Search

Background/objectives: A better understanding of adipose tissue biology is crucial to tackle insulin resistance and eventually coronary heart disease and diabetes, leading causes of morbidity and mortality worldwide. GALNT2, a GalNAc-transferase, positively modulates insulin signaling in human liver cells by down-regulating ENPP1, an insulin signaling inhibitor. GALNT2 expression is increased in adipose tissue of obese as compared to that of non-obese individuals. Whether this association is secondary to a GALNT2-insulin sensitizing effect exerted also in adipocytes is unknown. We then investigated in mouse 3T3-L1 adipocytes the GALNT2 effect on adipogenesis, insulin signaling and expression levels of both Enpp1 and 72 adipogenesis-related genes. Methods: Stable over-expressing GALNT2 and GFP preadipocytes (T0) were generated. Adipogenesis was induced with (R+) or without (R−) rosiglitazone and investigated after 15 days (T15). Lipid accumulation (by Oil Red-O staining) and intracellular triglycerides (by fluorimetric assay) were measured. Lipid droplets (LD) measures were analyzed at confocal microscope. Gene expression was assessed by RT-PCR and insulin-induced insulin receptor (IR), IRS1, JNK and AKT phosphorylation by Western blot. Results: Lipid accumulation, triglycerides and LD measures progressively increased from T0 to T15R- and furthermore to T15R+. Such increases were significantly higher in GALNT2 than in GFP cells so that, as compared to T15R+GFP, T15R- GALNT2 cells showed similar (intracellular lipid and triglycerides accumulation) or even higher (LD measures, p < 0.01) values. In GALNT2 preadipocytes, insulin-induced IR, IRS1 and AKT activation was higher than that in GFP cells. GALNT2 effect was totally abolished during adipocyte maturation and completely reversed at late stage maturation. Such GALNT2 effect trajectory was paralleled by coordinated changes in the expression of Enpp1 and adipocyte-maturation key genes. Conclusions: GALNT2 is a novel modulator of adipogenesis and related cellular phenotypes, thus becoming a potential target for tackling the obesity epidemics and its devastating sequelae.

Objective Transient neonatal diabetes mellitus (TNDM) is caused by activating mutations in ABCC8 and KCNJ11 genes (KATP/TNDM) or by chromosome 6q24 abnormalities (6q24/TNDM). We wanted to assess whether these different genetic aetiologies result in distinct clinical features. Design Retrospective analysis of the Italian data set of patients with TNDM. Methods Clinical features and treatment of 22 KATP/TNDM patients and 12 6q24/TNDM patients were compared. Results Fourteen KATP/TNDM probands had a carrier parent with abnormal glucose values, four patients with 6q24 showed macroglossia and/or umbilical hernia. Median age at diabetes onset and birth weight were lower in patients with 6q24 (1 week; −2.27 SD) than those with KATP mutations (4.0 weeks; −1.04 SD) (P = 0.009 and P = 0.007, respectively). Median time to remission was longer in KATP/TNDM than 6q24/TNDM (21.5 weeks vs 12 weeks) (P = 0.002). Two KATP/TNDM patients entered diabetes remission without pharmacological therapy. A proband with the ABCC8/L225P variant previously associated with permanent neonatal diabetes entered 7-year long remission after 1 year of sulfonylurea therapy. Seven diabetic individuals with KATP mutations were successfully treated with sulfonylurea monotherapy; four cases with relapsing 6q24/TNDM were treated with insulin, metformin or combination therapy. Conclusions If TNDM is suspected, KATP genes should be analyzed first with the exception of patients with macroglossia and/or umbilical hernia. Remission of diabetes without pharmacological therapy should not preclude genetic analysis. Early treatment with sulfonylurea may induce long-lasting remission of diabetes in patients with KATP mutations associated with PNDM. Adult patients carrying KATP/TNDM mutations respond favourably to sulfonylurea monotherapy.

The high mortality risk of patients with type 2 diabetes mellitus may well be explained by the several comorbidities and/or complications. Also the intrinsic genetic component predisposing to diabetes might have a role in shaping the risk of diabetes-related mortality. Among type 2 diabetes mellitus SNPs, rs1801282 is of particular interest because (i) it is harbored by peroxisome proliferator-activated receptor-γ2 (PPARγ2), which is the target for thiazolidinediones which are used as antidiabetic drugs, decreasing all-cause mortality in type 2 diabetes mellitus, and (ii) it is associated with insulin resistance and related traits, risk factors for overall mortality in type 2 diabetes mellitus. We investigated the role of PPARγ2 P12A, according to a dominant model (PA + AA vs. PP individuals) on incident all-cause mortality in three cohorts of type 2 diabetes mellitus, comprising a total of 1672 patients (462 deaths) and then performed a meta-analysis of ours and all available published data. In the three cohorts pooled and analyzed together, no association between PPARγ2 P12A and all-cause mortality was observed (HR 1.02, 95 % CI 0.79-1.33). Similar results were observed after adjusting for age, sex, smoking habits, and BMI (HR 1.09, 95 % CI 0.83-1.43). In a meta-analysis of ours and all studies previously published (n = 3241 individuals; 666 events), no association was observed between PPARγ2 P12A and all-cause mortality (HR 1.07, 95 % CI 0.85-1.33). Results from our individual samples as well as from our meta-analysis suggest that the PPARγ2 P12A does not significantly affect all-cause mortality in patients with type 2 diabetes mellitus.

In cells and tissues resistin affects IL-1β, IL-6, IL-8, IL-12 and TNF-α expression, thus suggesting the existence of a multi-cytokine “resistin pathway”. We investigated whether such pathway does exist in humans and, if so, if it is associated with cardiovascular risk factors and with major adverse cardiovascular events (MACE). Serum cytokines were measured in 280 healthy subjects from the Gargano Study 2 (GS2) whose BMI, waist circumference, HOMAIR, triglycerides, HDL-cholesterol, systolic and diastolic blood pressure data were available and in 353 patients with type 2 diabetes and coronary artery disease from the Gargano Heart Study (GHS)-prospective design (follow-up 5.4 ± 2.5 years; 71 MACE). In GS2, cytokines mRNA levels in white blood cells were also measured. In GS2, resistin mRNA was correlated with all cytokines expression (all p

Obesity is a major predisposing risk factor for the development of insulin resistance. Recently, a sexual dimorphism in the relationship between body mass index (BMI) and insulin resistance has been reported, with the deleterious effect of increasing BMI being more evident in men than in women [1]. To the best of our knowledge, the biology underlying this sex-specific association has never been addressed. By inhibiting insulin signaling, TNF-a overexpression is the main mechanism responsible for the deleterious effect of obesity on insulin resistance [2]. It is, therefore, reasonable to address the role of TNF-a as a mediator of the sex-specific response of insulin sensitivity to the insult of increasing BMI. Accordingly, we measured both the HOMA-IR insulin resistance index and TNF-a mRNA levels (in peripheral whole blood cells; PWBC) from healthy subjects with a wide range of BMI and then looked for BMI-by-sex interaction in modulating these two variables.

Objective : We have previously reported the combined effect of SNPs perturbing insulin signaling ( ENPP1 K121Q, rs1044498; IRS1 G972R, rs1801278; TRIB3 Q84R, rs2295490) on insulin resistance (IR), type 2 diabetes (T2D) and cardiovascular events. We here investigated whether such a combined effect affects also all-cause mortality in a sample of 1851 Whites of European ancestry. Methods : We investigated a first sample of 721 patients, 232 deaths, 3389 person-years (py). Replication was assessed in two samples of patients with T2D: the Gargano Mortality Study (GMS) of 714 patients, 127 deaths, 5426 py and the Joslin Kidney Study (JKS) comprising 416 patients, 214 deaths, 5325 py. Results : In the first sample, individuals carrying 1 or ≥2 risk alleles had 33% ( p = 0.06) and 51% ( p = 0.02) increased risk of mortality, as compared with individuals with no risk alleles. A similar, though not significant, trend was obtained in the two replication samples only for subject carrying ≥ 2 risk alleles. In a pooled analysis, individuals carrying ≥2 risk alleles had higher mortality rate as compared to those carrying 0 risk alleles (HR = 1.34, 95%CI = 1.08–1.67; p = 0.008), and as compared to those carrying only one risk allele (HR = 1.41, 95%CI = 1.13–1.75; p = 0.002). This association was independent from several possible confounders including sex, age, BMI, hypertension and diabetes status. Conclusion : Our data suggest that variants affecting insulin signaling exert a joint effect on all-cause mortality and is consistent with a role of abnormal insulin signaling on mortality risk.

Mortality rate of diabetic patients is twice as much that of non-diabetic individuals. The role of obesity on mortality risk in patients with type 2 diabetes is controversial. Aim of our study was to address the relationship between obesity and all-cause mortality in a real-life set of white patients with type 2 diabetes from central-southern Italy from the Gargano Mortality Study (GMS). In addition, we used genetic data from genome-wide association studies (GWAs)-derived single nucleotide polymorphisms (SNPs) firmly associated with body mass index (BMI), in order to investigate the intrinsic nature of reduced mortality rate we, in fact, observed in obese patients. Study subjects with type 2 diabetes (n = 764) are part of the GMS, which is aimed at unraveling predictors of incident all-cause mortality. Time-to-death analyses were performed by Cox regression. Association between genotype risk score and obesity was tested by logistic regression. Of the 32 SNPs firmly associated with BMI, we investigated those with BMI β value ≥0.10 kg/m(2) and allele frequency ≥10 %. Genotyping was performed by KBioscience (http://www.lgcgenomics.com/). In GMS, obesity predicted a 45 % reduction in all-cause mortality. Individuals with high "obesity genetic load" (i.e., those carrying >9 risk alleles) were 60 % more likely to be obese as compared to individuals with low "obesity genetic load." Most importantly, mortality rate was not different in individuals with high and low "obesity genetic load," thus indicating no role of obesity genes on all-cause mortality and speaking against a cause-effect relationship underlying the association between obesity and reduced mortality rate.

Ectonucleotide pyrophosphatase phosphodiesterase 1 (ENPP1) inhibits insulin signaling and action. Understanding the mechanisms underlying ENPP1 expression may help unravel molecular mechanisms of insulin resistance. Recent data suggest a role of ENPP1-3'untraslated region (UTR), in controlling ENPP1 expression. We sought to identify trans-acting ENPP1-3'UTR binding proteins, and investigate their role on insulin signaling. By RNA pull-down, 49 proteins bound to ENPP1-3'UTR RNA were identified by mass spectrometry (MS). Among these, in silico analysis of genome wide association studies and expression profile datasets pointed to N-acetylgalactosaminyltransferase 2 gene (GALNT2) for subsequent investigations. Gene expression levels were evaluated by RT-PCR. Protein expression levels, IRS-1 and Akt phosphorylation were evaluated by Western blot. Insulin receptor (IR) autophosphorylation was evaluated by ELISA. GALNT2 down-regulation increased while GALNT2 over-expression reduced ENPP1 expression levels. In addition, GALNT2 down-regulation reduced insulin stimulation of IR, IRS-1 and Akt phosphorylation and insulin inhibition of phosphoenolpyruvate carboxykinase (PEPCK) expression, a key neoglucogenetic enzyme. Our data point to GALNT2 as a novel factor involved in the modulation of ENPP1 expression as well as insulin signaling and action in human liver HepG2 cells.

Background and aim Insulin resistance and increased left ventricular mass (LVM) characterize patients with essential hypertension. Some genetic polymorphisms play a role in the modulation of both insulin resistance and LVM. The aim of this work was to investigate whether the PC-1 and ACE genes exert a polygenic control of insulin resistance and LVM in hypertensive patients. Methods and results In 158 never-treated hypertensive patients, we evaluated insulin resistance by HOMA index [insulin (μU/mL)×glucose (mmol/L)]/22.5 and LVM by echocardiograms. Genetic polymorphisms were obtained by polymerase chain reaction. PC-1 X121Q genotype carriers (K121Q+Q121Q, n =46) had higher HOMA (3.14±1.28 vs. 2.49±1.25; p =0.002) and LVM (137±34 vs. 127±24g/m 2 ; p =0.02) than K121K patients ( n =112). Similarly, ACE DD carriers ( n =56) showed higher HOMA (3.94±1.13 vs. 1.98±0.72; p 2 ; p =0.00004) than XI (ID+II, n =102) patients. When considering both PC-1 and ACE polymorphisms, HOMA ( p p =0.00003) progressively increased from K121K/XI to X121Q/XI, K121K/DD and X121Q/DD patients. The association of both gene polymorphisms with LVM was no longer significant after adjusting for HOMA values. As compared to K121K/XI patients (i.e. no at risk alleles), X121Q/DD patients had a significantly increased risk (OR: 4.4, 95% C.I. 1.4–14.0, p =0.011) to have left ventricular hypertrophy. Conclusions In hypertensive patients PC-1 K121Q and ACE I/D polymorphisms have an additive deleterious effect on insulin resistance and, consequently, on LVM, thus increasing the global cardiovascular risk. Identification of carriers of the at-risk genotypes may help set up prevention strategies to be specifically targeted at these patients.

Our aims were to investigate in several large samples, with a wide range of adiposity, whether: (1) the effect of BMI on insulin sensitivity is different between sexes; (2) also waist circumference plays a sex-specific role on insulin sensitivity; and (3) serum adiponectin and resistin are mediators of such sex-dimorphic effect. Samples used were: Gargano study 1 (GS1), GS2 and Catania study (CS) comprising 3274 individuals. Adiponectin and resistin were measured by ELISA. Associations between variables were tested by linear models. In all samples, relationship between BMI and HOMAIR was steeper in males than in females (BMI-by-sex interaction p = 0.04–0.0007). No interaction was observed on serum adiponectin and resistin (p = 0.40–059), which are therefore unlikely to mediate the sex-dimorphic effect of BMI on insulin resistance. Relationship between waist circumference and HOMAIR was similar between sexes in GS1 and GS2 but not in CS (waist-by-sex interaction p = 0.01), comprising much heavier individuals. This suggests that a sex-dimorphic effect of abdominal adiposity on insulin resistance is observable only in the context of high BMI. Our findings represent a proof of concept that BMI and insulin sensitivity are associated in a sex-specific manner. This may explain why females are protected from diabetes and cardiovascular disease, compared to males of similar BMI.

Diabetes mellitus is a highly heterogeneous disorder encompassing several distinct forms with different clinical manifestations including a wide spectrum of age at onset. Despite many advances, the causal genetic defect remains unknown for many subtypes of the disease, including some of those forms with an apparent Mendelian mode of inheritance. Here we report two loss-of-function mutations (c.1655TA [p.Leu552(∗)] and c.280GA [p.Asp94Asn]) in the gene for the Adaptor Protein, Phosphotyrosine Interaction, PH domain, and leucine zipper containing 1 (APPL1) that were identified by means of whole-exome sequencing in two large families with a high prevalence of diabetes not due to mutations in known genes involved in maturity onset diabetes of the young (MODY). APPL1 binds to AKT2, a key molecule in the insulin signaling pathway, thereby enhancing insulin-induced AKT2 activation and downstream signaling leading to insulin action and secretion. Both mutations cause APPL1 loss of function. The p.Leu552(∗) alteration totally abolishes APPL1 protein expression in HepG2 transfected cells and the p.Asp94Asn alteration causes significant reduction in the enhancement of the insulin-stimulated AKT2 and GSK3β phosphorylation that is observed after wild-type APPL1 transfection. These findings-linking APPL1 mutations to familial forms of diabetes-reaffirm the critical role of APPL1 in glucose homeostasis.

Maturity-onset diabetes of the young (MODY; MIM# 606391) is a genetically and clinically heterogeneous form of diabetes, accounting for 1–2% of all diabetes cases (1). MODY is characterized by mild hyperglycemia or overt diabetes usually detected in three consecutive generations, with onset before the age of 25 years and absence of type 1 diabetes autoantibodies. Among the thirteen MODY genes identified, two subtypes, GCK -MODY and HNF1A -MODY, account for most of cases (1). The prevalence of GCK -MODY has been reported higher in Southern Europe (2), while HNF1A -MODY is the most common MODY subtype in Northern Europe (3). This difference might be attributable to the clinical setting in which genetic screening is performed, especially when pediatric and adult diabetes clinics are distinct entities. We addressed this issue by investigating MODY patients identified in the pediatric or in the adult diabetes clinics of the same research-based …

Impaired insulin action plays a major role in the pathogenesis of type 2 diabetes, a chronic metabolic disorder which imposes a tremendous burden to morbidity and mortality worldwide. Unraveling the molecular mechanisms underlying insulin resistance would improve setting up preventive and treatment strategies of type 2 diabetes. Down-regulation of GALNT2, an UDPN-acetyl-alpha-D-galactosamine polypeptideN-acetylgalactosaminyltransferase-2 (ppGalNAc-T2), causes impaired insulin signaling and action in cultured human liver cells. In addition, GALNT2 mRNA levels are down-regulated in liver of spontaneously insulin resistant, diabetic Goto-Kakizaki rats. To investigate the role of GALNT2 in human hyperglycemia, we measured GALNT2 mRNA expression levels in peripheral whole blood cells of 84 non-obese and 46 obese non-diabetic individuals as well as of 98 obese patients with type 2 diabetes. We also measured GALNT2 mRNA expression in human U937 cells cultured under different glucose concentrations. In vivo studies indicated that GALNT2 mRNA levels were significantly reduced from non obese control to obese non diabetic and to obese diabetic individuals (p

The aim of this study was to deeper investigate the mechanisms through which ENPP1, a negative modulator of insulin receptor (IR) activation, plays a role on insulin signaling, insulin secretion and eventually glucose metabolism. ENPP1 cDNA (carrying either K121 or Q121 variant) was transfected in HepG2 liver-, L6 skeletal muscle- and INS1E beta-cells. Insulin-induced IR-autophosphorylation (HepG2, L6, INS1E), Akt-Ser(473), ERK1/2-Thr(202)/Tyr(204) and GSK3-beta Ser(9) phosphorylation (HepG2, L6), PEPCK mRNA levels (HepG2) and 2-deoxy-D-glucose uptake (L6) was studied. GLUT 4 mRNA (L6), insulin secretion and caspase-3 activation (INS1E) were also investigated. Insulin-induced IR-autophosphorylation was decreased in HepG2-K, L6-K, INS1E-K (20%, 52% and 11% reduction vs. untransfected cells) and twice as much in HepG2-Q, L6-Q, INS1E-Q (44%, 92% and 30%). Similar data were obtained with Akt-Ser(473), ERK1/2-Thr(202)/Tyr(204) and GSK3-beta Ser(9) in HepG2 and L6. Insulin-induced reduction of PEPCK mRNA was progressively lower in untransfected, HepG2-K and HepG2-Q cells (65%, 54%, 23%). Insulin-induced glucose uptake in untransfected L6 (60% increase over basal), was totally abolished in L6-K and L6-Q cells. GLUT 4 mRNA was slightly reduced in L6-K and twice as much in L6-Q (13% and 25% reduction vs. untransfected cells). Glucose-induced insulin secretion was 60% reduced in INS1E-K and almost abolished in INS1E-Q. Serum deficiency activated caspase-3 by two, three and four folds in untransfected INS1E, INS1E-K and INS1E-Q. Glyburide-induced insulin secretion was reduced by 50% in isolated human islets from homozygous QQ donors as compared to those from KK and KQ individuals. Our data clearly indicate that ENPP1, especially when the Q121 variant is operating, affects insulin signaling and glucose metabolism in skeletal muscle- and liver-cells and both function and survival of insulin secreting beta-cells, thus representing a strong pathogenic factor predisposing to insulin resistance, defective insulin secretion and glucose metabolism abnormalities.

Objective— Insulin resistance induces increased pulse pressure (PP), endothelial dysfunction (ED), and reduced bioavailability of endothelium-derived nitric oxide (NO). The genetic background of these 3 cardiovascular risk factors might be partly common. The ENPP1 K121Q polymorphism is associated with insulin resistance and cardiovascular risk. Methods and Results— We investigated whether the K121Q polymorphism is associated with increased PP in white Caucasians and with ED in vitro. In 985 individuals, (390 unrelated and 595 from 248 families), the K121Q polymorphism was associated with PP ( P =8.0×10 −4 ). In the families, the Q121 variant accounted for 0.08 of PP heritability ( P =9.4×10 −4 ). This association was formally replicated in a second sample of 475 individuals ( P =2.6×10 −2 ) but not in 2 smaller samples of 289 and 236 individuals ( P =0.49 and 0.21, respectively). In the individual patients’ data meta-analysis, comprising 1985 individuals, PP was associated with the Q121 variant ( P =1.2×10 −3 ). Human endothelial cells carrying the KQ genotype showed, as compared to KK cells, reduced insulin-mediated insulin receptor autophosphorylation ( P =0.03), Ser 473 -Akt phosphorylation ( P =0.03), and NO synthase activity ( P =0.003). Conclusions— Our data suggest that the ENPP1 Q121 variant is associated with increased PP in vivo and reduced insulin signaling and ED in vitro, thus indicating a possible pathogenic mechanism for the increased cardiovascular risk observed in ENPP1 Q121 carriers.

Ectonucleotide pyrophosphatase phosphodiesterase 1 (ENPP1) inhibits insulin-receptor (IR) signaling and, when over-expressed, induces insulin resistance in vitro and in vivo. Understanding the regulation of ENPP1 expression may, thus, unravel new molecular mechanisms of insulin resistance. Recent data point to a pivotal role of the ENPP1 3'UTR, in modulating ENPP1 mRNA stability and expression. We sought to identify trans-acting proteins binding the ENPP1-3'UTR and to investigate their role on ENPP1 expression and on IR signaling. By RNA electrophoresis mobility shift analysis and tandem mass spectrometry, we demonstrated the binding of heat shock protein 70 (HSP70) to ENPP1-3'UTR. Through this binding, HSP70 stabilizes ENPP1 mRNA and increases ENPP1 transcript and protein levels. This positive modulation of ENPP1 expression is paralleled by a reduced insulin-induced IR and IRS-1 phosphorylation. Taken together these data suggest that HSP70, by affecting ENPP1 expression, may be a novel mediator of altered insulin signaling.

Insulin resistance is believed to be under the control of several genes often interacting each other. However, whether genetic epistasis does in fact modulate human insulin sensitivity is unknown. In 338 healthy unrelated subjects from Sicily, all nondiabetic and not morbidly obese, we investigated whether two gene polymorphisms previously associated with insulin resistance (namely PC-1 K121Q and PPARgamma2 P12A) affect insulin sensitivity by interacting. PC-1 X121Q subjects showed higher level of fasting glucose, lower insulin sensitivity (by both the Matsuda insulin sensitivity index and M values at clamp, the latter performed in a subgroup of 113 subjects representative of the overall cohort) and higher insulin levels during the oral glucose tolerance test (OGTT) than PC-1 K121K subjects. In contrast, no difference in any of the measured variables was observed between PPARgamma2 P12P and X12A individuals. The deleterious effect of the PC-1 X121Q genotype on each of these three variables was significant and entirely dependent upon the coexistence of the PPARgamma2 P12P genotype. Among PPARgamma2 P12P carriers also fasting insulin and glucose levels during OGTT were higher in PC-1 X121Q than in K121K individuals. In contrast, no deleterious effect of the PC-1 X121Q genotype was observed among PPARgamma2 X12A carriers; rather, in these subjects a lower body mass index and consequently lower fasting insulin level was observed in PC-1 X121Q than in K121K carriers. Overall, a significant interaction between the two genes was observed on body mass index, insulin levels (both fasting and after OGTT) and both insulin sensitivity (i.e., insulin sensitivity index and M value) and insulin secretion (i.e., HOMA-B%) indexes.

GRB10 encodes for an inhibitor of insulin receptor signaling [1] and is therefore a candidate for type 2 diabetes (T2D). In a preliminary study, the minor allele (MA) of GRB10 rs4947710 was associated with a reduced T2D risk in Whites from Italy, whereas a trend in the opposite direction, albeit not significant, was observed in Whites from the US [2], making the overall observation uncertain. A different GRB10 single nucleotide polymorphism (SNP) (rs2237457) has been recently associated with T2D among Amish but not in other populations [3], suggesting the possibility of allelic heterogeneity. To further investigate the role of GRB10 variability in modulating susceptibility to T2D in Whites, we genotyped rs4947710 and rs2237457 in a total of 3,433 diabetic cases (1899 males/1534 females; age=61.4±9.2 yrs; BMI=31.5±6.1 Kg/m2) and 2,660 non-diabetic controls (1153 males/1507 females; age=45.5±16.0 yrs; BMI=27.8±6.1 Kg/m2) of European origin included in the “GENetics of T2D in Italy and United States (GENIUS) Consortium”. The clinical characteristics of these subjects have been previously reported in details [4]. All samples, which included those from the previous smaller study (2), were genotyped by TaqMan allelic discrimination assay. The average agreement rate of duplicate samples was >99%. Failure rate of genotyping was