Enter the nose (literally). In tomorrow's Nature Neuroscience, Jan Born and colleagues, at the Universities of Luebeck and Marburg in Germany, and Texas Tech University in Amarillo, report that they detected three different peptides in the CSF of 36 healthy human volunteers as early as 10 minutes after these people had sniffed them.

The researchers used melanocortin, vasopressin, and insulin, and then measured each of these peptide's concentration in CSF and serum for the next 80 to 120 minutes. Within 80 minutes, all three accumulated in CSF to statistically significant proportions. Total increases varied from person to person, but generally, concentrations rose within ten minutes after nasal administration, peaked between 30 and 80 minutes, and had not fully returned to baseline after 120 minutes.

This study is new in that it validates, in humans, earlier animal data suggesting that intranasally delivered peptides can penetrate the brain directly (Illum, 2000). The authors write that the rapid time course suggests the peptides were not transported inside neurons but rather diffused through intracellular clefts in the olfactory epithelium and into the subarachnoid space. Molecular weight, as well as lipophilic and ionic properties presumably affect the peptides' speed and reach into the brain.

This is a small study that used high doses of peptide. It does not establish intranasal administration as more effective than intravenous administration. It does, however, suggest it may be worthwhile exploring nasal drug delivery to achieve biologically significant brain concentrations of neuropeptides that have unwanted effects in the periphery.-Gabrielle Strobel.

Reference:
Born J, Lange T, Kern W, McGregor GP, Bickel U, Fehm HL. Sniffing neuropeptides: a transnasal approach to the human brain. Nat Neurosci. 2002 Jun;5(6):514-6. No abstract available. Abstract

Q&A with Jan Born:

Q: Do you know some of the delivered peptide was not inhaled and may have reached the brain through the blood?
A: Certainly some amount of peptide following nasal delivery is absorbed into the blood stream via the nasal, bronchial and alveolar epithelium. However, we measured peptide concentrations after nasal administration concurrently in blood and found that, except for vasopressin, the concentrations were negligible compared with those in CSF.

Q: You mentioned potential usefulness of this delivery method for Alzheimer's. Which compounds could be tested? Are you working toward this goal?
A: We are performing one study with elderly people who show slight memory deficits. This study investigates whether cognitive performance can be enhanced in these people by intranasal administration of insulin. However, it is clearly too early for any conclusions based on our very preliminary (though positive) data. I know of other labs running similar studies. I can imagine that peptides other than insulin (e.g. neuronal growth factors) will turn out also to be important. Our own research in this regard is directed mainly towards peptide effects on weight regulation and sleep.

Q: Does your method also work with other types of peptides, for example antibody fragments against pathogenic targets in AD?
A: The nasal delivery should work with all peptides. However, limits are probably set by molecule size. We found less access to CSF for insulin than for MSH/ACTH4-10, which is a very small molecule. So with even larger molecules like BDNF or NGF, as well as antibody fragments, the passage to the brain might turn out to be biologically insignificant, unless huge amounts are administered. On the other hand, animal studies provided evidence that even larger proteins may reach the brain after intranasal administration (perhaps via axonal transport).

Q: Has this never been shown in humans before?
A: To my knowledge, our study is the first providing this direct evidence in humans.

Note by ARF:

At the recent conference "7th Neurodegenerative Disorders: Common Molecular Mechanisms" in Montego Bay, Jamaica, Perry, a neuropharmacologist, presented data suggesting that the insulinotropic peptide hormone GLP-1, which is currently in clinical trials for the treatment of type 2 diabetes mellitus, has neuroprotective effects on cultured neurons in vitro and can limit neurodegeneration in a rat model of an excitotoxic lesion that produces a basal forebrain cholinergic deficit.-Gabrielle Strobel