Herbal Medicines Targeting the Improved β-Cell Functions and β-Cell Regeneration for the Management of Diabetes Mellitus

There is an increasing trend of investigating natural bioactive compounds targeting pancreatic β-cells for the prevention/treatment of diabetes mellitus (DM). With the exploration of multiple mechanisms by which β-cells involve in the pathogenesis of DM, herbal medicines are gaining attention due to their multitasking ability as evidenced by traditional medicine practices. This review attempts to summarize herbal medicines with the potential for improvement of β-cell functions and regeneration as scientifically proven by in vivo/in vitro investigations. Furthermore, attempts have been made to identify the mechanisms of improving the function and regeneration of β-cells by herbal medicines. Relevant data published from January 2009 to March 2020 were collected by searching electronic databases “PubMed,” “ScienceDirect,” and “Google Scholar” and studied for this review. Single herbal extracts, polyherbal mixtures, and isolated compounds derived from approximately 110 medicinal plants belonging to 51 different plant families had been investigated in recent years and found to be targeting β-cells. Many herbal medicines showed improvement of β-cell function as observed through homeostatic model assessment-β-cell function (HOMA-β). Pancreatic β-cell regeneration as observed in histopathological and immunohistochemical studies in terms of increase of size and number of functional β-cells was also prominent. Increasing β-cell mass via expression of genes/proteins related to antiapoptotic actions and β-cell neogenesis/proliferation, increasing glucose-stimulated insulin secretion via activating glucose transporter-2 (GLUT-2) receptors, and/or increasing intracellular Ca2+ levels were observed upon treatment of some herbal medicines. Some herbal medicines acted on various insulin signaling pathways. Furthermore, many herbal medicines showed protective effects on β-cells via reduction of oxidative stress and inflammation. However, there are many unexplored avenues. Thus, further investigations are warranted in elucidating mechanisms of improving β-cell function and mass by herbal medicines, their structure-activity relationship (SAR), and toxicities of these herbal medicines.


Introduction
DM is a chronic metabolic disorder characterized by hyperglycemia which is caused by deficiency of insulin secretion, insulin action, or both [1]. DM has become a major global health issue since it causes premature mortality and lower quality of life due to diabetes-related complications. According to the latest data provided by the International Diabetes Federation, the global adult population (20-79 years) with DM was approximately 463 million (9.3% of the global adult population) in the year 2019 and is estimated to rise up to 700 million by the year 2045. DM and its complications caused 4.2 million deaths worldwide in the year 2019. e prevalence of DM among the Sri Lankan adult population is reported as 8.7% [2]. DM accounts for 9% mortality of all deaths in Sri Lanka [3]. e majority of DM cases can be categorized into two broad categories, i.e., type 1 diabetes mellitus (T1DM) and type 2 diabetes mellitus (T2DM). T1DM is caused due to autoimmune β-cell destruction which leads to an absolute deficiency of insulin secretion whereas T2DM resulted from a combination of resistance to insulin and inadequate compensatory insulin secretion [4]. At present, around 90% of diabetic cases are of T2DM with the highest proportion in low and middle-income countries [2].
Pancreatic β-cells which secrete the anabolic hormone insulin have an important role in glucose homeostasis hence in the pathogenesis of DM [5]. Insulin is a polypeptide consisting of two chains of amino acids known as A and B chains, linked by three disulfide bridges. Initially, a peptide chain known as preproinsulin is synthesized by mRNA coded for insulin which gets converted to proinsulin in the endoplasmic reticulum. Proinsulin is then shifted to the Golgi apparatus and assembled into granules. e peptide segment connecting the A and B chains, the connecting peptide (C-peptide), is detached in the granules before insulin secretion making equimolar amounts of insulin and C-peptide in the granules. C-peptide levels can be measured, and its level in the blood provides an index of β-cell function in patients receiving exogenous insulin [6].
An overview of mechanisms involved in β-cell damage and dysfunction in the pathogenesis of T1DM and T2DM is presented in Figure 1. Autoimmune destruction of islets of Langerhans/β-cells leads to the development of T1DM. It has been found that human leukocyte antigen (HLA) encodes cell surface proteins of the islets that interact with immune cells. Activated cytotoxic CD8+ T lymphocytes via the aforementioned HLA genetic factor cause damage to β-cells. Furthermore, inflammatory stress caused by cytokines secreted by T lymphocytes also causes damage to β-cells. e presence of multiple autoantibodies can be seen in T1DM; however, their contribution to the pathogenesis of T1DM is not confirmed [7]. T2DM occurs due to insulin resistance associated with a defect of compensatory insulin secretion. Insulin resistance is multifactorial and commonly develops with obesity and aging. Genetic factors as well as lifestyle changes affect insulin resistance and then T2DM. Decreased muscle glucose uptake and impaired suppression of hepatic glucose production due to insulin resistance are the main causes of hyperglycemia observed in T2DM. Toxicity exerted by chronic exposure of the β-cell to glucose and fatty acids is believed to be leading to β-cell damage and dysfunction in T2DM subjects. Chronic hyperglycemia also leads to glycation reactions, and the production of reactive oxygen species causes β-cell damage in T2DM [8,9]. e exact mechanisms of alteration of β-cell mass and function in the pathogenesis of DM are still unclear. However, it has been proven that β-cell mass and function begin to alter in both T1DM and T2DM even before the symptoms are shown. In T1DM, autoimmune destruction initially reduces β-cell mass, and exhaustion of remaining β-cells in compensating increased insulin requirement results in subsequent β-cell dysfunctions. In T2DM, initially, β-cell dysfunction occurs due to exhaustion of β-cells in compensating increased insulin requirement imposed by insulin resistance. is leads to a reduction of the β-cell mass due to cytotoxicity exerted by hyperglycemic conditions. Hence, in general, therapies targeted at increasing β-cell mass and therapies targeted at improving β-cell function could be considered as the main treatment and prevention targets of the disease for T1DM and T2DM, respectively [10].
Pancreatic β-cell mass and abnormalities of the structure can be assessed by histopathological examination or immunohistochemical analysis of the pancreatic tissue [10,11]. e histopathological examination involves morphometric analysis such as islet size and density, fractional β-cell area, and the number of apoptotic β-cells. Immunohistochemical staining using anti-insulin antibodies is important in the assessment of functional β-cell mass whereas anti-Ki67 antibodies are used in the assessment of β-cell proliferation [12]. Homeostatic model assessment-β-cell function (HOMA-β) is a mathematical model used for the estimation of β-cell function from measurements of fasting plasma glucose and insulin or C-peptide. HOMA-β is calculated using the fasting plasma glucose, and insulin hence can be used to evaluate β-cell function in the fasting state. For detailed investigation of defects of β-cell, dynamic tests such as intravenous glucose, arginine, or glucagon are also used in combination with a hyperglycemic clamp. e β-cell function during daily-life conditions is evaluated using an oral glucose tolerance test (OGTT) or a mixed meal tolerance test in combination with an estimation of insulin secretion rates and mathematical modeling [13]. Assessment of gene expressions and protein profiling is also used in assessing β-cell regeneration and function. None of the aforementioned methods on its own is considered the gold standard for assessing β-cell function. Hence, a combination of several methods is needed for a better conclusion [13].
β-Cell regeneration can occur via replication of existing β-cell or conversion of other pancreatic cells into β-cells. Examples for investigations on experimental drugs derived from the natural origin for their ability to improve β-cell regeneration are summarized in Table 1. β-Cell replication is mediated by the inhibition of multiple mitogenic signaling pathways. Experimental drugs such as harmine and aminopyrazine compounds have shown β-cell regenerative potential in this manner [14,15]. Modification of distinct genes which are associated with the development of β-cell during embryogenesis has a prominent role in converting other cells in the pancreas to β-cells. Drugs such as c-aminobutyric acid and artemether act as potential activators of conversion of other cells to β-cells [16,17]. Furthermore, there is a possibility of generating β-cells from progenitor cells, including human embryonic and induced pluripotent stem cells derived from somatic cell populations [18].
A β-cell function can be restored by stimulating insulin secretory pathways. Sulfonylureas (e.g., glimepiride and glipizide) are believed to stimulate insulin release through activating the K + /ATP/sulfonylurea receptor channels. ough these drugs initially reduce hyperglycemia, eventually β-cells may get damaged due to exhaustion. Incretin peptides amplify insulin release in a glucose-dependent manner. Glucagon-like peptide 1 (GLP-1) receptor agonists (e.g., polypeptide compounds: liraglutide, exenatide) are effective in promoting biphasic insulin release and rarely cause hypoglycemia or body weight gain. Dipeptidyl peptidase 4 (DPP-4) inhibitors (e.g., vildagliptin and sitagliptin) also contributed to improving β-cell function by prolonging the half-life of incretins via prevention of their rapid 2 Evidence-Based Complementary and Alternative Medicine degradation. It has been suggested that promoting insulin sensitization is a better approach for glycemic control rather than increasing insulin secretion alone which may cause exhaustion of β-cells [19]. Available allopathic drugs have not been able to restore β-cell function or regeneration and are also associated with side effects such as weight gain, weight loss, hypoglycemia, and abdominal discomforts [20]. Apart from the lifestyle interventions, stem cell treatments and pancreatic islet transplants are also utilized for the management of DM but none has been able to totally manage the disease. Hence, medicinal plants with a history of use in traditional medicine practices and as evident by ethnopharmacological knowledge have gained attention to be used as antidiabetic therapeutics to restore β-cell functions and mass due to their multitasking ability as in targeting different aspects of the pathways in β-cell regeneration and function [21]. ere are limited reviews on these natural products targeting β-cells. Oh and Semwal et al. provided comprehensive details on investigations conducted on plant-derived natural products targeting β-cells [22,23]. However, they lack the emphasis on cellular and molecular mechanisms of restoring β-cell function and β-cell mass, their structure-activity relationships, and toxicity concerns. Ghorbani et al. discussed the cellular and molecular mechanisms of restoring β-cell function and β-cell mass but the focus was confined to flavonoids [24]. is review attempts to identify the mentioned aspects and elaborate unexplored avenues for future research. We believe that with the given information it would be possible to fill the gap between existing knowledge and unexplored avenues in herbal medicines targeting β-cells.

Description of Study Selection
Online electronic databases "PubMed," "ScienceDirect," and "Google Scholar" were searched with the keywords "medicinal plant extracts," "phytoconstituent," "β-cell regeneration," "β-cell function," and "diabetes mellitus" in the whole text. A combination of these keywords with relevant Boolean operators was used in searching. Data were collected from January 2009 to March 2020. Studies on the effects of plant crude extracts/fractions/isolated compounds on β-cell structure and function using in vitro cell culture studies and/or in vivo models of T1DM and T2DM were selected for this review. Studies on commercial oral hypoglycemic agents other than plant-derived drugs and synthetic drugs with natural product moiety were excluded. Studies with no direct evidence on β-cell function and mass but indirectly suggested improvement of those activities via improving insulin resistance and other antidiabetic parameters were also excluded. Accordingly, in the PubMed database, a total of 81 hits were obtained. Out of them, 39 articles were excluded being not relevant to the scope of this review. e remaining 42 articles were selected for the study. In the ScienceDirect database, under the category of research articles, a total of 753 hits were obtained and 24,100 hits were obtained via Google Scholar. First, the most relevant 100 articles were selected from each database, i.e., ScienceDirect and Google Scholar. Out of 100, 58 articles in ScienceDirect were excluded for being not within the scope, and four articles were excluded being duplications. e remaining 38 articles were selected for the review. Out of 100 articles in Google Scholar, 64 articles were excluded for being not within the scope and seven articles were excluded being duplications. e remaining 29 articles were selected for the review. A total of 109 articles were studied for the review. A summary of the study selection process is shown in Figure 2.

Results
Plant-derived natural products, i.e., crude plant extracts, fractions, and isolated compounds, have been investigated for antidiabetic activity and are described in the following sections. Among these, some are reported as having potential for improving β-cell regeneration and function as evidenced by in vitro cell culture studies and animal and human studies. However, only a few natural products have been investigated for detailed mechanisms on regeneration and function of β-cell in the above models. In vitro cellular models and animal models have been applied widely in elucidating these mechanisms. In many instances, polyphenol compounds including flavonoids that possess pronounced antioxidant activities were found to be improving β-cell regeneration and function. e following sections are focused on the investigations conducted on herbal medicines with the potential for improving regeneration and function of β-cell.

Single Herbal Extracts Targeting β-Cells. Extracts of
3.1.1. Artemisia dracunculus L. Ethanol (80% v/v) extract of seeds of Artemisia dracunculus was investigated in vitro using islets of Langerhans isolated from mouse, NIT-1 cells (a pancreatic β-cell line established from a transgenic NOD/ Lt mouse), and human islets. Mouse islets and NIT-1 cells were exposed to 1 mM glucose, and human islets were exposed to 2.8 mM glucose prior to treatment with the extract. A significant (p < 0.05) increase in glucose-stimulated insulin release was observed in the aforementioned cells when treated with 10 μg/mL of the extract. MTT ((3-(4, 5-dimethylthiazol-2-yl)-2, 5-diphenyltetrazolium bromide)) assay revealed that the extracts are not cytotoxic at tested concentrations. Furthermore, it was revealed that extracts do Evidence-Based Complementary and Alternative Medicine 3 not increase β-cell mass in any of the concentrations. Western blot analysis of protein extract of NIT-1 cells treated with plant extract showed activation of the AMPactivated protein kinase (AMPK) followed by acetyl-CoA carboxylate (ACC) and protein kinase B (PKB/Akt) which was suggested as a possible mechanism for glucose-stimulated insulin secretion [25]. However, the AMPK activity on β-cell function and mass is still under debate, and controversial findings are reported among studies conducted to date. ere is growing evidence toward the concept that AMPK activation in β-cells may cause decreased insulin secretion [26,27]. ere are contradictory theories on the effect of activation of AMPK on β-cell apoptosis/survival. However, the activation of Akt has been found to inhibit the AMPK-activated apoptosis of β-cells [28]. Hence, the suggested mechanism should be further investigated to ascertain the mechanism of increasing glucose-stimulated insulin secretion and survival of β-cells by the investigated plant extract.

Centaurium erythraea Rafn.
Centaurium erythraea is a traditional medicinal plant in Serbia used to treat DM. Methanolic extract of dried aerial parts of C. erythraea has been investigated for its mechanisms of improving β-cell regeneration and function. In vivo studies were carried out using male albino Wistar rats weighing 220-250 g with DM induced by multiple intraperitoneal (i.p.) injections of a low dose of streptozotocin (STZ) (40 mg/kg). Oral administration of the extract at a dose of 100 mg/kg for four weeks resulted in significant improvement of glycemic control as evidenced by significant (p < 0.01) reduction of blood glucose level (33%) and increased insulin level (74%) in diabetic rats. Furthermore, hematoxylin and eosin (H&E) stained pancreatic sections showed that the shape, size, and number of islets and the number of β-cells returned back to normal upon treatment. Immunohistochemical staining of the pancreatic tissue with anti-insulin antibodies showed an increase in the number and distribution of β-cells.
Staining of pancreatic islets for GLUT-2 has shown positive GLUT-2 staining with a uniform distribution of GLUT-2 within the islets. e GLUT-2 is responsible for glucose entry into β-cells and thereby stimulating insulin synthesis. Immunohistochemical analysis of the pancreatic tissue of diabetic rats revealed the presence of activated protein kinase B (p-Akt) uniformly distributed throughout the islets. Akt plays an important role in the regulation of pancreatic β-cell growth and survival [29]. It has been found that overexpression and activation of Akt in β-cells result in an increase in the β-cell mass, cell size, and proliferation [30].
Further studies on pancreatic islet tumor Rin-5F cells revealed that cells induced with 12 mM STZ have a significant increase of 12% in cell viability in the presence of 0.25 mg/mL of extract. Furthermore, treatment with the extract showed a 77% increase in insulin 1 (Ins1) gene (i.e., the gene encoding insulin) expression and a 90% increase in insulin secretion and protection of DNA from damage.

Cornus officinalis Sieb. et Zucc.
Aqueous extract of dried fruits of Cornus officinalis at a dose of 500 μg/mL was investigated in a T1DM model using human 1.1B4 pancreatic β-cells.
e extract accelerated β-cell regeneration following cytokine-induced death. Furthermore, the treatment showed a 2-fold increase in cell proliferation. e plant extract was found to be rich in iridoid glycosides. It was suggested that the effects might be due to the observed twofold increase in the gene expression of the calciumdependent transcription factor, nuclear factor of activated T cells, and cytoplasmic 2 (NFATC2) [32]. NFATC2 is considered the key transcriptional factor which regulates transcription of many genes associated with DM thereby promoting β-cell proliferation and function [33].
3.1.4. Gynura divaricata (L.) DC. Gynura divaricata is a traditional Chinese medicinal herb with antidiabetic activity. Methanolic extract of aerial parts of G. divaricata has been investigated in ICR mice (three weeks of age) with T2DM induced by a high-fat diet followed by i.p. injection of STZ (100 mg/kg). It was found that the crude extract contains polyphenolic compounds including 3caffeoylquinic acid, 4,5-dicaffeoylquinic acid, 3,4-dicaffeoylquinic acid, and 3,5-dicaffeoylquinic acid. e incorporation of 10% G. divaricata to the diet of the diabetic mice for four weeks caused a marked recovery of pancreatic islets with an increase of β-cell counts, and an improvement of islet structure which was observed on H&E staining. An immunohistochemical study using anti-insulin antibodies showed an increased number of insulin-positive β-cells reflecting β-cell regeneration. A significant (p < 0.01) increase in the expression of GLUT-2, glucokinase (GK), v-maf musculoaponeurotic fibrosarcoma oncogene family protein A (MafA), and pancreatic duodenal homeobox-1 (PDX-1) was observed reflecting increased functioning of pancreatic β-cells. PDX-1 and MafA are important transcription factors which regulate gene expressions in β-cells. PDX-1 is essential for the development of pancreatic cells including β-cells and involves insulin gene transcription. MafA binds to the promoter region of the insulin gene promoting insulin expression in response to glucose [34]. An increase in the expression of antiapoptotic protein and B-cell lymphoma-2 (Bcl-2) and a decrease in the expression of apoptotic proteins, Bcl2associated X (Bax), and cysteinyl aspartate specific proteinase-3 (caspase-3) suggesting β-cell survival were also observed. Accordingly, G. divaricata crude extract showed pronounced hypoglycemic effects by inhibiting islet cell apoptosis and improving pancreatic function [35].  Artemether Mouse β-cell line (Min6) induced with doxycycline (1 μg/mL) for overexpression of ARX transcription factor 10 μM ARX repression thereby increase β-cell turnover and conversion of α-cells to β-cells [17] Zebrafish larvae induced with dimerizer AP20187 3.1.7. Myrica rubra Sieb. and Zucc. Myrica rubra (bayberry) fruit extract rich in cyanidin-3-glucoside (4) (Figure 3) was investigated in vitro using pancreatic β-cells and in vivo using mice induced with DM by i.p. injection of STZ (200 mg/kg). Pretreatment of β-cells exposed to hydrogen peroxide (H 2 O 2 ) with the extract at a dose of 15 μg/g resulted in increased cellular viability and decreased mitochondrial reactive oxygen species production and prevented cell death and cell necrosis. Furthermore, the extract upregulated PDX-1 gene expression, contributing to increased gene transcription of insulin-like growth factor 2 (IGF-2) and insulin in INS-1 cells (rat insulinoma cell line, a β-cell line) [38]. expressions of cyclin D2, PDX-1, and insulin receptor substrate 2 (IRS-2) which were suggested as the mechanism of β-cell proliferation. Ginsenosides (steroid glycosides) present in the extract would be causing the observed biological effects [39].
3.1.9. Tamarindus indica Linn. Aqueous extract of seeds of Tamarindus indica was investigated in rats induced with DM by i.p. injection of STZ (90 mg/kg). Oral administration of 240 mg/kg extract for four weeks resulted in a slight decrease in the number of insulin-positive granules compared to the control group. Furthermore, islet atrophy was not detected compared to the control group indicating protective effects against β-cell damage. e anti-inflammatory activity of the extract was represented by a significant (p < 0.05) reduction of tumor necrosis factor-α (TNF-α) level. An increased insulin secretion via increasing (p < 0.001) cytosolic Ca 2+ levels was observed in islets isolated from rats upon treatment with 100 μg/mL extract [40].

Teucrium polium L.
Hydroalcoholic extract of aerial parts of Teucrium polium was investigated for its antidiabetic activity in rats induced with DM by single i.p. injection of STZ (45 mg/kg). e administration of 0.5 g/kg extract to diabetic rats for 42 days resulted in increased β-cell mass. Treatment with T. polium extract also caused a reduction in the MDA (64%) level and increment of the pancreatic superoxide dismutase (SOD) (EC 1.15.1.1) (45%), CAT (EC 1.11.1.6) (52%), and GSH (105%) contents indicating amelioration of oxidative stress. Assessment of transcription factors showed higher levels of PDX-1 in diabetic rats while the level of FOXO-1 decreased. c-Jun N-terminal kinase (JNK) activation under oxidative stress, which leads to FOXO-1 formation, is a molecular mechanism related to the destruction of β-cells. Nuclear accumulation of FOXO-1 is believed to reduce the PDX-1 expression thereby reducing insulin gene expression. PDX-1 is also important in the restoration and proper function of the insulin-producing β-cells [41].

Uncaria tomentosa (Willd.) DC. Uncaria tomentosa
is a native species to the Amazon rain forest and is widely used as traditional medicine for DM. Ethanol (50% v/v) extract of stem bark of U. tomentosa was investigated in sixweek-old C57BL/6 male mice with DM induced by five i.p. injections of STZ (40 mg/kg). Oral administration of U. tomentosa extract (400 mg/kg) for 21 days showed a higher number of intact islets and a significant inhibition of destructive insulitis in histopathological examination of the pancreatic tissue. Furthermore, the immunohistochemical evaluation showed an increase of insulin-positive β-cells.
e phenotypic analysis indicated that treatment with higher doses (100-400 mg/kg) resulted in CD4(+) and CD8(+) T-cell values similar to healthy animals. Same doses also increased the number of CD4(+), CD25(+), and Foxp3(+) regulatory T cells. Furthermore, the extract modulated the production of 1 and 2, with increased levels of IL-4 and IL-5. Altogether, results suggest the immunomodulatory action of extract and protection of β-cells from immune reactions.

Ginseng
Oligopeptides. Panax ginseng Meyer (Family: Araliaceae) is a traditional Chinese medicinal herb used to treat DM. Administration of ginseng oligopeptides (0.5 g/kg) for 52 days to Sprague-Dawley rats (5 weeks old, 130-170 g) with DM induced by high carbohydrate/high-fat diet followed by alloxan (105 mg/kg) injection resulted in an increase (p < 0.05) in the HOMA-β values. e increase in insulin level and the decrease in expression of NF-κB and Bcl-2 family in pancreatic islets were also detected by Western blot analysis. Histopathological examination of the pancreas of treated animals showed amelioration of pancreatic damage. Furthermore, the β-cell survival time and rate were significantly longer in treated animals [46].  μM) followed by exposure to high glucose (33 mM) were protected against cytotoxicity exert by high glucose concentration. Vitexin improved insulin signaling as analyzed by the levels of functional proteins including GLUT-2, insulin receptor (IR), IRS-1, and IRS-2 and causing glucosestimulated insulin secretion. Vitexin improved the high glucose-induced nuclear transcription factor system by suppressing Rel A, Rel B, IκB, and P50/p105 expression resulting in decreased cell apoptosis, further confirmed by the reduction in the percentage of Annexin-V (i.e., a fluorescent stain for detect apoptotic cells) positive cells. Furthermore, vitexin activated proteins including NF-κB and Nrf2 in β-cells regulating apoptosis [48]. insulin secretion. e treatment also increased the expression of IRS-2 and increased the phosphorylation of Akt protein and decreased the protein level of FOXO-1. Furthermore, saponins increased the level of PDX-1. Saponins also increased glucose-stimulated insulin secretion and intracellular insulin content. All these expressions were reversed by phosphoinositide 3-kinase (PI3K) indicating that saponins may increase insulin secretion via the phosphatidylinositol 3-kinase (PI3K)/Akt/FOXO-1 signaling pathway [49].

3.2.6.
Andrographolide-Lipoic Acid Conjugate. Andrographolide (14)  3.2.7. Geniposide. Geniposide (15) ( Figure 5), an iridoid glycoside isolated from Gardenia jasminoides Ellis (Family: Rubiaceae), at a dose of 100 mg/kg given for 56 days was found to be causing β-cell regeneration in investigations conducted using C57BL/6J mice with DM induced by feeding high-fat diet. Further investigations using Min6 cells revealed that geniposide (20 μM) resulted in increased expression of T-cell factor 7-like 2 (TCF7L2), a transcription factor of Wnt/β-catenin signaling which involves in β-cell survival and regeneration. Geniposide promotes β-cell survival by increasing proliferation of β-cell and decreasing β-cell apoptosis in cultured mouse islets after being challenged with diabetic stimuli (glucose/inflammatory cytokines). e compound further activated Akt, enhanced the expressions of GLP-1R inhibited GSK3β-activity, and promoted β-catenin nuclear translocation. Geniposide induced small islet-like cell clusters formation as a result of β-cell neogenesis from ductal epithelium [51].  i.p. injection of STZ (60 mg/kg) were treated with asiatic acid (25 mg/kg) for two weeks. Treatment caused improved Akt and Bcl-xL expression in the pancreatic islets of rats as evaluated by Western blot methods. Immunohistochemical staining revealed preservation of insulin-producing β-cells in the pancreatic islets of diabetic rats. Furthermore, asiatic acid resulted in prosurvival Akt kinase activation and BclxL expression in the pancreatic islets of diabetic rats [52].
3.2.9. Angelica sinensis Polysaccharide (ASP). ASP derived from the roots of Angelica sinensis (Oliv.) Diels (Family: Apiaceae) was investigated on pancreatic islets of BALB/c mice with T2DM induced by feeding a high-fat diet followed by i.p. injection of STZ (60 mg/kg). Treatment with ASP (100 mg/kg) for eight weeks showed antiapoptotic function on β-cells via increasing the Bcl-2/Bax ratio thereby blocking caspase-9-caspase-3 cascade. Simultaneous suppression of caspase-8-driven extrinsic apoptotic pathway further inhibited the activation of proteins such as poly (ADP-ribose) polymerase (PARP) involved in β-cell destruction. Immunohistochemistry and immunocytochemistry indicated that ASP could increase intracellular insulin. ASP also promoted the secretion of insulin by stimulating the insulin gene expression [53]. (an indicator of β-cell neogenesis) of pancreatic tissue, and both were not detected in this case. In contrast, terminal deoxynucleotidyl transferase dUTP nick end labeling (TUNEL) staining which is used to detect DNA breaks formed during the final phase of cell apoptosis revealed suppressed apoptosis in PPAG-treated obese mice. In vitro studies conducted using isolated β-cells from mice and INS-1E cell lines showed that PPAG protected β-cells from palmitate-induced (i.e., a model of lipotoxicity) apoptosis. Exposure of PPAG-treated cells to different stresses revealed that PPAG protected β-cells against endoplasmic reticulum (ER) stress. Further Western blot analysis revealed that this protection is conferred by increasing the expression of antiapoptotic Bcl2 protein in β-cells without affecting proapoptotic signals [54].
In another study, PPAG has investigated in vivo using BALB/c mice induced with DM by i.p. injection of STZ (200 mg/kg) and in vitro using INS-1E β-cells and human pancreatic islet cells. Treatment with the PPAG (10 mg/kg) protected β-cells from apoptotic cell death and prevented loss of expression of antiapoptotic protein Bcl2. In vitro studies showed that PPAG protected INS-1E β-cells from STZ-induced apoptosis in a Bcl2-dependent and independent way, respectively, depending on glucose concentration. PPAG also protected human pancreatic islets against the cytotoxic action of palmitate [55]. . Sangguayin was investigated using diabetic mice and mouse insulinoma 6 (MIN6) cells for its antiapoptotic action. Treatment with sangguayin (175 mg/kg) for nine weeks reduced pancreatic pathological changes and islet β-cell apoptosis in db/db diabetic mice. Pretreatment with sangguayin (25-100 μg/ml) resulted in decreased palmitate-induced apoptosis in MIN6 cells. Decreased expression of cleaved PARP, cleaved caspase-3, and Bax, and increased Bcl-2 reflected β-cell survival upon treatment. Furthermore, the treatment suppressed ER stress which was reflected by decreased ER stress pathway-related proteins (Bip/XBP1/IRE1α/CHOP/Caspase-12) and autophagy as indicated by downregulation of autophagy-related protein expressions (LC3/p62/Atg5) [56].

Polyherbal Mixtures with Potential for β-Cell
Several other polyherbal mixtures were found in the literature as having potential for improving β-cell regeneration and function. However, the mechanisms of their actions are not yet investigated.
A polyherbal mixture consisting of 95% ethanol extract of seeds of Nigella sativa L. (Family: Ranunculaceae) and bark of Cinnamomum cassia (L.) D. Don (Family: Lauraceae) was investigated in Wistar rats with DM induced by i.p. injection of STZ (45 mg/kg). Oral administration of the extract at a dose of 100-200 mg/kg for 30 days resulted in reversal of STZ-induced pancreatic cell damage as observed from histopathological studies [58].
A polyherbal mixture comprised of 80% ethanol extract of leaves of Vernonia amygdalina Del. (Family: Asteraceae) and Gongronema latifolium (Utazi) (Family: Apocynaceae) was investigated in albino rats with DM induced by i.p. injection of STZ (65 mg/kg). Oral administration of the extract at a dose of 200 mg/kg for 28 days resulted in regeneration of pancreatic β-cells as evident by histopathological studies [59].

Clinical Studies on Herbal Medicines with Potential for β-Cell Regeneration and Function.
Only a few clinical studies were reported highlighting the potential for β-cell regeneration and function by herbal medicines. However, the detailed mechanisms of action of these medicines are yet to be investigated.
Diabetes tea ™ , a black tea originating from Camellia sinensis (L.) Kuntze var. sinensis (Family: eaceae) which is supplemented with 12 other medicinal plants, has been investigated in patients with T2DM in a double-blind clinical trial. Administration of 2.5 g of tea powder for 12 weeks resulted in a significant (p < 0.05) decrease in HbA1C and LDL. Furthermore, the treatment caused suppression of CD4+ T-cell expression of IL-1β-and IL-8 and upregulation of the expression of IL-10 and the Treg/IL-17 ratio. Suppression of these proinflammatory cytokines is beneficial for the protection of β-cells from cytokine-mediated damage [60].
A 70% ethanol extract of aerial parts of Melissa officinalis L. (Family: Lamiaceae) was investigated in patients with T2DM in a double-blind clinical trial. Rosmarinic acid (17) ( Figure 5) was found to be the main component of this extract. Administration of a capsule containing 700 mg of extract for 12 weeks resulted in improvement of β-cell activity as evident by significant (p < 0.05) improvement of HOMA-β- [61].
A crude drug combination used in traditional Chinese medicine has been investigated in patients with T2DM in a double-blind clinical trial. e crude drug combination consists of Coptis chinensis-rhizome (50%) and Shen-Ling-Bai Oral administration of 9 g of the above-mentioned combination daily for 12 weeks resulted in increased pancreatic β-cell function as evident by increased HOMA-β-value by 68.9% [62].
Several other plant extracts and isolated compounds exerted antidiabetic effects via targeting β-cells. However, the mechanisms have not yet been investigated in depth.
ese are included in Tables 2 and 3, respectively. Western blot analysis and immunohistochemistry to identify transcriptional factors related to gene expressions and other proteins in β-cell were found to be commonly used in elucidating mechanisms of action of herbal medicines identified in this review. Increasing β-cell mass via the expression of genes/proteins related to antiapoptotic actions (e.g., Angelica sinensis polysaccharide, vitexin, aqueous extract of Hibiscus rosa sinensis, methanolic extract of Woodfordia fruticosa, salidroside, and sangguayin) and β-cell neogenesis (e.g., geniposide) and proliferation (e.g., aqueous extract of Cornus officinalis, PPAG) identified as mechanisms of β-cell regeneration in some herbal products. Apoptosis could occur when the proapoptotic Bcl-2 proteins (Bad, Bid, Bik, and Bax) exceed antiapoptotic proteins of the Bcl family (Bcl-2 and Bcl-xL) in the mitochondrial membrane [139]. Increased expression of antiapoptotic proteins with decreased expression of proapoptotic proteins was observed in herbal medicines with antiapoptotic action. Increasing glucose-stimulated insulin secretion via activating GLUT-2 receptors (e.g., methanolic extract of Centaurium erythraea) and increasing intracellular Ca 2+ levels (e.g., aqueous extract of Tamarindus indica) were also observed upon treatment with some herbal medicines. Most of the studies suggested that reduction of oxidative stress by natural compounds with antioxidant potential protects β-cells from damage from reactive oxygen species. Herbal medicines with the ability to reduce proinflammatory cytokines also exerted protective effects on β-cells. Furthermore, the herbal medicines act on various insulin signaling pathways including activation of PIK3/Akt/ FOXO1 pathways, increasing expression of transcriptional factors AMPK, PDX1, and NFkB, and activation of GLP-1R. However, there are knowledge gaps and unexplored avenues on mechanisms of action of herbal medicines on pancreatic β-cells which should be further investigated in order to develop effective therapy against DM.

Discussion
ough the exact cellular and molecular mechanisms of β-cells causing pathogenesis of diabetes mellitus are still under investigation, certain therapeutic targets in β-cell for the bioactive compounds have already identified such as nuclear receptors (e.g., peroxisome proliferator-activated receptors (PPARs)), cell membrane receptors (e.g., K + /ATP channels, Na + /Ca 2+ exchanger, GLP-1R), transcription factors (e.g., PDX1, Ngn3), and intracellular enzymes (e.g., glucokinase) [140,141]. Bioactive compounds specifically acting on the aforementioned therapeutic targets could be an effective approach in developing therapies against DM. However, a limited number of publications are available on the structure-activity relationship (SAR) of natural products targeting β-cells; hence, this aspect needs to be addressed.
It was observed that the majority of the natural bioactive compounds responsible for β-cell regeneration and improvement of β-cell function are phenolic compounds including polyphenols such as flavonoids.
e SAR of polyphenols as antidiabetic agents is still unclear. However, the occurrence of multiple-OH groups attached to the aromatic ring and the arrangement of these hydroxyls in the ortho-dihydroxy conformation are supposed to be significant structural characteristics for high radical scavenging activity [142]. Most of the polyphenols share those characteristics and hence possess radical scavenging activity. However, the pharmacophore related to antioxidant activity may differ among compounds and hence should be investigated for each compound separately [143]. e amphiphilic nature of most of the polyphenols allows interactions with membrane components at different levels thereby generating cellular responses [144]. In the case of flavonoids, the total number and the configuration of hydroxyl groups are found to be important in regulating antioxidant and antidiabetic properties. C-2-C-3 double bond and C-4 ketonic group are two essential structural features for the antidiabetic property of flavonoids [145].
A limited number of studies have been conducted on SAR of isolated compounds found in this review. Salidroside (11) (Figure 5) is a phenylethanoid glycoside of which the aglycone is a tyrosol. Upon in vivo administration, salidroside gets metabolized into p-tyrosol (31) (Figure 9) which accounts for the bioactivity of salidroside [146]. Small peptides (molecular weights of less than 1 kDa) such as Panax ginseng oligopeptides showed higher antioxidant capacity and strong immunomodulatory action. However, the SAR of oligopeptides is still unclear [147]. Free hydroxyl groups on the B ring are found to be responsible for the Evidence-Based Complementary and Alternative Medicine 13 Increase in HOMA-β Improvement in pancreatic morphology [86] 14 Evidence-Based Complementary and Alternative Medicine Evidence-Based Complementary and Alternative Medicine 15  Regeneration of β-cell [117] Evidence-Based Complementary and Alternative Medicine antioxidant properties of C-glycosylflavones such as vitexin (13) ( Figure 5) [148]. C-glycosides such as vitexin and Oglycosides such as salidroside (11) exert similar biological effects. However, both vitexin and O-glycosides have poor water solubility. Hence, modification of glycosidic bonds to improve water solubility would lead to improved bioavailability [149]. SAR studies reported on several other compounds are also noteworthy to identify pharmacophores targeting β-cells. A SAR study on oleuropein (32) (Figure 9) (a secoiridoid glycoside) isolated from leaves of Olea europaea (Family: Oleaceae) reported that 3-hydroxytyrosol moiety of oleuropein is the main entity responsible for amyloid inhibition. However, the entire structure scaffold of the molecule was found to be needed for its glucose-stimulated insulin secretion effect [150]. SAR study of a cinnamic acid derivative, (E)-3-(3-phenylbenzo[c]isoxazol-5-yl)acrylic acid (33) (Figure 9), with β-cell regeneration ability has shown the presence of both carboxylate group and the intact isoxazolyl ring required for its β-cell regeneration ability via cAMP/PKA/mTOR signaling pathway by inhibiting IKK and Ikb selectively in β-cells [151]. SAR optimization of a 2,4-diaminoquinazoline (34) (Figure 9) compound showed antiapoptotic action and increased insulin secretion which were mediated by regulating genes related to signaling pathways of caspase-3 and PDX1/MafA, respectively. ese effects were optimum when 4-methoxy and benzylamine groups are present in the molecule [152]. Apart from the above, attempts have been made to identify potential β-cell targets of isolated compounds from medicinal plants through molecular docking studies. Data of molecular docking studies could be used in selecting potential antidiabetic agents for further in vivo/in vitro investigations. Herbacetin (35) (Figure 9) and sorbifolin (36) (Figure 9) isolated from Ficus species are capable of binding to amino acid residues Leu1002, Met1079, and Asp1150 of insulin receptor [153]. Glycogen synthase kinase-3 (GSK-3) is an important target in the regulation of β-cell mass and regeneration. Docking study on (4Z, 12Z)-cyclopentadeca- 4,12-dienone (37) (Figure 9) isolated from Grewia hirsuta against target protein GSK-3 has shown that the oxygen atom of the isolated compound reacts with LYS-85 residue of GSK-3 and forms hydrogen bond through a strong hydrophobic interaction [154]. Quercetin (22) (Figure 6), iristectorigenin A (38), 4-prenylresveratrol (39), moracin H (40), moracin C (41), isoramanone (42), moracin E (43), and moracin D (44) (Figure 10) isolated from leaves of Morus alba Linne have shown good affinity to AKT-1 target protein [155]. Molecular docking and SAR studies conducted on compounds of different chemical categories are also reported, and these data could be useful in predicting activities of natural compounds with related structures. One such SAR study on benzamide derivatives reported that 3-(N-piperidinyl) methylbenzamide derivative protects β-cells against ER stress-induced dysfunction and death [156]. Virtual screening of thiadiazine compounds against dual-specificity tyrosine-phosphorylation-regulated kinase 1A (DYRK1A) enzyme has shown to be capable of inducing β-cell replication [157]. SAR optimization of harmine analogs revealed that derivatives of harmine with C-1 position substitutions act on DYRK1A promoting human β-cell proliferation at doses of 3-30 μM [158].

Toxicity Concerns of Herbal Medicines Targeting β-Cells
Herbal medicines with a long history of use for DM are considered to be safe for therapeutic applications. Single herbal extracts and polyherbal mixtures as used by traditional medicine practitioners are generally considered to be time tested for safety and hence are allowed to be used directly in clinical applications [159]. However, fractionation or isolation of compounds from plant extracts may lead to toxicities [160]. e intention of coadministration of herbs or herbal compounds in traditional medicine practices is to have a synergistic therapeutic effect while ameliorating toxicities exerted by individual herbs. Scientific investigations in favor of combination therapy of different herbs to reduce toxicity exerted from individual herbs are also reported. Such attenuations in toxicities are usually caused by herb-herb interactions in polyherbal mixtures or mixtures of herbal compounds which leads to alterations of pharmacokinetics, i.e., absorption, distribution, metabolism, and elimination of constituents [161].  [162]. However, there is a possibility of occurring undesirable alterations of pharmacokinetics upon concomitant administration of different herbs especially in the case of introducing novel herbal mixtures. is could lead to unexpected toxicities by means such as accumulation of herbal constituents in the body or formation of toxic metabolites. For example, herbal compounds with cytochrome p450 inhibitory activity may diminish the metabolism of another coadministered herbal compound metabolizing via the same enzyme leading to toxicity from the latter [163]. For example, pyrrolizidine alkaloids are reported to exert hepatotoxicity and genotoxicity by metabolic activation via cytochrome p450 enzyme and modulate other hepatic metabolizing enzymes and hence could produce unexpected toxicities when coadministered with other herbal compounds with hepatic enzyme modulatory effect [164]. Most of the herbal medicines discussed in this review were tested using in vitro cellular models and rats for their toxicity and were proven to be safe. However, natural products usually have multiple bioactivities which may include unwanted/toxic effects when it comes to clinical applications. For example, asiatic acid found in this review also possesses anticancer activity where it inhibits antiapoptotic proteins causing cell death [165]. Extensive toxicity studies including in vivo acute and repeated dose toxicity, in vitro cytotoxicity, genotoxicity, carcinogenicity, teratogenicity, and drug interaction studies are required to ensure the safety of traditional herbal medicines [166]. Hence, in-depth long-term toxicity studies using in vivo models are warranted to ascertain the safety of herbal medicines as therapeutics. Apart from that, contamination of herbal medicines with heavy metals such as lead, arsenic, mercury, and cadmium which may lead to toxicity in consumers is a matter of concern [167]. Hence, standardization of herbal medicines using limited tests for these heavy metals is also needed.

Challenges and Future Perspectives
Herbal medicines due to their multitasking ability and low toxicity (at prescribed doses) compared to synthetic medicines could be useful in addressing therapeutic needs of multifactorial diseases such as DM that have not yet been met by synthetic medicines. ere is a vast knowledge of herbal medicines with antidiabetic effects residing among different nations and ethnic groups. Retrieval of this information would pave the way to develop better therapeutic agents for DM particularly targeting the β-cell functions.

24
Evidence-Based Complementary and Alternative Medicine standardization are major limitations of herbal medicines. Encapsulation of herbal medicines into biocompatible nanomaterials is gaining attention to overcome these limitations. An attempt has been made to nanoencapsulated curcumin in polylactide-co-glycolide. Oral administration of 50 mg/kg nanocurcumin followed by intraperitoneal injection of STZ (60 mg/kg) showed β-cell preservation as evident by TUNEL and H&E staining of pancreatic tissue, an effect which was not observed in the plain curcumin-treated group [168]. Hence, nanoformulations could be useful in the targeted delivery of bioactive compounds to β-cells with improved solubility and increased bioavailability.

Conclusions
A number of single herbal extracts, polyherbal mixtures, and isolated compounds from plant extracts as well as combinations of natural bioactive compounds targeting pancreatic β-cells have been investigated as preventive/treatment strategies for DM in recent years. Many herbal products reported an improvement of β-cell function as observed through HOMA-β. Pancreatic β-cell regeneration is observed prominently in terms of increase in size and number of pancreatic islets, the number of insulin-secreting cells, and the number of proliferating β-cells in histopathological/ immunohistochemical studies. Increasing β-cell mass via the expression of genes/proteins related to antiapoptotic actions and β-cell neogenesis/proliferation was identified as a mechanism of β-cell regeneration in some herbal medicines. Increasing glucose-stimulated insulin secretion via activating GLUT-2 receptors and/or increasing intracellular Ca 2+ levels was also observed upon treatment with some herbal medicines. Most of the studies suggested that reduction of oxidative stress inflammation by natural products has protective effects on β-cells. which should be further investigated in order to develop effective therapies against the management of DM.

Data Availability
No data were used to support this study.

Conflicts of Interest
e authors declare that there are no conflicts.