PYCR1 is required to maintain skin integrity in humans and mice.

(A) Schematic representation of the mitochondrial proline metabolic pathway. P5CS mediates the conversion of glutamate into Δ1-pyrroline-5-carboxylate (P5C), which is then reduced into L-proline by PYCR1 or PYCR2 in a NAD(P)H dependent manner. The reverse reaction is catalyzed by PRODH and P5CDH. (B) Pedigrees of affected individuals from three consanguineous families and one non consanguineous family segregating with autosomal recessive cutis laxa type II. Square, male; circle, female; black shading, affected proband; double lines, consanguineous marriages; diagonal line, deceased. The patients in family 1 carry a homozygous change c.3G>A in the start codon of PYCR1. Family 2 carries an intronic single base change c.633+5G. The patients in families 3 and 4 carry missense mutations: c.721G>A and c.197T>G respectively. (C) Pictures from affected individuals from 2 to 13 years old show the typical progeroid triangular face (F1-II:3), thin and translucent skin in the abdomen (F2-II:2, F4-II:1), wrinkled skin affecting primarily the hands and feet (F1-II:3, F2-II:2, F4-II:1), prognathism and growth retardation (F3-II:3). (D) Schematic view of the PYCR1 gene and protein monomer indicating that p.Met1Ile and p.Leu66Pro are located in the NAD/NADH binding domain, whereas the p.Ala181_Leu211del and p.Ala241Thr are located in the dimerization domain. (E) Western blot showing that c.3G>A leads to a PYCR1 protein-null allele with concomitant reduction of PYCR2 protein levels. (F) Western blot showing that the p.Ala181_Leu211del mutation leads to a significant loss of full length PYCR1 protein (top band, lane 2), and the formation of smaller isoform PYCR1Δe5 detected by a commercial antibody directed against residues 1-171 of PYCR1 (bottom band, lane 2) and a custom antibody specific to PYCR1Δe5 (bottom band, lane 2). β-tubulin protein serves as a loading control. (G) Western blot of primary dermal fibroblasts shows that PYCR1 is absent in Pycr1-/- mice compared to WT cells. PYCR2 level is not affected by deletion of PYCR1. β-tubulin serves as a loading control. (H) Immunohistochemistry staining with PYCR1 (brown) antibody and hematoxylin (blue) illustrates that there is no PYCR1 expression in the dermis of the dorsal skin of Pycr1-/-animals at 80 days. Signal observed in the epidermis is background. Scale bar, 100 μm (I) Kaplan-Meier survival curves over 50 weeks show normal longevity of Pycr1-/- (n = 16 males, red line) compared to WT (n = 8 males, blue line) littermates. (J) Hematoxylin and Eosin (H&E) staining illustrates that the dermis of Pycr1-/- mice at day 80 is visibly thinner compared to that of WT, whereas the epidermis appears unchanged. Quantification shows that the dermis thickness of Pycr1-/- compared to WT mice is significantly reduced in the dorsal skin. No significant change in epidermal thickness can be documented. Scale bar, 100 μm. Lengths were calculated by averaging three measurements in each dorsal section (n = 6 mice/genotype). Two-tailed Student’s t test, ns p>0.05 ***p<0.001. (K) Masson Trichrome staining shows reduced number of collagen fibers (green) in Pycr1-/- mice (muscles are stained red and nuclei black). Quantification confirms that the number of collagen fibers is significantly reduced in Pycr1-/- compared to WT mice. Scale bar, 100 μm. Stainings were analyzed by measuring intensity using ImageJ. Images from three independent skin samples were analyzed and averaged in each group. Two-tailed Student’s t test, **p<0.01.

PYCR1 levels are reduced in human senescent fibroblasts and aged dermis.

(A) Western blot data show overexpression (o/e) of the transgenes P16, TP53 and BRAFV600E upon 4 days of doxycycline treatment. Levels of LB1 and DCR2 which are established senescence markers, decreased and increased respectively. PCNA, an established proliferation marker, is reduced upon doxycycline treatment. Endogenous PYCR1 and PYCR2 protein levels are reduced while PYCR3, P5CS and PRODH do not change significantly. (B) Senescence-induced by oxidative stress (H2O2) leads to LB1, PCNA, PYCR1 and P5CS protein level reduction, and no changes in PYCR2 levels. GAPDH serves as loading control. (C) Three independent primary dermal fibroblast lines derived from three young healthy individuals were serially cultured to replicative senescence. Western blots on the three dermal primary fibroblast lines show that endogenous PYCR1, P5CS and LB1 protein levels decrease with loss of replicative capacity in human primary fibroblasts. GAPDH levels are unchanged and serve as a loading control. (D) β-galactosidase staining quantification shows that the percentage of blue stained fibroblasts in the late PDs is significantly higher compared to the early PDs. Error bars indicate mean ± SEM. Two-tailed Student’s t test ***p<0.001, ****p<0.0001 (n = 6). (E) Representative immunohistochemistry images illustrating PYCR1 expression (brown) and hematoxylin (blue) on skin samples from young (15 to 30 years old) and old donors (70 to 80 years old). Black arrows indicate PYCR1 positive staining in the dermis. Pairwise comparison was performed according to gender (M, male; F, female) and anatomical region of the skin where the biopsy was taken. Scale bar, 50 μm. (F) Staining quantification for pairs 1 and 2 reveals that PYCR1 level is lower in the dermis of old compared to young individuals (n=5 pictures/pair). Percentages of PYCR1 positive cells were calculated in 5 independent pictures from each sample. In each picture 3 sections of 20 cells were measured and averaged. Error bars indicate mean ± SEM. Two-tailed Student’s t test, *p<0.05, **p<0.01, ***p<0.001, ****p<0.0001. (G) Graph demonstrates that the combined dermis quantification of PYCR1 positively stained cells is significantly lower in old compared to young individuals (n = 6 pairs). Error bars indicate mean ± SEM. Two-tailed Student’s t test, *p<0.05. (H) Western blot illustrates that PYCR1 and P5CS protein levels reduce in WT cells upon replicative senescence, like LB1, whose reduction indicates that senescence is induced. After overexpression of PYCR1, LB1 level reduces, suggestive of no changes in senescence. GAPDH serves as loading control. (I) Schematic representation summarizing the findings of this study. Congenital loss of PYCR1 results in premature aging, while normal aging leads to PYCR1 reduction in the dermal lineage.