The Nicotinamide/Streptozotocin Rodent Model of Type 2 Diabetes: Renal Pathophysiology and Redox Imbalance Features

Diabetic nephropathy (DN) is a common complication of diabetes mellitus. While there has been a great advance in our understanding of the pathogenesis of DN, no effective managements of this chronic kidney disease are currently available. Therefore, continuing to elucidate the underlying biochemical and molecular mechanisms of DN remains a constant need. In this regard, animal models of diabetes are indispensable tools. This review article highlights a widely used rodent model of non-obese type 2 diabetes induced by nicotinamide (NA) and streptozotocin (STZ). The mechanism underlying diabetes induction by combining the two chemicals involves blunting the toxic effect of STZ by NA so that only a percentage of β cells are destroyed and the remaining viable β cells can still respond to glucose stimulation. This NA-STZ animal model, as a platform for the testing of numerous antidiabetic and renoprotective materials, is also discussed. In comparison with other type 2 diabetic animal models, such as high-fat-diet/STZ models and genetically engineered rodent models, the NA-STZ model is non-obese and is less time-consuming and less expensive to create. Given that this unique model mimics certain pathological features of human DN, this model should continue to find its applications in the field of diabetes research.


Introduction
Diabetic nephropathy (DN), also known as diabetic kidney disease (DKD) [1][2][3], is a severe complication of diabetes mellitus [4,5]. Approximately 30% of diabetic patients can develop DN [6][7][8], which is also a chronic kidney disease and can progress to end-stage renal failure [9][10][11]. The hallmarks of DN are kidney hypertrophy [12][13][14], mesangial cell proliferation and mesangial matrix accumulation [15][16][17], glomerulosclerosis, and persistent levels of proteinuria [18,19]. Despite the great advancement in our understanding of the pathogenesis of DN and numerous approaches that have been tested to slow down DN development and progression, no effective therapeutics are currently available for the treatment of DN. This is because the specific mechanisms underlying the pathogenesis of DN are yet to be fully elucidated. Therefore, there are unmet needs in treating or halting DN. In this regard, animal models of diabetes [20], be they genetically engineered or chemically or dietary induced, are indispensable tools in DN research.
In this brief review, I focus on one particular type 2 diabetes animal model, which is created by using nicotinamide (NA, Figure 1A) and streptozotocin (STZ, Figure 1B) [21][22][23]. This non-obese type 2 diabetes animal model was initially established in rats [24] but has since been extended to include mice with modifications of the original protocol. In comparison with the high-fat-diet (HFD)-STZ-induced type 2 diabetes animal models [25][26][27][28] and genetically engineered diabetic animal models, such as db/db mice [29] and zsf1 obese rats [30], the NA/STZ model is time-saving and less expensive. Therefore, this model can be equally used as a platform for not only exploring the pathogenesis of DN but also screening and testing potential antidiabetic agents [31] or renoprotective compounds for their therapeutic effects [21].

Mechanisms Underlying NA/STZ Diabetes Induction
Masiello P. et al. initially developed this non-obese type 2 diabetes in rats in 1998 [24]. Since then, this model has been widely used to test a variety of antidiabetic materials for their beneficial effects on diabetes and diabetic complications, including DN. The establishment of this model takes advantage of the contradictory effects of the two chemicals on β cells as STZ is β cell cytotoxic while NAD is globally cytoprotective. Therefore, STZinduced β cell damage can be blunted by nicotinamide [23]. Consequently, a certain percentage of β cells are viable and respond to glucose stimulation to release insulin [24]. It should be noted that the percentage of β cells that can survive really depends on the doses of the two chemicals. For a fixed dose of STZ, if the NA is too low, there will be no blunting effect from the NA and all β cells can be destroyed by STZ. On the other hand, if the NA is too high, the blunting or protective effects of the NA could be too high. In fact, the blunting or protective effects of the NA could reach 100% and no diabetes would be induced. For NA to play a protective role, NA is often given before STZ administration. Nonetheless, that NA is given shortly after STZ ingestion has also been reported in the literature [32][33][34][35]. In these cases, however, whether there are any differences between NA being given first or STZ being given first in diabetes induction and the severity of kidney injury has not been investigated. Notwithstanding, based on the observation that NA given immediately after STZ is equally protective [36], any difference should be minimal when the NA is administered right after STZ administration.
STZ is a nitrosourea compound that has a component similar to glucose ( Figure 1B) [37]. Hence, STZ is also known as a glucose analog [37]. Because of this structural similarity to glucose, the STZ enters into β cells via the glucose transporter-2 (Glut2) [37] that is abundantly expressed on the β cell surface [38]. Once inside the β cells, the nitrosoamide moiety of STZ can attack DNA and causes DNA alkylation and is thus responsible for STZ genotoxicity and cytotoxicity [23]. STZ-caused DNA damage can activate poly (ADP-ribose) polymerase-1 (PARP-1) that can then repair damaged DNA using NAD+ as a substrate [39]. As a result, NAD+ could be potentially depleted by the activated PARP-1 [23], thereby leading to cell death. When NA is administered prior to STZ administration, the damaging effect of STZ is greatly mitigated. This mitigating effect is thought to be due to two establishments. One is that NA is a direct inhibitor of PARP-1 [23], the other is that NA is a precursor of NAD+ [23]. Hence, the STZ cytotoxic effects can be greatly blunted

Mechanisms Underlying NA/STZ Diabetes Induction
Masiello P. et al. initially developed this non-obese type 2 diabetes in rats in 1998 [24]. Since then, this model has been widely used to test a variety of antidiabetic materials for their beneficial effects on diabetes and diabetic complications, including DN. The establishment of this model takes advantage of the contradictory effects of the two chemicals on β cells as STZ is β cell cytotoxic while NAD is globally cytoprotective. Therefore, STZinduced β cell damage can be blunted by nicotinamide [23]. Consequently, a certain percentage of β cells are viable and respond to glucose stimulation to release insulin [24]. It should be noted that the percentage of β cells that can survive really depends on the doses of the two chemicals. For a fixed dose of STZ, if the NA is too low, there will be no blunting effect from the NA and all β cells can be destroyed by STZ. On the other hand, if the NA is too high, the blunting or protective effects of the NA could be too high. In fact, the blunting or protective effects of the NA could reach 100% and no diabetes would be induced. For NA to play a protective role, NA is often given before STZ administration. Nonetheless, that NA is given shortly after STZ ingestion has also been reported in the literature [32][33][34][35]. In these cases, however, whether there are any differences between NA being given first or STZ being given first in diabetes induction and the severity of kidney injury has not been investigated. Notwithstanding, based on the observation that NA given immediately after STZ is equally protective [36], any difference should be minimal when the NA is administered right after STZ administration.
STZ is a nitrosourea compound that has a component similar to glucose ( Figure 1B) [37]. Hence, STZ is also known as a glucose analog [37]. Because of this structural similarity to glucose, the STZ enters into β cells via the glucose transporter-2 (Glut2) [37] that is abundantly expressed on the β cell surface [38]. Once inside the β cells, the nitrosoamide moiety of STZ can attack DNA and causes DNA alkylation and is thus responsible for STZ genotoxicity and cytotoxicity [23]. STZ-caused DNA damage can activate poly (ADPribose) polymerase-1 (PARP-1) that can then repair damaged DNA using NAD+ as a substrate [39]. As a result, NAD+ could be potentially depleted by the activated PARP-1 [23], thereby leading to cell death. When NA is administered prior to STZ administration, the damaging effect of STZ is greatly mitigated. This mitigating effect is thought to be due to two establishments. One is that NA is a direct inhibitor of PARP-1 [23], the other is that NA is a precursor of NAD+ [23]. Hence, the STZ cytotoxic effects can be greatly blunted by NA and the blunting magnitude is known to be NA-concentration-dependent [23]. In rats, although different investigators would use a different dosage combination of NA and STZ, the initially established dosages of NA and STZ (230 mg/kg and 65 mg/kg, respectively) still seem to prevail in the literature (Figure 2), though the application of lower concentrations of NA and STZ has been reported. In mice, the ingested concentrations of the two chemicals also vary widely. Nevertheless, it should be noted that the concentration of STZ for mice can be higher than that for rats. It appears that the use of 240 mg/kg NA and 100 mg/kg STZ in a mouse model is a prevailing approach in the literature [40]. It should also be noted that when a mouse is used as a model, multiple daily injections of NA and STZ (up to a week) may be conducted [41]. Certain investigators have also reported using high-fat-diet (HFD) feeding followed by NA and STZ ingestions [40]. Regardless of whether rats or mice are being modeled, the key point is that a given investigator should stick to their own protocol of NA and STZ administrations, such as the dosages and routes of chemical ingestions [21,23], so that a reproducibility and data comparison could potentially be achieved. It should also be noted that the severity of the diabetic disorders depends on how long the animals are kept after diabetes induction by the two chemicals. In the absence of any interventions, diabetic disorders will progress, mimicking various stages of clinical practice in humans. Diagrams showing representative flow charts of non-obese type 2 diabetes animal models induced by nicotinamide and streptozotocin. Renoprotective materials can be tested for their beneficial effects by this model, which is also outlined in the diagram. Note that for mice to be used as a model, more than one NA and STZ injection may be performed. Depending on the objective of a given study, the mouse model of NA-STZ diabetes induction may also involve HFD feeding for weeks before NA and STZ administrations (please see the text for a detailed discussion).
Additionally, based on a modified mouse model of type 2 diabetes induced by combining the HFD-NA-STZ treatments [42], investigations of DN created by HFD feeding in conjunction with NA and STZ administrations have also been reported [43,44]. It should be noted that when an HFD is involved, the creation of such a model would take longer than when only NA and STZ are used.

Renal Pathophysiology in this NA/STZ Animal Model
It has been reported that when diabetes was induced in mice by injection of 230 mg/kg NAD along with 50 mg/kg or 65 mg/kg STZ, the kidney organ index (kidney weight vs. body weight) for both STZ doses showed an increase when compared with that of the controls [45]. For a six-week duration of testing, urinary and serum parameters, such as creatinine, urea, and uric acid, were enhanced in the NA-STZ diabetic animals. In the presence of NAD, mice lived longer than those that received only STZ administration [45]. Such a result further demonstrates the blunting effects of NA on STZ cytotoxicity.
It should be noted that while DN can be classified into five stages ( Table 1), none of the NA-STZ-involved animal studies published so far have systematically addressed the five stages of diabetic kidney injury. Therefore, future studies on the progression of DN from stage 1 to stage 5 in the NA-STZ animal model need to be conducted. Moreover, numerous diabetic kidney injury biomarkers, such as those recently reported by Pelle et al.
[46] and Natesan et al. [11], have also not been systematically and comprehensively evaluated in this NA-STZ animal model. Most studies use popular kidney injury parameters [47] such as blood urea nitrogen (BUN), serum cystatin C, creatinine, uric acid, or/and estimated glomerular flow rate (eGFR) for the evaluation of diabetic kidney injury after NA-STZ injections. The histopathological staining of kidney is also frequently used for the analysis of kidney injury in this NA-STZ animal model. Figures 3 and 4 [48,49] show a typical staining of the kidney tissues by the periodic acid-Schiff and Masson trichrome, respectively. As can be observed from these histochemical stainings, the pathophysiological changes are obvious in the NA-STZ diabetic kidneys. Table 2 summarizes the renal pathophysiological measurements in the NA-STZ diabetic animal models in the absence of any interventions.

Application of this Model in DN Research
As mentioned above, this NA-STZ diabetes animal model is a non-obese type 2 diabetes model [23]. The pathogenesis of diabetes in this model may be different from that of HFD/STZ or genetically engineered models, such as the db/db mouse model and the zsf1 obese rat model [30, [94][95][96]. Nevertheless, the NA-STZ model may provide a unique platform for the study of non-obese diabetes and diabetic complications [43,97]. With respect to DN research, this model has been widely used for testing the therapeutic effects of nu- Rat (100/55) 28 days Increased serum Cre and urea, glomerular architecture deranged [93] Abbreviations: BUN, blood urea nitrogen; Cre, creatinine.

Application of this Model in DN Research
As mentioned above, this NA-STZ diabetes animal model is a non-obese type 2 diabetes model [23]. The pathogenesis of diabetes in this model may be different from that of HFD/STZ or genetically engineered models, such as the db/db mouse model and the zsf1 obese rat model [30, [94][95][96]. Nevertheless, the NA-STZ model may provide a unique plat-   Rat (100/55) 28 days Increased serum Cre and urea, glomerular architecture deranged [93] Abbreviations: BUN, blood urea nitrogen; Cre, creatinine; HFD, high fat diet.

Application of this Model in DN Research
As mentioned above, this NA-STZ diabetes animal model is a non-obese type 2 diabetes model [23]. The pathogenesis of diabetes in this model may be different from that of HFD/STZ or genetically engineered models, such as the db/db mouse model and the zsf1 obese rat model [30, [94][95][96]. Nevertheless, the NA-STZ model may provide a unique platform for the study of non-obese diabetes and diabetic complications [43,97]. With respect to DN research, this model has been widely used for testing the therapeutic effects of numerous antidiabetic or renoprotective materials ( Table 3). Most of these materials are natural products derived from plants, such as herbs, trees, teas, and vegetables. Table 3 lists the representative materials that exhibit renoprotective effects in DN in the NA-STZ rodent model of type 2 diabetes. It should be noted that as oxidative stress has been established as one of the major mechanisms underlying DN, many of the listed materials in Table 3 thus have antioxidant properties. Figure 5 summarizes the major mechanisms of the renoprotective materials tested by this NA-STZ animal model. It should also be noted that the renoprotective effects of many of the tested materials are in a dose-dependent manner, such as reported in reference [98] and others. Grain amaranth Rat (120/60) Improving renal calcium metabolism [72] Glycosin Rat (120/60) Decreasing blood urea and creatinine [73] Hypericum perforatum Rat (85/65) Antioxidative stress [74]  Biomolecules 2022, 12, x FOR PEER REVIEW 8 Figure 5. Diagram summarizing the representative renoprotective mechanisms of the mate listed in Table 3, using the NA-STZ non-obese type 2 diabetes animal models. AGEs stands for vanced glycation end products".   Table 3, using the NA-STZ non-obese type 2 diabetes animal models. AGEs stands for "advanced glycation end products".

Redox-Related Mechanisms That Remain to Be Elucidated in this NA-STZ Model
The non-tissue-specific mechanisms involved in cellular injury are thought to be implicated in the development of diabetic nephropathy [5,50]. These mechanisms, as shown in Figure 6, include the activation of the polyol pathway [105] and protein kinase C signaling, the hexosamine pathway, and the increased formation of the advanced glycation products [5,106]. However, what has been lacking is the underlying pathological mechanisms of DN in this unique NA-STZ model, in particular, redox signaling and the mitochondrial mechanisms of NA-STZ-induced DN. In fact, numerous aspects remain to be investigated in detail. These include mitochondrial redox imbalance [39]; sources of mitochondrial reactive oxygen species [107]; proteomics of mitochondrial protein oxidation [108,109]; mitochondrial abnormalities such as the derangement of mitochondrial metabolic pathways, including the Krebs cycle and electron transport chain [29]; fatty acid oxidation [110,111]; mitochondrial fusion and fission [112,113]; and mitophagy and the mitochondrial unfolded protein response [114][115][116][117][118][119] (Figure 6). The changes in redox signaling during the progression of DN in this animal model also remain to be comprehensively studied. Nephron segment-specific investigations of targeted genes [120,121] as well as the role of epigenetics [122,123] in this DN model also remain to be fully conducted. Delineating the mechanisms of these biological processes in the diabetic kidney may provide comprehensive insights into the underpinnings of DN. Additionally, this model could also provide a platform for testing the therapeutic effects of stem cells and gene therapy on DN [11]. For studying multiple kidney disease-causing risk factors, this model could also be combined with other kidney disease animal models, such as those induced by folic acid [47, [124][125][126], cisplatin [127,128], cadmium [129][130][131], lipopolysaccharide [132,133], and hypoxia or ischemia reperfusion [134][135][136][137][138][139][140][141].

Summary
The NA-STZ induction of a type 2 diabetic animal model is a useful tool for both studying the mechanisms of DN and screening renoprotective materials for diabetic kidney disease. The model is less time-consuming and less expensive than that created by genetic engineering or high-fat-diet feeding. The establishment of this model is based on the fact that NA can partially protect pancreatic β cells against STZ cytotoxicity, leading to the incomplete destruction of β cells and thus development of non-insulin-dependent type 2 diabetes mellitus [21,23,24]. This unique animal model should continue to serve as a utility for studying the non-obese type 2 diabetes that is highly prevalent in East Asian diabetic patients [157].