Loss of SPRED3 Causes Primary Hypothyroidism and Alters Thyroidal Expression of Autophagy Regulators LC3, p62, and ATG5 in Mice

Abstract

Sprouty-related proteins with enabled/vasodilator-stimulated phosphoprotein homology 1 (EVH1) domain (SPREDs) are negative regulators of the Ras/MAPK signaling pathway and are known to modulate developmental and endocrine processes. While the roles of SPRED1 and SPRED2 are increasingly understood, the physiological relevance of SPRED3 remains elusive. To elucidate its function, we generated SPRED3 knockout (KO) mice and performed phenotypic, molecular, and hormonal analyses. SPRED3-deficient mice exhibited growth retardation and a non-Mendelian genotype distribution. X-Gal staining revealed Spred3 promoter activity in the thyroid, adrenal gland, pituitary, cerebral cortex, and kidney. Hormonal profiling identified elevated thyroid-stimulating hormone (TSH) and reduced thyroxine (T₄) levels, indicating primary hypothyroidism. Thyroidal extracellular signal-regulated kinase (ERK) signaling was mildly reduced in SPRED3 KO mice, and immunoblotting revealed altered expression of autophagy regulators, including reduced sequestosome 1 (p62), increased autophagy-related gene 5 (ATG5), as well as an elevated microtubule-associated protein 1 light chain 3 (LC3) II/I ratio and a decreased pBeclin/Beclin ratio in SPRED3 KO mice. Our findings indicate that SPRED3 is involved in thyroidal homeostasis and plays a regulatory role in autophagy processes within the thyroid gland.

Keywords: KO mice; SPRED; SPRED3; autophagy; hypothyroidism.

Figure 1

Generation of the SPRED3 KO mouse line, as well as genotyping of SPRED3 WT, Het, or KO mice by PCR analysis. (a) The gene trap vector (PRPGS00122_A_D01) includes elements like the reporter gene lacZ, a neomycin (neo) resistance gene, FRT sites, as well as loxP sites. The flanking upstream splice acceptor and downstream transcriptional termination sequences (polyadenylation sites) ensure that transcription is terminated and that the target gene is no longer transcribed. Insertion of this vector was conducted via homologous recombination, enabling precise genetic modification. Ultimately, the resulting protein consists solely of β-galactosidase (β-gal). For our purposes, it was not mandatory to make use of the conditional potential of the KO construct. Created in BioRender; (b) Representative X-Gal staining of cultured EPD0481_1 D11 embryonal stem cells demonstrating functionality of the SA, necessary for efficient SPRED3 knockout; (c) Genotyping of mice via PCR. The PCR products were separated by agarose gel electrophoresis to distinguish between WT, Het, or KO genotypes. WT mice displayed a single band at 368 bp, whereas KO mice showed a single band at 379 bp. Het mice exhibited both bands, confirming the presence of a WT and a KO allele. FRT = Flippase recognition target, loxP = locus of X-over P1, SD = splice donor, SA = splice acceptor, pA = polyadenylation site, Ex = exon. PC = positive control (heterozygous mouse), NC = negative control (H₂O), M = marker.

Figure 2

Transcriptional and translational verification of SPRED3 knockout mouse line. (a) Comparative Cq values for GAPDH and SPRED3 in WT and KO samples after 45 cycles. The Cq value is plotted on the y-axis, where a lower value indicates a higher initial amount of target cDNA. While GAPDH (left) showed nearly identical Cq values in both WT and KO samples, SPRED3 (right) was exclusively detected in WT mice. Data are presented as mean ± SEM (WT and KO n = 5); (b) Expression of SPRED3 in adrenal gland lysates of SPRED3 KO mice. Representative Western blot showing SPRED3 expression in adrenal gland lysates of SPRED3 D11 KO mice in comparison to WT control.

Figure 3

Impact of SPRED3 deficiency on growth dynamics and genotypic distribution. (a,b) Nonlinear regression curves representing body weight development over 800 days in male (left) and female (right) WT and SPRED3 KO mice (WT male n = 28, KO male n = 37, WT female n = 24, KO female n = 35). SPRED3 KO males and females exhibit distinctively reduced body weight compared to WT littermates throughout development. Data are represented as weight in grams [g] plotted against age in days; (c) Genotypic distribution of offspring from heterozygous crosses, showing the expected (white bars) versus observed (filled bars) numbers of WT, Het, and SPRED3 KO mice (WT expected n = 135, Het expected n = 294, KO expected n = 135, WT observed n = 147, Het observed n = 295, KO observed n = 100). A significant reduction in SPRED3 KO mice was observed compared to Mendelian expectations. Data are presented as absolute counts of mice per genotype. * p < 0.05.

Figure 4

X-Gal staining of SPRED3 KO mouse tissue indicates Spred3 promoter activity in the brain, kidney, adrenal gland, pituitary, and thyroid gland. (a) Coronal brain sections depict strong staining within the hippocampal formation, cerebral cortex, and amygdala; (b) In the kidney, prominent promoter activity was predominantly observed in the cortical region. Additionally, intense staining was detected in a nerve section (arrow); (c) In the adrenal gland, promoter activity was exclusively detected in the adrenal medulla; all other regions of the adrenal gland showed no detectable Spred3 promoter activity; (d) The pituitary gland exhibited the strongest staining in the pars intermedia; (e) In the thyroid gland, high Spred3 promoter activity was observed in the thyroid follicles. CTX = cerebral cortex, HPF = hippocampal formation, AMY = amygdala, SCTX = subcortical areas, M = medulla, C = cortex, AH = adenohypophysis, NH = neurohypophysis, THY = thyroid gland, T = trachea.

Figure 5

Hormonal analyses reveal primary hypothyroidism in SPRED3-deficient mice, characterized by elevated TSH and reduced T₄ levels. (a–d) Serum concentrations of adrenaline (WT n = 6, KO n = 9), noradrenaline (WT n = 5, KO n = 9), dopamine (WT n = 6, KO n = 9) and corticosterone (WT n = 12, KO n = 17) remain unchanged between SPRED3 WT and KO samples; (e) GH serum levels of SPRED3 WT and KO mice older than 120 days indicate significant decrease within KO samples (WT n = 8, KO n = 3); (f,g) Hormonal analyses of the thyroid axis reveal primary hypothyroidism in SPRED3 KO mice with significantly increased TSH (WT n = 8, KO n = 5) and markedly reduced T₄ levels (WT n = 21, KO n = 22). (h) Exemplary hematoxylin/eosin-stained [5] tissue section comprising thyroid gland (left), and detail enlargement (right) as used for the estimation of the ratio of colloid area (red) to thyroid area. (i) Increased relative colloid area in SPRED3-deficient mice (WT n = 11 sections, KO n = 8 sections). * p < 0.05, ** p < 0.01. n.s. = not significant.

Figure 6

SPRED1 compensates for SPRED3 deficiency in mice, causing a significantly reduced p-ERK/ERK ratio. (a,b) Representative Western blots illustrating the expression of p-ERK1/2, ERK1/2 (WT/KO n = 10), and SPRED1 (WT n = 10, KO n = 9) in thyroid gland lysates from SPRED3 KO mice and WT controls. GAPDH was used as a loading control to ensure equal protein loading; (c–f) Densitometric quantification of protein levels normalized to GAPDH. The phosphorylation ratio of p-ERK1/2 to total ERK1/2 was also calculated. The quantification indicates a significant decrease in the p-ERK 1/2 to ERK 1/2 ratio in SPRED3 KO mice compared to WT controls. (g) Best model of structure and interaction prediction by ChimeraX. SPRED3 blue, SPRED1 magenta, distance below 5 Ångström in yellow, distance above 5 Ångström in red. The error plot on the right side shows an exact prediction of the EVH1 and Sprouty domains (yellow and blue colors) as well as a possible interaction between the EVH1 and Sprouty domains (yellow color). Data are presented as mean ± SEM. Dots represent the number of used mice (WT p-ERK 1/2 and ERK 1/2 n = 26, KO p-ERK 1/2 and ERK 1/2 n = 24; WT SPRED1 n = 10, KO SPRED1 n = 9). * p < 0.05, ** p < 0.01, *** p < 0.001. **** p < 0.0001.

Figure 7

SPRED3 KO mice exhibit altered expression profiles of key autophagic regulator LC3. (a) Sequence alignment of murine SPRED1 and SPRED3 (mSPRED3; mSPRED1) within the SPR domain containing the LIR motifs (rectangles) highlight salient conservation of amino acids essential for efficient binding of LC3; (b) Representative Western blots illustrate the expression of LC3 (WT n = 5, KO n = 5), in thyroid gland lysates from SPRED3 D11 KO mice and WT controls. GAPDH was used as a loading control to ensure equal protein loading; (c) Densitometric quantification of protein levels. Quantification analysis shows an increased LC3-II/I ratio in SPRED3 KO thyroid lysates relative to WT controls (WT n = 15, KO n = 15). (d) Best model of structure and interaction prediction by ChimeraX. SPRED3 blue, LC3 magenta, distance below 5 Ångström in yellow, distance above 5 Ångström in red. The error plot on the right side shows an exact prediction of the EVH1 and Sprouty domain of SPRED3 and an exact prediction of LC3 structure (yellow and blue colors), as well as a possible interaction between the Sprouty domain, containing the LIR and LC3 (yellow color). Data are presented as mean ± SEM. Dots represent the number of used mice. * p < 0.05.

Figure 8

SPRED3 KO mice exhibit altered expression profiles of key autophagic regulators involving p62, Beclin, and ATG5. (a,b) Best models of structure and interaction predictions by ChimeraX. SPRED3 blue, p62 magenta (a), SPRED3 blue, ATG5 magenta (b), distance below 5 Ångström in yellow, distance above 5 Ångström in red. The error plots on the right side shows an exact prediction of the EVH1 and Sprouty domain of SPRED3 and an exact prediction of the N-terminal domain of p62 and ATG5 structure (yellow and blue colors) as well as a possible interaction between the Sprouty domain and p62, and the EVH1 domain with ATG5 (yellow and blue); (c,d) Co-immunoprecipitation with anti-SPRED3 co-precipitated p62 (c) and ATG5 (d). IP: precipitating antibody, irrel.: irrelevant, IB: antibodies used in subsequent immunoblot. (e) Representative Western blots illustrate the expression of p62, pBeclin, Beclin, and ATG5 in thyroid gland lysates from SPRED3 KO mice and WT controls. GAPDH was used as a loading control to ensure equal protein loading. (f–h) Densitometric quantification of protein levels. Quantification analysis revealed an increase in the pBeclin/Beclin ratio and a significant decrease in p62 levels in SPRED3 KO thyroid lysates relative to WT controls. ATG5 was significantly upregulated in the KO group. Data are presented as mean ± SEM. Dots represent the number of used mice (WT pBeclin/Beclin n = 8, KO n = 8; WT p62 n = 31, KO p62 n = 29; WT ATG5 n = 45, KO ATG5 n = 41). * p < 0.05, ** p < 0.01.