Group 1	BG + insulin	BG (~1 μL)	BG + insulin
	(~40 μL)		(~40 μL)
Group 2	BG + insulin	BG (~1 μL)	BG + insulin +
	(~40 μL)		exposure
			(~200 μL)
Group 3	—	BG + insulin	BG + insulin
		(~40 μL)	(~40 μL)

Plasma samples were kept at −80° C. until insulin measurement (using ELISA method).

A sample of blood (˜200 μL) was obtained just prior to the clamp procedure for eachGroup 2 animal, processed to plasma (˜60 μL), and maintained at −80° C. for subsequent determination of Ab14 concentrations utilizing a Meso Scale Discovery (MSD) ELISA platform.

Vehicle and Ab14 were administered by i.v. route 48 h before the TO of clamp procedure (at 02:00 pm two days before the clamp).

Metformin was administered by p.o. route 24 h and 4 h before the TO of clamp procedure (at 02:00 pm the day before the clamp and at 10:00 am the day of the clamp).

The rats were fasted 6 hours before the start of the clamp (at ˜8 h, just after the blood collection).

The hyperinsulinemic-euglycemic clamp was performed using³H-glucose as a tracer and 0.3 U/kg/h insulin infusion from 02:00 pm (TO) to 05:00 pm (T+3 h). A glucose solution was infused in parallel and the infusion rate was adjusted to reach the steady state (˜100+/−10 mg/dL). Blood glucose was measured from the tip of the tail using glucometers every 10 minutes. Blood was collected (10 μL) from the tip of the tail during the last hour (steady state) and the following parameters were assessed: Glucose infusion rate; Whole body glucose utilization rate; Hepatic glucose production rate; Whole body glycogen and glycolytic rates.

To determine the individual tissue glucose utilization rate, a bolus injection of 100 μCi per rat of deoxy-D-glucose 2-¹⁴C (¹⁴C-2-DOG) through the femoral vein was performed 60 min before the end of the D-[3-³H]-glucose infusion. Plasma¹⁴C-2-DOG disappearance and glucose concentration were determined in 10 μL drops of blood sampled from the tip of the tail vein at 0, 5, 10, 15, 20, 25, 30, 45, and 60 minutes after the injection. At the end of the experiment, vastus lateralis (VL), extensor digitorum longus (EDL) and soleus muscles, epididymal and inguinal white adipose tissues, heart apex, and skin (as negative control) were dissected, flash frozen and kept at −80° C. A piece of each tissue was dissolved in 1M NaOH and then neutralized with 1M HCl. D-2-¹⁴C deoxyglucose C deoxyglucose 6-phosphate. D-2-¹⁴C deoxyglucose was differentially precipitated by the use of a zinc hydroxide (0.3M) solution or a perchloric acid solution (6%). Both radioactivity contents were measured to evaluate the glucose uptake expressed as ng/mg/min.

Plasma insulin level was measured at the end of the clamp.

Statistical analyses were performed using GraphPad prism software. Histograms were analyzed using an ANOVA one way with a Dunnett's post test and curves were analyzed using an ANOVA two ways with a Bonferroni's post test. A difference was considered significant when p value was <0.05. NS: not significant.

Results and Discussion

Blood Glucose and Insulin Measurement.

In fed conditions, blood glucose level was not affected 42 h after treatment withAb14 100 mg/kg and 18 h after treatment withmetformin 500 mg/kg compared to the vehicle group (FIG. 1A). At the same time, Ab14 and metformin decreased plasma insulin level by 11% and 18% respectively (non significant,FIG. 1B). Then the index of insulin resistance HOMA-IR was decreased in a similar manner compared to the vehicle group (FIG. 1C). On the other hand, in 6 hours fasting condition, metformin significantly decreased blood glucose by 17% 30 h after treatment (FIG. 1D).

Whole Body Glucose Fluxes.

Hyperinsulinemic euglycemic clamps were performed in 6 hours fasting conditions 48 h after the single administration ofAb14 100 mg/kg, and 4 h after the last administration ofmetformin 500 mg/kg.

Metformin significantly increased GIR evolution from 60 minutes after the start of the infusion compared to vehicle group. Ab14 had a trend to increase the GIR evolution mostly after 130 minutes infusion (FIG. 2A).

The plateau of glucose infusion rate was reached from 140 min in all groups. The blood glucose level was similar in all groups during this steady state (FIG. 2B). Plasma insulin level at the end of the clamp was also similar in all groups (FIG. 2C). The plasma insulin level reached at the end of the clamp was almost the same as that measured in fed conditions, i.e., the dose of insulin used to obtain hyperinsulinemia was physiological.

Glucoses fluxes were then calculated from 140 minutes to 180 minutes of infusion (FIG. 3). Ab14 and metformin increased the glucose infusion rate by 18% (NS) and 27% (p<0.05) respectively, as well as the glucose turn over by 18% (NS) and 30% (p<0.05) respectively. The hepatic glucose production was totally inhibited by this supraphysiological dose of insulin in the 3 groups. Ab14 did not affect glycolysis rate whereas metformin increased it by 43% (NS). Ab14 increased glycogen synthesis rate by 23% similarly to metformin (NS).

Individual tissue glucose utilization rates were also determined using a bolus injection of 100 μCi per rat of deoxy-D-glucose 2-¹⁴C (¹⁴C-2-DOG) through the femoral vein introduced 60 min before the end of the D-[3-³H]-glucose infusion. Both radioactivity contents were measured to evaluate the glucose uptake expressed as ng/mg/min. As shown inFIG. 4A-C, metformin increased glucose utilization in the mixed vastus lateralis muscle (VL, 49% p<0.05), in the glycolytic extensor digitorum longus muscle (EDL, 19% NS) and decreased glucose utilization in the heart (−39%, p<0.01), which is a known effect of metformin thought to be due to the stimulation of myocardial fatty acid oxidation. Ab14 tended to increase glucose infusion rate and whole body glucose turn over (NS) and had a stronger effect than metformin on glucose utilization in the VL (70%, p<0.01), in EDL (26%, NS), and in the oxidative soleus muscle (27%, NS). It also tended to increase glucose utilization in the heart (21%, NS). Similar to metformin, Ab14 did not affect glucose utilization rate in white adipose tissues (deep and subcutaneous).

Ab14 Plasma Concentration Analysis.

The Ab14 plasma concentrations forGroup 2 animals undergoing the clamp procedure ranged from 642 to 797 μg/mL and supported up to 48-hours of systemic exposure.

Conclusion.

Acute treatment with Ab14 slightly decreased plasma insulin and tended to increase whole body glucose utilization by increasing glycogen synthesis and muscle glucose utilization.

Example 2

This example assesses the ability of Ab14 to improve insulin sensitivity in a rat model of insulin resistance. In this model, rats were fed with a high fat (69%) and high fructose (14%) diet (HFD) for 6 weeks to induce glucose intolerance, with the plasma insulin level becoming significantly increased and glycaemia becoming slightly increased compared to control animals fed normal chow. The antibody used in this example consisted of the light and heavy polypeptide chains of SEQ ID NOs 132 and 134.

Six-week HFD fed rats were treated for 2 weeks with Ab14, and a 2-step hyperinsulinemic euglycemic clamp was performed to assess insulin sensitivity. A physiological dose of insulin was used during the first step and a pharmacological dose of insulin was used during the second step to assess the effects of peripheral and hepatic insulin sensitivity, respectively, under these two conditions.

Summary

After 6 weeks of HFD rats were randomized to treatment groups according to their glucose intolerance (AUC calculation during an oral glucose tolerance test (OGTT)) and their HOMA-IR (insulin resistance index). HFD rats were treated for 15 days (i.v., two administrations one week apart) with Ab14 at 10, 30 and 100 mg/kg/week, or daily withmetformin 200 mg/kg/day in drinking water.

Body weight and food consumption were measured 3 times per week until 10 days of treatment. HOMA-IR was measured on

day

10 and 15. A 2-step clamp (5 mU/kg/min and then 15 mU/kg/min insulin) was performed on

day

15 or 16. Glucose Turnover (GTO) was assessed using³H-glucose tracer infusion during the clamp procedure.

At the end of the study, when compared to control rats fed normal chow, HFD rats were observed to exhibit significant increases in body weight, fasting blood glucose, plasma insulin, and C peptide, as well as a significant decrease in glucose infusion rate (GIR) during the 2-step hyperinsulinemic euglycemic clamp.

When compared to the HFD plus vehicle control group, Ab14 treatment had no effects upon body weight or food consumption while the metformin group was significantly decreased in both parameters.

Ab14 at 100 mg/kg significantly decreased HOMA-IR by 38% after 15 days of treatment (by decreasing fasting blood glucose as well as plasma insulin). Metformin had a non-significant (ns) decreasing effect on HOMA-IR onday 15. C-peptide was also significantly decreased by Ab14 treatment (by 30% with 10 mg/kg and 29% with 100 mg/kg).

Increased GIR (ns) was observed in Ab14 treated groups when compared to the HFD vehicle control group during the first step of the clamp, while metformin had an increasing effect upon GIR that was comparatively less in magnitude. Ab14 at 30 or 100 mg/kg and metformin treatment significantly increased GIR during the second step of the clamp (by 36, 28, and 27% respectively).

Ab14 at 30 mg/kg tended to increase GTO when compared to the HFD vehicle control group during the first step (by 17%, ns) of the clamp procedure. Ab14 at 30 mg/kg had a slight increasing effect (ns) upon glycolysis and glycogen synthesis during both clamp steps.

Hepatic glucose production (HGP) was slightly but not significantly decreased by Ab14 or metformin treatment during the first clamp step (between 11-20%). During the second step, HGP was non-significantly decreased by Ab14 at 10 mg/kg (by 78%) when compared to the HFD vehicle control group, and HGP was completely inhibited by Ab14 at 30 or 100 mg/kg and by metformin.

In conclusion, improved insulin resistance (mainly hepatic) was observed following intravenous administration with Ab14 in the HFD rat model.

Methods

82 Sprague Dawley rats (8 weeks old at start of study, average weight about 250 grams) were housed in housing cages (904 cm²×23 cm) throughout the experimental phase. Animals' cages litters were changed 3 times per week. They were housed in groups of 2-3 animals during acclimation, HFD and treatment period. Then rats were individually housing after surgery until the clamp procedure. The rats were housed with an inverted 12 hour light cycle (at 08:00 am lights off), with temperature maintained at 22±2° C. and 55±10% relative humidity. At least 5 days of acclimation period was provided before commencement of HFD feeding. During the acclimation phase, standard diet (RM1 (E) 801492, SDS) and tap water were provided ad libitum.

After the acclimation phase, 10 rats were fed with normal chow (NC) whereas 72 rats were fed with HFD (RD1, SAFE) throughout the experiment.

The high fat diet composition was as follows (Kcal %): Protein: 17.3%; Carbohydrate (fructose): 14%; Fat (lard)): 68.7%; cholesterol 1.65%, cholic acid 0.65%.

After 6 weeks of HFD feeding, the rats were fasted for 6 hours, and a glucose tolerance test was performed. The rats presenting the lowest AUC (˜17%) were excluded from the study. The remaining rats were then randomly allocated to the different groups according to their AUC (glucose tolerance index) and HOMA-IR (insulin resistance index).

Ab14 (10, 30, 100 mg/kg) and the vehicle were weekly administered via i.v. route (via the caudal vein, under isoflurane anaesthesia), in the morning, onday 1 andday 8 of treatment.

Metformin (200 mg/kg/day) was administered in drinking water for ˜2 weeks until the clamp procedure. Rats treated with metformin were treated with vehicle onday 1 and day 8 (via the caudal vein).

The test groups were as follows:

Group 1: NC+vehicle i.v. (n=10)

Group 2: HFD+vehicle i.v. (n=10)

Group 3: HFD+Ab14 10 mg/kg i.v. (n=10)

Group 4: HFD+Ab14 30 mg/kg i.v. (n=10)

Group 5: HFD+Ab14 100 mg/kg i.v. (n=10)

Group 6: HFD+metformin 200 mg/kg in drinking water+vehicle i.v. (n=10)

During the last week of HFD feeding, before the screening, water consumption was measured 3 times in the week to evaluate the metformin dilution in tap water.

After 6 weeks of normal chow and HFD, the 82 rats were fasted from 08:00 am to 02:00 pm (6 hours). A glucose bolus was administered (2.5 g/kg) at 02:00 pm (t0). Blood glucose was measured (using glucometer, in a blood drop collected from the tail tip) on t-30, 0, 15, 30, 60, 90, 120, 150 min. Blood was collected from the tip tail (40 μL on EDTA) on t-30 to measure plasma insulin (ELISA method).

Area under the curve (AUC) was calculated. The 12 HFD fed rats presenting the highest AUC were considered as the less glucose intolerant and were excluded from the study. The 60 remaining rats were randomly allocated to the 6 groups, according to homogeneous AUC, HOMA-IR, and body weight.

Body weight was measured once a week during the first six week of HFD. Body weight was measured 3 times per week during the first 10 days of treatment. Food consumption was measured over 48 h or 72 h, just before treatment and 3 times per week during treatment until surgery procedure (day 11).

Before the start of the treatment (the day of OGTT) and onday 10 of treatment, all rats were fasted from 08:00 am. At 01:30 pm, blood was collected from the from the tip tail (404 on EDTA). Blood glucose (using glucometers) and plasma insulin (ELISA method) were measured.

On day 11, rats were anesthetized (isoflurane) and a catheter was implanted in the femoral vein. A recovery period was followed for 4 days before the clamp procedure.

The morning of the clamp, rats were fasted 6 hours (from 8:00 am to 02:00 pm). A sample of blood (˜160 μL, on EDTA) was collected from the tip of the tail, just prior to the clamp procedure (at ˜01:00 pm) from each rat in

group

3, 4, 5 and 6, processed to plasma (˜60 μL), and maintained at −80° C. until assessment of Ab14 concentrations. Although the control and metformin groups were not analyzed for antibody concentration, a similar amount of blood was collected (and discarded) from the rats of

15 or 16, the 2-step hyperinsulinemic-euglycemic clamp was performed using³H-glucose as a tracer (except in the normal chow group), and 5 mU/kg/min insulin infusion from 02:00 pm (TO) to 04:00 pm (T+2 h), followed by 15 mU/kg/min insulin infusion from 04:00 pm to 05:30 pm (t+3.5 h). A glucose solution was infused in parallel and the infusion rate was adjusted to reach the steady state (100+/−10 mg/dL). Blood glucose was measured from the tip of the tail using glucometers every 10 minutes. Blood was collected (104) regularly from the tip of the tail during the steady states of each step.

The following parameters were assessed: Glucose infusion rate (in all groups); Whole body glucose utilization rate (except in the normal chow group); Hepatic glucose production rate (except in the normal chow group); Whole body glycogen and glycolytic rates (except in the normal chow group).

Moreover, except in the normal chow group, 1 hour before the end of the clamp experiment, a bolus injection of¹⁴C-2DOG was performed and samples of the following tissues were collected at the end of the clamp and retained for further evaluation:

Vastus lateralis (VL) muscle; Extensor digitorum longus (EDL) muscle; Soleus muscle heart apex; epididymal white adipose tissue inguinal white adipose tissue skin (negative control).

Plasma insulin and C peptide levels were measured just before the infusion starts (˜T-30 min.), at the end of the steady state of step 1 (T2 h) and step 2 (T3.5 h). For that, blood collection was performed from the tip of the tail (˜100 μL, on EDTA).

Statistical analyses were performed using GraphPad prism software. Curves were analyzed using ANOVA two ways with a Bonferroni's post-test. Histograms were analyzed using a t-test to compare the HFD plus vehicle control group and the normal chow plus vehicle control group. Histograms were analyzed using an ANOVA one way with a Dunnett's post-test to compare Ab14 and metformin groups against the HFD plus vehicle control group. A difference was considered significant when the p-value was <0.05. NS: not significant.

Results

Animal Model and Screening.

8-week old rats were fed with a fructose enriched high fat diet (HFD, 69% fat and 16% fructose) for 7 weeks before the start of the treatment. Body weight was 522±5 g in the HFD group vs 448±13 g in control group fed normal chow. There was then a 16% increase of body weight (p<0.001 with a t-test onday 42 under HFD (FIG. 5). After 7 weeks, body weight was increased by 187±3 g in HFD population vs 158±9 g in the control group fed normal chow (p<0.001 with a t-test onday 42,FIG. 6).

During the 7th week of HFD, an oral glucose tolerance test was performed to assess glucose intolerance in the HFD population. The blood glucose level remained higher until 150 min after the glucose administration in the HFD population (not shown). The AUC calculated relatively to the TO was significantly higher (9%) in HFD rats compared to control chow-fed rats (not shown).

The HOMA-IR (insulin resistance index) was calculated on t-30 of OGTT. The rats presenting the higher AUC and the higher HOMA-IR, were randomly allocated to the 6 groups. AUC was higher (˜9%, not shown) in HFD groups compared to the control chow-fed group, as well as the HOMA-IR (34%, not shown) and the body weight (˜17%, p<0.001,FIG. 7)

Body Weight and Food Intake Follow-Up.

Body weight was followed for 10 days of treatment. Ab14 treatment had no effects upon body weight. Body weight of control chow-fed rats remained significantly lower than HFD vehicle rats (FIG. 7). Ab14 (30 mg/kg) slightly decreased (ns) andMetformin 200 mg/kg significantly decreased body weight gain from the second day of treatment onward.

Food consumption was lower in HFD vehicle than in control chow-fed group as expected. Ab14 treatment had no effect on the follow-up food intake (FIG. 8A) or on cumulative food intake (FIG. 8B not shown). Metformin significantly decreased cumulative food consumption by 25%. The fasting of animals prior to surgical procedures disrupted the food consumption measurements between

day

9 and 10.

Biochemical Parameters.

Fasting blood glucose increased in the HFD vehicle group when compared to the control chow-fed group by 5% (ns), 11% (ns), and 20% (p<0.001) on

days

0, 10 and 15 respectively. Treatment with Ab14 or metformin had no effect onday 10. Treatment with Ab14 at 100 mg/kg had a significant decreasing effect on fasting blood glucose onday 15 when compared to the HFD vehicle group (FIG. 9).

Fasting plasma insulin was increased by 33% (ns), 49% (ns) and 67% (p<0.01) in HFD vehicle group when compared to the control chow-fed group on

days

0, 10 and 15 respectively. Ab14 treatment had no effect onday 10 whereas metformin decreased plasma insulin by 37% (ns). Ab14 treatment at 10, 30 or 100 mg/kg decreased plasma insulin by 26, 16, or 18% (ns), respectively, after 15 days of treatment, and metformin decreased plasma insulin by 11% (ns,FIG. 10).

As expected, the plasma C-peptide level profile onday 15 was similar to the plasma insulin level, but the effects were more marked and less variable. The C-peptide was significantly increased by 67% in HFD vehicle group as compared to the control chow-fed group. Ab14 treatment at 10, 30, or 100 mg/kg decreased C-peptide level by 30% (p<0.05), 23% (ns), and 29% (p<0.05), respectively, and metformin decreased C-peptide by 13% (ns) (FIG. 11, lower left panel).

HOMA-IR, an index of insulin resistance, was increased in the HFD vehicle group as compared to the control chow-fed group by 36% on day 0 (ns), by 42% on day 10 (ns) and by 98% on day 15 (p<0.01). Compared to HFD vehicle, Ab14 had no effect after 10 days of treatment whereas metformin had a decreasing effect (ns) by 36%. After 15 days of treatment, Ab14 at 10, 30 or 100 mg/kg decreased HOMA-IR by 33% (ns), 17% (ns) and 38% (p<0.05), respectively, and metformin tended to decrease HOMA-IR by 18% (ns,FIG. 12).

Hyper-Insulinemic Clamp.

FIG. 13 shows the glucose infusion rate (GIR) over time during the 2-step hyper-insulinemic clamp. During the first step (5 mU/kg/min insulin), hepatic glucose production (HGP) was incompletely inhibited and the glucose infusion rates (GIR) were lower than during the second step (15 mU/kg/min insulin) when hepatic glucose production was inhibited.

GIR for the control chow-fed group was higher than the HFD vehicle group during both of the clamp steps, and confirmed that HFD rats had an insulin-resistant phenotype after 8-9 weeks of diet. Metformin had no effect on GIR during the first clamp step, while the GIR plateau was slightly higher (ns) in Ab14 treated groups. All treated groups were observed with GIR plateaus higher than the HFD vehicle group during the second clamp step, with significant differences observed for metformin and Ab14 at 30 or 100 mg/kg (FIG. 13). Statistical significance was evaluated using a two-way ANOVA with Bonferroni's post test versus HFD. During the first clamp step, the GIR was significantly different only for the normal chow control, vehicle treated rats at the 50 and 60 minute time points (p<0.01 and p<0.05 respectively). During the second clamp step, the GIR was significantly different for the HFD rats treated with 30 mg/kg Ab14 at the 160-210 minute time points (p<0.05 at 160 minutes and p<0.01 for the 170-210 minute time points), for the HFD rats treated with 100 mg/kg Ab14 at the 170-210 minute time points (p<0.01 at 190 minutes and p<0.05 for the 170-180 and 200-210 minute time points), and for the HFD rats treated with metformin at the 170-210 minute time points (p<0.01 at 180 and 190 minutes, and p<0.05 at the 170 and 200-210 minute time points).

The GIR means were calculated for each plateau (FIG. 14). GIR was significantly decreased in HFD vehicle group compared to control chow-fed group by 32% (p<0.05) and 17% (p<0.01) during the first and the second steps, respectively. Ab14 at 10, 30, or 100 mg/kg increased GIR (ns) during the first step (by 26, 37, and 29% respectively), and metformin had also an increasing (ns) effect by 11%. During the second step, all treatments increased GIR as compared to the HFD vehicle group:

Ab14

10, 30, or 100 mg/kg by 19% (ns), 36% (p<0.01), and 28% (p<0.05), respectively, and metformin by 27% (p<0.05).

Blood glucose means during the two clamp steps corresponded to a euglycemic state as expected. Although there was a significant difference between control chow-fed and the HFD vehicle group during the first clamp step, the glycaemia remained in a normal range, and the biological state was the same in both groups (FIG. 14).

Plasma insulin was measured during the clamp procedure. As expected the insulin level was similar between all groups at the end of the two clamp steps. During the first clamp step the insulin concentration was approximately 140 μU/mL, and was a physiological level expected during fed conditions. The insulin concentration after the second clamp step was approximately 490 μU/mL, which was a pharmacological level (FIG. 11, upper panel).

C-peptide was also measured during the clamp procedure (FIG. 11, lower right panel). During euglycemic conditions the insulin secretion by beta cells was inhibited, and the plasma C-peptide levels were therefore low and not interpretable.

³H-glucose was infused with insulin during the clamp procedure in all HFD groups (not in the control chow-fed group). The whole body glucose fluxes were then calculated. During the first clamp step, the glucose turn over (GTO) was similar in all groups, excluding that Ab14 at 30 mg/kg tended to increase GTO as compared to the HFD vehicle group (17%, ns). Glycolysis and glycogen synthesis tended to increase, 15 and 16%, respectively (ns,FIG. 15), following treatment with Ab14 at 30 mg/kg.

During the second clamp step, GTO, glycolysis, and glycogen synthesis were similar in all treated groups, with a slight increase of glycogen synthesis observed in theAb14 30 mg/kg treated group when compared to HFD vehicle group (by 10%, ns). Ab14 at 30 mg/kg (p<0.05) and 100 mg/kg (ns) completely inhibited HGP, as did metformin (ns), and Ab14 treatment at 10 mg/kg decreased HGP by 78% (ns,FIG. 16).

Conclusion.

Treatment with Ab14 decreased HOMA-IR by decreasing fasting blood glucose as well as plasma insulin levels in the HFD rat model. Furthermore, liver insulin sensitivity was markedly improved by Ab14, whereas an effect on whole body peripheral insulin sensitivity was not clearly observed. As expected, metformin treatment also improved liver insulin sensitivity.

Example 3

This example assesses the effect of Ab14 on glucose metabolism and on glycemic control in a model a rat model of diabetes, the Zucker diabetic fatty (ZDF) rat. The effects of chronic administration of Ab14 on glucose control was evaluated in ZDF rats that were progressing from a prediabetic (hyperinsulinemic, normoglycemic) state to an overtly diabetic (hypoinsulinemic, hyperglycemic) state. These animals develop prediabetes, characterized by marked hyperinsulinemia to compensate for their developing insulin resistance, but with little to no hyperglycemia, by seven weeks of age. This rapidly progresses to overt diabetes, characterized by hypoinsulinemia, as a result of pancreatic beta cell failure, and marked hyperglycemia by 10-12 weeks of age. The antibody used in this example consisted of the light and heavy polypeptide chains of SEQ ID NOs 132 and 134.

Methods

81 ZDF fa/fa rats (Charles River Laboratories, France) and 10 lean ZDF ?/+rats (controls) were housed in ventilated and enriched housing cages in groups of 1-2 animals on a normal 12 hours light cycle (at 08:00 pm lights off), 22±2° C. and 50±10% relative humidity. The rats were 7 weeks of age at delivery and were acclimated for one week prior to study commencement. The rats were fed the standard diet for ZDF rats (Purina 5008, Charles River) and tap water were provided ad libitum. All animals were monitored at least once daily for any signs of ill health, adverse reactions to treatment, or morbidity throughout the study.

Eight week old male ZDF fa/fa rats were hyperinsulinemic and mildly diabetic. Due to the variability of the blood glucose and insulin levels at this state, the ZDF rats were screened and selected according to their HOMA-IR.

For the groups treated with AB14, the antibody was administered once weekly via the caudal vein (i.v., 5 mL/kg) on

days

1, 8, 15, and 22 at twodifferent doses 20 mg/kg/week (groups 3 and 7) or 60 mg/kg/week (groups 4 and 8). All other groups were treated once weekly with vehicle 1 (i.v., 5 mL/kg). The intravenous treatments were performed in the morning on

day

1, 8, 15, and 22 while under isoflurane anaesthesia. The volume of administration was individually adapted according to the most recent body weight.

Metformin (Met) and pioglitazone (PIO) were administered once daily for 28 days, via per os route (p.o., 5 mL/kg) between 8:00 and 10:00 am, except that on the day of OGTT or after intravenous treatments, some per os treatments were completed after 10:00 am.

Group

5, 7, and 8 were treated with 200 mg/kg/day metformin, andgroup 6 was treated with 10 mg/kg/day pioglitazone. All other groups were treated daily with vehicle 2 (p.o., 5 mL/kg) for 28 days. The most recent body weight was used to calculate the average volume of administration in each group.

The test groups are summarized in the following Table. Met: metformin.


		Treat-
Groups	n	ment	dose	route	frequency

Group
1	10	Vehicle	—	i.v.	Day	1, 8, 15, 22
(Lean		Vehicle	—	p.o.	Daily
Zucker rats)
Group 2	10	Vehicle	—	i.v.	Day	1, 8, 15, 22
(ZDF rats)		Vehicle	—	p.o.	Daily
Group
3	10	AB14	20 mg/kg/week	i.v.	Day	1, 8, 15, 22
(ZDF rats)		Vehicle	—	p.o.	Daily
Group
4	10	AB14	60 mg/kg/week	i.v.	Day	1, 8, 15, 22
(ZDF rats)		Vehicle	—	p.o.	Daily
Group
5	10	Vehicle	—	i.v.	Day	1, 8, 15, 22
(ZDF rats)		Met	200 mg/kg/day	p.o.	Daily
Group
6	10	Vehicle	—	i.v.	Day	1, 8, 15, 22
(ZDF rats)		PIO	10 mg/kg/day	p.o.	Daily
Group
7	10	AB14	20 mg/kg/week	i.v.	Day	1, 8, 15, 22
(ZDF rats)		Met	200 mg/kg/day	p.o.	Daily
Group
8	10	AB14	60 mg/kg/week	i.v.	Day	1, 8, 15, 22
(ZDF rats)		Met	200 mg/kg/day	p.o.	Daily

After 1 week of acclimation period, all rats were weighted and fasted for 6 hours (from ˜8:00 am to ˜2:00 pm). At ˜2:00 pm, blood was collected (˜150 μL, EDTA) from the tip of the tail. Blood glucose (glucometer) and plasma insulin (ELISA) were measured, and the HOMA-IR (insulin resistance index) was calculated. The 11 ZDF fa/fa rats presenting extreme HOMA-IR values were excluded from the study but were kept housed for 28 days for plasma collection at the end of the study. Then the 70 remaining rats were randomly allocated to 7 treatment groups according to their HOMA-IR and body weight. The lean rats were all kept ingroup 1 and treated with vehicle only.

Body weight was measured twice a week during the 4 weeks of treatment.

Food consumption was measured just before the screening procedure and then twice weekly over 24 h during the first three weeks of treatment. Food consumption was measured once weekly during the week of OGTT (week 4).

A fasting (6 hours onday 0 from 8:00 am to 02:00 pm, and overnight on

day

12, 19 and 26 from ˜6:00 pm to ˜8:00 am) was performed before each blood collection. Blood was collected at ˜2:00 pm, from the tip of the tail, on day 0 (before screening, 150 μL, potassium EDTA), and at ˜8:00 am prior to dosing on days 12 (110 μL, potassium EDTA), 19 (150 μL, potassium EDTA) and 26, (110 μL, potassium EDTA).

Fasting blood glucose (glucometers) was measured on

days

0, 12, 19, and 26. Fasting plasma insulin, peptide-C(ELISA method), free fatty acids, triglycerides, total cholesterol (colorimetric method), and HDL-cholesterol (phosphotungstate precipitation, colorimetric method) were measured onday 0, and prior to dosing on

days

12, 19, andday 26. Non HDL-cholesterol was calculated as total cholesterol—HDL-cholesterol. Fructosamine was measured on

days

0, 19 and 28. HbA1c (DCA 2000) was measured on

days

0 and 28.

The oral glucose tolerance test (OGTT) was performed as follows. Onday 25, rats were fasted at ˜6:00 pm and an oral glucose tolerance test was performed the day after (on day 26). At ˜8:00 am (T-60) a blood collection (110 μL, EDTA) was performed for biochemical parameters measurements. One hour after (˜09:00 am), an oral glucose bolus (1.5 g/kg) was administered (TO). Blood glucose was measured (glucometer or colorimetric method in case of high glycaemia) on T-60, TO, T15, T30, T60, T90, T120, and T180 minutes. Area under the curve (AUC) was calculated based on the blood glucose values measured at TO. Plasma insulin and C-peptide were measured (ELISA method) on T-60, T15 (˜40 μL of blood, EDTA), and T30 minutes (˜40 μL of blood, EDTA).

After 2 hours of food restriction (from 8:00 am to 10:00 am) 80 rats from the study were anesthetized onday 28 with isoflurane. Blood was collected (˜3000 μL from abdominal vein, on K2-EDTA) for the determination of AB14 plasma concentrations. Plasma (3 aliquots of ˜2004) was kept at −80° C. until testing. The pancreas tissue was then excised. Rats were euthanized by incision of the abdominal vein and aorta.

Each pancreas was divided into 2 parts (longitudinal cut). One piece was fixed in 10% formalin solution for histopathological processing. The other piece of pancreas was flash frozen and kept at −80° C. for determination of insulin and proinsulin levels.

The 11 ZDF fa/fa rats excluded from the study were sacrificed after 28 housing days. They were anesthetized with isoflurane. Blood was collected from abdominal vein (maximum volume on potassium EDTA). Plasma samples (2 aliquots of 1 mL each) were frozen for further testing. Rats were euthanized by incision of the abdominal vein and aorta.

Each pancreas sample was homogenized in an acid buffer, and insulin and proinsulin content were measured using ELISA kits in the following groups:

Group 1: Lean rats+vehicles (n=10)

Group 2: ZDF rats+vehicles (n=10)

Group 4: ZDF rats+AB14 60 mg/kg/week (n=10)

Group 5: ZDF rats+metformin 200 mg/kg/day (n=10)

Group 8: ZDF rats+AB14 60 mg/kg/week+metformin 200 mg/kg/day (n=10)

Each pancreas sample was fixed in 4% formalin during 24-48 hours maximum; the volume of formalin was 5-10 times higher than the sample volume to assure appropriate fixation. After 48 h, the samples were placed in 70% ethanol. Samples were then included in paraffin for histological process in the following groups:

Group 1: Lean rats+vehicles (n=10)

Group 2: ZDF rats+vehicles (n=10)

Group 4: ZDF rats+AB14 60 mg/kg/week (n=10)

Group 5: ZDF rats+metformin 200 mg/kg/day (n=10)

Group 8: ZDF rats+AB14 60 mg/kg/week+metformin 200 mg/kg/day (n=10)

After delineating the islets of Langerhans, surface and intensity of the insulin labelling were quantified by image analysis of the labeled (brown) and non-labeled (blue) areas.

The means of vehicle ZDF rats and lean rats were compared using a student test when the Fisher test did not show significant differences in variances. If not, the non-parametric Mann Whitney test was used.

The means of the treated ZDF rats were compared to the vehicle ZDF rats using a 1-way ANOVA+Dunnett's post-test. If the Bartlett's test showed significant differences in variances, the non-parametric Kruskall Wallis+Dunn's post-test was used.

The means ofAB14 20 mg/kg alone was compared tometformin 200 mg/kg alone or in combination withAB14 20 mg/kg using a 1-way ANOVA+Newman-Keuls post-test.

The means ofAB14 60 mg/kg alone was compared tometformin 200 mg/kg alone or in combination withAB14 60 mg/kg using a 1-way ANOVA+Newman-Keuls post-test.

The curves were analyzed using a 2-way ANOVA+Bonferroni's post-test.

Rats were excluded from analysis if they were an outlier in all or almost all parameters. This resulted in exclusion of four rats, each from a different group.

Results

As expected in 8-week old ZDF rats, HOMA-IR was strongly increased as compared with lean rats (˜111 vs 3.5,FIG. 18A). ZDF rats were mildly hyperglycemic (˜180 vs 113 mg/dL,FIG. 18B) and hyperinsulinemic (˜250 vs 12.6 μU/mL,FIG. 18C). Body weight was slightly increased in ZDF rats (FIG. 18D).

Compared with lean rats, the body weights of ZDF rats remained higher over the entire treatment period (FIG. 19A), while body weight gain was similar between lean and ZDF rats (FIG. 19B).

Pioglitazone significantly increased body weight compared to the ZDF vehicle rats, from 8 days of treatment and body weight gain was 3-fold higher at the end of the treatment (FIGS. 19A and B). All other drug treatments had no significant effect on body weight compared to vehicle ZDF rats.AB14 60 mg/kg+metformin 200 mg/kg combination significantly increased body weight gain from 22 days of treatment (FIG. 19B).

Food intake was ˜2-fold increased in vehicle ZDF rats compared to the lean rats (significant on day 13). Rats treated with pioglitazone showed a trend to higher food consumption as compared with ZDF vehicle rats (significant on

day

15, 20 and 22,FIG. 20A). Cumulative food intake was increased by 94% in the vehicle ZDF group as compared with the lean group (p<0.01), and by 14% in pioglitazone group as compared with the vehicle ZDF group (NS,FIG. 20B). Other treatments had no effect on food intake as compared with vehicle-treated ZDF rats.

Fasting blood glucose remained in a normal range during the 26 days of treatment in lean rats, either after 6 hours or one night of fasting (FIG. 21A). This was correlated with normal insulin levels (FIG. 21B). In vehicle ZDF rats, overnight fasting blood glucose reached 362±32 mg/dL on day 12 (at about 10 weeks of age) and remained significantly higher than lean rats until the end of the treatment (FIG. 21A). This was correlated with decreasing plasma insulin levels (49.2±6.7, 41.2±4.8, and 36.6±2.7 μU/mL−p<0.001 vs. lean rats,FIG. 21B) and with decreasing plasma C-peptide levels (2813±249, 2472±195, 2156±165 pM−<p<0.001 vs lean rats (FIG. 21D) measured on

days

12, 19, and 26. The evolution of the HOMA-IR in lean and vehicle ZDF rats reflected the change in blood glucose and plasma insulin levels (FIG. 21C).

Pioglitazone significantly decreased overnight fasting blood glucose levels to a normal level from 12 days of treatment (p<0.001,FIG. 21A). At both doses AB14 decreased by about 15% the blood glucose after 12, 19 or 26 days of treatment (n.s,FIG. 21A). Compared to AB14,metformin 200 mg/kg had similar effect onday 12, but this effect was not observed at

day

19 and 26. Compared with ZDF rats treated with vehicle,AB14 20 mg/kg+metformin combination slightly reduced blood glucose on day 12 (12%, ns), and showed no effect onday 19 and day 26 (FIG. 21A). In contrast, theAB14 60 mg/kg combination with metformin significantly reduced blood glucose levels day 12 (38%, p<0.01 vs vehicle treated ZDF rats). Although not statistically significant, the blood reduction was still observed onday 19 and 26 (22% and 27% respectively,FIG. 21A).

Pioglitazone seemed to have no protective effect on insulin secretion as it showed no effect on plasma insulin and C-peptide levels on

days

12, 19 and 26 compared to vehicle ZDF rats (FIGS. 21B and D). Hence the reduction in blood glucose levels was related to the insulin sensitizing effect of pioglitazone, which reduced HOMA-IR by 67%, 62% and 54%, on

days

12, 19 and 26 respectively, as compared with vehicle ZDF rats (FIG. 21C).

Compared with vehicle-treated ZDF rats,AB14 20 mg/kg did not change plasma insulin and C-peptide levels, as well as HOMA-IR on

days

12, 19 and 26 (FIG. 21B-D). Meanwhile,AB14 60 mg/kg increased plasma insulin levels on

days

12, 19 and 26 by 74%, 21% and 19%, respectively (ns vs. vehicle ZDF rats,FIG. 21B).

AB14 60 mg/kg increased plasma C-peptide levels onday 12 by 10% (ns), and had no effect ondays 19 and 26 (FIG. 21D).

Compared with vehicle-treated ZDF rats, metformin increased plasma insulin levels on

days

12, 19 and 26 by 79%, 55% and 48%, respectively (ns,FIG. 21B). Metformin increased plasma C-peptide levels on

days

12, 19 and 26 by 23%, 21%, and 9%, (NS vs ZDF rats treated with vehicle,FIG. 21D).

Compared with vehicle-treated ZDF rats,AB14 20 mg/kg+metformin combination increased plasma insulin levels on

days

12, 19 and 26 by 2-fold (NS,FIG. 21B).AB14 20 mg/kg+metformin combination increased plasma C-peptide levels on

days

12, 19 and 26 by 21%, 23% and 25%, respectively (NS,FIG. 21D).

Compared with vehicle-treated ZDF rats,AB14 60 mg/kg+metformin combination significantly increased plasma insulin levels on

days

12, 19 and 26 by a factor 2.5, 2.3 and 2.7 respectively.AB14 60 mg/kg+metformin combination significantly increased plasma C-peptide levels fromday 12 by 45% (day 12), 48% (day 19), and 52% (day 26) (p<0.05 vs vehicle-treated ZDF rats,FIG. 21D).

In this model where insulin secretion was reduced over time, the increase in HOMA-IR was reflecting an improvement of insulin secretion. Thus an increase of HOMA-IR was observed in metformin alone or in combination with the AB14 treated groups, as compared with vehicle ZDF groups, and this increase was maintained on

days

12, 19 and 26 (FIG. 21C).

Compared with lean rats, fructosamine was significantly higher (66%) in 8-week old vehicle-treated ZDF rats (208±6 vs 144±2 μM, p<0.001). Fructosamine levels remained in a similar range in lean rats during the treatment period, but increased in vehicle ZDF rats after 19 (253±5 μM, p<0.001) and 28 (234±6 μM, p<0.001) days of treatment (FIG. 22). As expected, pioglitazone significantly reduced fructosamine levels from day 19 (30% on

day

19 and 25% onday 28, p<0.001).

AB14

20 and 60 mg/kg had no effect on fructosamine levels. Compared with vehicle-treated rats, metformin showed a trend towards lower fructosamine levels only on day 19 (6%, ns). Compared with vehicle-treated ZDF rats,AB14 20 mg/kg+metformin combination showed a non significant trend towards lower fructosamine levels on

days

19 and 28 by 10% and 8%, respectively. As well theAB14 60 mg/kg+metformin combination showed a non significant trend towards lower fructosamine levels (9%) on day 28 (FIG. 22).

Compared with lean rats, HbA1c was higher in 8-week old ZDF rats (4.3±0.1% vs3.1±0.04%), although these values were in a normal range.

In 12-week old ZDF rats, HbA1c reached a pathological value of 8.8±0.2% onday 28, (p<0.001 ZDF vs lean rats,FIG. 23). Compared with vehicle-treated ZDF rats,

AB14

20 and 60 mg/kg had no effect on HbA1c levels after 28 days of treatment. After 28 days of treatment, pioglitazone and metformin significantly decreased HbA1c by 44 and 15%, respectively (FIG. 23). The combination of metformin withAB14 20 mg/kg and with 60 mg/kg significantly reduced HbA1c by 11 and 19% respectively (FIG. 23).

Compared with lean rats, plasma triglycerides levels were strongly increased in 8-week old ZDF rats (˜8 mM vs ˜0.7 mM,FIG. 24A).

Compared with vehicle, pioglitazone strongly decreased plasma triglycerides levels from the 12th day of treatment.AB14 20 mg/kg slightly decreased plasma triglycerides levels ondays 12 and 19 (by 15% and 7% respectively, ns), and had no effect onday 26.AB14 60 mg/kg slightly decreased plasma triglycerides levels on

days

12, 19 and 26 by 14%, 9% and 12%, respectively (ns). Metformin increased plasma triglycerides levels on

days

12, 19 and 26 by 26%, 40%, and 49%, respectively (significant from day 19). The metformin+AB14 20 mg/kg combination showed a trend towards higher plasma triglycerides ondays 19 and 26 (by 13% and 23%, respectively, ns) as compared with the vehicle ZDF group. Metformin+AB14 60 mg/kg combination showed a trend towards higher plasma triglycerides on

days

12, 19 and 26 (by 9%, 48% and 43% respectively, significant fromday 19,FIG. 24A).

After 6 hours of fasting, plasma free fatty acids levels were higher in 8-week old ZDF rats than in lean rats (˜0.85 mM vs. ˜0.59 mM). After an overnight fasting (maximal lipolytic conditions), free fatty acids levels were similar (˜1.3 mM) at 10 and 11 weeks. At 12 weeks of treatment, the lipolytic capacity of ZDF rats was decreased, as shown by lower free fatty acids levels, as compared with lean rats (1.05±0.06 vs 1.38±0.03 mM,FIG. 24B). Compared to vehicle ZDF rats, rats treated with pioglitazone showed lower plasma free fatty acids levels by 35% on day 12 (p<0.001), 17% on day 19 (ns) and 30% on day 26 (p<0.05).

AB14

20 and 60 mg/kg had no effect. Metformin increased free fatty acids levels by 14%, 25% and 8% on

days

12, 19 and 26 but not significantly when compared to vehicle ZDF rats. The effect of metformin alone or in combination withAB14 20 mg/kg was similar. On the other hand, when combined withAB14 60 mg/kg, metformin showed an increasing effect only onday 19 when compared with the vehicle ZDF group (by 20%, NS,FIG. 24B).

Compared with lean rats, plasma total cholesterol and HDL-cholesterol levels were higher in 8-week old ZDF rats and gradually increased over the 4 following weeks (FIG. 25A-B,FIG. 26A-B). Plasma non-HDL cholesterol levels were similar in lean and ZDF rats at 8 weeks of age, but increased over time in ZDF rats vs lean rats from 10 weeks (FIG. 25C andFIG. 26C). As there was significant difference in total cholesterol and HDL-cholesterol between the ZDF groups at day 0 (FIG. 25), the results were expressed in relative values from day 0 (FIG. 26). As shown inFIG. 26A, pioglitazone tended to prevent the increase in plasma total cholesterol levels overtime.AB14 20 mg/kg and metformin had no effect.AB14 60 mg/kg increased total cholesterol by 8%, 14%, and 15% on

days

12, 19, and 26 respectively compared to the vehicle ZDF group. When combined with metformin,AB14 20 mg/kg increased total cholesterol by 15% and 10% on

days

12 and 26 respectively, whereasAB14 60 mg/kg increased total cholesterol by 24%, 21% and 13% on

days

12, 19 and 26 respectively compared to the vehicle ZDF group. Compared with the vehicle, pioglitazone increased plasma HDL-cholesterol by 38%, 17% and 19% on

days

12, 19 and 26 respectively. Metformin alone had no effect.AB14 20 mg/kg and 60 mg/kg, alone or in combination with metformin increased plasma HDL-cholesterol levels by 11 to 22% after 12, 19 and 26 days of treatment. Plasma non HDL-cholesterol levels (FIG. 26C) were similar in all ZDF groups after 12 days of treatment. Compared with vehicle,AB14 20 mg/kg,AB14 60 mg/kg, metformin, alone or in combination with AB14 had no effect on non HDL-cholesterol levels. Only pioglitazone significantly decreased plasma non HDL-cholesterol levels by 49% and 47% onday 19 andday 26, respectively.

An oral glucose tolerance test was performed after 26 days of treatment. As compared with vehicle-treated lean rats, blood glucose levels were expectedly higher in vehicle-treated ZDF rats before and after glucose load (FIG. 27A). Compared with vehicle-treated ZDF rats, only pioglitazone-treated ZDF rats showed significantly reduced blood glucose levels at all time points. Compared with vehicle,AB14 60 mg/kg and metformin combination tended to reduced blood glucose levels at t-60 minutes (FIG. 27A), while other drug treatments showed no significant effect. Compared with lean rats, blood glucose area under the curve (AUC) was significantly increased by 3.7-fold in vehicle-treated ZDF rats. Compared with vehicle-treated ZDF rats, rats treated with pioglitazone showed a significant 54% reduction in blood glucose AUC.AB14 20 mg/kg,AB14 60 mg/kg and metformin alone or in combination with 20 mg/kg or 60 mg/kg AB14 showed a non significant reduction on AUC (7%, 11%, 6%, 7% and 17%, respectively). TheAB14 60 mg/kg+metformin combination was slightly more effective in reducing AUC when compared with AB14 or metformin alone (FIG. 27B).

Plasma insulin and C-peptide levels were measured at 15 and 30 minutes after the glucose load. The concentration versus time profiles were similar for both insulin and C-peptide. Insulin and C-peptide levels were similar between the vehicle,AB14 20 mg/kg andAB14 60 mg/kg treated groups, slightly increased in the metformin and pioglitazone treated groups, and more increased in a dose-dependent manner in groups treated withAB14 20 mg/kg andAB14 60 mg/kg combined with metformin (FIG. 28A-B). The capacity of insulin or C-peptide secretion in response to the glucose charge was evaluated by expressing the results in relative values calculated from T-60 minutes. As expected, vehicle ZDF rats had significantly lost their capacity to secrete insulin and C-peptide in response to the glucose charge as compared with lean rats (FIG. 29A-B). When compared with vehicle-treated ZDF rats, rats treated with pioglitazone had increased insulin secretion by 20% (p<0.05) and 5% at time T15 and T30 respectively. All other treatments had no effect on insulin secretion at time T15. Metformin decreased insulin secretion at time T30 by 26% (p<0.01).AB14 20 mg/kg alone had no effect at time T30, and showed a trend to decrease insulin secretion (by 14%, ns) in combination with metformin.AB14 60 mg/kg alone or combined with metformin showed a trend to decrease insulin secretion by 19% and 18%, respectively (FIG. 29A). As compared to lean rats, C-peptide secretion in response to glucose load (FIG. 29B) was significantly reduced in vehicle-treated ZDF rats by ˜40% at time T15 and T30. When compared to vehicle-treated ZDF rats, only pioglitazone significantly increased C-peptide secretion at time T15 and T30 by 21% and 22% respectively.

As expected, pancreas proinsulin (FIG. 30A) and insulin (FIG. 30B) levels were significantly lower in 12-week old ZDF rats when compared with lean rats.AB14 60 mg/kg and metformin completely prevented reductions in proinsulin andAB14 60 mg/kg combined with metformin significantly increased proinsulin levels (p<0.05 vs. vehicle) (FIG. 30A).AB14 60 mg/kg slightly increased pancreatic insulin levels, and metformin orAB14 60 mg/kg combined with metformin significantly increased insulin levels (p<0.05 vs. vehicle) (FIG. 30B). Compared with lean rats proinsulin/insulin ratio was significantly increased in ZDF rats, while no change was observed with drug treatments (FIG. 30C).

No evidence for toxicity was reported in this study focusing on microscopic changes in pancreas of lean or ZDF rats treated with either vehicle, AB14, metformin or AB14/metformin combination.

A higher incidence and severity of focal to multifocal large/giant islet(s)—corresponding to islet hyperplasia—and islet fibrosis were noted in ZDF rats given vehicle control (Table 1 and Table 2).

TABLE 1

Incidence of Histophathological Observations. All treatment
groups were ZDF rats except where indicated otherwise.

		Vehicle		Ab14	60		Ab14 60
	Sever-	only (ZDF	Vehicle	mg/kg/		mg/kg/
Observation	ity	Lean rats)	only	wk.	Met.	wk. + Met.

Islet Fibrosis	0	6	—	—	—	—
	1	2	—	4	3	7
	2	2	2	3	7	1
	3	—	7	2	—	1
Brown	0	4	—	—	—	—
pigment-	1	4	6	7	7	8
laden	2	2	3	2	3	1
	3	—	—	—	—	—
Mononuclear	0	4	—	—	—	—
cell	1	5	7	9	10	8
infiltration	2	1	2	—	—	1
	3	—	—	—	—	—
Islet Cell	0	10	—	—	1	1
Vacuolation	1	—	1	6	4	7
	2	—	8	3	5	1
	3	—	—	—	—	—
Islet cell	0	9	3	5	7	3
single cell	1	1	6	4	3	6
necrosis	2	—	—	—	—	—
	3	—	—	—	—	—
Islet cell	0	10	1	5	2	1
anisocytosis/	1	8	4	8	8
anisokaryosis	2	—	—	—	—	—
	3	—	—	—	—	—
Islet	0	8	6	5	6	7
granulocytic	1	2	3	4	4	2
infiltration	2	—	—	—	—	—
	3	—	—	—	—	—
Islet	0	8	7	7	8	7
hemorrhage	1	2	2	2	2	2
	2	—	—	—	—	—
	3	—	—	—	—	—
Islet cell	0	9	6	7	5	8
mitosis	1	1	3	2	5	1
	2	—	—	—	—	—
	3	—	—	—	—	—
Interstitial	0	2	2	3	4	2
hemorrhage	1	7	5	4	6	5
	2	1	2	2	—	2
	3	—	—	—	—	—
Interstitial	0	5	5	8	9	8
fibroplasia/	1	5	3	1	1	1
fibrosis	2	—	1	—	—	—
	3	—	—	—	—	—
Arteries,	0	10	6	6	9	7
medial/	1	—	2	3	1	2
intimal	2	—	1	—	—	—
hypertrophy	3	—	—	—	—	—
Pancreatic	0	9	3	9	5	8
ductule	1	1	6	—	5	1
hyperplasia	2	—	—	—	—	—
	3	—	—	—	—	—
Exocrine cell	0	8	8	8	10	7
vacuolation	1	2	1	1	—	2
	2	—	—	—	—	—
	3	—	—	—	—	—
Eosinophilic	0	9	6	8	10	5
granulocyte	1	1	3	1	—	3
infiltration	2	—	—	—	—	1
	3	—	—	—	—	—
Large/Giant	0	5	1	—	1	—
Islets	1	5	2	4	1	1
	2	—	5	3	3	5
	3	—	1	2	5	3

Met.: metformin 200 mg/kg/day.

TABLE 2

Histopathological analysis: group mean scores of elementary
findings. All treatment groups were ZDF rats except where indicated otherwise.

		Vehicle only				Ab14 60
		(ZDF Lean		Ab14	60		mg/kg/wk. +
Organ	Findings	rats)	Vehicle only	mg/kg/wk.	Met.	Met.

Fat tissue	Fat tissue necrosis	0	0	0	0	0
Lymph node	Brown pigment-	0	1	0	0	1
	laden macrophages
	Eosinophilic	1	1	0.5	0.5	0
	granulocyte
	infiltration
	Erythrophagocytosis-	1	1.5	1	0	1
	sinusoidal
	Pink protein rich	0	0	0	0	0
	sinusoidal lymph
	Sinusoidal
	1	0.5	0	0	1
	hemorrhage
	Sinusoidal mast
	0	0	0	0	0
	cells
Pancreas	Arteries, medial/	0	0	0	0	0
	intimal hypertrophy
	Brown pigment-	1	1	1	1	1
	laden macrophages
	Eosinophilic	0	0	0	0	0.5
	granulocyte
	infiltration
	Exocrine cell	0	0	0	0	0
	vacuolation
	Interstitial	0.5	0	0	0	0
	fibroplasia/fibrosis
	Interstitial
	1	1	1	1	1
	hemorrhage
	Islet cell
	0	1	0.5	1	1
	anisocytosis/
	anisokaryosis
	Islet cell mitosis	0	0	0	0.5	0
	Islet cellsingle cell	0	1	0.5	0	1
	necrosis
	Islet cell
	0	2	1	1.5	1
	vacuolation
	Islet fibrosis
	0	3	2	2	1
	Islet granulocytic	0	0	0	0	0
	infiltration
	Islet hemorrhage
	0	0	0	0	0
	Large/giant islet(s)	0.5	2	1.5	2.5	2
	Mononuclear cell	1	1	1	1	1
	infiltration
	Pancreatic ductule	0	1	0	0.5	0
	hyperplasia

Met.:metformin 200 mg/kg/day.

Compared with lean rats, vehicle treated ZDF rats had slight islet cell vacuolation and increased incidence and severity of islet fibrosis. There was a consistent trend towards a decrease in vacuolization and islet fibrosis severity with all drug treatments,AB14 60 mg/kg, metformin, or combined metformin with AB14. These effects were more pronounced when the AB14 and metformin were combined.

As expected, pancreas insulin measured by immunohistochemistry showed a reduction in insulin labelling in ZDF rats. As observed from the insulin content measurement (FIG. 30B), drug treatments slightly prevented this insulin labeling reduction with a better effect when AB14 and metformin were combined (FIG. 31).

DISCUSSION

Eight week old ZDF rats that were markedly insulin-resistant and severely hyperinsulinemic but only mildly hyperglycemic were given Ab14 intravenously at doses of 0 mg/kg (vehicle), 20 mg/kg, or 60 mg/kg once a week for 4 weeks. During this time-period, the vehicle-treated controls progressed to overt diabetes and were severely hypoinsulinemic and markedly hyperglycemic byday 12 of the study, consistent with complete pancreatic beta cell failure by 10 weeks of age. This was confirmed at the end of the study by direct measurement of the pancreatic insulin and proinsulin levels, both of which were substantially reduced, and through immunohistochemical assessment of pancreatic insulin labeling, which was also dramatically lowered. In addition, histological analysis conducted at the end of the study also demonstrated an increased incidence and severity of islet vacuolation, islet hyperplasia (large/giant islet(s)), and islet fibrosis in these animals, consistent with diabetic pancreatic islet pathology.

By contrast, the rise in fasting blood glucose in the vehicle-treated controls onday 12 of the study was partially prevented by both doses of Ab14, and this partial prevention was also observed on

days

19 and 26 of the study. In addition, the high dose of Ab14 also partially prevented the reduction in plasma insulin and C-peptide levels observed in the vehicle-treated controls onday 12 of the study. This partial prevention was also observed but to a lesser extent on

days

19 and 26 of the study, indicative of a modest delay in disease progression (pancreatic beta cell failure) and suggestive of partial pancreatic beta cell protection by the compound. Indeed, the high dose of Ab14 also completely prevented the reduction in pancreatic proinsulin levels observed in the vehicle-treated controls when pancreas tissue was obtained at the end of the study (day 28) and also partially prevented the reduction in pancreatic insulin levels when measured either directly or through immunohistochemical analyses. Furthermore, Ab14 consistently decreased the islet vacuolation, islet fibrosis, and islet hyperplasia noted in the vehicle-treated animals upon histological evaluation at the end of the study, further indicating a favorable impact on diabetic pancreatic islet pathology.

As demonstrated in Example 2, when compared to the vehicle-treated control group, Ab14 had no effect on food consumption or body weight, indicating that effects of Ab14 on the parameters evaluated above were not a result of caloric restriction or weight loss.