Pnpla5-knockout rats exhibit reduced expression levels of proteins involved in steroid metabolism and wound healing compared to wild-type rats

Abstract

Background

Patatin-like phospholipase domain containing 5 (PNPLA5) is a newly-discovered lipase. Although the PNPLA family plays critical roles in diverse biological processes, the biological functions of PNPLA5 mostly unknown. We previously found that the deletion of Pnpla5 in rats causes a variety of phenotypic abnormalities. In this study, we further explored the effects of Pnpla5 knockout (KO) on male rats.

Results

The body weight and testicular or epididymal tissue weight of three to six 3-month-old Pnpla5 KO or wild-type (WT) male Sprague–Dawley rats were measured. The protein expression levels were also measured via western blotting and iTRAQ (isobaric tags for relative and absolute quantitation) analyses. No significant difference between Pnpla5 KO and WT rats, regarding body weight, testicular or epididymal tissue weight, or hormone levels, were found. However, the relative testicular tissue weight of the KO (Pnpla5^−/−) rats was higher (P < 0.05) than that of WT rats. Significant increases in apoptotic cells numbers (P < 0.001) and BAX and Caspase-9 expression levels were observed in the testicular tissue of Pnpla5^−/− rats. Moreover, iTRAQ analysis revealed that the levels of proteins involved in steroid metabolism and wound healing were significantly decreased in Pnpla5^−/− rats.

Conclusion

This study revealed that Pnpla5 knockout induced apoptosis in rat testes. We also ascertained that Pnpla5 plays an important role in lipid metabolism, wound healing, and affects reproductive organs negatively, providing new target genes and pathways that can be analyzed to unravel the biological function of Pnpla5.

Background

The patatin-like phospholipase domain-containing (PNPLA) protein family is a newly discovered type of lipases [1]. Ten PNPLA proteins have been identified and characterized in the human genome [2]. The biochemical functions and underlying molecular mechanism of Pnpla5 remain elusive. Although most tissues of rats [3], mice, and human [4] contain low levels of PNPLA5, expression patterns significantly vary among different species. In rats, the mRNA expression of Pnpla5 is abundant in the skin, testes, epididymis, uterus, and ovaries [3], while in mice, Pnpla5 is highly expressed in the lungs, epididymis and brown adipose tissues [4]. The expression pattern of murine Pnpla5 is highly correlated with that of Pnpla3, which located upstream of the Pnpla5 gene in the mouse genome. Both these genes are inhibited by fasting, up-regulated in the process of adipocyte differentiation and highly activated in the liver of obese (ob/ob) mice [4]. Furthermore, Pnpla5 mRNA expression level increases in Pnpla3 KO mice, and it can be up-regulated to reach a level that equates untampered Pnpla3 level [5], when high sucrose lipogenic diets are administered. This indicates that, in mice, Pnpla5 partially compensate for the biological deficiencies caused by the absence of Pnpla3.

PNPLA5 exhibits triglyceride (TG) hydrolase activity [6, 7], further, the over expression of Pnpla5 decreases the cellular TG content [4]. Whole-exome sequencing in human also revealed that Pnpla5 is associated with low density lipoprotein cholesterol (LDL-C) [8]. In our previous study, we bred Pnpla5 KO (Pnpla5^−/−) Sprague–Dawley (SD) rats [3] and found that they have increased serum total cholesterol (TC), TG and high density lipoprotein cholesterol (HDL-C) levels, but reduced LDL-C level. Interestingly, we also found that Pnpla5^−/− rats exhibited abnormal bleeding [3] and reduced fertility [9]. These findings imply that Pnpla5 not only plays a key role in lipid metabolism, but also affects the fertility of male rats. To further explore the biological functions of Pnpla5, the body weight, testicular and epididymal tissue weight, and reproductive hormone levels of Pnpla5^−/− male rats were measured. We detected germ cell apoptosis in rat testicular tissues using terminal deoxynucleotidyl transferase-mediated dUTP nick end labeling (TUNEL) and western blotting analyses. We also performed iTRAQ (Isobaric tags for relative and absolute quantitation) proteomic analysis of testes to identify target genes and pathways that may be regulated by Pnpla5.

Results

Knockout of Pnpla5 does not influence the body weight, reproductive organs weight or serum hormone levels of rats

Pnpla5 plays an important role in lipid metabolism, and its deletion causes disorders of lipid metabolism in rats [3]. Lipid metabolism plays an important role in the reproductive organ-development and spermatogenesis. Vendramini et al. (2014) found that obesity could cause a reduction in testicular weight in male rats [10], and high-fat diet administration alters the testicular morphology of rats [11]. We also reported that Pnpla5 knockout significantly reduce the fertility of male rats and the curvilinear velocity of their sperm [9]. Therefore, we assessed the body and reproductive organs (testis and epididymis) weight, and serum hormone levels of Pnpla5^−/− and WT (Pnpla5^+/)⁺ male rats (Fig. 1).

Fig. 1

The body weight, testicular and epididymal tissue weight and serum hormone levels of Pnpla5^−/− and Pnpla5^+/+ male rats. A No significant differences in the body weight between Pnpla5^−/− and Pnpla5^+/+ male rats. P = 0.49. B No significant differences in the testes (P = 0.58) and epididymis (P = 0.31) weight between Pnpla5^−/− and Pnpla5^+/+ male rats. C Significant differences in the relative weight of testicular tissue (P = 0.02) but not epididymal tissue (P = 0.20) were observed * P < 0.05. D No significant differences in serum hormone levels between Pnpla5^−/− and Pnpla5^+/+ male rats. T: Testosterone (P = 0.16). FSH: Follicle-stimulating hormone (P = 0.43). LH: Luteinizing hormone (P = 0.11). ns: not statistically significant. All data are presented as the means ± standard deviation (SD)

The rats exhibited no significant differences (P = 0.49) in their body weight (Fig. 1A). Although the testicular and epididymal weights of Pnpla5^−/− rats increased compared to the control rats, the differences were not significant (P = 0.58 for testes, P = 0.31 for epididymis, Fig. 1B). We also calculated the relative weight of testicular and epididymal tissues, i.e., the tissue to body weight ratio. In Pnpla5^−/− rats, the relative testicular weights increased significantly (P = 0.02), but the relative epididymal weights were not significantly different (P = 0.20, Fig. 1C). In addition, there are no significant differences in the serum level of testosterone (T), follicle-stimulating hormone (FSH) or luteinizing hormone (LH) between Pnpla5^−/− and Pnpla5^+/+ male rats either (Fig. 1D). The calculated P-value of these hormones was 0.16, 0.43 and 0.11, respectively. Interestingly, the TG content of testes was no significantly different (P = 0.36, Fig. S 2), despite Pnpla5 knockout resulted in increased serum TG levels [3].

Fig. 2

HE and TUNEL staining of testes and apoptotic protein expression in Pnpla5^−/− and Pnpla5^+/+ male rats. A HE and TUNEL staining of testicular tissue. Pnpla5^−/− testes appeared loose, disarranged histological structure, with detached epithelial cells. Red arrows indicate apoptotic cells. Scale bar = 50 μm. The TUNEL staining slides were counterstained with hematoxylin. B The number of apoptotic cells was significantly increased in testicular tissue of Pnpla5^−/− male rats. *** P < 0.001. C The expression levels of apoptosis-related proteins in the testicular tissue were measured by western blotting. β-actin was used as the internal reference. All data are presented as means ± standard deviation (SD)

Knockout of Pnpla5 induces cell apoptosis in testicular tissue of rats

Previously, our group reported that Pnpla5 KO significantly reduce the fertility of male rats [9]. Compared to the Pnpla5^+/+ rats, the impregnating rate, and sperm cell curvilinear velocity and progressive motility percentage of their KO counterparts, were significantly decreased. Additionally, Pnpla5^−/− testes appeared loose, and had an disarranged histological structure, with detached epithelial cells (Fig. 2A) [9]. To further determine the Pnpla5 gene’s role in testicular apoptosis regulation, TUNEL staining and western blotting assays were performed. The results showed that Pnpla5 deletion caused an abnormal increase in the number of apoptotic cells in the rat testicular tissue (Fig. 2A). Statistically, the number of apoptotic cells were expected to significantly increase (Fig. 2B, P < 0.001). In addition, the expression levels of pro-apoptotic genes like Bax, Caspase 9 increased while those anti-apoptotic genes, such as Bcl2, decreased (Fig. 2C). This is consistent with the TUNEL assay results. The mRNA expression level of Nlrp3 was also measured using quantitative PCR (qPCR), and no significant differences were found (Fig. S3).

Pnpla5 KO significantly influences the expression level of proteins involved in steroid metabolism and wound healing process

To explore the effect of Pnpla5^−/− on spermatogenesis and sperm maturation at the proteomic level, we used gel-free iTRAQ and liquid chromatography-tandem mass spectrometry (LC–MS/MS) quantitative proteomic analyses to compare Pnpla5^−/− with Pnpla5^+/+ rat testes. With false discovery rate (FDR) < 1.0% and specific peptides identified per protein ≥ 2, a total of 5118 proteins were identified using the UniProt rat protein database. Proteins that had an absolute value of log2 of KO/WT ≥ 0.26 and P-values < 0.05 (via t-test) were identified as differentially expressed proteins (DEPs). Forty nonredundant DEPs were identified in Pnpla5^−/− rat testis (Table 1), of which 35 were down-regulated and 5 were up-regulated.

Table 1 Differentially expressed proteins in Pnpla5^−/− rat testis

To explore the biological functions of these identified DEPs, Gene ontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG) enrichment analysis were performed. The GO enrichment analysis (Fig. 3A, Table S1) showed the DEPs significantly enriched in the following GO terms; wound healing, protein kinase B signaling, steroid metabolic process, acute inflammatory response, protein activation cascade and oxygen transport. Interestingly, several steroid metabolism-related GO terms were enriched, including steroid metabolic process (biology process class), steroid dehydrogenase activity (molecular function class), and protein-lipid complex (cellular component class). This enriched ‘protein-lipid complex’ is also the parental term of high-density lipoprotein particles, plasma lipoprotein particles and lipoprotein particles. These terms were enriched by GO enrichment analysis and are highly correlated with the abnormal lipid level found in Pnpla5^−/− rats [3], indicating that Pnpla5 interacts with genes enriched in these terms to regulate steroid metabolism in rats. DEPs, such as Apoa2 (Apolipoprotein A2), Pon1 (Paraoxonase 1), Hsd17b3 (Hydroxysteroid (17-beta) dehydrogenase 3), Akr1c19 (Aldo–keto reductase family 1, member C19) and Hsd3b3 (Hhydroxy-delta-5-steroid dehydrogenase, 3 beta- and steroid delta-isomerase 3) are involved in these steroid metabolic processes, and are all down-regulated in Pnpla5^−/− rat testis (Table 1). Using gene-concept network analysis, we depicted the linkages of DEPs and enriched GO terms (Fig. 3B). Apoa2 and Pon1 were enriched into multiple GO terms, and may play key roles in the process of lipids metabolism and wound healing. Further studies on these genes could uncover the underlying mechanisms of the phenotypic abnormalities induced by Pnpla5 KO.

Fig. 3

GO enrichment analysis of DEPs. A Dotplot of GO enrichment analysis for DEPs. GeneRatio: ratio of the number of differentially expressed genes to the number of total genes involved in the corresponding GO term; p.adjust: adjusted P value; Count: the number of genes that were enriched into the corresponding GO term. B Gene-concept network of GO enrichment analysis for DEPs. Pon1 and Apoa2 were involved in multiple GO terms. Size: gene number; Fold change: Log2 (KO/WT)

Previously, we reported uncontrollable bleeding and a longer blood coagulation time in Pnpla5^−/− rats [3]. GO enrichment analysis revealed that several DEPs were enriched for wound healing and acute inflammatory response terms (Fig. 3A, Table S1), including Dcn (Decorin), Fmod (Fibromodulin), F10 (Coagulation factor X), Eng (Endoglin), C6 (Complement C6), Apoa2, Kng1 (Kininogen 1), and Kng2 (Kininogen 2). Dcn, Fmod, Kng2, Kng1, F10, and Eng were enriched in the wound healing GO term and C6, Apoa2, Kng1, and Kng2 were enriched in the acute inflammatory response GO term. Moreover, C6/Kng1/Kng2 and F10 were also enriched in the complement and coagulation cascades pathways (Fig. 4). F10, C6, and KNG play essential roles in wound healing and acute inflammatory response. F10 can activate F2 (also known as thrombin) [12], which indirectly interacts with the C5 precursor, which is proteolytically processed to generate C5b. One unit of complements 5b, 6, 7, and 8 and several units of complement 9 form a membrane attack complex (MAC) [13] (Fig. 4). MAC can cause cell lysis and death. It forms trans-plasma membrane channels on the surface of pathogenic bacteria [13], neutralizing their threating. KNG is essential for blood coagulation and assembly of the kallikrein-kinin system which is involved in hemostasis [14]. Moreover, F10, C6 and KNG were down-regulated in Pnpla5^−/− rats, which blocked MAC-induced cell lysis and inhibited hemostasis. Therefore, Pnpla5 KO induced down-regulation of F10, C6, and KNG may contribute to the abnormal bleeding in Pnpla5^−/− rats. Further biological and clinical studies are required to fully elucidate the underlying molecular mechanism.

Fig. 4

F10, C6 and KNG were enriched in complement and coagulation cascades pathways (modified from an image obtained by KEGG [15,16,17]). Proteins down-regulated in Pnpla5^−/− rats are highlighted in green. Red arrows indicate activation, and the dotted red arrow indicates indirect activation

Discussion

Although Pnpla5 is minimally expressed in rat tissues [3], a variety of phenotypic abnormalities arise from its absence. Lipid metabolism disorders, abnormal bleeding, and reduced male fertility induced by Pnpla5 deletion have been reported [3, 9, 18]. In this study, we further explored the effect of Pnpla5 knockout on male rats to augment the knowledge of this gene’s biological functions.

The body and reproductive organs weight and serum hormone levels of Pnpla5^−/− and Pnpla5^+/+ rats were measured. No significant differences were identified between the two groups (Fig. 1), despite Vendramini et al. (2014) reporting that testicular weight loss in male rats can be caused by lipid metabolism disorder [10]. Meanwhile, the relative weight (ratio of tissue to body weight) of testicular tissue was significantly higher in Pnpla5^−/− rats, necessitating the further study of underlying mechanism. Moreover, quantitative proteomic analysis revealed that genes involved in the steroid metabolic processes and protein-lipid complexes (including high-density lipoprotein particles, plasma lipoprotein particles and lipoprotein particles) were significantly enriched. This is consistent with a previous report that Pnpla5^−/− rats have abnormal serum TC, TG, HDL-C and LDL-C levels, indicating that Pnpla5 may play an important role in protein-lipid complex formation.

Quantitative proteomic analysis revealed several genes and potential pathways that may be involved in the molecular mechanisms underlying abnormal bleeding in Pnpla5^-/- rats. Dcn, Fmod, Kng2, Kng1, F10, C6, Apoa2, and Eng were enriched in the wound healing and acute inflammatory response GO terms, and may contribute to the uncontrollable bleeding and longer blood coagulation time observed in Pnpla5^-/- rats [3]. Notably, F10, C6, Kng1, and Kng2 were also enriched in the complement and coagulation cascades pathway by KEGG enrichment. The down-regulation of F10/C6 directly blocks MAC formation [12], which is crucial for neutralizing bacteria [13]. Therefore, it is reasonable to assume that Pnpla5^-/- rats may also suffer from recurrent bacterial infections.

Previous reports noted the decreased pregnancy rates in female rats that mated with Pnpla5^−/− male rats, indicative of reduced fertility and sperm motility, and supported by the disarranged histological structure of Pnpla5^−/− male rats testes [9]. In this study, apoptosis in the testicular tissue was identified by TUNEL assay (Fig. 2), and confirmed by western blotting of apoptosis related proteins. A high-fat diet not only alters the testicular morphology, with reduced seminiferous epithelium height and tubule diameter but it also decreases the proliferation of spermatogenic cell [11]. Therefore, the disorders of lipid metabolism caused by Pnpla5 knockout may be one of the reasons for abnormalities in testicular tissue.

In summary, Pnpla5^−/− rats were used as the animal model to study the biological functions of Pnpla5. The data showed that this gene’s deletion in rats caused no significant difference in body, testicular or epididymal tissue weight, or hormones levels. However, Pnpla5 KO significantly increases testicular apoptosis level, indicating that the deletion of Pnpla5 adversely affects the reproductive organs. Moreover, iTRAQ proteomic analysis revealed that DEPs, involved in steroid metabolism and wound healing, were significantly decreased in Pnpla5^−/− rats. Our study sheds new light, but does not fully address, the underlying mechanisms of phenotypic abnormalities induced by Pnpla5 knockout. Several questions remain unanswered at present. The fact that, Pnpla5 deletion had no effect on the TG content of testes, yet increased the TG serum levels, is a discrepancy that needs to be resolved in the future. Much work is yet to be done to fully characterize the biological functions of the Pnpla5 gene.

Materials and methods

Animals

The Pnpla5 KO SD rats used in this study were obtained as previously described [3]. All rats were 3-month-old and specific pathogen-free (SPF) and were housed in standard cages and provided with food and water, ad libitum. All animal experiments performed in this study were approved by the Animal Care and Use Committee of the Institute of Animal Sciences, Chinese Academy of Agricultural Sciences (IAS2017-8).

Serum hormone analysis

After pentobarbital sodium (Sigma-Aldrich, Missouri, USA) anesthesia (15–40 mg/kg, intraperitoneal injection), blood samples were collected from the orbital of Pnpla5^−/− and Pnpla5^+/+ male rats. Whole blood was centrifuged at 4000 × g for 10 min at room temperature to obtain the serum. Then serum testosterone, follicle-stimulating hormone, and luteinizing hormone levels were measured using a Hitachi 7080 Biochemical Automatic Analyzer (Hitachi, Tokyo, Japan).

qPCR

Total RNA was extracted from testes of rats using TRIzol (Invitrogen, California, USA), and reverse transcribed to cDNA using the RevertAid First Strand cDNA Synthesis kit (Thermo Fisher Scientific, Massachusetts, USA). qPCR was performed using TB Green® Premix Ex Taq II (TaKaRa, Tokyo, Japan) following the manufacturer’s instruction. Gapdh was used as a control, and the relative gene expression level was calculated using the 2 ^− ΔΔCt method. The primers used are listed in Table S3.

Tissue TG Content Assay

The TG content in the Pnpla5^−/− and Pnpla5^+/+ testes were measured by a Tissue TG Content Assay Kit (Cat. No. E1013), following the manufacturer’s instructions.

Western blotting

For western blotting, the tissues were homogenized, and the supernatants were collected after centrifugation. Total protein concentration was quantified using a BCA kit (Beyotime, Shanghai, China). Equal amounts of total protein were separated by 10% sodium dodecyl sulfate–polyacrylamide gel electrophoresis (SDS_PAGE, EpiZyme, Shanghai, China) and transferred to a polyvinylidene difluoride membrane (Millipore, Massachusetts, USA). After blocking with 5% non-fat milk at 4 °C overnight, the membrane was incubated with primary antibodies at 4 °C overnight, washed three times with Tris-buffered saline with tween 20 (TBST) buffer (EpiZyme, Shanghai, China), and then incubated with secondary antibodies at room temperature for 2 h. The membrane was washed three times with TBST buffer, and the blots were stained using an enhanced chemiluminescence (ECL) method on a Tanon-5200 Chemiluminescent Imaging System and ImageJ (National Institutes of Health, Maryland, USA). Immunoblotting used the following primary antibodies: BAX (Cat. No. 2772, Cell Signaling Technology, Massachusetts, USA), BCL2 (Cat. No. Ab182858, Abcam, Cambridge, UK), anti-caspase 9 (Cat. No. 9508, Cell Signaling Technology, Massachusetts, USA), and β-actin antibody (Cat. No. 4970 s, Cell Signaling Technology, Massachusetts, USA). We used anti-rabbit IgG (Cat. No. 7074, Cell Signaling Technology, Massachusetts, USA) as the secondary antibody.

Histological analysis of the testis

Fixed rat testis was embedded in paraffin in vertical position. Then embedded testis-tissue was sliced with a microtome into 5‐7 µm thick sections and adhered to the slides. After incubated in a xylene bath and rehydration, the slides were stained with hematoxylin and eosin (HE). These HE stained slides were used for morphological analysis of the testicular structure.

TUNEL assay

To evaluate the status of the testis cells, we preformed TUNEL assays, using the In Situ Cell Death Detection Kit (Cat. No. 11684817910, Roche, Basel, Switzerland). Briefly, testicular tissue samples from Pnpla5^−/− and Pnpla5^+/+ rats were fixed in 4% formaldehyde for 24 h and embedded in paraffin. Then, 5‐7 μm paraffin sections were adhered to the slides. These section slides were incubated in a xylene bath, at room temperature, for 10 min. The slides were then rehydrated by being transferred through a graded ethanol series: 3 min in 90, 80, and 70% ethanol, each, followed by 3 min in double-distilled water. The sections were covered by pipetted proteinase K solution (20 mg/mL) and incubated for 15 min at room temperature. Thereafter, the slides underwent two 5-min PBS washing cycles. The sections were then covered with 50 µL TUNEL reaction buffer and converter-POD at 37 °C, for 1 h and 30 min, respectively, as well as 50 µL IDAB substrate for 10 min at room temperature. Thereafter, the slides were washed with PBS and counterstained with hematoxylin for 10 min. Then stained slides were then subjected to a decreasing alcohol series. After incubation for 20 min in xylene, the slides were mounted. Tissue sections were examined by microscopy. At least ten fields were randomly selected for each sample, the number of TUNEL-positive cells in each field was counted, and the average number of apoptotic cells in each sample was calculated.

Protein extraction, digestion, and labeling with iTRAQ reagents

For the proteomics quantitative analysis, two groups of 3-month-old rats (n = 3), Pnpla5^+/+ andPnpla5^−/− were analyzed. The rats were euthanized by cervical dislocation under deep anesthesia, and the testes were recovered and stored in liquid nitrogen. Total protein was isolated as previously described [19]. Protein concentrations were measured by the Bradford assay using bovine serum albumin (BSA) as the standard. For trypsin digestion and iTRAQ labeling, 200 µg of protein from each sample was reduced with 25 mM DTT (Bio-Rad, California, USA) at 37 °C for 1 h, alkylated with 50 mM iodoacetamide (Bio-Rad, California, USA) for 30 min at room temperature in the dark, and then digested with 4 µg trypsin (Promega, Wisconsin, USA) at 37 °C overnight. The digested samples were labeled with iTRAQ reagents according to the manufacturer’s instructions. In brief, labels 114–116 and labels 117–119 were used for the relative abundance of three Pnpla5^−/− and Pnpla5^+/+testicular tissues, respectively. The labeled samples were then mixed and vacuum-dried.

LC–MS/MS analysis, peptide identification and quantification

The dried peptides were reconstituted in 10 µL of 0.1% formic acid and loaded onto a low-pH reversed-phase C18 column (75 μm × 100 mm, 3 μm), interfaced with an Orbitrap Q-Exactive-plus (Thermo Fisher Scientific, Massachusetts, USA). Three independent biological replicates were performed, and the resulting peptides were independently analyzed using LC–MS/MS. The raw data files, produced by mass spectrometry (MS) and protein quantification, were identified using Mascot (Version 2.5.1) search against the rat protein database (downloaded from UniProt). The parameters used for database searching were as follows: trypsin selected as the enzyme, a maximum of two missed cleavages were allowed, precursor mass tolerance was set to 10 ppm, fragment mass tolerance was set to 0.05 KDa, carbamidomethylation of cysteine was set as a fixed modification, and methionine oxidation and acetyl at the N-terminal and lysine side chains were allowed as dynamic modifications. The FDR was set to 1% for protein identification. Proteins with P < 0.05 and absolute value of log2 of KO/WT ≥ 0.26 were considered DEPs.

Bioinformatics analysis

To better understand the biological functions of DEPs, Gene Ontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG) enrichment analyses were performed using the clusterProfiler (version 4.2.1) R package [20]. GO terms with a corrected P < 0.05 were considered significantly enriched by differential expressed genes. KEGG pathways with corrected P < 0.05 were considered significantly enriched. GO and KEGG enrichment results were visualized using the ggplot2 (version 3.3.5), pathview (version 3.1.4) [21], and ggpubr (version 0.4.0) R packages.

Statistical analysis

All data are presented as means ± standard deviation (SD). Data from the two groups were compared using an unpaired t-test. Data were analyzed using GraphPad Prism version 8 (GraphPad Software, California, USA). P values less than 0.05, 0.01, and 0.001 are indicated by *, **, and ***, respectively. Each rat served as an experimental unit.