Effects of long-term experimental diabetes on adrenal gland growth and phosphoribosyl pyrophosphate formation in growth hormone-deficient dwarf rats

Summary

The availability of growth hormone (GH)-deficient dwarf rats with otherwise normal pituitary function provides a powerful tool to examine the relative role of hyperglycaemia and the reordering of hormonal factors in the hypertrophy-hyperfunction of the adrenal gland that is seen in experimental diabetes. Here, we examine the effects of long-term (6 months) experimental diabetes on the growth of the adrenal glands; their content of phosphoribosyl pyrophosphate (PRPP); and the activity of the PRPP synthetase, G6P dehydrogenase and 6PG dehydrogenase enzymes in GH-deficient dwarf rats compared to heterozygous controls. These parameters were selected in view of the known role of PRPP in both de novo and salvage pathways of purine and pyrimidine synthesis and in the formation of NAD, and in view of the role of the oxidative enzymes of the pentose phosphate pathway in both R5P formation and the generation of the NADPH that is required in reductive synthetic reactions. This study shows that GH deficiency prevents the increase in adrenal gland weight, PRPP synthetase, PRPP content and G6P dehydrogenase and 6PG dehydrogenase. This contrasts sharply with the heterozygous group that showed the expected increase in these parameters. The blood glucose levels of the groups of long-term diabetic rats, both GH-deficient and heterozygous, remained at an elevated level throughout the experiment. These results are fully in accord with earlier evidence from studies with somatostatin analogues which showed that the GH-insulin-like growth factor I (IGF-I)-axis plays a key role in the adrenal diabetic hypertrophy-hyperfunction syndrome.

The adrenal gland hypertrophy-hyperfunction that accompanies early experimental diabetes has been well documented (De Nicola et al. 1976, 1977; Rhees et al. 1983; by Kunjara et al. 1992, 2012), and there is further evidence for a role for insulin-like growth factor I (IGF-I) in adrenal cell function and steroidogenic response from studies on cultured bovine adrenal fasciculata cells (Penhoat et al. 1988). The question arises as to the part played in diabetic adrenal hypertrophy by hyperglycaemia and/or the reordering of hormonal factors. In this regard, the extensive studies on renal hypertrophy and temporal parallelisms with early adrenal growth are pertinent (Kunjara et al. 1986, 1992; Flyvbjerg et al. 1988; Steer et al. 1988). Recent studies have shown that treatment with somatostatin analogues ab initio effectively abolished the adrenal growth, the increase in phosphoribosyl pyrophosphate (PRPP) formation and the rise in PRPP synthetase activity in the initial period following the induction of diabetes without having any significant effect on the blood glucose values. This supports the hypothesis that growth factors or hormones regulated by somatostatin analogues play a significant role in adrenal hypertrophy (Kunjara et al. 2012). However, the multifunctional properties of somatostatin include (in different clinical and experimental conditions) the depression of growth hormone (GH), IGF-I levels, IGF binding proteins, glucagon, TSH, circulating ACTH and cortisol, and this pleiotropism makes it difficult to determine the precise mode of action that is responsible for the changes (Gill et al. 1982; Masuda et al. 1989; Itoh et al. 1990; Ørskov et al. 1992; Boehm & Lustig 2002; Pokrajac et al. 2009; Rutter 2009).

The discovery and characterization of a new mutant dwarf rat by Charlton et al. (1988), with a selective reduction in pituitary GH synthesis and storage; a growth retardation of 40% less than the normal litter mates; importantly, with concentrations of anterior pituitary hormones (luteinizing hormone, thyroid-stimulating hormone and prolactin) within the normal range; and with growth hormone-releasing factor present in the hypothalamus, opened up a new era in the examination of the endocrinology of growth in a range of tissues. The GH/IGF-1-deficient rats have only 5–10% of normal pituitary GH and low circulating levels of IGF-1. This model of GH deficiency has been of particular interest in relation to the changes in organ size and function in both short- and long-term experimental diabetes in the rat (Flyvbjerg et al. 1992; Grønbæk et al. 1997). Measurements have been made of somatic and longitudinal bone growth in dwarf rats (Charlton et al. 1988) and, recently, of the effect of different patterns of GH administration on these parameters in this model of GH deficiency (Westwood et al. 2010). It has been established that there is normal spermatogenic function within smaller testes in the dwarf rat and that the reproductive axis is only partially influenced by GH deficiency (Bartlett et al. 1990), again highlighting the specific nature of the lesion in this model.

The early studies exploring the effects of GH on renal hypertrophy in short- and long-term diabetes showed that over the first seven days of streptozotocin (STZ) diabetes, dwarf rats had attenuated kidney IGF-1 accumulation, less increase in glomerular volume and a slower rate of renal enlargement when compared with normal diabetic control rats. The renal changes in longer-term experimental diabetes of six month duration in the dwarf rat model include a decrease in renal and glomerular hypertrophy and reduction in the increase in urinary albumin excretion compared with diabetic rats of the same age with a normal GH/IGF-1 axis. The mean blood glucose levels were between 19 and 28 mmol/l throughout the six month period of diabetes and were similar in the diabetic GH/IGF-1 dwarf rats and the intact diabetic control group, indicating that hormonal factors rather than the blood glucose levels per se were key to the renal changes in diabetes. In more recent studies relating to GH regulation of body weight, deletion of the mouse GH receptor results in severe decrease in body weight, including decreased kidney weight, and IGF-1 levels. However, there was an increase in lifespan (Coschigano et al. 2003, 2010).

In this study, we have used this model to examine the effect of long-term diabetes of six months duration on the growth of the adrenal gland, the concentration of PRPP and PRPP synthetase, and of the enzymes of ribose formation via the pentose phosphate pathway in GH-deficient rats compared with their heterozygous controls; these parameters were selected for their role in nucleotide and nucleic acid synthesis, and in NAD and NADP formation.

Materials and methods

Animals

Male dwarf rats, homozygous for the dwarf trait, were obtained from Harlan Olac (Blackthorn, Bicester, UK). The background to this strain was described previously [Flyvbjerg et al. (1992) and Grønbæk et al. (1997)]. The GH/IGF-I-deficient dwarf rats have only 5–10% of normal pituitary GH content and low circulating levels of IGF-I, but otherwise a normal pituitary function, with initial mean body weight of 125 g. Age-matched heterozygous rats with normal GH/IGF-I production and mean initial body weight of 230 g were used as controls.

Diabetes was induced in dwarf rats and heterozygous controls by intravenous injection of streptozotocin (STZ, Upjohn, Kalamazoo, MI, USA) in a dose of 23 mg/kg body weight in acidic 0.154 mol/l NaCl (pH 4.5) after a 12-h fast. Body weight and blood glucose were measured in the initial period from one to seven days and thereafter monthly. The mean blood glucose values were between 19 and 28 mmol/l throughout the study period of six months, and blood glucose levels were similar in the two diabetic groups. The two non-diabetic control groups had similar mean blood glucose levels between 4.5 and 5.5 mmol/l. The rats were allowed free access to standard rat chow and water throughout the six month period. After completion of the six month period, the rats were anaesthetized with sodium barbital (50 mg/kg body weight i.p.), and the adrenals removed, trimmed and snap-frozen in liquid nitrogen. They were stored at −80 °C until required for the measurement of either PRPP content or assay of enzyme activity.

For the purposes of clarity and brevity, the four groups have been designated:

• Dwarf rats control (GH/IGF-I-deficient rats – no treatment).

• Dwarf rats diabetic (GH/IGF-I-deficient rats – STZ treatment).

• Heterozygous control (GH/IGF-I-normal rats – no treatment).

• Heterozygous diabetic (GH/IGF-I-normal rats – STZ treatment).

Quality control values for results presented in Tables 1–3 were of control adult male albino rats, from the same stock as those used in the study on the effect of long-acting somatostatin analogues on adrenal growth and PRPP formation in experimental diabetes (Kunjara et al. 2012), which were stored for the same period of time in liquid nitrogen. These parallel measurements were made to ensure that enzyme activities were comparable between the normal and heterozygous control groups. The quality control values (from five animals) were as follows: weight of one adrenal gland, 22.5 ± 0.8 mg; PRPP, 168 ± 4 nmol/g; PRPP synthetase, 6.5 ± 0.4 units/g; G6P dehydrogenase, 4.8 ± .3 units/g; 6PG dehydrogenase, 4.1 ± 0.2 units/g.

Table 1.

Long-term experimental diabetes – control vs. GH-deficient dwarf rats at six months

Group of rats	Weight of gland (mg)	PRPP (per/g) (nmoles)	PRPP (per two glands) (nmoles)
Dwarf control (5)	14.5 ± 0.42 NS	28.6 ± 3.02 NS	0.809 ± 0.09 NS
Dwarf diabetic (5)	13.6 ± 1.08	28.5 ± 4.23	0.762 ± 0.12
Hetero. control (5)	21.8 ± 0.81 NS	31.12 ± 1.64 <0.05	1.35 ± 0.08 <0.001
Hetero. diabetic (5)	24.4 ± 1.61	43.0 ± 3.44	2.07 ± 0.03
Fisher’s P values
Controls: Dwarf vs. heterozygous	<0.001	NS	<0.01
Diabetic: Dwarf vs. heterozygous	<0.001	<0.05	<0.001

Table 3.

The effect of long-term experimental diabetes on pentose phosphate pathway dehydrogenases of the adrenal gland in GH-deficient dwarf rats

Group of rats	Weight of gland (1 gland) (mg)	G6P dehydrogenase (per g) (μmol/g/min)	G6P dehydrogenase (per two glands) (μmol/g/min)	6PG dehydrogenase (per g) (μmol/g/min)	6PG dehydrogenase (per two glands) (μmol/g/min)
Dwarf control	14.7 ± 0.52	2.32 ± 0.23	0.068 ± 0.007	3.05 ± 0.31	0.090 ± 0.009
Dwarf diabetic	14.5 ± 1.20	2.27 ± 0.28	0.066 ± 0.008	3.26 ± 0.65	0.095 ± 0.019
Heterozygous control	21.5 ± 1.22	3.29 ± 0.14	0.141 ± 0.006	3.33 ± 0.24	0.143 ± 0.010
Heterozygous diabetic	26.5 ± 1.2	4.26 ± 0.16	0.226 ± 0.008	4.56 ± 0.39	0.242 ± 0.020
P values
Controls:	<0.001	<0.01	<0.001	NS	<0.001
Diabetic:	<0.001	<0.001	<0.001	NS	<0.001

PRPP content

For the measurement of PRPP content, one frozen adrenal was taken and extracted as described by Kunjara et al.(1992) using 1.2 ml of 100 mM KH2PO4/1 mM EDTA, pH 7.4, and heated for 5 min at 95 °C in a heating block. The PRPP content was measured as described by Kunjara et al. (1986).

Enzyme assays

The second adrenal gland was homogenized, 1:10 in 10 mM KH2PO4/1 mM mercaptoethanol, pH 7.4, and centrifuged at 27,000 g for 20 min. The supernatant was dialysed for 2 h against the homogenizing medium at 4 °C and used for enzyme assays as described below.

PRPP synthetase (EC 2.7.6.1)

This was measured as described for adrenal glands by Kunjara et al. (1992) using 20 μl of the above extract. A unit of activity is 1μmol of ¹⁴CO2 formed from [carboxyl-¹⁴C]-orotic acid/h at 37 °C.

Glucose 6-phosphate dehydrogenase (EC 1.1.1.49) (G6PDH) and 6-phosphogluconate dehydrogenase (EC 1.1.1.44) (6PGDH)

These activities were also measured by methods described for adrenal glands by Kunjara et al. (1992) using 10 μl of the above extract. A unit of activity is 1 μmol of substrate converted per minute at 25 °C.

Statistics

Comparison between the experimental groups was shown by Student’s t test, with 2-tailed values of <0.05 being considered statistically significant. The instat 2.01 program for Mackintosh was used.

Results

Adrenal gland growth

The changes in adrenal gland weights over the six month period are given in Tables 1–3. There were no significant changes in weight between the dwarf diabetic and the dwarf control groups (five animals per group). There was an increase (+23%, P < 0.001) in adrenal gland weight in the heterozygous diabetic rats relative to the control group. This was further confirmed in a larger group of animals (Tables 2 and 3) (11 rats per group).

Table 2.

Long-term experimental diabetes – control vs. GH-deficient dwarf rats at six months. Effects on adrenal PRPP synthetase

Group of rats	Weight of gland (1 gland) (mg)	PRPP synthetase (per g) (nmoles)	PRPP synthetase (per 2 glands) (nmoles)
Dwarf control (11)	14.7 ± 0.52 NS	5.56 ± 0.71 NS	0.168 ± 0.026 NS
Dwarf diabetic (5)	14.5 ± 1.20	4.85 ± 0.77	0.146 ± 0.035
Hetero. control (11)	21.5 ± 1.22 <0.01	5.27 ± 0.52 <0.001	0.229 ± 0.029 <0.001
Hetero. diabetic (11)	26.5 ± 1.2	9.99 ± 0.76	0.509 ± 0.039
Fisher’s P values
Controls: Dwarf vs. heterozygous	<0.001	NS	NS
Diabetic: Dwarf vs. heterozygous	<0.001	<0.001	<0.001

The heterozygous rats, control and diabetic, shown in Table 1 did not show statistically significant changes. The differences between the gland weights of dwarf vs. heterozygous rats were significant by comparing both the two control groups (P < 0.001) and the two diabetic groups (P < 0.001).

PRPP content

Table 1 showed that there were no significant differences in either PRPP (nmol/g adrenal) or PRPP (nmol per two glands) between the dwarf control and dwarf diabetic groups. In contrast, significant increases were seen in the heterozygous diabetic rats relative to their control groups. Their difference was marked particularly when calculated as PRPP per two glands (P < 0.001).

PRPP synthetase

Phosphoribosyl pyrophosphate synthetase activity of the four groups is shown in Table 2. The profile of change mirrors that of the PRPP content reported in Table 1. There were no significant differences between the dwarf control and dwarf diabetic groups. In the heterozygous diabetic group, there was an approximately twofold rise in PRPP synthetase activity.

G6P dehydrogenase and 6PG dehydrogenase

In the dwarf rats, diabetes had no apparent effect on the oxidative enzymes of the PPP. In contrast, in the heterozygous group, diabetes led to increase in the activities of both G6P dehydrogenase (+30%) and 6PG dehydrogenase (+39%) samples per adrenal gland (Table 3).

Discussion

The present study showed that dwarf rats deficient in GH and IGF-I but with otherwise normal pituitary function, including ACTH (Flyvbjerg et al. 1992; Grønbæk et al. 1997), did not respond to long-term STZ diabetes with enhanced adrenal gland growth. This contrasts with the adrenal glands of the heterozygous diabetic rats in which significant adrenal growth was observed. Similarly, the PRPP content, PRPP synthetase, G6P dehydrogenase and 6PG dehydrogenase were all unresponsive in the dwarf GH-IGF-I-deficient rats to long-term STZ diabetes. This contrasts with significant increases in the enzymes profile in the long term diabetic heterozygous group. These data highlight further the importance of GH/IGF-I in adrenal changes in diabetes and support the extensive evidence that hormonal regulation, rather than elevated blood glucose per se, is the driving force for both renal and adrenal growth, and function in untreated experimental diabetes (Flyvbjerg 1997, 2004; Flyvbjerg et al. 1989, 1991; Grønbaek et al. 2002; Kunjara et al. 2012).

There is a significant disparity between published values for the PRPP content of adrenal glands in control rats (168 ± 4 nmol/g) and STZ-treated rats (at seven days) (318 ± 31nmol/g) (Kunjara et al. 1992), and the present study where the heterozygous control is reported as having a PRPP content/g adrenal gland of 31 nmol/g and the heterozygous diabetic groups is reported as having a PRPP content/g adrenal gland of 31 nmol/g.

There are two possible explanations for this major difference: (i) These diabetic dwarf rats and their heterozygous controls were used six months after treatment with STZ, that is after a considerably longer time period than in any previous studies. Long term changes in adrenal structure and function may be a major causal factor. This was supported by our previous study of the changes in PRPP at three and seven days and at six weeks after induction of STZ diabetes (Kunjara et al. 1992). A marked decline from a peak at seven days post-STZ of (318 nmol/g) to a level of 176 nmol/g at six weeks post-STZ was observed. It might be anticipated that a further decline would be found at six months. Sricharoenvej et al. (2009) examined the morphological and microvasculature changes in the adrenal glands in STZ-induced long-term diabetic rats and reported that 20 weeks after induction of diabetes the cells of the zona glomerulosa had become atrophied and degenerative changes were found including lymphocyte infiltration, macrophages and amyloid deposits in the adrenal medulla. These reported changes further strengthen the possibility that long-term diabetes may also have a marked effect on PRPP synthesis and that parallel long-term heterozygous controls might differ from the normal control users in our experiments that were used mainly at six to eight weeks of age. (ii) As shown in Tables 2 and 3, the profile of change in PRPP synthetase and oxidative enzymes of the PPP (G6P dehydrogenase and 6PG dehydrogenase) follows the pattern seen in STZ diabetic rats, and diabetic rats treated with somatostatin analogues (Kunjara et al. 2012) indicated that specific factors might be involved in diminished PRPP concentration in the GH-deficient dwarf rats.

There are multiple regulatory factors that can restrict the activity of PRPP synthetase (Fox & Kelly 1972; Becker et al. 1979) including AMP, ADP, NADPH and several nucleotide di- and tri-phosphates; moreover, the supply of R5P is a critical factor. The changes in these parameters with long-term diabetes in GH-deficient dwarf rats remain to be established, although it may be noted that the non-oxidative pathway of R5P formation via transketolase shows a significant increase in early STZ diabetes (3 days) but falls to the control value after six weeks of diabetes (Kunjara et al. 1992). Thus, changes in substrate and/or inhibitor levels cannot be ruled out as relating to the significantly reduced PRPP observed in the present experiments.

Studies from this laboratory on the effects of GH on pathways of R5P formation in liver have some bearing on the present results and, in particular, on the response of the adrenal gland of GH-deficient dwarf rats to long-term diabetes. Measurements of hepatic changes in hypophysectomized rats showed that transketolase, the rate-limiting enzyme of the non-oxidative route of R5P formation, was decreased significantly and restored to normal by treatment with GH (Novello et al. 1969). A different approach to the study of the role of GH in the regulation of pathways leading to R5P formation and thus to PRPP synthesis was provided by the use of rats bearing a transplantable GH hormone-secreting pituitary tumour, thus providing an animal exposed to a continuously high concentration of the hormone (MacLeod et al. 1968; Gumaa et al. 1969). The examination of both the oxidative and non-oxidative segments of the PPP showed only marginal changes in the activities of G6P dehydrogenase and 6PG dehydrogenase, contrasting with the non-oxidative route where striking increases were observed, in the case of transketolase up to +70%.

Growth hormone and adrenal glucocorticoids were identified as ‘physiological insulin antagonists’ many years ago, and it is known that hypophysectomy and adrenalectomy correct some defects of glucose metabolism in experimental diabetes. The present data showing that GH-deficient dwarf rats fail to show adrenal hypertrophy in long-term experimental diabetes are in accord with the role of GH in the diabetic syndrome.

Funding

This study was supported by grants from the Basil Samuel Charitable Trust, the British Diabetic Association, the Association for International Cancer Research, the Danish Diabetes Association and the Danish Medical Research Council.

Declaration of interest

The authors declare that there is no conflict of interest that could be perceived as prejudicing the impartiality of the research reported.

Animal experimentation approval statement

All animals received proper care in compliance with the principles of laboratory care and were in accordance with the Home Office Animal (Scientific Procedures) Act 1986, U.K. requirements.