Engineering of long-acting human growth hormone-Fc fusion proteins: Effects of valency, fusion position, and linker design on pharmacokinetics and efficacy

Abstract

Fc fusion proteins, formed by fusing an active protein to the Fc region of immunoglobulin G, are a validated strategy for extending the half-life of therapeutic proteins. Human growth hormone (hGH) Fc fusion proteins exhibit longer circulation half-lives than hGH, reducing injection frequency and improving convenience for hGH replacement therapy. Most approved Fc fusion proteins involve directly attaching the active protein to the hinge region of IgG Fc; however, few reports have described the effects of structural variations on these characteristics in detail. We analyzed pharmacokinetic and pharmacodynamic properties of various hGH-Fc fusion constructs differing in linker type, hGH valency, and fusion position to investigate the structure-function relationships of these proteins in cell-based assays and animal models, including normal and hypophysectomized rats. Monovalent hGH-Fc fusion variants and those with hGH fused to the C-terminal of IgG Fc exhibited higher in vitro and in vivo activity than bivalent hGH-Fc. However, these variants also exhibited accelerated clearance in rat pharmacokinetic experiments. The linker connecting the hGH moiety to the Fc domain significantly influenced in vitro activity and pharmacokinetics. Constructs with a rigid alpha-helical A(EAAAK)5A linker showed greater in vitro activity than those with a flexible (GGGGS)3 linker but exhibited accelerated clearance in rats. To a lesser extent, linker length influenced activity and pharmacokinetics. Bivalent hGH-Fc constructs with shorter linkers (0-1 GGGGS repeats) exhibited higher in vivo exposure (AUC) but lower in vitro activity than those with longer linkers (2-3 repeats). In vitro activity did not correlate linearly with linker length, as constructs with no linker (n = 0) showed reduced activity, while no consistent trend was observed for n = 1-3. These findings provide valuable insights into the design of hGH-Fc fusion proteins, offering a framework for systematically improving their potency and longevity and supporting the development of long-acting hGH therapies.

Fig 1. Schematic structure of human growth hormone Fc fusion proteins.

(A) Human growth hormone (hGH) fused to the hinge region of one Fc chain (monovalent hGH-Fc). (B) hGH fused to the C-terminal end of one Fc chain (monovalent Fc-hGH). (C) hGH fused to the hinge regions of both Fc chains (bivalent hGH-Fc). (D) hGH fused to the C-terminal ends of both Fc chains (bivalent Fc-hGH). Table 1 shows the various fusion proteins constructed in each category.

Fig 2. Expressed and purified hGH-Fc fusion proteins.

(A) Anti-Fc western blot analysis of hGH-Fc fusion proteins and byproducts in transient Chinese hamster ovary (CHO) cell expression supernatants. Lane 1: monovalent hGH-Fc with wild-type CH3 Fc domains. Lane 2: monovalent hGH-Fc with knob and hole mutations in the CH3 Fc domains. Lane 3: bivalent hGH-Fc. The expected band identities are indicated on the side of the blot. (B) Anti-Fc western blot analysis of hGH-Fc fusion proteins expressed by co-transfection with two plasmids at various ratios. The ratios of the co-transfected plasmids are indicated above the blot. (C) Sodium dodecyl-polyacrylamide gel electrophoresis (SDS-PAGE) of bivalent fusion proteins purified using protein A chromatography and monovalent hGH-Fc fusion proteins purified using hGH affinity chromatography followed by hydrophobic interaction chromatography. Original blot and gel images are provided in S1 raw image.

Fig 3. In vitro Nb2 cell-based activity assay results for various hGH-Fc fusion proteins.

Schematic diagrams to the right of each sample name represent the protein structure of the corresponding construct, with hGH moieties (circle) and Fc regions (rectangular) connected by linkers (wavy line). (A) Comparison of fusion position and valency effects: Di-hGH-(FL3)-Fc (bivalent, N-terminal fusion), Fc-(FL3)-di-hGH (bivalent, C-terminal fusion), and Mono-hGH-(FL3)-Fc (monovalent, N-terminal fusion). (B) Comparison of linker rigidity and valency effects: Di-hGH-(RL)-Fc (bivalent with rigid linker), Di-hGH-(FL3)-Fc (bivalent with flexible linker), and Mono-hGH-(FL3)-Fc (monovalent with flexible linker). (C) Comparison of optimized constructs: Fc-(FL3)-di-hGH (bivalent, C-terminal fusion), Fc-(FL3)-mono-hGH (monovalent, C-terminal fusion), and Mono-hGH-(RL)-Fc (monovalent, N-terminal fusion with rigid linker). Data are presented as the mean ± standard deviation (SD).

Fig 4. Pharmacokinetics of recombinant human growth hormone and hGH-Fc fusion proteins in Sprague–Dawley rats.

Recombinant human growth hormone (rhGH; 0.2 mg/kg) or hGH-Fc fusion proteins (1 mg/kg) were administered by single-dose subcutaneous injections. Serum concentrations were determined using a commercial hGH ELISA kit with custom standard curves for each fusion protein. Pharmacokinetic parameters were calculated using non-compartmental analysis in WinNonlin software. The pharmacokinetic parameters of (A), (B), (C), and (D) correspond to Experiments A, B, C, and D in Table 2, respectively. Data are presented as the mean ± standard deviation (SD).

Fig 5. Body weight gain of hypophysectomized rats treated with vehicle, rhGH, or hGH-Fc fusion proteins.

Rats received daily dosing of the vehicle or rhGH (days 0–13) or weekly dosing (days 0 and 7) of hGH-Fc fusion proteins for 2 weeks. Data are presented as the mean ± standard error of the mean (SEM). All hGH or hGH-Fc fusion protein-treated groups showed significant (p < 0.001) increases in body weight compared with the vehicle (PBS) group (⏺). Monovalent hGH-Fc administered at 350 µg/head/week (○) showed significantly (p < 0.05) higher body weight gain than the rhGH group (⏹).

Fig 6. Tibial growth plate widths of hypophysectomized rats.

After evaluating body weight gain, the hypophysectomized rats were euthanized, and tibial growth plate widths were measured. The numbers in the graph indicate the mean values for each group. All hGH or hGH-Fc fusion protein-treated groups showed significant (p < 0.01) increases in growth plate width compared to the vehicle (PBS) group. Asterisks (*) indicate significantly lower mean values (p < 0.05) compared with those of the rhGH group. Di-hGH-(FL3)-Fc (230 ug/week) and mono-hGH-(FL3)-Fc (350 ug/week) treatment group showed no significant differences with the rhGH group. No significant differences were observed among the hGH-Fc fusion proteins.

Fig 7. In vitro Nb2 cell-based activity of bivalent hGH-Fc fusion proteins with various linker lengths.

Constructs tested include Di-hGH-(NL)-Fc (no linker), Di-hGH-(FL1)-Fc (one GGGGS unit), Di-hGH-(FL2)-Fc (two GGGGS units), and Di-hGH-(FL3)-Fc (three GGGGS units). Data are presented as mean ± standard deviation (SD).

Fig 8. Pharmacokinetics of bivalent hGH-Fc fusion proteins with various linker lengths in Sprague–Dawley rats.

Bivalent hGH-Fc fusion proteins (1 mg/kg) were administered by single-dose subcutaneous injections. Data are presented as the mean ± standard deviation (SD).

Fig 9. Body weight gain of hypophysectomized rats treated with hGH-Fc fusion proteins with various linker lengths.

Rats received weekly dosing (days 0 and 7) of hGH-Fc fusion proteins with various linker lengths or vehicle control (PBS). Data are presented as the mean ± standard error of the mean (SEM). Asterisks (*) indicate a significant (p < 0.05) increase in body weight of the Di-hGH-(NL)-Fc compared to other hGH-Fc constructs. All hGH-Fc fusion protein-treated groups showed a significant (P < 0.001) increase in body weight compared to the vehicle (PBS) group.