In Vitro Impact of FSH Glycosylation Variants on FSH Receptor-stimulated Signal Transduction and Functional Selectivity

Abstract

FSH exists as different glycoforms that differ in glycosylation of the hormone-specific β-subunit. Tetra-glycosylated FSH (FSH²⁴) and hypo-glycosylated FSH (FSH^18/21) are the most abundant glycoforms found in humans. Employing distinct readouts in HEK293 cells expressing the FSH receptor, we compared signaling triggered by human pituitary FSH preparations (FSH^18/21 and FSH²⁴) as well as by equine FSH (eFSH), and human recombinant FSH (recFSH), each exhibiting distinct glycosylation patterns. The potency in eliciting cAMP production was greater for eFSH than for FSH^18/21, FSH²⁴, and recFSH, whereas in the ERK1/2 activation readout, potency was highest for FSH^18/21 followed by eFSH, recFSH, and FSH²⁴. In β-arrestin1/2 CRISPR/Cas9 HEK293-KO cells, FSH^18/21 exhibited a preference toward β-arrestin-mediated ERK1/2 activation as revealed by a drastic decrease in pERK during the first 15-minute exposure to this glycoform. Exposure of β-arrestin1/2 KO cells to H89 additionally decreased pERK1/2, albeit to a significantly lower extent in response to FSH^18/21. Concurrent silencing of β-arrestin and PKA signaling, incompletely suppressed pERK response to FSH glycoforms, suggesting that pathways other than those dependent on Gs-protein and β-arrestins also contribute to FSH-stimulated pERK1/2. All FSH glycoforms stimulated intracellular Ca²⁺ (iCa²⁺) accumulation through both influx from Ca²⁺ channels and release from intracellular stores; however, iCa²⁺ in response to FSH^18/21 depended more on the latter, suggesting differences in mechanisms through which glycoforms promote iCa²⁺ accumulation. These data indicate that FSH glycosylation plays an important role in defining not only the intensity but also the functional selectivity for the mechanisms leading to activation of distinct signaling cascades.

Keywords: biased agonism; follicle-stimulating hormone; follicle-stimulating hormone receptor; functional selectivity; glycosylation; macroheterogeneity; signal transduction.

Figure 1.

Typical glycans attached to human pituitary FSH, human recombinant FSH produced by Chinese hamster ovary cells (recFSH) glycoforms, and equine FSH (eFSH). The bars indicate the common-α and hormone-specific FSHβ subunits. N-glycosylation sites are indicated by the numbers below the bars. The eFSHα subunit has 4 additional residues at the N-terminus, accounting for the difference in numbering. The glycan at position β7 in eFSH (dotted area) is absent from 90% of the molecules present in highly purified preparations [34]. Glycans attached to recFSH were taken from the report by Mastrangeli et al [104]. Note that glycans at the α subunit of FSH^18/21 are in fact smaller (i.e. biantennary) than those present in FSH²⁴, which account for the different migration profiles of FSH^18/21 shown in Fig. 2D.

Figure 2.

(A) Superdex 75 chromatography of immunopurified pituitary hFSH. Three 1 × 30 cm Superdex 75 columns were connected in series and equilibrated with 0.2 M ammonium bicarbonate containing 20% acetonitrile. The chromatogram was developed with the same buffer using a flow rate of 0.4 mL/min at 25°C. Fractions under the FSH heterodimer peak were collected and 1-µg samples evaluated by Western blot using anti-FSHβ antibody 15-1.E3.E5 [45] diluted 1:2000 (inset). The solid bar indicates fractions pooled to obtain FSH²⁴ and the open bar indicates fractions pooled to obtain FSH^18/21. (B-D) FSH glycoform preparations. Samples of FSH²⁴ (VB-II-277A) and FSH^18/21 (VB-II-277B) were subjected to SDS-PAGE under reducing conditions and FSH was detected by (B) Coomassie Blue staining, (C) FSHβ Western blot, or (D) FSHα Western blot. (B) Lane 1, 5 µg FSH²⁴; lane 2, 2.5 µg FSH²⁴; lane 3, BioRad Precision Plus MW markers; lane 4, 2.5 µg FSH^18/21; lane 5, 5 µg FSH^18/21. (C,D) Western blots with anti-FSHβ antibody 15-1.E3.E5 [45] diluted 1:2000 and anti-FSHα antibody 15-2.C3.C5 [36, 105] diluted 1:1000, respectively. Lane 1, 1 µg FSH²⁴; lane 2, 0.5 µg FSH²⁴; lane 3, BioRad Precision Plus MW markers; lane 4, 0.5 µg FSH^18/21; lane 5, 1 µg FSH^18/21. Differences in migration between the α-subunits of FSH²⁴ (lanes 1 and 2) and FSH^18/21 (lanes 4 and 5) in panel D are due to the higher content of smaller, biantennary glycans at both Asn⁵² and Asn⁷⁸ in the latter, which led to a more rapid migration in the blot.

Figure 3.

(A) Equilibrium dissociation constant (KD) of FSH glycoforms as assessed by plasmon surface resonance (Biacore). Each bar represents the mean ± SD from 4 independent experiments. *P < 0.001 vs recFSH; **P < 0.01 vs eFSH. (B) Dose-response curves of total (intra- plus extracellular) cAMP production by HEK293-hFSHR⁺ cells exposed to the different FSH glycoforms for 3 hours. Data are presented as mean ± SD from 3 independent experiments. (Upper inset) Nonlinear regression curve of data shown in the main graph with the corresponding ED₅₀ values for each glycoform (*P < 0.01 eFSH vs all others; **P < 0.05 FSH^18/21 vs FSH²⁴; ***P ≤ 0.02 recFSH vs FSH^18/21 and FSH²⁴). (Lower inset) E_max of each glycoform from the data shown in the main graph (*P < 0.001 eFSH and recFSH vs FSH^18/21 and FSH²⁴; **P = 0.01 FSH^18/21 vs FSH²⁴). (C) Kinetic curves of total (intra- plus extracellular) cAMP production by HEK293-hFSHR⁺ cells exposed to 50 ng/mL (eFSH, FSH^18/21, and FSH²⁴) or 100 ng/mL (recFSH) concentrations of FSH glycoforms. (Inset) Linear regression of the curves shown in the main graph, using only points until cAMP production reached maximum values (*P < 0.01 eFSH vs all other glycoforms; **P = 0.03 recFSH vs FSH^18/21; P = N.S. recFSH vs FSH²⁴, and FSH^18/21 vs FSH²⁴). (D) Dose-response curves for pSOMLuc expression by HEK293-hFSHR⁺ cells transiently transfected with the cAMP-sensitive pSOMLuc reporter plasmid and exposed to different FSH glycoforms. Each point represents the mean ± SEM of 3 independent experiments. *P < 0.02 eFSH vs FSH²⁴ and recFSH vs FSH24; **eFSH vs FSH24. The levels of significance shown in the right side of the figure correspond to the differences between the FSH glycoform curves considering all doses “en bloc.”

Figure 4.

ERK1/2 phosphorylation levels stimulated by exposure of HEK293-hFSHR⁺ cells to FSH glycoforms. (A) Dose-response curves for ERK phosphorylation stimulated by increasing concentrations of each FSH glycoform. Results are expressed as the pERK/total ERK ratio calculated by densitometric analysis of the blots. Representative immunoblots from a single experiment are shown at the right of the graph. Comparisons among doses and dose-response curves: *P < 0.03 FSH^18/21 vs FSH²⁴ at 10, 30, 100, and 300 ng/mL doses; dose-response curves (all doses “en bloc”): **P < 0.04 FSH^18/21 vs recFSH and FSH²⁴; ***P < 0.01 FSH²⁴ vs recFSH and eFSH. (B) ERK1/2 phosphorylation levels during 120 minutes of exposure of HEK293-hFSHR⁺ cells to 50 ng/mL of each FSH glycoform and quantified by densitometric analysis of the corresponding immunoblots (see Figs. 5A and 5B for statistical comparisons among areas under the pERK1/2 curve and pERK levels at 5 minutes). Representative immunoblots are shown to the right of the graph. Data were normalized to the maximal recFSH-stimulated ERK phosphorylation at 5 minutes, arbitrarily chosen as 100%. Graphs shown in (A) and (B) are the means ± SD of 3 independent experiments. C. Representative immunoblot of β-arrestins 1 and 2 in hFSHR⁺ and hFSHR^-control HEK293 cells (lanes 1 and 2, respectively) and in CRISPR β-arr1/2 KO cells (lane 3).

Figure 5.

pERK levels at 5 min and areas under the curves (AUC) for the initial (0–15 minutes) and late (15–120) pERK responses to FSH glycoforms. (A) AUC of pERK (total area corresponding to the 0 to 120 minute period as well as to the early (0–15 minutes) and late (15–120 minutes) pERK responses in lysates of HEK293-hFSH⁺ cells exposed to 50 ng/mL of FSH glycoforms (see Fig. 4B). *P < 0.01 vs all others; **P < 0.01 vs eFSH and recFSH; ***P < 0.01 vs recFSH; ****P = 0.05 vs recFSH. (B) Percent of pERK1/2 normalized to the maximal response of recFSH at 5 minutes, arbitrarily chosen as 100%. *P < 0.01 vs all others; **P < 0.01 vs eFSH. (C) Percent of pERK1/2 in CRISPR β-arr1/2 HEK293-hFSHR⁺ KO cells, relative to control β-arr1/2 non-KO HEK293-hFSHR⁺ cells. *P < 0.01 vs all others; **P ≤ 0.05 vs all others. (D) Percent of pERK1/2 in CRISPR β-arr1/2 HEK293-hFSHR⁺ KO cells in the presence of H89, relative to CRISPR β-arr1/2 HEK293-hFSHR⁺ KO cells in the absence of H89. *P < 0.01 vs FSH²⁴ and recFSH, and P < 0.05 vs eFSH; **P < 0.01 vs recFSH; ***P < 0.01 vs FSH²⁴ and recFSH, and P = 0.05 vs eFSH; ****P ≤ 0.05 vs recFSH and eFSH.

Figure 6.

ERK1/2 phosphorylation in control (open symbols) and CRISPR β-arr1/2 HEK293-hFSHR⁺ KO cells (colored symbols). Each panel (A–D) shows the kinetics of pERK response to 50 ng of each glycoform. Results are expressed as the pERK/total ERK ratio calculated by densitometric analysis of the blots; data were normalized to the maximal ERK phosphorylation response in control cells at 5 minutes, arbitrarily chosen as 100%. The insets show representative immunoblots for each glycoform in control and KO HEK293 cells. Because the pERK signal for FSH²⁴ was weak (see Fig. 4A), the blot shown in the inset was overexposed to better appreciate the differences between control and β-arr1/2 KO HEK293-hFSHR⁺ cells. See Fig. 5C for statistical differences between the glycoforms.

Figure 7.

Intracellular iCa²⁺accumulation in HEK293-hFSHR + cells expressed as the area under the FSH-stimulated curve calculated from the data presented in the digital research materails repository [68]. Areas under the curve were calculated considering the first significant change in fluorescence (1/n mean fluorescence intensity) following addition of FSH and the last value of the fluorescence descending curve before reaching baseline levels (see curves in the digital research materails repository [68]). Each panel corresponds to different conditions (A and B) presence or absence of Ca²⁺ in the incubation buffer, (C) addition of EGTA, (D) or pretreatment with thapsigargine (TG) before FSH addition. Each symbol represents the normalized result of an independent experiment (see Fig. S2); horizontal and vertical lines indicate the mean ± SD from 4 to 8 independent experiments, respectively. Differences among FSH glycoform responses are shown on the top of each graph. N.S.: not significant. Different scales in the Y-axes were used to allow better appreciation of the differences among glycoforms.

Figure 8.

Desensitization of the hFSHR in HEK293-hFSHR⁺ cells exposed to different FSH glycoforms. (A) Cells were stimulated for 2 hours with increasing amounts of each glycoform in the presence of IBMX, and cAMP in the incubation media was measured by RIA. *All treatments were significantly different from each other at P < 0.01. (B) After stimulation with FSH, cells were washed twice and then rechallenged with a saturating (1200 ng/mL) dose of each FSH glycoform in the presence of IBMX; total (intra- and extracellular) cAMP was then determined. *P < 0.01 eFSH and recFSH vs FSH^18/21 and FSH²⁴ in cells preexposed to 10 and 25 ng/mL FSH; **P < 0.01 eFSH vs all other glycoforms in cells preexposed to 50, 75, and 100 ng/mL FSH; ***P < 0.01 recFSH vs FSH^18/21 and FSH²⁴ in cells preexposed to 50 ng/mL; ****P < 0.01 FSH^18/21 vs FSH²⁴ and recFSH at the 100 ng/mL preexposed FSH dose. Inset: Nonlinear regression of the data shown in the main graph. Level of significant differences among the glycoform curves in the inset graph (all doses “en bloc”): **P < 0.01 eFSH vs all other glycoforms; ***P < 0.01 recFSH vs FSH^18/21 and FSH²⁴.