Leveraging neonatal Fc receptor (FcRn) to enhance antibody transport across the blood brain barrier

Abstract

The blood-brain barrier (BBB) restricts efficient penetration of systemically delivered therapeutic antibodies into the brain, limiting the development of this class of drugs to treat neurodegenerative diseases. Here we demonstrate that the neonatal Fc receptor (FcRn), which is highly expressed at the BBB, can be used to facilitate IgG transport to the brain. Engineering of the antibody Fc region to promote binding to FcRn at neutral pH enhances antibody transcytosis in a cellular model. In vivo, these modifications improve brain penetration, as well as brain target engagement and activity, of systemically administered antibodies in both mice and non-human primates. This engineering approach can be broadly implemented to enhance central nervous system (CNS) exposure of antibody- and Fc-based drugs, improving the clinical potential of biotherapeutics for the treatment of human brain diseases.

Fig. 1. Antibodies with improved mFcRn affinity at neutral pH enhance brain penetration and target engagement in wild-type mice.

A–F A single dose (50 mg/kg) of control (anti-gD), anti-BACE1, or anti-BACE1-YTE hIgG1 antibodies was delivered by IV injection to wild-type (C57BL/6 J) (A–C) or hFcRn transgenic (Tg32) mice (D–F). Plasma antibody concentrations (A, D), brain antibody concentrations (B, E) and brain Aβ concentrations (C, F) were monitored for 7 days following dosing. In wild-type mice, anti-BACE1 YTE shows faster clearance (A), improved brain uptake (B) and improved pharmacodynamic activity (C) as compared to WT anti-BACE1. In hFcRn mice, improved serum exposure was noted for anti-BACE YTE over WT (D), while no differences in brain uptake or activity are seen (E, F). n = 4 biologically independent animals per group per time point. (G, H) A single IV injection (20 mg/kg) of anti-Aβ hIgG4 or anti-Aβ hIgG4-YTE was administered to PS2APP transgenic mice. 5 days later, animals were euthanized and brains harvested and assessed for target engagement (G, H). Antibody binding to Aβ in the mossy fiber hippocampal tract (G) and to peri-plaque associated halos in the prefrontal cortex (H) was visualized by immunostaining with anti-hIgG-Alexa594 antibody (red). Plaques are labeled with methoxy-X04 (blue). Scale bar: 500 µm (G), 200 µm (H). n = 4 biologically independent animals per group. I–K Non-binding antibodies bearing Fc mutations were administered to wild-type (SCID) mice by single dose IV injection (40 mg/kg). I Affinities (K_D) of mIgG2a variants to mFcRn at pH 6.0 and pH 7.4 as measured by SPR. Serum antibody concentration (J) and brain antibody concentration (K) were monitored for 42 days following dosing. n = 3 biologically independent animals per group per time point. mFcRn variants M252Y and M252Y.Q309T show faster clearance, while all variants show enhanced brain uptake through 10 days post-dose. (Statistical significance between Fc variant treated groups and respective WT treated group was evaluated over time by 2-way ANOVA or mixed effect analysis with Tukey or Dunnett’s multiple comparisons test using Graphpad Prism 9.5.1; *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001).

Fig. 2. Engineering of human IgG1 antibodies for improved FcRn affinity at neutral pH leads to increased transcytosis in vitro.

A Table showing the 9 residue positions in the CH2 and CH3 regions that were substituted, either individually or as a combination of 2, 3 or 4 positions. B Affinities (K_D) of the engineered variants to hFcRn at pH 6.0 and 7.4, as measured by SPR. C Schematic of the transcytosis assay: MDCK-II cells expressing hFcRn were cultured in a transwell plate. After addition of antibodies in the upper well, the plate was incubated for 24 h to allow transcytosis to the lower well, where antibodies are collected and their concentrations measured. D Transcytosis of all engineered Fc variants, normalized to WT hIgG1 Fc, compared to hFcRn affinity at pH 7.4. Transcytosis overall trended positively with affinity. Figure 2C was created in BioRender. Lafrance-Vanasse, J. (2025) https://BioRender.com/bc0teil.

Fig. 3. Human IgG1 antibodies with Fc engineered for improved FcRn affinity at neutral pH demonstrate increased serum clearance, enhanced brain uptake and partitioning, and improved target engagement in hFcRn mice.

A–C A single IV injection (50 mg/kg) of engineered human IgG1 Fc region variants was administered to hFcRn transgenic mice. Serum antibody exposure (AUC) over 7 days (A), brain antibody concentration at day 3 (B) and % brain:serum ratio at day 3 (C) are plotted as correlations to hFcRn affinity at pH 7.4 for each variant. n = 4 biologically independent animals per group per time point. YTE results from Fig. 1D–F are included to represent an antibody with improved hFcRn affinity at pH 6.0 but not 7.4. (Note: YY data points are derived from data in Fig. 3D, E). D–F A single IV injection (50 mg/kg) of anti-BACE1 hIgG1 WT or FcRn 7.4 variants YY or YQAY was administered to hFcRn transgenic mice. The Fc variant antibodies demonstrated faster serum clearance (D), higher brain antibody concentrations (E), and reduced brain Aβ concentrations (F) over 7 days post-dose. n = 5 biologically independent animals per group per time point. (Statistical significance between Fc variant treated groups and respective WT treated group was evaluated over time by 2-way ANOVA or mixed effect analysis with Tukey or Dunnett’s multiple comparisons test using Graphpad Prism 9.5.1; *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001).

Fig. 4. Anti-BACE1 hIgG1 variants with improved affinity to primate FcRn at neutral pH demonstrate faster serum clearance and improved brain uptake and enhanced CNS pharmacodynamic response in cynomolgus monkeys.

A single dose (50 mg/kg) of anti-BACE1 hIgG1 or hIgG1 Fc variants with enhanced binding to human and cynomolgus monkey FcRn at neutral pH (YY, YEY, YQAY, YPY) was delivered by IV injection to cynomolgus monkeys (n = 4 per group). Fc variants show reduced serum antibody concentrations compared to WT hIgG1 over the first 7 days post-dose (A), indicative of faster clearance. Serum antibody concentrations continue to drop over 28 days post-dose (B). Brain tissue was collected from animals dosed with anti-BACE1-WT, anti-BACE1-YY or anti-BACE1-YQAY at 2 or 7 days post-dose, and average brain antibody concentrations were determined (2 animals per time point) (C). BACE1 antibodies with modified Fcs were present at higher concentrations in brain than the anti-BACE1 WT Fc antibody. The ratio of the concentration of antibody in the brain relative to the concentration of antibody in the serum (D) shows that the modified Fcs (YY, YQAY) resulted in an increase in the proportion of anti-BACE1 antibody in the brain. Mean CSF antibody concentrations at 1, 2 or 7 days post-dose were also measured, and are reported as either absolute concentration (E) or as a ratio of the concentration in CSF relative to the concentration in the serum (F). Anti-BACE1 antibodies with modified Fcs showed a trend for greater CSF concentrations and partitioning as compared to anti-BACE1 with a WT Fc. Anti-BACE1 hIgG1 Fc variants showed a reduction in sAPPβ/ɑ ratio in the CSF (G) and brain (H). The extent of reduction was comparable in brain (open symbols) and CSF (closed symbols) (I). sAPPβ/α ratio decreased with higher brain concentrations of the Fc variant antibodies (J). Anti-BACE1 hIgG1 Fc variants showed a decrease in circulating endogenous levels of IgG (K). The number of biologically independent animals per group is: n = 4 for serum measurements, except Day 7 YY n = 2; n = 4 for CSF measurements, except Day 2 YY, YQAY and Day 7 YY n = 2; n = 2 for all brain measurements.. (Statistical significance between Fc variant treated groups and respective WT treated group was evaluated over time by 2-way ANOVA or mixed effect analysis with Tukey or Dunnett’s multiple comparisons test using Graphpad Prism 9.5.1; *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001).

Fig. 5. Anti-Aβ hIgG4 variants with improved affinity to primate FcRn at neutral pH demonstrate faster serum clearance, improved brain uptake and enhanced CNS target engagement in cynomolgus monkeys.

A single dose (50 mg/kg) of anti-Aβ hIgG4 or hIgG4 Fc variants with enhanced binding to FcRn at neutral pH (YY, YEY, YQAY), was delivered by intravenous injection to cynomolgus monkeys (n = 4 per group). Two animals each were euthanized at either 2 or 7 days post-dose to determine antibody exposure and target engagement in the brain and CSF. Fc variants show reduced serum antibody concentrations compared to WT hIgG1 over the first 7 days post-dose (A). Serum antibodies concentrations continue to drop over 28 days post-dose (B). Brain tissues were collected and homogenized to determine average brain antibody concentrations. C Anti-Aβ antibodies with modified Fcs (YY, YEY, YQAY variants) were present at higher concentrations in brain than the anti-Aβ WT Fc antibody. Mean CSF antibody concentrations at 1, 2 or 7 days post-dose were also measured (E). Anti-Aβ antibodies with modified Fcs (YY, YEY and YQAY) showed a trend for higher CSF concentrations compared to anti-Aβ with a wild-type Fc. The ratio of the concentration of anti-Aβ antibody in the brain (D) and CSF (F) to the concentration of anti-Aβ antibody in the serum shows that the modified Fcs (YY, YEY, YQAY) resulted in an increase in the proportion of anti-Aβ antibody in the CNS. Anti-Aβ hIgG4 variants showed increased target engagement in the brain compared to WT, as reflected by increased brain Aβ concentrations (G). Target engagement increased with an increase in brain concentration of the Fc variants (H). Anti-Aβ hIgG4 Fc variants showed a decrease in circulating endogenous levels of IgG (I). The number of biologically independent animals per group is: for serum and CSF measurements n = 7 at Days 0-2, and n = 5 at Days 4 and 7; n = 2 for all brain measurements. (Statistical significance between Fc variant treated groups and respective WT treated group was evaluated over time by 2-way ANOVA or mixed effect analysis with Tukey or Dunnett’s multiple comparisons test using Graphpad Prism 9.5.1; *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001).

Fig. 6. Relationship of FcRn affinity at neutral pH with serum clearance, reduction of circulating endogenous IgG levels, and partitioning to the brain and CSF.

FcRn affinity at pH 7.4 for the various antibodies tested in hFcRn mouse and NHP studies above is plotted against averages for normalized serum exposure (A), reduction in circulating endogenous IgG (B), fraction of antibody partitioning to brain (C), and fraction of antibody partitioning to CSF (D). Higher affinity antibodies were associated with lower serum exposure and a greater reduction in circulating endogenous IgG, as well as greater partitioning into brain and CSF. Model of cellular trafficking pathways at the BBB impacted by FcRn (E). For WT IgG, most antibodies are recycled back to the apical side (blood), with only a small proportion crossing the BBB by transcytosis (left side). Improved antibody interaction with FcRn at pH 7.4 promotes IgG transcytosis across the BBB (right side). Panel (E) was created in BioRender. Lafrance-Vanasse, J. (2025) https://BioRender.com/r1lvvuy.