Brain uptake pharmacokinetics of incretin receptor agonists showing promise as Alzheimer's and Parkinson's disease therapeutics

Abstract

Among the more promising treatments proposed for Alzheimer's disease (AD) and Parkinson's disease (PD) are those reducing brain insulin resistance. The antidiabetics in the class of incretin receptor agonists (IRAs) reduce symptoms and brain pathology in animal models of AD and PD, as well as glucose utilization in AD cases and clinical symptoms in PD cases after their systemic administration. At least 9 different IRAs are showing promise as AD and PD therapeutics, but we still lack quantitative data on their relative ability to cross the blood-brain barrier (BBB) reaching the brain parenchyma. We consequently compared brain uptake pharmacokinetics of intravenous ¹²⁵I-labeled IRAs in adult CD-1 mice over the course of 60 min. We tested single IRAs (exendin-4, liraglutide, lixisenatide, and semaglutide), which bind receptors for one incretin (glucagon-like peptide-1 [GLP-1]), and dual IRAs, which bind receptors for two incretins (GLP-1 and glucose-dependent insulinotropic polypeptide [GIP]), including unbranched, acylated, PEGylated, or C-terminally modified forms (Finan/Ma Peptides 17, 18, and 20 and Hölscher peptides DA3-CH and DA-JC4). The non-acylated and non-PEGylated IRAs (exendin-4, lixisenatide, Peptide 17, DA3-CH and DA-JC4) had significant rates of blood-to-brain influx (Ki), but the acylated IRAs (liraglutide, semaglutide, and Peptide 18) did not measurably cross the BBB. The brain influx of the non-acylated, non-PEGylated IRAs were not saturable up to 1 μg of these drugs and was most likely mediated by adsorptive transcytosis across brain endothelial cells, as observed for exendin-4. Of the non-acylated, non-PEGylated IRAs tested, exendin-4 and DA-JC4 were best able to cross the BBB based on their rate of brain influx, percentage reaching the brain that accumulated in brain parenchyma, and percentage of the systemic dose taken up per gram of brain tissue. Exendin-4 and DA-JC4 thus merit special attention as IRAs well-suited to enter the central nervous system (CNS), thus reaching areas pathologic in AD and PD.

Keywords: Alzheimer’s disease; Blood–brain barrier; Incretin receptor agonists; Parkinson’s disease; Pharmacokinetics.

Fig 1.

Clearance of ¹²⁵I-IRAs from serum. The level of radioactivity in serum over 60 min clock time is graphed in (A). The linear distribution phase during the first 15 min when represented as log(%Inj/mL) was used to calculate the clearance (B). The half-time clearance from blood for each IRA is listed in Table 3.

Fig. 2.

Multiple-time regression analysis of ¹²⁵I-IRAs across the BBB. The brain to serum (B/S) ratios of radioactivity, corrected for vascular space, present in whole brain over the entire time curve are graphed in (A). The linear first 10 min clock time of the experiments are shown in (B) and are used to calculate the unidirectional influx rate, K_i for Fig. 3 and Table 3.

Fig. 3.

Rate constants of ¹²⁵I-IRAs transport into whole brain. The unidirectional influx rates, K_i (slope) and V_i (y-intercept) are listed in Table 3. n=10-14/IRA

Fig. 4.

Percent of the iv injected dose of ¹²⁵I-IRAs taken up per gram of brain tissue (%Inj/g) corrected for the initial level of vascular binding (V_i). The values for all the tested IRAs other than exendin-4 plateaued by 30 min. n=1-2/time point/IRA

Fig. 5.

Characterization of exendin-4 influx into brain. The influx of ¹²⁵I-exendin-4 was enhanced by WGA about 45% in the time span of the study. Results are expressed in terms of tissue to serum (T/S) ratios.

Fig. 6.

Characterization of liraglutide efflux from brain. Brain-to-blood efflux of ¹²⁵I-liraglutide was not inhibited by 1 μg unlabeled liraglutide, consistent with a non-saturable efflux mechanism. Results are expressed as % Transport defined in section 2.12. Data are presented as mean ± SEM with n = 10 per group.