Is nasal delivery a viable option for trophic factor delivery to the brain?

Neurotrophic factors are naturally occurring proteins that promote the development, growth and/or survival of brain cells, making them ideal candidates to halt the progression and perhaps even reverse the course of neurodegenerative diseases in ways not possible with current symptomatic therapies.  For several decades, there has been great interest in using neurotrophic factors as neuroprotective or restorative agents to treat Parkinson's, Alzheimer's and other central nervous system diseases but clinical success has not yet been achieved, largely due to the vexing challenges associated with effectively delivering these proteins to target sites in the brain (see Thorne and Frey, 2001).  

Delivering proteins or gene therapy vectors to the central nervous system has been limited by the blood-brain barrier, which normally restricts nearly all but the smallest (<0.5 kDa), lipophilic substances from passing into the brain from the bloodstream after systemic administration (Pardridge, 2002).  Methods to bypass the blood-brain barrier have often relied upon surgically invasive procedures to deliver proteins or gene therapy vectors into the cerebrospinal fluid or brain parenchyma (Thorne and Frey, 2001).  Non-invasive methods targeting large, hydrophilic substances to the brain and spinal cord are greatly needed, particularly for chronic conditions where it may be necessary to repeat dosing over time.  The intranasal route has many advantages for clinical use (Costantino et al, 2007): non-invasiveness, ease of application/termination, avoidance of hepatic first-pass elimination, and a growing record of experience with approved formulations (e.g. nasal spray of the 3.5 kDa polypeptide hormone calcitonin has been used to treat postmenopausal osteoporosis for many years). 

Figure 1Intranasal administration of radiolabeled insulin-like growth factor-I bypasses the blood-brain barrier to reach the rat nervous system. (A) Autoradiographs of a sagittal brain section and a transverse trigeminal nerve section showing high signal in the olfactory bulbs and trigeminal nerve. (B) Schematic of olfactory (red) and trigeminal (blue) pathways for entry of macromolecules into the brain and spinal cord following nasal application. (C) Hypothetical mechanisms conveying substances from the submucosa (lamina propria) into the central nervous system following nasal application. (Adapted from Thorne et al. Neuroscience, 2004.)The first reports showing intranasal administration could target potentially therapeutic levels of macromolecules, including nerve growth factor (NGF; 26.5 kDa), to the brain were described over a decade ago by Frey and coworkers (Thorne et al, 1995; Frey et al, 1997. Drug Delivery 4:87-92).  Since that time, a rapidly growing number of published studies have demonstrated a variety of peptides and proteins, including many different neurotrophic factors, bypass the blood-brain barrier to reach or have effects in the central nervous systems of mice, rats, monkeys and human beings following intranasal application (Hanson and Frey, 2008; Hallschmid et al, 2008; Thorne et al, 2008; Thorne et al, 2004).  Studies in rodents have shown radiolabeled NGF (Chen et al, 1998), insulin-like growth factor-I (IGF-I; 7.6 kDa; Thorne et al, 2004), insulin (5.8 kDa; Francis et al, 2009; Francis et al, 2008) and the cytokine interferon-β1b (18.5 kDa; Ross et al, 2004) are rapidly transported to the olfactory bulb, brainstem and many other brain and spinal cord areas after intranasal application; intranasal IGF-I and NGF have also been effective in rodent models of ischemic stroke (Liu et al, 2001) and Alzheimer's disease (Capsoni et al, 2002; De Rosa et al, 2005), respectively.   Intranasal fibroblast growth factor-2 (17 kDa) has been shown to increase neurogenesis in the olfactory bulb and subventricular zone of normal adult mice (Jin et al, 2003) and in the subventricular zone and hippocampus of rats subjected to transient focal ischemia (Wang et al, 2008).  In adult cynomolgus monkeys, radiolabeled interferon-β1b is transported to many different brain areas within an hour following intranasal administration (Thorne et al, 2008).  Importantly, a number of studies suggest this administration method may also be used to target the human central nervous system.  Intranasal insulin and melanocortin (960 Da) are detectable in human CSF less than 30 minutes following administration, with no elevation in serum levels (Born et al., 2002).  Intranasal insulin has also been reported to improve memory in normal adults as well as memory-impaired older individuals (Benedict et al, 2004; Reger et al, 2006; Reger et al., 2008; Benedict et al, 2008) and to improve motor development, cognitive function and spontaneous activity in a small clinical trial involving six young children with developmental delay due to 22q13 deletion syndrome (Schmidt, 2009).  Other clinical trials are in progress to test whether intranasal administration of an eight amino acid peptide, NAPVSIPQ (NAP; see Gozes et al, 2008), derived from a larger neurotrophic factor, is beneficial in patients with Alzheimer's disease, schizophrenia-related cognitive impairment and frontotemporal dementia.

Figure 2Intranasal administration of radiolabeled interferon-beta1b targets basal ganglia components in cynomolgus monkeys. (A) Autoradiograph of coronal brain section showing highest signal in the putamen and globus pallidus. (B) Autoradiographs of transverse midbrain sections (two successive rostral levels) showing highest signal in the substantia nigra. (Adapted from Thorne et al. Neuroscience, 2008.)How does it work?  The precise mechanisms involved in direct nose-to-brain transport are not yet clear.  However, the central distribution and kinetics of [¹²⁵I]-labeled protein entry in rats and monkeys suggests rapid, extracellular flow occurs along olfactory and trigeminal nerve components in the nasal epithelium to the olfactory bulb and brainstem, respectively, where dispersion to other CNS areas may be possible via pulsatile flow within the perivascular spaces of cerebral blood vessels (Figure 1; Thorne et al, 2004; Thorne et al, 2008).  The olfactory epithelium contains olfactory sensory neurons whose dendritic processes are exposed to the external environment, creating potential intracellular and extracellular pathways to the underlying submucosa and brain.  The respiratory nasal epithelium contains free trigeminal nerve endings close to the epithelial surface, allowing similar possibilities for transport.

Can nasally applied proteins target regions of the brain important for Parkinson's disease?  Intranasal application of the cytokine interferon-β1b to cynomolgus monkeys has been demonstrated to yield highest levels within the olfactory bulbs and trigeminal nerves less than an hour after administration, along with significant targeting of the caudate, putamen, globus pallidus, substantia nigra and nucleus accumbens (Thorne et al, 2008).   Importantly, basal ganglia components exhibited the highest levels among central nervous system regions other than the olfactory bulbs, dramatically illustrated by autoradiographic analysis of sections through the brain and brainstem (Figure 2).

Although much work remains to establish the precise mechanisms and pathways responsible for the direct transport of peptide and protein drugs into the central nervous system from the nasal passages, many published studies suggest nasal application may be a viable strategy for delivering neurotrophic factors to the brain.