Atherosclerosis and atherosensitivity in two southwest Algerian desert rodents, Psammomys obesus and Gerbillus gerbillus, and in Rattus norvegicus.

Cardiovascular disease, including atherosclerosis, is the leading cause of death in patients with diabetes worldwide; thus, it is a major medical concern. The endothelium contributes to the control of many vascular functions, and clinical observations show that it is a primary target for diabetic syndrome. To get better insight into the mechanisms underlying atherosclerosis, we studied the interspecific differences in the arterial metabolisms of two, Psammomys obesus and Gerbillus gerbillus, as well as Rattus norvegicus (Wistar rat), well known for its atheroresistance. Twenty-two enzymatic activities and six macromolecular substances were histochemically compared in the two desert species and in Wistar aortas (abdominal and thoracic) and arteries (femoral and caudal) embedded in a common block. In the healthy adult rodents, enzyme activities were very intense. They demonstrated that aortic myocytes are capable of various synthesis and catabolism processes. However, considering the frequency of atherosclerosis and its phenotypes, significant differences appeared between the species studied. Our comparative study shows that aortic atherosensitive animals have several common metabolic characteristics, which are found in Psammomys rich in metachromatic glycosaminoglycans (involved in the inhibition of lipolysis and in calcification of the organic matrix), reduced activity in enzymes related to the Krebs cycle (weakening energetic power), and low lipolytic enzyme, adenosine triphosphatase, and adenosine diphosphatase activities. However, the most fundamental pathophysiological difference is the low lipolytic power of the aorta of Psammomys when compared to Wistar rats. This characteristic determines its atherosensitivity and makes this animal model more applicable to the experimental development of atherosclerosis.

Introduction

Cardiovascular disease is the leading cause of death in patients with type 1 or type 2 diabetes mellitus (T2D) worldwide; thus, it is a major medical concern. The endothelium contributes to the control of many vascular functions, and clinical observations show that it is a primary target for diabetic syndrome (DS). In humans, abnormalities in endothelium-dependent vasodilatation have been demonstrated in diabetics in all vascular areas, and the presence of endothelial dysfunction is an independent predictive factor of the occurrence of cardiovascular events. Integrated experimental work on animal models demonstrates that endothelial dysfunctions are responsible for cardiac and vascular remodeling, eventually leading to the loss of coherence in the hemostasis system, which is responsible for thrombosis.1 However, the underlying causes of diabetic macrovascular pathogenesis have not yet been well established.2 Animal models, in which some features of diabetes and atherosclerosis diseases have been induced, are therefore useful and advantageous in biomedical studies because they promise new insights into human diabetes. Unfortunately, it is well known that no animal model completely mimics the human pathology since most rodents used develop only small vascular lesions, sometimes limited to the stage of simple fatty streaks.3 It is only through streptozotocine injection that diabetes is induced, and a few weeks later, more complex and extensive lesions resembling those described in humans develop. However, the specific abnormalities of T2D are directly related to chronic hyperglycemia,4 and more particularly, to the glycation of amine endings of the proteins of the basement membrane; the sudden change from a healthy state to an insulin-dependent and sclerotic state does not allow us to identify the different entities, often unstable, generated during the intermediate stages in the pathway of the Maillard reaction.5 Yet, although no model completely mimics the human pathology, some species of Gerbillinaes living in extreme conditions, like Gerbillus gerbillus (Gerbillus) and Psammomys obesus (Psammomys), have often served as experimental models in the domain of comparative physiology for diseases related to DS. The phenotypes of histological disorders observed in the target organ tissues of these animals affected by one of the consequences of DS indeed closely resemble those described in humans.6–9 In fact, recent works show that Psammomys is the most appropriate model animal for experimentation in diet-induced obesity and diabetes; it is also very often used as a model to study other human diseases, such as neuronal degeneration and cerebral ischemia and its consequences,10–12 as well as in hearing studies.10–12

All these gerbil species live in comparable semi-desert biotopes of North Africa, the Eastern Mediterranean, and southern Asia. They are generally the only stocking of rodents in the dune sands. Among these gerbils, Gerbillus gerbillus (Olivier 1801) is the predominant species. It occupies all the covered areas of a steppe scattered with shrubs (Atriplex halimus) and with perennial grass, but may be also found in rocky terrain or saline marsh areas. Psammomys obesus (Cretzs 1828) is a rarer species, closely related to the semi-desert steppe. They live in burrows that are often located under the bushes in which the rats forage, such as the salty bushes of the Chenopodiaceae family, rich in sodium chloride and low in calories like Traganum nudatum, Suaeda mollis, and Salsola foetida. These plants are also well known for their antioxidant properties due to the phenolic compounds contained in their leaves, as has been recently confirmed.13 These animals, living in drastic climatic and nutritional conditions, develop comparable adaptive, behavioral, and physiological strategies that allow them to live and develop normally in such biotopes. However, these gerbils, even though they have adapted to the same desert environment, do not have the same metabolic characteristics. The observation of the frequency and the morphological aspects of atherosclerosis in laboratory animals show that there are considerable differences from one species to another. This different behavior cannot be fully explained by the involvement of environmental, nutritional, or serum factors. Many histochemical and biochemical works suggest that the characteristics of the arterial metabolism of different species are determining factors in their atherosensitivity.14 These considerations prompted us to examine the arterial metabolisms of the desert rodents, Psammomys and Gerbillus, and compare them with that of Wistar rats (Rattus norvegicus), which are considered an atheroresistant animal model.

Material and methods

Animals

In this study, we used adult (from 3 to 6 months) Psammomys (50) and Gerbillus (25) gerbils of both sexes, trapped in the semi-desert region of Beni-Abbes-Abadla (900 km south-west of Algiers) and adapted to laboratory conditions (25°C, 70% hygrometry, and a 12-hour light–dark cycle), and 10 Wistar rats. Upon their arrival in the laboratory, the two gerbil species were numbered, weighed, and transferred to separate suitable cages. After a two-week acclimation period, during which they were fed halophytes, Traganum nudatum, Suaeda mollis, and Salsola foetida (50 g/d/animal, equivalent to 30–32 kcal), they were randomly distributed into two groups: a control group of 10 Psammomys, maintained with a natural diet (ND); and the remaining Psammomys (40), Wistar (10), and gerbils (25), which were fed a standard laboratory chow (15–20 g/d/animal, equivalent to 52–70 kcal) and salty water (0.9% NaCl) ad libitum.

This diet is highly caloric (HCD) for the Psammomys, but not for the Gerbillus or the Wistar rats. Table 1 shows the composition of each diet. Halophyte plants are characterized by a very high content of water and salt (particularly NaCl) and low calorific value (about 450 kcal/kg of fresh plant). Due to the peculiarity of its diet, and particularly effective kidneys, which allow it to significantly reduce water loss15 Psammomys never drink water in their biotope.

Table 1

Natural diet (ND) and hypercaloric diet (HCD)

ND	Composition (%)	HCD	Composition (%)
Water	80.79	Water	9.0
Minerals	6.86	Minerals	7.1
Fats	0.40	Fats	7.5
Proteins	3.53	Proteins	25
Carbohydrates	8.42*	Total sugars	47.43**
Total sugars	0.18	Cellulose	4.0
Hemicellulose	2.62
Cellulose	2.23
Lignine	1.12
Undetermined	2.27

Notes:

The carbohydrates (total sugars, lignine, cellulose, hemicelluloses, and undetermined quantities) constitute 8.42% of the total composition;

rate of useful carbohydrate used = 100 − rate of: crude protein + fat + ash + cellulose + humidity.

All our animals were treated according to Directive 86/609/EEC, which regulates the use of animals for experimental and other scientific purposes, and which was adopted in the EU in 1986.16

The histoenzymologic methods

Methodology

After being anesthetized with nembutal, the animals were bled, and their arteries quickly dissected. Fragments of the upper thoracic aorta and the abdominal aorta and the femoral and caudal arteries were removed, and juxtaposed in the same block so that several samples from the three species could be compared on the same cup. The common blocks were immersed in liquid nitrogen for 10 seconds, then cryostat sections of 8 microns were prepared using a PMV Cryo- Microtome 2550 (LKB Co, Stockholm, Sweden).

For the histochemical studies, fragments adjacent to the fragments collected for the study were set in Dubosq-Brasil in paraffin and cut at 5 microns. The following macromolecular substances and enzymatic activities were studied after the application of the appropriate histochemical methods. To avoid enzyme leakage and prevent decline in activity and false location, various methods were used:

– After a period in pure acetone and rinsing with an appropriate buffer, the sections were incubated in conventional media for the detection of dehydrogenase (DH) and esterase.

– For redox reactions, nicotinamide adenine dinucleotide (NAD) and nicotinamide adenine dinucleotide phosphate (NADP) (Boehringer, Mannheim) were the intermediate electron acceptors, while Nitro Blue tetrazolium III (nitro-BT III) and tetra-nitro BT (Sigma, St Louis, MO) were the final electron acceptors. To reduce the diffusion of soluble enzymes in the medium, we added an aliquot of 6% agar. The incubation time at 17°C was prolonged (20 min). For each histoenzymatic reaction, a “witness” cut was incubated in the absence of a specific substrate. The intensity of these reactions was expressed by indices of positivity: 0, 1, 2, 3, 4, and 5, indicating, respectively, barely positive, positive, weak, moderate, intense, and very intense reactions.

Macromolecular substances

Glycoproteins and other carboxyhydrates: periodic acid Schiff ’s reaction (PAS) with and without the digestion test with alpha- and beta-amylase;17 glycogen: was demonstrated by the dimédon PAS technique;18 glycosaminoglycans (GAG): toluidine blue (Merck) 0.2% pH = 2, 4, and 6; ribonucleoproteins Unna-Pappenheim method with Brachet’s test;17 fatty acids and neutral lipids: Nile blue sulfate “Gurr” 1% Sudan black staining.

Enzymatic activities

Oxydoreductases

1,1,1,8-Glycerol-3-phosphate DH, 1,1,1,30,3-hydroxybutyrate dehydrogenase, 1,6,4,3-nicotinamide adenine dinucleotide dehydrogenase (NADH DH), 1,6,99,1-nicotinamide adenine phosphate dinucleotide dehydrogenase (NADPH DH), and 1,4,1,4-glutamate DH;19 1,1,1,27-lactate DH;20 Isocitrate DH;21 1,1,1,49-glucose-6-phosphate dehydrogenase (G6PD);22 1,3,99- succinate DH;23 1,4,3,4-monoamine oxidase (MAO).24

Transferases

2,4,1,1-Glycogen phosphorylase;25 2,4,1,11-Uridine diphosphate glucose (UDPG)-glycogen synthase.26

Hydrolases

3,1,1,1-α-Naphthyl acetate esterase;23 3,1,1,8-cholinesterase;27 3,1,3,1-alkaline phosphatase and 3,1,3,2-acid phosphatase;28 3,1,3,5,5′-nucleotidase;29 3,2,1,31-beta-glucuronidase;30 Mg2+- ATPase and Ca2+-ATPase;31 3,6,1,7-arylsulfatase;32 adenosine polyphosphatase:33 3,4,1,1-leucine aminopeptidase.34

Results

Histochemistry and histoenzymology of the arterial tissue of the three species

To search for interspecific differences in the arterial metabolism in Psammomys, Gerbillus, and Wistar rats, four arterial segments of each of these species were juxtaposed in the same block, sectioned, histochemically processed together, and then observed and compared on the same section. Thus, any differences that would appear would be characterized by a sure value. The results are shown in Table 2 and in Boards 1 to 3 of Figure 1. These results provide information about the metabolic behavior of arterial tissue of adult Psammomys and they highlight the differences between this animal and the two other rodent species.

Table 2

Histochemical and histoenzymologic parameters of abdominal and thoracic aortas, and femoral and caudal arteries of Psammomys obesus, Gerbilllus gerbillus, and Wistar rats

Parameters	Aortas						Arteries
	Thoracic			Abdominal			Caudal	Femoral
	P(40)	G(25)	R(10)	P(40)	G(25)	R(10)	P(40)	G(25)	R(10)
Oxydoreductases
Glucose G6-PDH	2.5	1.5	2	2.5	1.5	2	2.5	2.5	2.5
α-Glycerol 3-PDH	3.5	3.5	0.5	3	2.5	1	3	2	1.5
Lactate DH	3	3	2.5	3	3	2.5	3.5	3	3
β-H Butyrate DH	1	1	1	1	1	1	1.5	1.5	1.5
Isocitrate DH	2.5	2	1.5	2.5	2.5	1.5	3	3	3
Succinate DH	3	2.5	2	3	3	1.5	3.5	3.5	3
Glutamate DH	2.5	2.5	2	3	3	2.5	3.5	3.5	3
NADPH2 DH	3	2.5	2	3	2.5	2	3	3	3
NADH DH	3	2.5	2	3	3	2.5	3	3	3
Monoamine oxidase	2	2	1	2	2	1	2	2	2
Transferase
UDPG-synthetase	0.5	0.5	0.5	0.5	0.5	0.5	1	1	1
Phosphorylase	1.5	1	1	1.5	1	1	1.5	1	1
Hydrolases
Alkaline phosphatase V.V	0	0	3	0	0	3	0	0	3
5′ nucleotidase	0.5	0.5	1.5	0.5	0.5	1.5	0.5	0.5	1.5
ATPase Ca++	3.5	3.5	3.5	3.5	3.5	3.5	3	3	3
Indoxylesterase	0.5	1	1.5	0	1	1.5	0	0	0.5
Naphtylesterase	1	1.5	2	0.5	1.5	2	0	0.5	0.5
Cholinesterase	0.5	1.5	2	1	1.5	2	0	0	0
Acid phosphatase	0.5	0.5	0.5	0.5	0.5	0.5	0.5	0.5	0.5
LA peptidase	1.5	1.5	0.5	0.5	0.5	0.5	0.5	0.5	0.5
Arylsulfatase	4	4	4	4	4	4	2	2	2
β-glucoronidase	1	1	1	1	1	1	0	0	0
Macromolecules
PAS+ substances	2	2	2	2	2	2	1.5	1.5	1.5
Glycogen	0	0	0	0	0	0	0	0	0
Glycosaminoglycans	2	2	2	2	2	2	1	1	1
Neutral lipids	0	0	0	0	0	0	0	0	0
Fatty acids	0	0	0	0	0	0	0	0	0
RNA	1	1	1	1	1	1	1.5	1.5	1.5

Notes: The number of animals studied is given in brackets. The numbers given in the table indicate the average of the “index of positivity” of the reaction, resulting in several observations.

Abbreviations: P, Psammomys obesus; G, Gerbillus gerbillus; R, Wistar rat; V.V, Vasa vasorum.

Figure 1

Interspecies differences of aortic enzymes of Psammomys, Gerbillus gerbillus, and Wistar rats.

Abbreviations: P, Psammomys; G, Gerbillus; R, Wistar rats.

Features of the metabolism of Psammomys obesus

As in the other species, the smooth muscle cells (SMCs) in the desert rodent media were, histoenzymatically, the most active parietal element. In the healthy adult animals, enzyme activities were very intense. They demonstrated that aortic myocytes can achieve various synthesis processes (glycosaminoglycans, glycoproteins, lipids), and various types of catabolism (glycolysis, esterolysis, lipolysis). The myocyte enzymes contributed to the maintenance of balance between humoral inputs and the metabolic systems of the aortic wall and a better adaptation of cells to biochemical alterations occurring with age.

In adult animals, the myocytes’ oxidative metabolism was dominated by glycolysis, lactate and α-glycerophosphate dehydrogenase being more intense than the oxidoreductases linked to the aerobic degradation of glucose in the pentose pathway glucose-6-phosphate-dehydrogenase (G6-PDH); and in the Krebs cycle, succinate dehydrogenase (SDH) and isocitrate dehydrogenase (ICDH). Glycogen metabolism was reflected by uridine diphosphoglucose UDPG-glycogen synthase (anabolic enzyme) and glycogen phosphorylase (catabolic enzyme). These two activities were very weakly positive. This suggests that glucose that penetrated the media of Psammomys was metabolized and used for energy needs, rather than transformed and stored as glycogen. The latter was not shown histochemically. The phosphate adenosine esterolysis was useless concerning adenosine monophosphate AMP and very intense for adenosine-5′-triphosphate ATP, ATPase being very active in myocytes. The lipolytic enzymes (acid esterase, cholinesterase) were weakly positive. The hydrolysis of carboxylic esters must be very small when referring to the extremely low activities of acid esterases and cholinesterase. Lysosomal enzymes were unequally active: acid phosphatase (a classic marker of lysosomes) was barely detectable. Aminopeptidase and beta-glucuronidase were better represented. Arylsulfatase was extremely active and aminopeptidase was very intense in the fibroblasts of the adventitia for which it revealed itself as a true enzyme marker. Monoamine oxidase (MAO) was clearly positive, suggesting that the cells in the Psammomys media were capable of oxidative deamination of vasoactive monoamines (catecholamines).

Contrary to what has been observed in rabbit aorta (which are atherosensitive), the enzymatic activities in the aerobic oxidative degradation path of glucose, the Krebs cycle (SDH, IDH, and GDH), and the pentose pathway (G6PDH) were higher in Psammomys than in Wistar rats.35

Concerning the macromolecules, the media free of lipids appeared moderately rich in GAGs and in PAS+ substances (glycoproteins). The myocyte cytoplasm was relatively poor in RNA. This suggests that the synthesis activities in the adult animal would be slowed down, and that the enzymatic exaltation observed in various pathways of intermediary metabolism is mainly related to the permanent tonic character of the arterial myocytes and to the naturally hypertensive character of Psammomys.

Some enzyme activities, such as the oxidoreductases, were especially higher in the femoral and caudal arteries than in the aorta. This difference must reflect the greater abundance of cells in the media of muscular artery, but also their more pronounced contractile ability.

Comparative study of aortic macromolecules in Psammomys obesus, Gerbillus gerbillus, and Rattus norvegicus

The differences between the three species were either purely “quantitative,” or “qualitative.” The first were phenotypic, and involved the aerobic oxidoreductases: NADH2 DH, NADPDH 2 DH, SDH, MAO, and alpha-G 6-PDH, as illustrated in Figure 1, boards 1 and 2. These images clearly show that the intensities of aerobic oxidoreductases in myocytes from the desert rodents were higher than in the Wistar rats.

The second differences were genetic. Some activities, weak in one species, could be severe in the other; alkaline phosphatase, which is considered a specific marker of endothelial plasma membrane, was strongly positive in the vasa vasorum and fibroblasts at the junction of the medial– adventitial borders of the rats, whereas it was totally absent in the desert rodents. Lipolytic activities (indoxyl-naphthyl- esterase-cholinesterase) were intense in the Wistar rats but were barely noticeable in Psammomys and Gerbillus as shown in images 3b and 3c of board 3. The 5′ nucleotidase, a biochemical marker of tissue calcification, and probably of early atherosclerosis, was high in the desert rodents and completely absent in the Wistar rats.

Whatever the exact functional significance of these interspecific variations, they suggest the existence of subtle molecular differences in the endothelial and myocyte cell plasma membranes between the Wistar rats and the desert rodents. Lysosomal leucine aminopeptidase, which was barely detectable in the rats, was much more intense in Psammomys and Gerbillus, and suggests the existence of intense remodeling in the media, and especially in the adventitia of the latter. But the most significant difference concerns the esterases and cholinesterase, present in the aorta of Wistar rats and Gerbillus and virtually absent in Psammomys, as the lipolytic capacity of the aortic media depends largely on the activity of these enzymes.33 The comparative activity of several relatively low carboxylic esterases was the most characteristic feature of the aorta of the atherosensitive animals.

Metabolic behavior of the aorta of Psammomys obesus after a diabetogenic diet

Metabolic status

To enter the response of the aortic tissue into diabetogenic aggression, an aorta of a Psammomys that was fed HCD was removed and juxtaposed in the same block as an aorta of an animal of the same age fed with halophytes. The two aortas were treated simultaneously to study their histochemical characteristics, and then studied comparatively on the same section. This examination was repeated nine times to ensure reproducibility of measurements. The results are shown in Figure 2, boards 4 and 5.

Figure 2

Comparison of aortic enzymes of Psammomys, healthy (N) and diabetic (D).

To determine the metabolic status of the animals in the experiment, we measured all the usual biometabolic parameters. The animals that were systematically fed the HCD became obese and developed hyperinsulinemia in an attempt to adapt to new living conditions. Of the 40 Psammomys submitted to HCD, 24 managed to regulate their blood sugar, while the other 16 animals developed significant hyperglycemia. From the sixth month, we observed an abrupt fall of plasma insulin rate in 11 of the animals in the latter group, and they became insulin-dependent. Glucose and blood lipid rates were very high. Ketoacidosis appeared, resulting in diabetic coma and death of the animal. Of the 16 diabetic animals selected for the histochemical study, 11 were insulin-dependent and five were not insulin-dependent.

Enzymatic changes and metabolic pathways

Enzymes attached to the pentose pathway (G6PDH), to the Krebs cycle (SDH, UDH), and to the redox of the nucleotides (NADH2 DH and NADPH2 DH) showed a trend toward activation, in the hyperglycemic animals (Figure 2, board 4). This phenomenon must be interpreted as an attempt by the SMCs of the arteries to adapt to a microenvironment enriched with glucose. However, the activity of enzymes attached to the parietal lipolysis (esterase and cholinesterase) was decreased in most animals with DS (Board V). Since, in most cases, there is a non-insulin-dependent diabetes mellitus with strong hyperinsulinemia, it can be assumed that inhibition of lipolysis in these cases is due more to excess insulin than to hyperglycemia. Neutral lipids appeared in the intima and in the internal quarter of the media in Psammomys subjected to the diabetogenic diet for nine months, and which were in a state of insulin-dependent diabetes mellitus (Figure 2, images 5c and 5d of board 5). In these animals, lipolytic enzyme activities were greatly reduced or zero.

The SMCs, of the most inner part of the media underlying the intima, had a vacuolar cytoplasm in the preparation previously dehydrated, and therefore, are devoid of lipids. Histologically, this is the typical aspect of “cell foam” loaded with lipid droplets. The “lipoidosis” was circumferential and involved almost the full circumference of the wall; this phenomenon was more marked in the thoracic aorta of Psammomys obesus.

Conclusion

Examination of the frequency of atherosclerosis and its phenotypes shows significant differences from one species to another. These differences cannot be fully explained by environmental factors, since the three rodent species considered live in the same habitat and eat exactly the same diet. Comparative studies show that susceptibility to aortic atherosclerosis has the following traits, which were found in Psammomys:

The great abundance of GAGs inhibits lipolysis, creating fixing points for lipoproteins, which penetrate the media, and induces calcification of the extracellular matrix.

The reduced activity of enzymes related to the Krebs cycle lowers the energy capacity and leads to the danger of hypoxia; the reduced activity of lipolytic enzymes leads to accumulation of lipids that have penetrated the aortic wall.

The activity of ATPase is also reduced. This enzyme controls the level of ATP, a factor which promotes the synthesis of long fatty acid chains, and thereby, sclerogenic lipids.35,36

The differences observed in the caudal and femoral arteries of the three species studied were especially of the histoenzymologic type. The very low pyroninophilicity of the myocyte cytoplasm reflects its low content of ribonucleoproteins. The activities of some oxidoreductases were higher in these arteries than in the aorta. This difference must reflect the greater cell density in the media of the muscular artery, but may also induce a better contractility. Finally, the arterial metabolism of Psammomys appears closer to that of Gerbillus than to that of Wistar rats. In our precedent contribution, we show that histological characterization demonstrates the typical alterations observed in atherosclerosis and that Psammomys obesus is an excellent animal model in which to study the progression of T2D and the development of atherosclerosis.36 The observations reported in this paper reveal that the atherosensitivity of Psammomys is essentially of enzymatic origin, as has been observed in other atherosensitive animals.35 Psammomys obesus is, then, very appropriate for the experimental study of the development of atherosclerosis and diabetes.