Intermittent vs Continuous Administration of Nerve Growth Factor to Injured Medial Septal Cholinergic Neurons in Rat Basal Forebrain

Abstract

Many medial septal neurons of the basal forebrain are dependent on nerve growth factor (NGF) from the hippocampus for survival and maintenance of a cholinergic phenotype. When deprived of their source of NGF by axotomy, medial septal neuronal cell bodies atrophy and lose their cholinergic markers. This is similar to what is observed in the basal forebrain during Alzheimer's disease (AD). In the present study, medial septal neurons were axotomized in female rats by way of a fimbria/fornix lesion. For fourteen days following axotomy, varying NGF doses (1 - 250 μg/ml) were administered to the lateral cerebral ventricle with either mini-osmotic infusion or daily injection. The responsiveness of medial septal neurons was evaluated with choline acetyltransferase immunohistochemistry. Within the mini-osmotic pumps, NGF activity diminished greatly during the first five days of implantation, but increased dramatically in the CSF after five days of infusion. The responsiveness of medial septal neurons to NGF was dose dependent and the ED₅₀ for NGF injection was determined to be 14.08 μg/ml compared to 27.60 μg/ml for NGF infusion. Intermittent injections at varying intervals were evaluated with 30 μg/ml NGF over a fourteen-day time period (2, 3, 6, or 12 injections). No differences occurred in the number of choline acetyltransferase neurons from rats that received weekly injections to those that received daily injections of NGF. NGF administration has been suggested as a therapy for AD. The results of these studies continue to highlight the need for NGF stability within the delivery system and AD patient CSF, the choice of delivery system, frequency of administration, and the NGF dose for maintaining basal forebrain cholinergic neurons during AD.

Keywords: Alzheimer’s; Choline Acetyltransferase; Fimbria; Fornix; Medial Septal Nucleus; Nerve Growth Factor.

Figure 1.

The NGF biological activity was evaluated both from mini-osmotic pumps and CSF of uninjured rats at 0, 1, 5, 10, and 14 days. The NGF activity (open squares) in the pumps was reduced by 70% at 5 days and was not detected by fourteen days. In comparison, the NGF activity in the CSF (closed squares) increased slowly over five days and rapidly increased from five to fourteen days.

Figure 2.

Choline acetyltransferase (ChAT) immunoreactive neurons (arrows) were identified in the medial septum with peroxidase immunohistochemistry. In A, ChAT immunoreactive neurons are largely absent (oval) in the medial septum fourteen days after fimbria-fornix (FF) lesion followed by cytochrome c administration to the lateral cerebral ventricle. In B, ChAT immunoreactive neurons (arrows) are retained (oval) in the medial septum fourteen days after fimbria/fornix (FF) lesion and continuous NGF infusion (125 μg/ml).

Figure 3.

Rats with FF lesions were injected with 30 μg/ml of NGF at the time of axotomy. The total number of 30 μg/ml NGF injections varied over a 14 day time period: 2, 3, 6, or 12. ChAT immunoreactive neurons ranged between 300 – 400 cells for the various injection regimens. No significant differences were seen among the groups with FF lesions and varying intermittent NGF injections (Figure 3).

Figure 4.

Animals received fimbria/fornix (FF) lesions and were administered NGF immediately. NGF was administered into the lateral cerebral ventricle by continuous infusion (open circles, full line) or intermittent injection (12 injections; open squares, dashed line) over a fourteen day period. ChAT immunoreactive neurons were totaled from five tissue sections of the MS. Unlesioned rats averaged approximately 600 neurons. An FF lesion with cytochrome c administration (control group) caused an 80% reduction of ChAT immunoreactive neurons compared to unlesioned rats. Administration of 100 μg/ml NGF by either intermittent injection or continuous infusion returned the number of ChAT neurons close to normal levels (>90%). MS neurons responded to NGF in a dose dependent manner (1 – 250 μg/ml), but there were differences between injection vs infusion application. The dose response curve for NGF injection was shifted to the left compared to the dose response curve for NGF infusion. The ED₅₀ for NGF injection was 14.08 μg/ml and 27.60 μg/ml for NGF infusion.