapeutically relevant genes has potential for treatment of neurological disorders such as Parkinson’s disease53 and epileptic seizures.54,55 This paper provides new evidence that liposome-mediated gene transfer can produce nine- to 12-fold increases of secreted transgene products in the CSF and that these increases are associated with reductions in neuronal damage following traumatic brain injury. Future studies should focus on further optimization of liposomemediated gene transfer in CNS, by increasing transfection efficiency, targeting delivery to specific cell types or regions and regulation of duration of transgene expression.

Materials and methods Sprague–Dawley rats were purchased from Harlan Laboratories (Indianapolis, IN, USA). DC-Chol liposomes were obtained from Dr Leaf Huang’s laboratory at the University of Pittsburgh. Alkaline phosphatase plasmid DNA was obtained from Genzyme (Framingham, MA, USA). Anti-ChAT antibodies were purchased from Chemicon International (Temecula, CA, USA). NGF and anti-NGF antibodies were purchased from Boehringer Mannheim (Indianapolis, IN, USA). Vectastain ABC reagent was purchased from Vector Laboratories (Burlingame, CA, USA). Alkaline phosphatase substrates, including NBT and BCIP, were purchased from Promega (Madison, WI, USA). Five-bromo-4-chloro-3-indolyl ␤-dgalactoside (X-gal), MgCl2, potassium ferricyanide, potassium ferrocyanide and other reagents were all purchased from Sigma (St Louis, MO, USA).

Production of NGF expression vector and liposome formulation formation The expression vector was prepared by inserting rat NGF cDNA into a pUC19-based plasmid containing a cytomegalovirus (CMV) promoter. The NGF cDNA gene (780 bp) was removed by digestion with EcoRI and BamHI from a pBluescribe (M13+) vector. After phenol and chloroform extraction and ethanol precipitation, NGF cDNA was treated with the Klenow fragment of E. coli DNA polymerase I to generate the plasmid backbone. The NGF insert was ligated to the pUC19-based pCMV promoter in the NotI site. The resultant plasmid contains the CMV immediate–early promoter/enhancer, SV40 splice donor and splice acceptor, and SV40 polyadenylation sequence. We used the DC-Chol liposome as a cationic lipid to condense the plasmid DNA. Both DC-Chol liposomes and plasmid DNA were allowed to complex at room temperature for 15 min before the in vitro gene transfection or in vivo stereotactic injection. Primary septo-hippocampal cultures The primary cell cultures were prepared as previously described.35,56,57 Hippocampi and septi were dissected from the brains of 18-day-old Sprague–Dawley rat fetuses.58 After washing, tissue was dissociated by repeated passage through a flame-constricted Pasteur pipette, collected by centrifugation, resuspended in Dulbecco’s modified Eagle’s medium (DMEM), supplemented with 10% fetal calf serum and plated on poly-

l-lysine coated 24-well plates (2.18 × 105 cells per well). Cultures were kept in a humidified CO2 incubator at 37°C. After 5 days, culture media was changed to DMEM plus B18 supplement.59 Subsequent media replacement was carried out three times a week.

Liposome-mediated ␤-galactosidase and NGF cDNA transfection Based on our previous studies of liposome-mediated gene transfection in septo-hippocampal cultures,34,60,61 we initially employed six different ratios of ␤-gal cDNA or NGF cDNA to DC-Chol liposomes for the transfection of primary septo-hippocampal cell cultures. One microgram of ␤-gal cDNA in 100 ␮l DMEM (serum free) was mixed with DC-Chol liposomes in ratios of 1:1; 1:2; 1:3; 1:4; 1:5 and 1:6 (DNA (␮g):liposome (␮l)) and overlaid on each 16 mm well of primary cultures. Following 6 or 12 h incubation, DMEM medium was replaced in the well. To determine the transgene expression, we used X-gal staining and ␤-gal activity assay to evaluate the ␤-gal gene expression. RT-PCR and ELISA were applied to determine the messenger RNA and protein expression of NGF. Calcium-dependent depolarization injury in cultures Calcium-dependent potassium-mediated depolarization injury in primary septo-hippocampal cultures was conducted as previously reported.35 To produce the injury, normal media were replaced with media containing 60 mm KCl and 5.8 mm CaCl2. After 6 min of treatment, this high potassium and high calcium media was replaced by normal media (which contains 5.3 mm K+ and 1.8 mm Ca2+). This injury produces significant neuronal injury including proteolytic degradation of cytoskeletal proteins62 and reduced ChAT activity in the absence of cell death. NGF transfection was assayed in septo-hippocampal cell cultures for protection against the injury. Exogenous NGF was used as positive control at 100 ng/ml, a dose higher than that secreted by the cell cultures. That is because exogenous NGF was added only once, it could be inactivated during the assay, whereas the transfected cell cultures could synthesize and secrete fresh NGF consistently. Furthermore, according to our experience, NGF freshly secreted is usually more active than a commercially available one. Cortical impact injury The controlled cortical impact rat injury model was used for in vivo studies. This model reproduces many features of severe human traumatic brain injury including neurological and cognitive deficits.62 Male Sprague–Dawley rats (250–350 g) were initially anesthetized with 4% isofluorane and N2O/O2 (2:1) in a vented anesthesia chamber. Following endotracheal intubation, rats were mechanically ventilated with 2% isofluorane and secured in a stereotactic frame. A midline incision was made, the soft tissues reflected, and a circular section of skull 8 mm in diameter was removed from the right margin of the skull midway between the frontal and occipital sutures, 2 mm medial to the temporal ridge. The dura was left intact. This craniotomy exposed the injury site in all animals. Injury was produced by a controlled lateral cortical impact model described in detail by Dixon et al.62 The impact velocity was adjusted to 6.0 m/s by controlled gas pressure and verified by a time-displacement curve that was measured by a linear variable differential trans-

1001


1002

former (Shaevitz model 500 HR) and a PC-based data acquisition system (Axoscope; Axon Instruments, Foster City, CA, USA).

Injection of liposome and cDNA complexes These studies employed intraventricular injections. The injection procedures were modified as previously described for intracranial injection.63 Male Sprague–Dawley rats (250–300 g) were anesthetized with sodium pentobarbital (54 mg/kg i.p.) and secured on a stereotactic frame. For intraventricular injection, the burr hole was opened at 1.0 mm posterior to the bregma and 1.6 mm lateral to the sagittal suture. The needle was inserted 3.6 mm below the dura. Liposome (36 ␮l) or liposome complexed with 12 ␮g cDNA encoding either ␤-gal or NGF were injected in the lateral ventricle. For rats undergoing cortical impact, the injections were performed at 1 day after injury. Histological preparation of rat brain Rats were deeply anesthetized with sodium pentobarbital and perfused transcardially with 200 ml PBS (0.1 m, pH 7.4), followed by 150 ml of fixative (4% paraformaldehyde, 0.05% glutaraldehyde and 0.2% picric acid in PBS). The brains were removed and postfixed in the same fixative for 1 day at 4°C, followed by 2 or 3 days cryoprotection in 30% sucrose in PBS until the tissues completely sank. Coronal sections (30 ␮m) were cut at −20°C in a cryostat. Alkaline phosphatase histochemical staining After cutting and washing with PBS the rat brain sections were incubated in PBS for 30 min at 65°C to inactivate endogenous alkaline phosphatase. The sections were subsequently transferred to Tris buffer (0.1 m, pH 9.5) containing 100 mm NaCl, 20 mm MgCl2 and 0.02% NP-40, and incubated overnight at 4°C with AP substrates (NBT/1:300 and BCIP/1:150) in the same Tris buffer. Incubation was stopped by replacing the substrate buffer with PBS containing 20 mm EDTA. The sections were examined under a light microscope. ␤-Gal activity assay ␤-Galactosidase activity was measured quantitatively with a Shimadzu (Columbia, USA) UV-visible recording spectrophotometer using o-nitro-phenol-␤-d-galactoside (ONPG) as the substrate. After the appropriate incubation time, the medium was removed and the cells were washed twice with PBS. PBS (1 ml) was added to each well and the cells were removed from the plate by scraping with a pipetteman and transferred into microfuge tubes. The cells were recovered by centrifugation at 15 000 g for 10 s, resuspended in 1 ml PBS, and recovered again by centrifugation. The pellet was resuspended in 50 ␮l 0.25 m Tris-HCl (pH 7.8), and cells were lysed by three cycles of freezing in dry ice and ethanol and thawing at 37°C. Cell debris was removed by centrifugation at 15 000 g for 5 min. A standard curve was constructed using E. coli ␤-galactosidase (grade VII, 250–500 units/mg protein). The standards consisted of 10 serial dilutions of ␤-galactosidase ranging from 1 unit/ml to 0.002 unit/ml. Each standard and sample assay contained 1.0 ml 100 mm sodium phosphate buffer (pH 7.2), 40 ␮l of 30 mm magnesium chloride, 40 ␮l of 3.36 m 2-␤-mercaptoethanol, 40 ␮l enzyme solution and 40 ␮l of 68 mm

ONPG. Assays were incubated at 37°C for 2 h and the reactions were stopped by adding 2 ml of 1 m sodium carbonate. Absorbancies were measured at 410 nm in the spectrophotometer.

Reverse transcription polymerase chain reaction (RTPCR) One day after DC-Chol liposome-mediated NGF gene transfection, cells in culture were lysed. After RNA extraction, the samples were digested with RNase-free DNase I, and cDNA was synthesized by reverse transcription. For PCR, one pair of forward and backward primers of NGF or one pair of primers of ChAT was used. The sequences of primers for NGF are GGCATGCTGGACCCAAGCTC (NGF/5) and GCGCTT GCTCCGGTCAGTCC (NGF/3). The sequences of primers for ChAT are TTAATTTCCGCCGTCTCAGTGAGG (ChAT/5) and TGCACCAGGACGATGCCATCGAAT (ChAT/3). We used actin as an internal control. Actin and NGF or ChAT co-PCR was carried out in a programmable heating block using cycles consisting of denaturation at 95°C for 1 min, followed by annealing at 55°C for 1 min, and DNA extension at 72°C for 2 min. After 30 cycles of PCR, samples were electrophoresed on 1.5% agarose gels. Gels were stained with ethidium bromide and photographed under UV light. To check for possible DNA contamination during RNA preparation, we included RNA samples without performing reverse transcription. The control studies confirmed the absence of DNA contamination. NGF ELISA For in vitro studies, septo-hippocampal cell culture medium was removed for ELISA at 2, 4 and 8 days after transfection. For in vivo studies, CSF samples were withdrawn from NGF transfected rats through the occipital foramen magnum. Samples were diluted 1:1 with extraction buffer containing 1 mm PMSF, 7 ␮g/ml aprotinin, and 4 mm EDTA. ELISA procedure was performed as described by protocol from Boehringer Mannheim. Briefly, plate wells were coated with 0.25 ␮g/ml antiNGF antibodies and incubated for 2 h at 37°C. The wells were washed four times with wash buffer, coated with blocking buffer, and incubated for 30 min at 37°C. A standard NGF dilution series was prepared. Diluted culture medium (100 ␮l) and standard NGF dilutions (100 ␮l) were added to the antibody-coated wells and incubated overnight at 4°C. After washing the plate with wash buffer, antibody-␤-gal-conjugate solution (100 ␮l) was added to each well and incubated (4 h at 37°C). After washing again with buffer, substrate solution (200 ␮l) was added to each well and incubated at 37°C until color was sufficiently developed. Plates were read at 570 nm. The NGF concentrations were determined from a standard concentration curve by a computer-assisted software program (Bio-Rad Laboratories, Hercules, CA, USA). NGF biological assays Rat pheochromocytoma (PC12) cells were used to assay the specific biological activity of the NGF in the medium of transfected cell cultures as previous described.64 Three hours after plating, control media of PC12 cells were removed and replaced with 0.5 ml of media conditioned by septo-hippocampal cell cultures transfected with NGF


cDNA with DC-Chol liposome mediation. NGF (20 ng/ml) was added to sister wells as a positive control. The dose was chosen mainly based on our preliminary experiments and others’ report.64 Thirty-three hours later, cells were examined for neurite outgrowth. PC12 cells are photographed and evaluated for morphological differentiation.

ChAT activity assays ChAT activity in the septo-hippocampal cell extracts was determined according to previous studies.32 In brief, the cultured rat septo-hippocampal cells were lysed in 100 ␮l of 10 mm sodium phosphate buffer containing 0.1% NP40 (pH 7.4) on ice. After addition of a 50 ␮l aliquot of reaction mixture containing 150 mm NaCl, 0.5 mm EDTA, 0.15 mm eserine, 0.2 mm [3H]-acetyl coenzyme A (specific activity 32 ␮Ci/␮mol, Amersham, Arlington Heights, IL, USA), 45 mm chloride, in 10 mm phosphate buffer (pH 7.4), the samples were incubated at 37°C for 40 min. The reaction was stopped by the addition of 2 ml of ice-cold PBS. The synthesized 3H-acetylcholine was extracted with 2 ml of toluene scintillation cocktail containing 2 mg/ml sodium tetraphenylboron, which was dissolved in acetonitrile and the magnitude of radioactivity was measured with a liquid scintillation spectrometer. ChAT activity was expressed as pmol/h/well. ChAT immunohistochemistry and cell count The frozen sections were washed in PBS and incubated with 0.5% H2O2 in PBS for 10 min to remove endogenous peroxide activity. Following three washes in PBS, the sections were blocked with 3.0% normal rabbit serum at room temperature for 1 h and then incubated overnight at 4°C with a 1:200 dilution of polyclonal goat ChAT antibody (primary antibody). The carrier solution was PBS containing 1.5% rabbit serum and 0.1% Triton X-100. After three washes in PBS, the sections were incubated in turn with rabbit anti-goat antibody and ABC reagents according to the instructions provided by Vector. The carrier solution for the second antibody was PBS containing 1.5% rabbit serum only. The resulting sections were developed for staining in freshly prepared DAB solution (in Tris-buffered solution) for 2 to 5 min. The sections were subsequently dried on Fisher (Pittsburgh, PA, USA) brand superfrost/Plus glass slides, dehydrated in graded ethanol, cleared in xylene and coversliped. Due to severe tissue damage in many animals after cortical impact injury, complete sampling and stereological estimation of total cell number became impossible. Only sections approximately 390 ␮m rostral to anterior commissure decussation (ACD) were examined and counted for the ChAT positive cells under a Nikon (Melville, USA) microscope (100× magnification) equipped with a 10 × 10 ocular grid. These sections were chosen because they provide a region densely populated with ChAT-positive neurons which can serve as a representative of medial septal neuron population.5 The medial septal area was defined dorsally and laterally by the distribution of stained neurons and ventrally by an imaginary line drawn through the center of the anterior commissure according to Hagg et al.5 The ChAT-positive neurons were defined as immunolabeled cell bodies, regardless of intensity of staining. These criteria were utilized to exclude counts of smaller portions of cells contained primarily in adjacent sections, but included any

small, atrophied, weakly immuno-positive neurons.11 The ChAT positive neurons were counted bilaterally in five to seven consecutive sections (serial section). Values were compared for injury alone versus injury with injection rats, as well as ipsilateral versus contralateral to the side of injury. These values in the injured rats were compared with intact rats, or in the ipsilateral side of the injured rats to the contralateral side.

Statistical analyses Data were subjected to analyses of variances followed by Student’s t tests to show statistical significance. All data were expressed as group mean ± standard error. A P value less than 0.05 was accepted as significance. All animal studies carefully conformed to the guidelines outlined in the Guide for the Care and Use of Laboratory Animals from the US Department of Health and Human Services.

Acknowledgements This work was supported by grants from the National Institutes of Health, RO1-NS35502 to K Yang, CA 64654 to L Huang and RO1-NS21458 to R Hayes. Texas Higher Education Coordinating Board grants ARP 011618–100 and ATP 004949–049 to K Yang and the Vivian L Smith Center for Neurologic Research.

References 1 Hefti F. Nerve growth factor promotes survival of septal cholinergic neurons after fimbrial transections. J Neurosci 1986; 8: 2155–2162. 2 Kromer LF. Nerve growth factor treatment after brain injury prevents neuronal death. Science 1987; 235: 214–216. 3 Williams LR et al. Continuous infusion of nerve growth factor prevents basal forebrain neuronal death after fimbria-fornix transection. Proc Natl Acad Sci USA 1986; 83: 9231–9235. 4 Rosenberg MB et al. Grafting genetically modified cells to the damaged brain: restorative effects of NGF expression. Science 1988; 242: 1575–1578. 5 Hagg T, Varon S. Neurotropism of nerve growth factor for adult rat septal cholinergic axons in vivo. Exp Neurol 1993; 119: 37–45. 6 Fischer W, Bjorklund A, Chen K, Gage FH. NGF improves spatial memory in aged rodents as a function of age. J Neurosci 1991; 11: 1889–1906. 7 Fischer W, Bjorklund A. Loss of AchE and NGFr-labeling precedes neuronal death of axotomized septodiagonal band neurons: reversal by intraventricular NGF infusion. Exp Neurol 1991; 113: 93–108. 8 Rylett RJ, Goddard S, Schmidt BM, Williams LR. Acetylcholine synthesis and release following continuous intracerebral administration of NGF in adult and aged Fischer-344 rats. J Neurosci 1993; 13: 3956–3963. 9 Dekker AJ, That L. Effect of delayed treatment with nerve growth factor on choline acetyltransferase activity in the cortex of rats with lesions of the nucleus basalis magnocellularis: dose requirements. Brain Res 1992; 584: 55–63. 10 Dekker AJ, Ray WJ, Tha LJ, Gage FH. Grafting of nerve growth factor-producing fibroblasts reduces behavioral deficits in rats with lesions of the nucleus basalis magnocellularis. Neuroscience 1994; 60: 299–309. 11 Dixon CE et al. Nerve growth factor attenuates cholinergic deficits following traumatic brain injury in rats. Exp Neurol 1997; 146: 479–490. 12 McDermott KL et al. Delayed administration of basic fibroblast growth factor (bFGF) attenuates cognitive dysfunction following parasagittal fluid percussion brain injury in the rat. J Neurotrauma 1997; 14: 191–198.

1003


1004

13 Lawrence MS et al. Overexpression of Bcl-2 with herpes simplex virus vectors protects CNS neurons against neurological insults in vitro and in vivo. J Neurosci 1996; 16: 486–496. 14 Lawrence MS et al. Herpes simplex viral vectors expressing Bcl2 are neuroprotective when delivered after a stroke. J Cereb Blood Flow Metab 1997; 17: 740–744. 15 Yoon SO et al. Adenovirus-mediated gene delivery into neuronal precursors of the adult mouse brain. Proc Natl Acad Sci USA 1996; 93: 11974–11979. 16 Naldini L et al. In vivo gene delivery and stable transduction of nondividing cells by a lentiviral vector. Science 1996; 272: 263– 267. 17 Naldini L et al. Efficient transfer, integration, and sustained long-term expression of the transgene in adult rat brains injected with a lentiviral vector. Proc Natl Acad Sci USA 1996; 93: 11382–11388. 18 Karpati G, Lochmuller H, Nalbantoglu J, Durham H. The principles of gene therapy for the nervous system. Trends Neurosci 1996; 19: 49–54. 19 Castel-Barthe MN et al. Direct intracerebral nerve growth factor gene transfer using a recombinant adenovirus: effect on basal forebrain cholinergic neurons during aging. Neurobiol Dis 1996; 3: 76–86. 20 Choi-Lundberg DL et al. Dopaminergic neurons protected from degeneration by GDNF gene therapy. Science 1997; 275: 838–841. 21 Baumgartner BJ, Shine HD. Targeted transduction of CNS neurons with adenoviral vectors carrying neurotrophic factor genes confers neuroprotection that exceeds the transduced population. J Neurosci 1997; 17: 6504–6511. 22 Yang GY, Zhao YJ, Davidson BL, Betz AL. Overexpression of interleukin-1 receptor antagonist in the mouse brain reduces ischemic brain injury. Brain Res 1997; 751: 181–188. 23 Felgner PL et al. Lipofection: a highly efficient, lipid-mediated DNA-transfection procedure. Proc Natl Acad Sci USA 1987; 84: 7413–7417. 24 Felgner PL. Improvements in cationic liposomes for in vivo gene transfer. Hum Gene Ther 1996; 7: 1791–1793. 25 Nabel GL. Direct gene transfer with DNA liposome complexes in melanoma: expression, biological activity, lack of toxicity in humans. Proc Natl Acad Sci USA 1993; 90: 11307–11311. 26 Zhu N, Liggitt D, Liu Y, Debs R. Systemic gene expression after intravenous DNA delivery into adult mice. Science 1993; 261: 209–211. 27 Roessler B, Davidson BL. Direct plasmid mediated transfection of adult murine brain cells in vivo using cationic liposomes. Neurosci Lett 1994; 167: 5–10. 28 Gao X, Huang L. Cationic liposome-mediated gene transfer. Gene Therapy 1995; 2: 710–722. 29 Li S, Huang L. In vivo gene transfer via intravenous administration of cationic lipid-protamine-DNA (LPD) complexes. Gene Therapy 1997; 4: 891–900. 30 Yang K, Clifton GL, Hayes RL. Gene therapy for central nervous system injury: the use of cationic liposomes: an invited review. J Neurotrauma 1997; 14: 281–297. 31 McIntosh TK, Juhler M, Wieloch T. Novel pharmacological strategies in the treatment of experimental traumatic brain injury: 1998. J Neurotrauma 1998; 15: 731–769. 32 Le W et al. Liposome-mediated NGF gene transfection increases ChAT activity in rat septo-hippocampal cell cultures. NeuroReport 1996; 7: 710–712. 33 Yang K et al. DC-Chol liposome-mediated gene transfer in rat spinal cord. NeuroReport 1997; 8: 2355–2358. 34 Yang K et al. Optimizing liposome-mediated gene transfer in primary rat septo-hippocampal cell cultures. Neurosci Lett 1994; 182: 287–290. 35 Whitson JS et al. Brief potassium depolarization decreases neurofilament proteins in CNS culture. Brain Res 1995; 694: 213–222. 36 Suhr ST, Gage FH. Gene therapy for neurologic disease. Arch Neurol 1993; 50: 1252–1268. 37 Doran SE et al. Gene expression from recombinant viral vectors

38

39

40

41

42

43

44

45

46

47

48

49

50

51

52

53

54

55

56

in the central nervous system after blood–brain barrier disruption. Neurosurgery 1995; 36: 965–970. Thorsell A, Lomqvist AG, Heilig M. Cationic lipid-mediated delivery and expression of prepro-neuropeptide Y cDNA after intraventricular administration in rat: feasibility and limitations. Reg Pept 1996; 61: 205–211. Caplen NJ et al. Liposome-mediated CFTR gene transfer to the nasal epithelium of patients with cystic fibrosis. Nature Med 1995; 1: 39–46. Fasbender A et al. Effect of co-lipids in enhancing cationic lipidmediated gene transfer in vitro and in vivo. Gene Therapy 1997; 4: 716–725. Hayes RL et al. Rescue of injury-induced loss by BDNF gene transfection in primary septo-hippocampal cell cultures. Neurosci Lett 1995; 191: 121–125. Frick KM, Price DL, Koliatsos VE, Markowska AL. The effects of nerve growth factor on spatial recent memory in aged rats persist after discontinuation of treatment. J Neurosci 1997; 17: 2543–2550. Martinez-Serrano A, Hantzopoulos PA, Bjorklund A. Ex vivo gene transfer of brain-derived neurotrophic factor to the intact rat forebrain: neurotrophic effects on cholinergic neurons. Eur J Neurosci 1996; 8: 727–735. Sinson G, Voddi M, McIntosh TK. Nerve growth factor administration attenuates cognitive but not neurobehavioral motor dysfunction or hippocampal cell loss following fluid-percussion brain injury in rats. J Neurochem 1995; 65: 2209–2216. Sinson G, Voddi M, McIntosh TK. Combined fetal neural transplantation and nerve growth factor infusion: effects on neurological outcome following fluid-percussion brain injury in the rat. J Neurosurg 1996; 84: 665–662. Batchelor PE et al. Nerve growth factor receptor and choline acetyltransferase colocalization in neurons within the rat forebrain: response to fimbria-fornix transection. J Comp Neurol 1989; 284: 187–204. Winn SR et al. Polymer-encapsulated cells genetically modified to secrete human nerve growth factor promote the survival of axotomized septal cholinergic neurons. Proc Natl Acad Sci USA 1994; 91: 2324–2328. Alderson RF, Alterman AL, Barde YA, Lindsay RM. Brain derived neurotrophic factor increases survival and differentiated functions of rat sepral cholinergic neurons. Neuron 1990; 5: 297–306. Scali C et al. Nerve growth factor increases extracellular acetylcholine levels in the parietal cortex and hippocampus of aged rats and restores object recognition. Neurosci Lett 1994; 170: 117–120. Tuszynski MH, Gage FH. Bridging grafts and transient nerve growth factor infusions promote long-term central nervous system neuronal rescue and partial functional recovery. Proc Natl Acad Sci USA 1995; 92: 4621–4625. Tuszynsk MH, Roberts J, Senut M-C, Gage FH. Gene therapy in the adult primate brain: intraparenchymal grafts of cells genetically modified to produce nerve growth factor prevent cholinergic neuronal degeneration. Gene Therapy 1996; 3: 305–314. Roner S et al. Effects of intraventricular transplantation of NGFsecreting cells on cholinergic basal forebrain neurons after partial immunolesion. J Neurosci Res 1996; 45: 40–56. Imaoka T, Date I, Ohmoto T, Nagatsu T. Significant behavioral recovery in Parkinson’s disease model by direct intracerebral gene transfer using continuous injection of a plasmid DNA-liposome complex. Hum Gene Ther 1998; 9: 1093–1102. Zhang LX, Wu M, Han JS. Suppression of audiogenic epileptic seizures by intracerebral injection of a CCK gene vector. NeuroReport 1992; 3: 700–702. Zhang LX et al. Lipofectin-facilitated transfer of cholecystokinin gene corrects behavioral abnormalities of rats with audiogenic seizures. Neuroscience 1997; 77: 15–22. Yang K et al. Liposome-mediated NGF gene transfection increases phosphorylated neurofilament proteins. Mol Brain Res 1996; 43: 13–20.


57 Qiu YH et al. Activation of phosphatidylinositol 3 kinase by BDNF gene transfection in septo-hippocampal cultures. J Neurosci Res 1998; 52: 192–200. 58 Banker GA, Cowan WM. Rat hippocampal neurons in dispersed cell culture. Brain Res 1977; 126: 397–425. 59 Brewer GJ, Cotman CW. Survival and growth of hippocampal neurons in defined medium at low density: advantages of a sandwich culture technique or low oxygen. Brain Res 1989; 494: 65–74. 60 Yang K et al. Sustained expression of functional nerve growth factor in primary septohippocampal cell cultures by liposomemediated gene transfer. Neurosci Lett 1994; 182: 291–294.

61 Yang K et al. In vitro studies of liposome-mediated gene transfection. In: Perez-Polo JR (ed). Paradigms in Neural Injury. Academic Press: San Diego, 1995, pp 290–297. 62 Dixon CE et al. A controlled cortical impact model of traumatic brain injury in the rat. J Neurosci Meth 1991; 39: 253–262. 63 Iwamoto Y et al. BDNF cDNA following liposome-mediated gene transfer in intact and traumatically injured rat brain. NeuroReport 1996; 7: 609–612. 64 Greene LA, Tischler AS. Establishment of a noradrenergic clonal line of rat adrenal pheochromocytoma cells which respond to NGF. Proc Natl Acad Sci USA 1976; 73: 2424–2428.

1005