Many similarities exist between type 2 diabetes mellitus in cats and in humans.1,2 Both conditions typically have an onset in middle age or later, are associated with obesity, and are characterized by defective insulin secretion, insulin resistance, loss of pancreatic β cell mass, and islet amyloid deposition. Islet amyloid is derived from IAPP (also called amylin), a normal secretory hormone product of β cells that is copackaged and cosecreted with insulin.3–7 The amino acid sequence of the midportion of IAPP is amyloidogenic in cats, humans, and macaques, and type 2 diabetes mellitus can develop in these species.8–10 Intracellular formation of small oligomeric aggregates of IAPP is an early step in amyloid formation. These aggregates are toxic to β cells and are associated with apoptosis; therefore, they may be linked to the loss of β cells that develops in cats and humans with type 2 diabetes mellitus.11–13 Results of an earlier study14 by our laboratory group indicated that cats with impaired glucose tolerance have increased β cell IAPP content and an increased incidence of islet amyloidosis, compared with that of clinically normal cats. However, it is not known whether obesity in cats is linked to increased circulating concentrations of IAPP or, if so, at what stage of obesity these concentrations increase. Circulating concentrations of insulin and IAPP are increased in overweight, glucose-intolerant human patients and in those with early type 2 diabetes mellitus, whereas these values are decreased in patients with advanced type 2 diabetes mellitus that require insulin treatment.15,16 Therefore, the purpose of the study reported here was to evaluate plasma concentrations of IAPP in nondiabetic cats classified according to BCS (as an estimate of obesity) and in cats with naturally acquired diabetes mellitus after food was withheld from all cats for 12 hours; we also sought to compare plasma concentrations of IAPP with serum concentrations of insulin and glucose in these same cats.
Materials and Methods
Animals—One hundred nine client-owned cats that were either nondiabetic or had naturally acquired diabetes mellitus were enrolled in the study with informed consent of the owners. Nondiabetic cats (n = 82) were determined to be healthy on the basis of history, physical examination, and results of analysis of a blood sample obtained after food was withheld for 12 hours (serum glucose < 152 mg/dL or serum fructos-amine < 340 μmol/L).17
Each diabetic cat was evaluated by means of physical examination, CBC, serum biochemical analysis, FeLV antigen anti-FIV antibody testing, measurement of serum thyroxine concentrations, urinalysis (via cys-tocentesis), and aerobic microbial culture of a urine sample. Glycosuria and ketonuria were evaluated via a dipstick test. Diabetes mellitus was diagnosed in affected cats ≤ 1 week prior to enrollment in the study; cats were excluded if they had received any diabetogenic drug treatment (eg, corticosteroids, progestogens, or anticonvulsants) within 3 months before diagnosis. Diagnosis of nonketoacidotic diabetes mellitus (DM group; n = 21 cats) was made on the basis of clinical signs (polyuria, polydipsia, and weight loss) with hyperglycemia and concurrent glycosuria on at least 2 days without ketonuria. Cats were excluded from this group if they were determined to have a history of treatment for diabetes (other than a change in diet) or any of the following conditions: ketonuria, dehydration, debilitation, poor appetite, or physical or biochemical abnormalities other than those attributable to untreated diabetes (eg, hepatomegaly and hypercholesterolemia). Diagnosis of ketoacidotic diabetes mellitus (DKA group; 6 cats) was made on the basis of clinical signs with concurrent hyperglycemia, acidosis (total CO2 < 12 mmol/L), glycosuria, and ketonuria. Ketoacidotic cats were excluded if they had a history of diabetogenic drug therapy (eg, corticosteroids or progestogens) within the preceding 3 months or if they had any clinically important abnormality other than conditions commonly associated with untreated diabetes such as urinary tract infection.
A BCS estimate was made for each cat by 1 investigator (MSH) according to a scale from 1 to 9 (1 = emaciated, 5 = ideal, and 9 = morbidly obese).18 For purposes of the present study, only cats with BCSs ≥ 5 were included. Nondiabetic cats were then further grouped according to BCS for various analyses. The study was approved by the University of Minnesota Animal Care and Use Committee and conducted in accordance with the US Animal Welfare Act and the NIH Principles for Utilization and Care of Vertebrate Animals Used in Testing, Research, and Training.
Experimental protocol—After food was withheld for 12 hours (with free access to water), a blood sample (10 mL) was collected from a jugular or medial saphenous vein for measurement of plasma concentration of IAPP and serum concentrations of glucose and insulin. For nondiabetic cats, food was withheld at home by the owners for the prescribed period of time. Samples from cats in the DM group were collected prior to any treatment with insulin or oral hypoglycemic agents. Samples from cats in the DKA group were collected after ketoacidosis was resolved by means of rehydration with intravenous fluid treatment (saline [0.9% NaCl] solution supplemented with potassium chloride and potassium phosphate) and regular insulin administration; insulin was then withheld for 12 hours prior to sample collection.
Sample handling and analysis—Blood samples were divided among 2 tubes. Samples for serum insulin and glucose analysis were allowed to clot for 15 minutes at room temperature (21°C) and then centrifuged for 10 minutes at 500 × g. Serum was transferred to polypropylene tubes and stored at −70°C until assay. Glucose measurements were performed by use of the glucose oxidase method.19 Blood samples for IAPP analysis were immediately transferred to chilled (0°C) 7-mL tubes containing EDTA and the protease inhibitor aprotinin (500 kallikrein U/mL)a and centrifuged at 4°C for 10 minutes at 500 × g. The plasma was then transferred to polypropylene tubes and frozen at −70°C until extraction.
Radioimmunoassay for insulin—Serum insulin concentrations were determined by use of a commercially available radioimmunoassay kitb for measurement of porcine insulin according to the manufacturer's instructions. All samples were assayed in duplicate and means reported. The minimum detectable insulin concentration that could be distinguished from 0 was 12 pmol/L. The intra-assay coefficient of variation (measurement precision) determined by means of repeated measurement (5 tests/pool) of insulin concentration in each of 3 pools of feline serum that contained 510, 306, or 90 pmol of insulin/L (determined by use of the same assay) was 2.6%, 5.1%, and 0.9%, respectively. The interassay coefficient of variation for insulin measurement determined by use of these same 3 pools in 6 separate assays was 3.6%, 3.9%, and 8.9%, respectively.
Assay linearity was demonstrated by measuring insulin concentrations in 2 samples of pooled feline serum that contained 330 and 486 pmol of insulin/L, respectively, prepared at various dilutions (undiluted, 3 parts serum: 1 part diluent, 1:1, and 1:3) with the manufacturer-provided zero standard used as diluent. Percentages of expected concentrations at the 4 dilutions were 100%, 93%, 96%, and 96%, respectively, for the first pooled sample (330 pmol/L) and 100%, 112%, 116%, and 132%, respectively, for the second pooled sample (486 pmol/L). The percentage of assay recovery was determined by adding various concentrations (0, 45.6, 50, and 252 pmol/L) of porcine insulin to feline serum (pooled sample from study cats) and measuring insulin concentration in each of these 4 test conditions. For this analysis, 4 separate aliquots of each test sample were assayed in duplicate. Mean percentage recovery for the test samples was 100%, 87%, 103%, and 114%, respectively.
Radioimmunoassay for IAPP—The IAPP assay was performed essentially as described elsewhere.20 Samples were analyzed in duplicate, and mean values were reported. For preparation of calibrators and controls, a 12.8-μg vial of synthetic feline IAPPc was reconstituted with water to make a 0.01 g/L stock solution. Calibrators were immediately prepared via further dilution of the stock solution with a freeze-drying solution (50 g of lactose/L, 2.5 g of bovine serum albumin/L, and 4.25 mL of formic acid/L in water) at 4°C, then aliquoted, lyophilized, and stored at −70°C. Each calibrator was used only once because 1 freeze-thaw cycle of IAPP in buffer (60 mmol of phosphate/L; pH, 7.4; with 0.1% Triton X-100 and 0.1% bovine serum albumind) resulted in a 39.7% reduction in measured peptide concentrations. The remaining stock solution was aliquoted and stored at −70°C. The concentration of IAPP in the stock solution was confirmed via amino acid analysis.e
Because of the large volume of plasma required for validation studies and assay controls, plasma from multiple healthy study cats was pooled; to make the matrices of the 3 pooled samples as different as possible, each pooled sample consisted of plasma from cats with particular BCSs; one contained plasma from cats with BCSs of 5 or 6, another contained plasma from cats with a BCS of 7, and the remaining pooled sample contained plasma from cats with a BCS of 8. Concentrated spiking solutions for controls and for recovery and dilution experiments were prepared from thawed stock solution (predicted concentration corrected for expected freeze-thaw loss) with buffer added at 4°C to maintain the same proportional volume and a spiking percentage < 4%. Spiked pooled samples were aliquoted and stored in polypropylene tubes at −70°C until extraction. For dilution experiments, independent fractional dilutions in buffer solution were made after reconstitution of the lyophilized samples immediately prior to assay. Solid-phase extraction and radioimmunoassays were performed as previously described.20 For calculations, plasma IAPP concentrations below the minimal detectable concentration of 2.5 pmol/L were recorded at the arbitrary level of half that concentration.
Statistical analysis—Mean ± SD circulating concentrations of IAPP, insulin, and glucose were summarized for nondiabetic cats grouped according to BCS and for diabetic cats grouped as ketoacidotic or nonketo-acidotic (DKA and DM groups, respectively) on the basis of clinicopathologic findings. Box-and-whiskers plots were generated for graphic evaluation of plasma IAPP and serum insulin and glucose concentrations for each group.
A Shapiro-Wilk test was used to test for normality of data, and ANOVA and F tests were performed to evaluate whether mean insulin, IAPP, and glucose concentrations were significantly different among groups. Because the variance of IAPP, insulin, and glucose concentrations was not constant across all groups, weighted least squares with weights that were inversely proportional to the variance in each group were used to yield the most precise estimates. To assess differences among means, multiple comparisons were performed with Tukey-Kramer adjustment for the P values and 95% confidence limits were calculated for the differences of means to account for multiple tests within a trait. Values of P ≤ 0.05 were considered significant.
Linear regression analysis was used to examine the relationship between circulating concentrations of IAPP and insulin. Spearman correlation coefficients for plasma IAPP and serum insulin concentrations were calculated for each group. Spearman correlation coefficients were also calculated to evaluate relationships between BCS and IAPP concentration and between BCS and insulin concentration.
Results
Physical and clinical characteristics of cats—Signalment was evaluated for nondiabetic cats grouped according to BCS and for diabetic cats assigned to DM or DKA groups on the basis of clinicopathologic findings (Table 1). Breeds of nondiabetic cats included domestic shorthair (n = 46), domestic medium-hair (7), domestic longhair (11), Siamese (14), Himalayan (1), Manx (1), and 2 cats for which breed was not recorded. All nondiabetic cats in the study were neutered with the exception of 1 sexually intact female that had a BCS of 7. All laboratory values were within reference limits for the 82 nondiabetic cats (data not shown).
Physical characteristics and mean ± SD concentrations of plasma IAPP and serum insulin and glucose for 82 nondiabetic cats grouped according to BCS and 27 diabetic cats with and without ketoacidosis (DKA and DM groups, respectively).
| Age | Sex | Concentration* | ||||||
|---|---|---|---|---|---|---|---|---|
| Group | No. of cats | Mean | Median (range) | M | F | IAPP (pmol/L) | Insulin (pmol/L) | Glucose (mg/dL) |
| Nondiabetic (any BCS) | 82 | 6.1 | 5 (2–14) | 45 | 37 | 9 ± 5 | 85 ± 58 | 112 ± 28g,h |
| BCS 5 | 24 | 7.0 | 7 (2–12) | 13 | 11 | 6 ± 3a | 55 ± 50b | 106 ± 25 |
| BCS 6 | 12 | 5.3 | 4 (2–13) | 5 | 7 | 9 ± 3c | 81 ± 72 | 115 ± 16 |
| BCS 7 | 19 | 5.0 | 4 (2–11) | 11 | 8 | 10 ± 4a,d | 78 ± 46 | 114 ± 16 |
| BCS 8 | 22 | 6.7 | 6 (1–14) | 11 | 11 | 10 ± 6e | 124 ± 53b | 117 ± 44 |
| BCS 9 | 5 | 5.6 | 5 (2–8) | 5 | 0 | 11 ± 6 | 98 ± 35 | 102 ± 19 |
| DM | 21 | 9.3 | 8.5 (6–14) | 15 | 6 | 11 ± 8f | 53 ± 21 | 269 ± 53g |
| DKA | 6 | 12.4 | 13 (8–18) | 6 | 0 | 3 ± 2c,d,e,f | 46 ± 9 | 368 ± 113h |
Only cats with a BCS ≤ 5 on a scale of 1 to 9 were included in the study. Nondiabetic cats were further grouped according to BCS for data analysis.
Samples were obtained after food was withheld for 12 hours.
F = Female. M = Male.
Superscripted letters indicate significant differences among values within a column:
P = 0.007,
P < 0.001, and
P = 0.002.
The 21 cats in the DM group were neutered males and females. Breeds included domestic shorthair (n = 13), domestic medium-hair (1), domestic longhair (5), Siamese (1), and Persian (1). Body condition scores for these cats were as follows: 5 (n = 1), 6 (4), 7 (4), 8 (10), and 9 (2). In addition to hyperglycemia and glycosuria, cats in the DM group had mild increases in serum concentrations of cholesterol (3/21 cats; mean, 288 ± 26.5 mg/dL; reference range, 119 to 264 mg/dL) and globulin (4/21 cats; mean, 5.75 ± 0.4 g/dL; reference range, 2.5 to 5.3 g/dL) and activities of alanine aminotransferase (7/21 cats; mean, 138 ± 15.7 U/L; reference range, 17 to 95 U/L) and aspartate aminotransferase (15/21 cats; mean, 56 ± 20.2 U/L; reference range, 13 to 29 U/L). All other laboratory variables were normal in these cats.
The 6 cats in the DKA group were neutered males. Breeds in this group included domestic shorthair (n = 3), domestic medium-hair (1), and domestic longhair (2). Body condition scores for these cats were as follows: 5 (n = 4), 6 (1), and 8 (1). In addition to clinicopathologic abnormalities associated with ketoacidosis, 1 cat had a urinary tract infection and 1 was recovering from perineal urethrostomy surgery. All cats in the DM and DKA groups tested negative for FeLV antigen and anti-FIV antibody, and all had serum thyroxine concentrations within the laboratory reference range.
IAPP, insulin, and glucose concentrations—Plasma concentrations of IAPP and serum concentrations of insulin and glucose were evaluated in samples obtained from all cats after food was withheld for 12 hours (Table 1). A significant (P = 0.007) difference in mean IAPP concentrations was detected between nondiabetic cats with BCSs of 5 (6 ± 3 pmol/L) and 7 (10 ± 4 pmol/L; Figure 1). A significant (P < 0.001) difference was also detected in mean insulin concentrations between nondiabetic cats with BCSs of 5 (55 ± 50 pmol/L) and 8 (124 ± 53 pmol/L; Figure 2). Although mean IAPP and insulin concentrations appeared to be increased in the remaining cats with BCSs ≥ 6, compared with those in cats with a BCS of 5, these differences were not significant. There were no significant differences in mean glucose concentrations among nondiabetic cats with various BCSs (Figure 3).
Box-and-whiskers plot of plasma IAPP concentrations in 82 nondiabetic cats grouped according to BCS (range, 5 to 9 on a scale of 1 to 9), and in 27 diabetic cats with and without ketoacidosis (DKA and DM groups, respectively). Samples were obtained after food was withheld for 12 hours. Whiskers indicate the most extreme data points in the group that lay within the fences. The upper fence was defined as the third quartile (represented by the upper edge of the box) plus 1.5 times the interquartile range. The lower fence was defined as the first quartile (represented by the lower edge of the box) minus 1.5 times the interquartile range. Box widths are proportional to group size. Observations outside the fences (suspected outliers) are identified with a circle. Horizontal lines within boxes represent group median. Intersecting lines within boxes represent group mean.
Citation: American Journal of Veterinary Research 72, 8; 10.2460/ajvr.72.8.1052
Box-and-whiskers plot of plasma IAPP concentrations in 82 nondiabetic cats grouped according to BCS (range, 5 to 9 on a scale of 1 to 9), and in 27 diabetic cats with and without ketoacidosis (DKA and DM groups, respectively). Samples were obtained after food was withheld for 12 hours. Whiskers indicate the most extreme data points in the group that lay within the fences. The upper fence was defined as the third quartile (represented by the upper edge of the box) plus 1.5 times the interquartile range. The lower fence was defined as the first quartile (represented by the lower edge of the box) minus 1.5 times the interquartile range. Box widths are proportional to group size. Observations outside the fences (suspected outliers) are identified with a circle. Horizontal lines within boxes represent group median. Intersecting lines within boxes represent group mean.
Citation: American Journal of Veterinary Research 72, 8; 10.2460/ajvr.72.8.1052
Box-and-whiskers plot of plasma IAPP concentrations in 82 nondiabetic cats grouped according to BCS (range, 5 to 9 on a scale of 1 to 9), and in 27 diabetic cats with and without ketoacidosis (DKA and DM groups, respectively). Samples were obtained after food was withheld for 12 hours. Whiskers indicate the most extreme data points in the group that lay within the fences. The upper fence was defined as the third quartile (represented by the upper edge of the box) plus 1.5 times the interquartile range. The lower fence was defined as the first quartile (represented by the lower edge of the box) minus 1.5 times the interquartile range. Box widths are proportional to group size. Observations outside the fences (suspected outliers) are identified with a circle. Horizontal lines within boxes represent group median. Intersecting lines within boxes represent group mean.
Citation: American Journal of Veterinary Research 72, 8; 10.2460/ajvr.72.8.1052
Box-and-whiskers plot of serum insulin concentrations in the same 109 cats in Figure 1. See Figure 1 for key.
Citation: American Journal of Veterinary Research 72, 8; 10.2460/ajvr.72.8.1052
Box-and-whiskers plot of serum insulin concentrations in the same 109 cats in Figure 1. See Figure 1 for key.
Citation: American Journal of Veterinary Research 72, 8; 10.2460/ajvr.72.8.1052
Box-and-whiskers plot of serum insulin concentrations in the same 109 cats in Figure 1. See Figure 1 for key.
Citation: American Journal of Veterinary Research 72, 8; 10.2460/ajvr.72.8.1052
Box-and-whiskers plot of serum glucose concentrations in the same 109 cats in Figure 1. See Figure 1 for key.
Citation: American Journal of Veterinary Research 72, 8; 10.2460/ajvr.72.8.1052
Box-and-whiskers plot of serum glucose concentrations in the same 109 cats in Figure 1. See Figure 1 for key.
Citation: American Journal of Veterinary Research 72, 8; 10.2460/ajvr.72.8.1052
Box-and-whiskers plot of serum glucose concentrations in the same 109 cats in Figure 1. See Figure 1 for key.
Citation: American Journal of Veterinary Research 72, 8; 10.2460/ajvr.72.8.1052
Cats in the DM and DKA groups had significantly (P < 0.001) higher mean serum glucose concentrations than did nondiabetic cats of any BCS (Table 1; Figure 3). Cats in the DM group had a mean plasma IAPP concentration similar to those of nondiabetic cats with a BCS of 8 or 9; mean serum insulin concentration was lower in cats of the DM group, compared with that of nondiabetic cats, but this difference was not significant. Cats in the DKA group had a significantly (P < 0.001) lower mean IAPP concentration (3 ± 2 pmol/L) than did nondiabetic cats with BCSs of 6 (9 ± 3 pmol/L), 7 (10 ± 4 pmol/L), or 8 (10 ± 6 pmol/L) and cats in the DM group (11 ± 8 pmol/L; P = 0.002). Although mean insulin concentration in cats of the DKA group (46 ± 9 pmol/L) was lower than in cats of the DM group (53 ±21 pmol/L), this difference was not significant. The mean glucose concentration was higher in cats of the DKA group (368 ±113 mg/dL) than in cats of the DM group (269 ± 53 mg/dL), but this difference was also nonsignificant.
Results of linear regression analysis with plasma IAPP concentration as the dependent variable and serum insulin concentration and group as independent variables indicated that, after adjusting for group, insulin concentrations were significantly (P < 0.003) associated with IAPP concentrations; as insulin concentration increased by 1 pmol/L, IAPP concentration increased by 0.032 pmol/L. In addition, calculation of Spearman correlation coefficients revealed significant correlations between plasma IAPP and serum insulin concentrations in nondiabetic cats with a BCS of 5 (P < 0.003) or 7 (P = 0.005) but not in those with a BCS of 6, 8, or 9 or in cats of the DM or DKA groups (Figure 4; Table 2). In nondiabetic cats overall, BCS and plasma IAPP concentration were significantly (P < 0.001) and positively correlated (r = 0.38; Figure 5). Similarly, BCS and serum insulin concentration were significantly (P < 0.001) and positively correlated in these cats (r = 0.54; Figure 6).
Scatterplot of serum insulin versus plasma IAPP concentrations in 109 cats. Samples were obtained after food was withheld for 12 hours. White circles represent values for nondiabetic cats with BCSs of 5 through 9 on a scale of 1 to 9. Black circles represent values for diabetic cats with and without keto-acidosis (DKAand DM groups, respectively).
Citation: American Journal of Veterinary Research 72, 8; 10.2460/ajvr.72.8.1052
Scatterplot of serum insulin versus plasma IAPP concentrations in 109 cats. Samples were obtained after food was withheld for 12 hours. White circles represent values for nondiabetic cats with BCSs of 5 through 9 on a scale of 1 to 9. Black circles represent values for diabetic cats with and without keto-acidosis (DKAand DM groups, respectively).
Citation: American Journal of Veterinary Research 72, 8; 10.2460/ajvr.72.8.1052
Scatterplot of serum insulin versus plasma IAPP concentrations in 109 cats. Samples were obtained after food was withheld for 12 hours. White circles represent values for nondiabetic cats with BCSs of 5 through 9 on a scale of 1 to 9. Black circles represent values for diabetic cats with and without keto-acidosis (DKAand DM groups, respectively).
Citation: American Journal of Veterinary Research 72, 8; 10.2460/ajvr.72.8.1052
Scatterplot of plasma IAPP concentrations versus BCS in 82 nondiabetic cats. Samples were obtained after food was withheld for 12 hours. Line represents results of linear regression analysis.
Citation: American Journal of Veterinary Research 72, 8; 10.2460/ajvr.72.8.1052
Scatterplot of plasma IAPP concentrations versus BCS in 82 nondiabetic cats. Samples were obtained after food was withheld for 12 hours. Line represents results of linear regression analysis.
Citation: American Journal of Veterinary Research 72, 8; 10.2460/ajvr.72.8.1052
Scatterplot of plasma IAPP concentrations versus BCS in 82 nondiabetic cats. Samples were obtained after food was withheld for 12 hours. Line represents results of linear regression analysis.
Citation: American Journal of Veterinary Research 72, 8; 10.2460/ajvr.72.8.1052
Scatterplot of serum insulin concentrations versus BCS in the same 82 cats in Figure 5. See Figure 5 for remainder of key.
Citation: American Journal of Veterinary Research 72, 8; 10.2460/ajvr.72.8.1052
Scatterplot of serum insulin concentrations versus BCS in the same 82 cats in Figure 5. See Figure 5 for remainder of key.
Citation: American Journal of Veterinary Research 72, 8; 10.2460/ajvr.72.8.1052
Scatterplot of serum insulin concentrations versus BCS in the same 82 cats in Figure 5. See Figure 5 for remainder of key.
Citation: American Journal of Veterinary Research 72, 8; 10.2460/ajvr.72.8.1052
Spearman correlation coefficients for plasma IAPP and serum insulin concentrations* for the same 109 cats in Table 1.
| Group | No. of cats | Correlation coefficient (r) | P value |
|---|---|---|---|
| Nondiabetic | 82 | ND | ND |
| BCS 5 | 24 | 0.59 | < 0.003 |
| BCS 6 | 12 | 0.52 | 0.085 |
| BCS 7 | 19 | 0.62 | 0.005 |
| BCS 8 | 22 | 0.40 | 0.062 |
| BCS 9 | 5 | 0.50 | 0.390 |
| DM | 19† | 0.09 | 0.720 |
| DKA | 6 | 0.10 | 0.870 |
Values of P < 0.05 were considered significant.
Because of laboratory error, no IAPP concentration was obtained in samples from 2 cats in the DM group. Therefore, these 2 cats were excluded from IAPP-insulin correlation analysis.
ND = Not determined.
See Table 1 for remainder of key.
Discussion
Although the pathogenesis of type 2 diabetes mellitus in humans and cats is far from being resolved, there is increasing evidence that pathological consequences of IAPP misfolding and subsequent aggregation into toxic oligomers and amyloid fibrils play an important role in the molecular pathogenesis of the disease.12,13 Evidence from studies12,13 of type 2 diabetes mellitus in transgenic rodents, humans, and cats has suggested an important role for increased synthesis and secretion of IAPP in this process. In this regard, evidence from human studies15,16 has shown that circulating insulin and IAPP concentrations are increased in overweight glucose-intolerant individuals and those with early type 2 diabetes, while they are decreased in patients with advanced type 2 diabetes who are being treated with insulin. Our group has also previously shown that cats with impaired glucose tolerance had increased IAPP content in their pancreatic β cells with early development of islet amyloid deposits.14
Results of the present study added to those earlier findings in cats and revealed that nondiabetic moderately obese cats (BCS, 7) had significantly increased plasma IAPP concentrations, compared with those of cats of ideal body weight (BCS, 5). Furthermore, a significant positive correlation was identified between BCS and plasma IAPP concentrations in nondiabetic cats. Although mean IAPP concentrations were also greater in the more severely obese groups of cats (BCS, 8 or 9) than in cats with ideal body weight, the differences among the groups were not statistically significant.
Because obesity is established as a predisposing factor in the development of type 2 diabetes, the findings of the present study indicate a potential link between obesity, with the associated increase in circulating IAPP concentrations, and the development of type 2 diabetes mellitus in cats. The pathogenetic connection between increased circulating IAPP concentrations and type 2 diabetes mellitus is linked to the fact that IAPP-derived amyloidosis is a characteristic feature of all species that develop type 2 diabetes mellitus, including humans, cats, and macaques.1,2,10 Furthermore, it has long been known that concurrent with the development of islet amyloidosis in these species there is significant loss of islet β cells and that this process occurs prior to the development of type 2 diabetes mellitus. Islet amyloid polypeptide amyloidogenesis has been linked to β-cell apoptosis via the generation of toxic intermediate-sized oligomers, which are formed early in the pathway and eventually lead to the formation of amyloid fibrils. A factor that has been implicated in the pathogenesis of islet amyloid is increased production of IAPP by β cells. The positive correlation between plasma concentrations of IAPP and BCS and the increased mean concentration of IAPP in cats of the DM group, compared with that of nondiabetic cats of ideal body weight in the present study, are therefore of pathogenetic interest because increased IAPP may potentially serve as a prelude to the eventual initiation of amyloidogenesis in susceptible cats.
Although obesity (or other causes of insulin resistance) appears generally to be necessary for the development of type 2 diabetes, obesity and increased IAPP values alone are not sufficient to explain the transition to diabetes. The factors that allow formation of toxic IAPP oligomers and subsequent development of type 2 diabetes mellitus in only a small proportion of obese cats, humans, or macaques are unknown and are the subject of considerable ongoing research.
In the study reported here, mean serum insulin and plasma IAPP concentrations measured in samples obtained after food was withheld for 12 hours were positively and significantly correlated in cats with ideal body weight (BCS, 5) consistent with a tightly regulated ratio of production and secretion between insulin and IAPP, as previously reported.7,21 However, correlation coefficients were reduced with increasing obesity in most remaining BCS groups, and a significant correlation was detected only for cats with a BCS of 7. Similarly, the correlation between serum insulin and plasma IAPP concentrations in cats of the DM and DKA groups was not significant. This was also reflected in the high mean IAPP concentrations detected in cats of the DM group, despite the fact that mean circulating insulin concentrations appeared to be decreased, compared with those in nondiabetic cats. Therefore, these data suggest that obesity and diabetes are associated with loss of correlation between circulating IAPP and insulin concentrations, which may highlight an underlying fundamental alteration in pancreatic β-cell function that begins prior to development of clinical diabetes and continues in the disease state. It should also be noted that cats at the time of diagnosis of type 2 diabetes mellitus typically have an approximately 50% reduction in β-cell mass,22 which implies that the remaining β cells must have extremely high IAPP production to maintain the described high circulating concentrations of IAPP. As mentioned, increased IAPP production has been hypothesized to be a contributing factor in the cascade of events leading to misfolding of the IAPP molecule and consequent formation of toxic IAPP oligomers ultimately leading to β-cell death.13
Another key feature in the pathogenesis of type 2 diabetes mellitus in cats and humans is the development of insulin resistance with attendant high insulin concentrations, which is most often secondary to obesity23–25 Results of the present study revealed that markedly obese cats (BCS, 8) had significantly higher serum insulin concentrations, compared with cats with an ideal BCS, despite normoglycemia, consistent with an insulin-resistant state. Furthermore, a significant positive correlation was detected between BCS and serum insulin concentrations, indicating that this metabolic state can occur early in the development of obesity. Thus, our data are consistent with previous findings in nondiabetic humans in which obesity was associated with hyperinsulinemia and high circulating concentrations of IAPP and also show that even mild obesity may be associated with perturbations in insulin and IAPP regulation in cats. These findings highlight the potential importance of maintaining normal body weight in the long-term health of domestic cats and other species at risk of development of type 2 diabetes mellitus.
ABBREVIATIONS
| BCS | Body condition score |
| IAPP | Islet amyloid polypeptide |
Boehringer Mannheim GmbH, Mannheim, Germany.
#PI-12K, Linco, St Charles, Mo.
Peninsula Laboratories, San Carlos, Calif.
Sigma Chemical Co, St Louis, Mo.
Beckman 7300 amino acid analyzer, Beckman Coulter, Fullerton, Calif.
References
- 1.
Johnson KH, Hayden DW, O'Brien TD, et al. Spontaneous diabetes mellitus-islet amyloid complex in adult cats. Am J Pathol 1986; 125: 416–419.
- 2.
Henson MS, O'Brien TD. Feline models of diabetes mellitus. Inst Lab Anim Res J 2006; 47: 234–242.
- 3.
Westermark P, Wernstedt C, O'Brien TD, et al. Islet amyloid in type 2 human diabetes mellitus and adult diabetic cats contains a novel putative polypeptide hormone. Am J Pathol 1987; 127: 414–417.
- 4.
Westermark P, Wernstedt C, Wilander E, et al. Amyloid fibrils in human insulinoma and islets of Langerhans of the diabetic cat are derived from a neuropeptide-like protein also present in normal islet cells. Proc Natl Acad Sci U S A 1987; 84: 3881–3885.
- 5.
Johnson KH, O'Brien TD, Hayden DW, et al. Immunolocalization of islet amyloid polypeptide (IAPP) in pancreatic beta cells using peroxidase antiperoxidase (PAP) and protein A-gold techniques. Am J Pathol 1988; 130: 1–8.
- 6.
Butler PC, Chou J, Carter WB, et al. Effects of meal ingestion on plasma amylin concentration in NIDDM and nondiabetic humans. Diabetes 1990; 39: 752–756.
- 7.
O'Brien TD, Westermark P, Johnson KH. Islet amyloid polypeptide and insulin secretion from isolated perfused pancreas of fed, fasted, glucose-treated, and dexamethasone-treated rats. Diabetes 1991; 40: 1701–1706.
- 8.
Betsholtz C, Christmansson L, Engstrom U, et al. Sequence divergence in a specific region of islet amyloid polypeptide (IAPP) explains differences in islet amyloid formation between species. FEBS Lett 1989; 251: 261–264.
- 9.
Betsholtz C, Svensson V, Rorsman F, et al. Islet amyloid polypeptide (IAPP):cDNA cloning and identification of an amyloidogenic region associated with the species-specific occurrence of age-related diabetes mellitus. Exp Cell Res 1989; 183: 484–493.
- 10.
O'Brien TD, Wagner JD, Litwak KN, et al. Islet amyloid and islet amyloid polypeptide in cynomolgus macaques (Macaca fascicularis): an animal model of human non-insulin-dependent diabetes mellitus. Vet Pathol 1996; 33: 479–485.
- 11.
Janson J, Ashley RH, Harrison D, et al. The mechanism of islet amyloid polypeptide toxicity is membrane disruption by intermediate-sized toxic amyloid particles. Diabetes 1999; 48: 491–498.
- 12.
Haataja L, Gurlo T, Huang CJ, et al. Islet amyloid in type 2 diabetes, and the toxic oligomer hypothesis. Endocr Rev 2008; 29: 303–316.
- 13.↑
Bonner-Weir S, O'Brien TD. Islets in type 2 diabetes: in honor of Dr. Robert C. Turner. Diabetes 2008; 57: 2899–2904.
- 14.↑
Johnson KH, O'Brien TD, Jordan K, et al. Impaired glucose tolerance is associated with increased islet amyloid polypeptide (IAPP) immunoreactivity in pancreatic beta cells. Am J Pathol 1989; 135: 245–250.
- 15.
Sanke T, Hanabusa T, Nakano Y, et al. Plasma islet amyloid polypeptide (amylin) concentrations and their responses to oral glucose in type 2 (non-insulin-dependent) diabetic patients. Diabetologia 1991; 34: 129–132.
- 16.
Enoki S, Mitsukawa T, Takemura J, et al. Plasma islet amyloid polypeptide concentrations in obesity, impaired glucose tolerance and non-insulin-dependent diabetes mellitus. Diabetes Res Clin Pract 1992; 15: 97–102.
- 17.↑
Reusch CE, Liehs MR, Hoyer M, et al. Fructosamine. A new parameter for diagnosis and metabolic control in diabetic dogs and cats. J Vet Intern Med 1993; 7: 177–182.
- 18.↑
LaFlamme DP. Development and validation of a body condition score system for cats: a clinical tool. Feline Pract 1997; 25 (5/6):13–18.
- 19.↑
Kadish AH, Hall DA. A new method for the continuous monitoring of blood glucose by measurement of dissolved oxygen. Clin Chem 1965; 11: 869–875.
- 20.↑
Hoenig M, Hall G, Ferguson D, et al. A feline model of experimentally induced islet amyloidosis. Am J Pathol 2000; 157: 2143–2150.
- 21.
Fehmann HC, Weber V, Goke R, et al. Co-secretion of amylin and insulin from isolated rat pancreas. FEBS Lett 1990; 262: 279–281.
- 22.↑
O'Brien TD, Hayden DW, Johnson KH, et al. Immunohisto-chemical morphometry of pancreatic endocrine cells in diabetic, normoglycemic glucose-intolerant and normal cats. J Comp Pathol 1986; 96: 357–369.
- 23.
Kirk CA, Feldman EC, Nelson RW. Diagnosis of naturally acquired type-I and type-II diabetes mellitus in cats. Am J Vet Res 1993; 54: 463–467.
- 24.
Hoenig M, Alexander S, Holson J, et al. Influence of glucose dosage on interpretation of intravenous glucose tolerance tests in lean and obese cats. J Vet Intern Med 2002; 16: 529–532.
- 25.
Appleton DJ, Rand JS, Sunvold GD. Plasma leptin concentrations are independently associated with insulin sensitivity in lean and overweight cats. J Feline Med Surg 2002; 4: 83–93.
