Contributo dell'iperglicemia alle complicanze diabetiche nel diabete di tipo 2 obeso

Introduction

Diabetes mellitus is one of the most common metabolic disorders, and the number of diabetic

patients has been increasing worldwide. The International Diabetes Federation (IDF) reported

that the 366 million patients with diabetes in 2011 will increase to 552 million by 2030

[ ]. Even worse, more than half of all diabetics

have one or more diabetic microvascular complications, such as diabetic nephropathy (DN),

diabetic peripheral neuropathy (DPN) or diabetic retinopathy (DR), all of which seriously

threaten quality of life. To clarify the pathogenetic mechanism of human diabetes and its

complications, and to develop drugs for diabetes, experimental diabetic animal models play

critical roles.

The spontaneously diabetic torii (SDT) fatty rat is a new model for obese type 2 diabetes

established by introducing the fa allele of the Zucker fatty rat into the

original (non-obese) SDT rat genome to defect leptin receptor signaling. Since SDT fatty

rats develop marked hyperglycemia with hyperinsulinemia, hyperlipidemia and hypertension

shortly after weaning [ , , , , , ], SDT fatty rats develop severe microvascular complications at an

early age [ , , ]. Therefore, this animal model is

useful for investigating diabetic complications and for evaluating new drugs. Previously, we

investigated diabetic microvascular complications in original SDT rats by controlling blood

glucose level with insulin treatment and showed that complications are caused by severe

hyperglycemia [ , ]. However, because of hyperinsulinemia associated with marked insulin

resistance, insulin treatment failed to control blood glucose level in SDT fatty rats

(unpublished data). Therefore in the present study, we investigated diabetic complications

by controlling blood glucose level with daily phlorizin (PZN) treatment. Phlorizin is a

natural compound originally isolated from apple trees [ ]. Its pharmacological mechanism is inhibiting sodium glucose co-transporters

(SGLTs) distributed in the proximal tubule brush border (SGLT2) and gastrointestinal tract

(SGLT1), leading to renal glucosuria and blocking intestinal glucose absorption, both of

which reduce hyperglycemia [ ]. In anticipation of

these mechanisms, we administered phlorizin to SDT fatty rats to control blood glucose level

and studied whether and how hyperglycemia causes diabetic microvascular complications in

this model.

Materials and Methods

### Animals and chemicals

Female SDT fatty rats from our colony were used in the study. At six weeks of age, SDT

fatty rats were divided into two groups (n=8); a phlorizin treated group and a vehicle

treated group. Age-matched female Sprague-Dawley (SD) rats (Charles River Laboratories

Japan, Yokohama, Japan) were used as control animals (n=8). All animal protocols used in

the study were in strict compliance with our own Laboratory Guidelines for Animal

Experimentation. Animals were housed in a climate-controlled room (temperature 23 ± 3°C,

humidity 55 ± 15%, 12 h lighting cycle) and allowed free access to basal diet (CRF-1,

Oriental Yeast, Tokyo, Japan) and water.

Phlorizin (Kanto chemical, Tokyo, Japan) was suspended in 20% propylene glycol and

injected subcutaneously once daily (100 mg/kg/day) to animals in the phlorizin treated

group for 23 weeks. Twenty % propylene glycol was administered to animals in the vehicle

treated group and control SD rats.

### Biochemical parameters

During the experimental period, biochemical parameters were monitored. Blood samples were

collected from the tail vein under fed condition. Glucose, HbA1c, triglycerides (TG), free

fatty acid (FFA) and total cholesterol (TC) were measured using commercial kits (Roche

Diagnostics, Basel, Switzerland) and an automatic analyzer (Hitachi 7180; Hitachi

High-Technologies, Tokyo, Japan). Commercial ELISA kits were used to measure plasma

insulin (Rat Insulin ELISA Kit; Morinaga Institute of Biological Science, Yokohama,

Japan).

### Evaluation of diabetic nephropathy

Urine samples were collected for 24 h using metabolic cages. During urine sampling,

animals were not limited to access to diet and water. Urinary glucose level was measured

as described above. Both urinary and plasma creatinine levels were measured with an

automatic analyzer to calculate creatinine clearance. Urinary albumin (Nephrat II;

Exocell, Philadelphia, PA, USA) and urinary 8-OHdG (New 8-OHdG Check ELISA; Japan

Institute for the Control of Aging, Nikken SEIL, Shizuoka, Japan) were measured using

commercial kits.

### Evaluation of peripheral neuropathy

Nerve conduction velocity was measured in accordance with previously described methods

[ ]. Briefly, the sciatic nerve was stimulated

at the sciatic notch and the Achilles tendon using adequate intensity under 37.5 mg/kg of

sodium pentobarbital (Kanto chemical) and 3 mg/kg of diazepam anesthesia. Action

potentials in the muscle were recorded via PowerLab through a needle electrode. Motor

nerve conduction velocity (MNCV) was calculated from the delta latency between M-wave

peaks divided by the distance of the nerve length measured. Sensory nerve conduction

velocity (SNCV) was also calculated from F-wave peaks.

### Evaluation of retinopathy and cataracts

Electroretinograms (ERGs) were performed as previously described [ ] with slight modifications. Briefly, rats were adapted to darkness

for at least 60 min and anesthetized with an intraperitoneal injection of 37.5 mg/kg

ketamine (Daiichi Sankyo Propharma, Tokyo, Japan) and 4.5 mg/kg xylazine hydrochloride (MP

Biomedicals, Santa Ana, CA, USA). A 40 J xenon lamp was flashed after pupillary mydriasis

with 0.5% tropicamide (Mydrin P; Santen, Osaka, Japan) and potential was recorded via the

PowerLab data acquisition system and software Scope (ADInstruments, Dunedin, New Zealand)

through a corneal contact lens electrode (Mayo, Aichi, Japan). Peak latencies of

oscillatory potentials (OP 1 , OP 2 and OP 3 ) were measured

and data was expressed as the sum of OP 1 to OP 3 (Σ

(OP 1 -OP 3 )).

Cataracts were evaluated after ERG measurements using a slit lamp (SL-14, Kowa, Tokyo,

Japan). Lens opacity was scored using three grades for each eye as follows: 0 (no opacity;

clear lens), 1 (partially clouded lens) and 2 (mature cataract; completely opaque lens).

The average score of both eyes was used as the individual cataract score.

### Histology and immunohistochemistry

At the end of the study, all animals were sacrificed at non-fasted condition by

exsanguination under isoflurane anesthesia and necropsy was performed. Kidneys and eyes

were fixed in 4% paraformaldehyde (PFA) and 4% glutaraldehyde/10% neutral-buffered

formalin, respectively. After resection, tissues were paraffin-embedded using standard

techniques and thin-sectioned (3 to 5 µ m). The sections were stained with

hematoxylin and eosin (HE) for histological evaluation. Intraepidermal nerve fiber density

(IENFD) was measured to evaluate small fiber neuropathy [ ]. The skin of hind limbs was dissected and fixed in 4% PFA, embedded in

paraffin, and sectioned (25 µ m). Nerve fibers were immunostained for

protein gene product 9.5 (PGP 9.5; rabbit polyclonal, 1:500; UltraClone, Isle of Wight,

U.K.) overnight at 4°C and Alexa Fluor ® mouse anti-rabbit IgG antibody 488

(1:1,000, Thermo Fisher Scientific, Waltham, MA, USA) for 30 min at room temperature. Five

fields from each section were randomly selected and Z-stack images were obtained using a

Nikon A1 confocal laser scanning microscope mounted on an inverted microscope (Eclipse

T i , Nikon, Tokyo, Japan). Nerve fibers with branching inside the

epidermis were considered one nerve. IENFDs were expressed as numbers of epidermal nerve

fibers per length of the epidermal basement membrane (fibers/mm).

### Statistical analysis

Results are expressed as the mean ± SD. Statistical analyses of differences between mean

values were performed using an F-test, followed by a Student’s t -test or

Aspin-Welch’s t -test. A Wilcoxon rank-sum test was used for cataract

scores. All statistical analyses were performed using the EXSUS statistical analysis

system for biological experiment data (CAC EXICARE, Tokyo, Japan). Differences were

accepted as significant at P <0.05.

Results

### Hypoglycemic effect of phlorizin on SDT fatty rats

To confirm the dose of phlorizin for the study, 100 mg/kg of phlorizin was subcutaneously

administered once to 18 week-old female SDT fatty rats under non-fasted conditions ( ). Prior to phlorizin treatment, the blood glucose level in SDT fatty rats was 370 ±

49 mg/dl. Six hours after dosing, the blood glucose level in the phlorizin treated group

decreased to an almost normal level (139 ± 32 mg/dl). The hypoglycemic effect of phlorizin

was sustained for 24 h (224 ± 42 mg/dl); although the effect of phlorizin was weakened at

24 h, normal blood glucose levels were expected with repeat dosing. Therefore, a single

dose of 100 mg/kg phlorizin is sufficient to reduce plasma glucose level for 24 h in

female SDT fatty rats.

### Effect of phlorizin on biochemical parameters of SDT fatty rats

Body weights of SDT fatty rats were significantly higher than SD rats in the experiment.

Phlorizin-treated SDT fatty rats were heavier than vehicle-treated SDT fatty rats after 12

weeks ( ). At the end of the study, food consumption of SDT fatty rats was almost twice that

of SD rats; however, there were no differences between the vehicle treated group and

phlorizin treated group ( ). Urinary

glucose considerably increased in SDT fatty rats, and phlorizin treatment significantly

decreased glucose excretion ( ).

After 2 weeks, blood glucose levels of SDT fatty rats increased to a range of 424 mg/dl

to 663 mg/dl ( ). Blood glucose levels were adequately controlled with phlorizin

treatment during the experiment and gradually decreased to near normal range (123 mg/dl to

167 mg/dl). Blood glucose levels of normal SD rats were in the

range of 104 mg/dl to 157 mg/dl. HbA1c levels reflected the change of blood glucose level

accurately ( ). Plasma insulin levels of

SDT fatty rats were significantly higher than SD rats at the beginning of the experiment

(6 weeks of age) and gradually decreased to normal level at 20 weeks. Phlorizin treatment

delayed insulin decreases and insulin was kept high even at the end of study ( ). Plasma TG levels ( ), FFA levels ( ) and TC levels ( ) in SDT fatty rats were higher than SD rats. TG levels and FFA

levels in the phlorizin treated group were significantly higher and TC levels were lower

than the vehicle treated group only at 7 weeks treatment.

### Effect of phlorizin on diabetic nephropathy (DN) of SDT fatty rats

After 20 weeks of treatment with phlorizin, urinary parameters were evaluated to assess

the effects of phlorizin on DN. Urinary albumin excretion (UAE; ) considerably increased in SDT fatty rats. Creatinine clearance ( ) and urinary 8-OHdG ( ) also increased significantly. Creatinine clearance

decreased significantly with phlorizin treatment; however, the effects on UAE and 8-OHdG

were limited.

show the typical

microphotographs of the kidneys of SD rats and SDT fatty rats. Slight glomerulosclerosis

and tubular regeneration, dilation, Armanni-Ebstein changes and inflammatory cell

infiltration in interstitial areas were found in SDT fatty rats. Phlorizin treatment

prevented Armanni-Ebstein changes ( );

however, other renal abnormal findings did not recover with controlled blood glucose

level. Moreover, tubular dilation and inflammatory cell infiltration in the urothelium

were frequently observed in phlorizin treated SDT fatty rats.

### Effect of phlorizin on diabetic peripheral neuropathy (DPN) of SDT fatty rats

At the end of the treatment, sciatic MNCV and SNCV were measured under anesthesia. Both

MNCV and SNCV were significantly delayed in SDT fatty rats compared with non-diabetic SD

rats. The MNCV of SDT fatty rats decreased to 82.0% of that in SD rats and SNCV decreased

to 82.4% ( ). These functional impairments in nerves were corrected to 95.6% and 98.3%

of those in normal rats, respectively, with 23-weeks of phlorizin administration.

To evaluate peripheral nerve density, skin biopsies with quantification of IENFD were

performed. IENFD from the skin of the hind paw foot of SDT fatty rats decreased

significantly at 29 weeks of age (SD rats; 27.1 ± 2.3 fibers/mm, SDT fatty rats; 14.8 ±

3.1 fibers/mm). 23 weeks of phlorizin treatment prevented the decrease of nerve fibers

(23.6 ± 3.2 fibers/mm) ( C–F).

### Effect of phlorizin on diabetic retinopathy (DR) and cataracts of SDT fatty

rats

At 29 weeks of age, SDT fatty rats showed prolongations of peak latencies of oscillatory

potential in ERGs compared with age-matched SD rats. There was a significant prolongation

of peak latency for each individual oscillatory potential (OP 1 -OP 3 ),

as well as that of the summed potential Σ (OP 1 -OP 3 ). Prolongation of

these peak latencies decreased significantly with 23 weeks of phlorizin administration

( ). Hisopathologically, retinal folding was observed in SDT fatty rats as previously

reported [ , ]. Retinal abnormalities were completely prevented with phlorizin ( ).

Cataracts progressed in SDT fatty rats from 8 weeks of age. At 29 weeks of age, all SDT

fatty rats showed mature cataracts via macroscopic observation. Cataracts in phlorizin

treated SDT fatty rats did not progress even at the end of experiments ( ).

Discussion

The SDT fatty rat has been developed as a new type 2 diabetes model with rapidly

progressing diabetic microvascular complications [ , , ]. SDT

fatty rats develop diabetes from 5 weeks of age, and the incidence of diabetes in both male

and female SDT fatty rats is 100%. Previously, we evaluated some hypoglycemic drugs such as

pioglitazone, metformin and DPP IV inhibitor, on male SDT fatty rats [ , ]. Despite that the female SDT

fatty rats also show severe diabetes and its complications, only few experiments have been

reported [ ]. Therefore, in the present study, we

used female SDT fatty rats to investigate the characteristics of diabetic complications in

this animal model by controlling blood glucose level with phlorizin. We have confirmed that

phlorizin treatment sufficiently reduce blood glucose level in male SDT rats (unpublished

data). Phlorizin is a non-selective SGLT inhibitor and recently-launched SGLT2 inhibitors

are treated as anti-hyperglycemic drugs with a novel mechanism of action. Although it is

difficult to distinguish the effect on SGLT1 in gut from that on SGLT2 in kidney, the

changes in body weight, blood glucose levels and HbA1c suggest the usefulness of SGLT

inhibitor phlorizin in the treatment of hyperglycemia. By using phlorizin, primarily

hypoglycemic effects are expected without affecting other biochemical parameters.

Pharmacological effects of phlorizin on diabetic animal models were reported previously. In

concurrence with our result, phlorizin treatment prevented hyperglycemia and preserved

insulin mRNA levels but failed to prevent hypertriglyceridemia in ZDF rats [ ].

Body weights of SDT fatty rats are obviously heavier than normal SD rats because of

hyperphagia due to the lack of leptin signaling [ , ]. With the progress of hyperglycemia, body

weight gain gradually slowed down. Repeated treatment with phlorizin completely prevented

the increase of plasma glucose levels during the experiment. Improvement of hyperglycemia

with phlorizin is considered to influence general condition, leading to further increases in

body weight.

For the kidneys of SDT fatty rats, we previously reported histopathological changes in

glomeruli (glomerulosclerosis, increased mesangial matrix and glomerular hypertrophy, and

nodular lesions) and renal tubules (glycogen deposition (Armanni-Ebstein lesions), fibrosis,

inflammatory cell filtration and tubular dilation) [ , , ]. In the present study, SDT fatty rats

showed significant increases in urinary glucose, UAE, and creatinine clearance. The primary

effect of phlorizin is inducing glucosuria; however, urinary glucose levels in the phlorizin

treated group were lower than the vehicle treated group. Lower glucosuria may reflect

reduced plasma glucose levels in phlorizin treated SDT fatty rats at these time points.

These data are consistent with previous report that hyperglycemia was decreased by phlorizin

in streptozotocin (STZ)-induced diabetes rats but significant diuresis and glucosuria remain

[ ]. Although creatinine clearance decreased

significantly, UAE and 8-OHdG were partially decreased with phlorizin treatment; these

parameters did not recover completely by controlling only blood glucose level. In addition

to hyperglycemia, abnormal lipid metabolism has been considered an important factor in the

pathogenesis of DN [ , , ]. Since phlorizin treatment failed

to reduce blood TG and TC levels, uncontrolled dyslipidemia may affect DN in SDT fatty rats.

In the report using STZ rats, phlorizin prevented proteinuria, hyperfiltration and whole

kidney hypertrophy, but not glomerular hypertrophy. Therefore, some part of renal impairment

in STZ rats is uncontrollable with decreasing blood glucose level alone [ ]. On the other hand, some histopathological findings

were observed in the phlorizin-treated group. Increased urine volume and susceptibility to

urinary tract infection by inhibiting SGLT2 in proximal renal tubules may lead to tubular

dilation and inflammatory cell infiltration in the urothelium of phlorizin treated SDT fatty

rats.

We reported that caudal MNCV in both male and female SDT fatty rats was delayed at 24 weeks

of age [ , ]. Histopathologically, at 40 weeks of age, significant decreases in sural nerve

fiber number due to atrophy were observed in male SDT fatty rats [ ]. In the present study, we evaluated IENFD in SDT fatty rats as a

hallmark of small fiber neuropathy. IENFD is considered a marker of DPN that reduces from

early stage diabetes [ ]. Similar to other diabetic

animal models [ , ], SDT fatty rats showed significantly decreased IENFD. Significant

histopathological and functional preservation of peripheral nerves were achieved with

phlorizin. As is the case with DN, effects of dyslipidemia and hypertension on DPN have been

pointed out [ , ]. In contrast to DN, these factors have insignificant effects on DPN in SDT

fatty rats because controlling only blood glucose level prevented histopathological and

functional nerve impairments.

As a marker of DR, retinal function was evaluated using ERGs. Previously, we reported the

ERG results in SDT fatty rats; prolongation of OPs was observed in both males (16 weeks of

age) and females (22 weeks of age) [ , ]. Delayed OPs were also observed in the present study

and phlorizin treatment prevented retinal dysfunction. Histopathological changes in the

lens, such as hyperplasia of the epithelium, vacuolation of fibers, and formation of

Morgagnian globules, coincide with previous reports [ , ]. Furthermore, retinal lesions, such

as folding and thickening, which are found in aged SDT fatty rats [ , ], were also observed. These

histopathological abnormalities were not observed in the eyes of SDT fatty rats treated with

phlorizin. Although dyslipidemia is a key factor of DR [ ], our findings suggest that hyperglycemia is the major cause of DR and other

ocular changes in SDT fatty rats.

In conclusion, diabetic complications in SDT fatty rats are caused by sustained severe

hyperglycemia. Notably, other factors than hyperglycemia, such as hyperlipidemia and

hypertension may be involved in diabetic nephropathy in SDT fatty rats. In addition, we

reported that enhanced vascular endothelial growth factor (VEGF) signaling also contributes

to microvascular dysfunction in SDT rats recently [ ]. Clarifying the mechanism of diabetic complications further enhances the

prospects for the usefulness of this animal model in developing new drugs and therapies for

diabetic microvascular complications.