The identification of leptin-derived peptides that are taken up by the brain

Abstract

We studied the brain uptake of leptin and of a set of peptides whose combined sequences spanned the entire mature human leptin protein. We compared their uptake to that of albumin and IgG. Two of these peptides, consisting of residues 1–33 and 61–90, demonstrated brain uptake on a par with leptin protein itself, and significantly higher than the uptake of albumin and IgG. Further investigation revealed a peptide, 12–32, with higher uptake than its parent peptide 1–33. Peptide 61–90 had the highest brain uptake, and this was shown to be saturable. Comparison of these brain-permeant peptides with the published structure of the leptin:leptin receptor complex revealed a high degree of correlation. All of the leptin residues that have been identified as important receptor-binding contacts appeared to have a role in brain uptake, indicating that receptor binding is an intrinsic part of transport across the blood–brain barrier. The effect of these peptides as leptin agonists or antagonists remains to be investigated. The newly identified peptides also have a potentially large role as carrier molecules for new brain therapeutics, since peptides can be readily coupled to other molecules.

Introduction

Leptin, a 146 amino acid polypeptide secreted into the bloodstream by adipocytes, acts centrally on leptin receptor-expressing cells located in the hypothalamus and other parts of the brain. Leptin acts as a satiety signal, leading to decreased food intake and retarding weight gain. Leptin has a molecular weight of 16 kDa, high enough to preclude the possibility of passive diffusion across the blood–brain-barrier, a process that provides limited entry for some relatively small peptides [1], [2]. The observations that peripherally administered leptin stimulates cells in the brain [3], [4], and that CSF leptin levels are strongly correlated to plasma leptin levels [5], therefore, imply the presence of a transport mechanism from blood to brain. A number of investigators have investigated the brain uptake of leptin, and all have reported that leptin is, indeed, taken up by the brain [6], [7], [8], [9], [10], [11]. Present evidence suggests that there may be at least two distinct transport mechanisms; a saturable one mediated by the leptin receptor [7], [12], and a mechanism that is independent of the leptin receptor but also saturable [13], [14]. It is apparent that leptin is taken up predominantly into the extracellular fluid of the brain, from which it diffuses into the CSF [8], although some direct uptake by the CSF also occurs [11].

The transport systems for leptin uptake are likely to be of low capacity. Evidence for this comes from studies showing that plasma leptin concentrations of the order of 0.15 to 5 ng/ml, much less than those present in obesity, are sufficient to competitively inhibit the brain uptake of exogenous leptin [15]. It is conceivable, therefore, that brain uptake may, in some circumstances, act as a limiting step in leptin signaling. Further, there is specific impairment of leptin transport in obesity. This impairment has been shown to be above and beyond the competitive inhibition due to high plasma leptin concentration in obese rats [16], [17]. This helps to explain the leptin resistance found in obesity. This impairment may also help explain why leptin has, so far, been unhelpful in the treatment of obesity in humans. Clearly, it is crucial that we learn more about the processes of brain uptake of leptin.

We undertook this study in an attempt to define the leptin sequences involved in uptake by the brain. We reasoned that certain peptide fragments of leptin may contain sequences important for binding to the leptin transporter, and that identifying these would be an important step towards understanding the process of brain uptake. It also seemed feasible that the smaller size of peptides could confer transport advantages, compared to the full-sized protein. An additional motivation for this study, therefore, was the possibility of developing leptin mimetics with superior brain uptake.

The in vivo measurement of brain uptake parameters after intravenous administration typically requires a large number of animals, as blood and brain levels must be measured at many time points. This requires that at least one, and perhaps several, animals be killed at each time-point [18]. This was not feasible for the present study, in which we aimed to screen the uptake properties of some 15 different peptides, as well as leptin itself and several control proteins. We therefore chose a simpler method; we administered an intravenous bolus of peptide, and measured the brain and plasma levels at a fixed interval after this. We then compared the brain-to-plasma levels of different peptides. This method has proven to be robust in previous studies of the brain uptake of peptides [19] and other compounds [20]. For the sake of comparison, we also measured uptake of the proteins albumin and IgG. Both of these proteins are known to be taken up into brain to a small extent, and it is thought that IgG may also be actively pumped out of the brain [22].