The effects of phytochemicals and herbal bio-active compounds on tumour necrosis factor-α in overweight and obese individuals: a clinical review

Obesity is abnormal fat accumulation in the body which acts as a risk factor for various cardiometabolic states. Adipose tissue in excess can release inflammatory factors, including TNF-α and IL-6, and suppress adiponectin production. TNF-α increases the levels of IL-6 and acute phase reactants such as C-reactive protein. Inflammation has a crucial role in developing and progressing various cardiometabolic diseases and a wide range of obesity-related complications. It has been shown that TNF-α has a significant role in the development of insulin resistance. Recently, a growing body of evidence has focused on herbal medicine, phytochemicals and natural bioactive compounds as inexpensive, relatively easy accessible agents with low adverse effects to reduce inflammatory markers such as TNF-α and simultaneously decrease insulin resistance, glucose intolerance, and dyslipidemia in obesity. The main focus of the current review is to summarize the results of the studies, which assessed the effects of phytochemicals and herbal bio-active compounds on serum TNF-α in subjects with overweight or obesity. This review suggests that herbal medicine have favorable effects on the reduction of TNF-α concentration; however, the results were not uniform for different products. Among the reviewed plants, ginger, ginseng, resveratrol, and flaxseed had more promising effects.


TNF-α
Tumour necrosis factor α IL And interleukin CRP C reactive protein NF-κB Nuclear factor kappa light chain enhancer of activated B cells CVDs Cardiovascular diseases NAFLD Non-alcoholic fatty liver diseases

Obesity and inflammation
Obesity is the abnormal fat accumulation that acts as a risk factor for various diseases (Ellulu et al. 2017;Karczewski et al. 2018). Adipose tissue is stimulated by extra macronutrients and release various inflammatory agents such as tumour necrosis factor α (TNF-α) and interleukin 6 (IL-6), as well as suppress adiponectin production, predisposing to oxidative stress and inflammation. TNF-α increases the level of IL-6. Enhancement in IL-6 results in the stimulation of the liver to increase acute-phase proteins synthesis and secretion such as C-reactive protein (CRP), fibrinogen, serum amyloid A, and α1-antichymotrypsin ( Fig. 1) (Ellulu et al. 2017;Karczewski et al. 2018;Tanaka et al. 2014). It has also been shown that IL-6 might increase free fatty acids (Popko et al. 2010). It is confirmed that inflammation has a substantial role as a risk factor in developing and progression of cardiovascular diseases (CVDs) such as coagulation, atherosclerosis, insulin resistance, hypertension, metabolic syndrome, type 2 diabetes and nonalcoholic fatty liver diseases (NAFLD) (Ellulu et al. 2017;Karczewski et al. 2018). Inflammation could also contribute to the development and progression of non-cardiovascular diseases, including psoriasis, rheumatoid arthritis, psychological disorders such as anxiety and depression, asthma, neurodegenerative diseases, cancer, and renal diseases (Ellulu et al. 2017). Chronic systemic inflammation and immune system activation are considered the main factors associated with developing and progressing these pathologies related to obesity (Ellulu et al. 2017;Shoelson et al. 2007). It has been shown that glucose and fat consumption cause a significant increase in the inflammatory factors by increasing oxidative stress and transcription factors, including nuclear factor (NF-κB), activating protein-1, and early growth response-1 (Shoelson et al. 2007).
On the other hand, calorie restriction, fasting or reduction in macronutrient intakes results in a significant decrease in oxidative stress and inflammatory markers (Shoelson et al. 2007). Reduction of inflammatory markers in obese individuals can decrease the risks of CVDs and poor outcomes mediated by obesity-related inflammation (Ellulu et al. 2017;Karczewski et al. 2018).

Fig. 1
Schematic summary of pathways of the effects of obesity, excess intake of fat and glucose on inflammatory response and its effects on induction of obesity-related diseases. TNF-α tumor necrosis factor α, IL-6 and interleukin 6, NF-κB nuclear factor kappa light chain enhancer of activated B cells, CRP C reactive protein, CVDs cardiovascular diseases, NAFLD non-alcoholic fatty liver diseases Obesity, immune function, TNF-α, free fatty acids and insulin resistance During obesity, in adipose tissue or liver, macrophage infiltration causes chronic systemic inflammation resulting in the activation of immune cells. Macrophages are classified into two kinds; (1) classically activated (M1) macrophages that are related to microbicidal activity, (2) alternatively activated (M2) macrophages that are related to antiparasitic and allergic responses (Tateya et al. 2013). M2 macrophages have a role in IL-4 and IL-10 secretion, which is associated with insulin sensitivity; however, proinflammatory cytokines are secreted through M1 macrophages causes insulin resistance ( Fig. 2) (Tateya et al. 2013). M2 macrophages are activated in a lean state by Treg cells, TH2 cells, natural killer T cells, or eosinophils through the secretion of IL-4 or IL-10. Conversely, neutrophils or mast cells, TH1 cells, and B cells play a significant role in inducing M1 macrophages activation in the obese state by the elevated levels of TNF-α and interferon-gamma (IFNγ). Hypertrophied adipocytes increase the section of TNF-α and free fatty acids, which lead to the activation of M1 macrophages. Insulin resistance in obesity through the infiltration of macrophage and the activation of the immune cells is well established, and the role of TNF-α in this pathway is well known. During the progression of obesity, TNF-α expression is increased in adipose tissue, although attenuation of insulin resistance is occurred in the TNF-α deactivation (Hotamisligil et al. 1993). It has also been shown that TNF-α inhibits insulin receptor tyrosine kinase activity, which leads to suppression of insulin signaling (Hotamisligil et al. 1996). Thus, it can be hypothesised that increases in the TNF-α concentration in adipose tissue, which defined as inflammation, is the foundation of systemic insulin resistance (Tateya et al. 2013). The infiltration of macrophages primarily produces TNF-α into the adipose tissue. TNF-α is the major pro-inflammatory cytokine, and as an adipokine activates proinflammatory signal cascades (Hotamisligil et al. 1993;Uysal et al. 1997). Likewise, TNF-α inhibits the signalling of the insulin receptor; altogether, it seems that this molecule is a primary mediator between adipose tissue inflammation and insulin resistance (Weisberg et al. 2003;Fig. 2 Macrophage infiltration into adipose tissue which induced by obesity and leads to insulin resistance. (i) In lean state, neutrophils or mast cells, TH1 cells, and B cells induce most resident macrophages are M2 macrophages that contribute to insulin sensitivity by secreting IL-10. (ii) Adipocyte hypertrophy which emerges in hyperphagia and lack of exercise induces MCP-1 secretion to the circulation, causing engagement of circulating monocytes to adipose tissues. Activated M1 macrophages emerge from the infiltrated monocytes differentiate, which induce the production and secretionproinflammatorytory cytokines including TNFα, IL-6, and MCP-1. As a result, low-grade inflammation occurs in the adipose tissue and subsequently adiponectin level decreases. Insulin resistance in the liver occurs due to the inflammatory cytokine activity 1 3 Xu et al. 2003). Moreover, free fatty acids are typically raised in obesity since pro-inflammatory cytokine TNFcauses an increase in lipolysis in the adipose tissue (Tateya et al. 2013). These free fatty acids serve as ligands for the toll-like receptor 4 (TLR4) complex, which is necessary for innate immune cells for intruding pathogens recognition and triggering a suitable immune response (Shi et al. 2006). Free fatty acid-induced TLR4 signaling activated by FFA causes activation of M1 through the transcriptional factors, including AP1, NF-κB, and interferon-regulatory factor (IRF) family members (Poltorak et al. 1998;Shi et al. 2006). Activation of the above transcriptional factors drives M1 activation, which all occurs by TNF-α (Tateya et al. 2013).

Reduction of TNF-α, why phytochemicals?
Lifestyle modification, mainly focusing on a healthy diet and physical activity, is considered a first-line strategy to prevent and treat obesity, resulting in reduced inflammatory factors. However, several challenges and poor adherence to this approach resulted in a lack of effectiveness of this therapy (Bagherniya et al. 2018). Another option is anti-inflammatory drugs which are the specific agents for reduction of TNF-α are not wholly acceptable due to several limitations such as their side effects (Antoni and Braun 2002;Rainsford 2007;Scheinfeld 2004;Tateya et al. 2013;Vane and Botting 2003). Thus, future studies are warranted to find an appropriate pharmacotherapy. Meanwhile, recently, a growing body of evidence focusing on herbal medicine, phytochemicals and natural bioactive compounds as inexpensive, readily available and accessible agents with low amounts of adverse effects to reduce inflammatory markers such as TNF-α and simultaneously decrease insulin resistance, glucose intolerance, as well as reduction of unfavorable blood lipids and apolipoproteins (Alikiaii et al. 2021;Alikiaii et al. 2020;Bagherniya et al. 2020;Das and Das 2007;Ghasemian et al. 2016;He et al. 2015;Li et al. 2016;Mahdavi et al. 2020;Talebi et al. 2020;Zareie et al. 2020). In addition, herbal medicine is being recognized as an alternative therapy for the prevention and treatment of non-communicable diseases including, hypertension (Houston 2005(Houston , 2010(Houston , 2014, diabetes mellitus (Bahadoran et al. 2013;Davì et al. 2010;McCarty 2005), NAFLD (Bagherniya et al. 2018), andCVD (Alissa andFerns 2012;Badimon et al. 2010;Ramaa et al. 2006;Zuchi et al. 2010). In most of this evidence related to obesity, inflammation reduction was considered the primary mechanism mediating the beneficial effects of medicinal plants on non-communicable diseases. It seems that these natural products have favorable effects on inflammatory markers, particularly TNF-α, among overweight and obese individuals. The main aim of the current review is to summarize the results of the previous studies, in which the effects of phytochemicals and herbal bio-active compounds on serum TNF-α among subjects who at baseline had a mean BMI of above 25 (kg/m 2 ) (Table 1.). The potential mechanisms of the effectiveness of herbal bio-active compounds on inflammatory markers and their relation to obesity-related diseases are shown in Fig. 3.

Propolis
Propolis or bee glue is a resinous mixture produced by honey bees from diverse plants. This product is a mixture of phenols such as aromatic compounds, flavonoid, and polyphenols (Silva-Carvalho et al. 2015). Propolis has considerable beneficial effects on different conditions, including diabetes mellitus, NAFLD, atherosclerosis, oral and dental diseases, dermatological problems, allergies, gastrointestinal disorders, gynecological and neurological diseases (Farooqui and Farooqui 2012;Pasupuleti et al. 2017). Propolis with phenolic acid and flavonoid compounds has unique antioxidant activities (Kurek-Górecka et al. 2014). Moreover, propolis is popular for its immune-modulatory and anti-inflammatory effects (Borrelli et al. 2002;Ramos and Miranda 2007). Among the very diverse constituents of the propolis, it is proposed that quercetin, caffeic acid, naringenin, and caffeic acid phenethyl ester (CAPE) are the main ingredients with anti-inflammatory properties (Mirzoeva and Calder 1996;Ramos and Miranda 2007). These compounds of propolis suppressed the synthesis of prostaglandins and leukotrienes in macrophages. Likewise, ornithine decarboxylase myeloperoxidase activity, tyrosine-protein-kinase and NADPH-oxidase effects were inhibited by these constituents of propolis (Miyataka et al. 1997;Ramos and Miranda 2007). Other ingredients of propolis, such as ferulic acid, salicylic acid, galangin, and apigenin also considered as ingredients with anti-inflammatory properties (Krol et al. 1996). Propolis also inhibits nitric oxide (NO) production by macrophages which is another mechanism showing the anti-inflammatory activity of propolis (Ramos and Miranda 2007). A recent meta-analysis indicated that propolis supplementation significantly reduced IL-6, CRP and TNF-α levels (Jalali et al. 2020;Shang et al. 2020). In a study, overweight and obese patients with breast cancer received 250 mg propolis/twice per day or a placebo for 3 months while being treated with chemotherapy. Results indicated that in the placebo group, serum TNF-α significantly increased, whereas, in the propolis group, there were no significant changes (Darvishi et al. 2020). In another study, 80 patients with type 2 diabetes received Brazilian green propolis 226.8 mg/day or placebo. After 8 weeks, there were no significant changes in TNF-α in the study groups (Fukuda et al. 2015). In another study conducted on patients with type 2 diabetes, Chinese propolis (900 mg/day) was used for 18 weeks, in which, at the end of the study, there were no significant differences in TNF-α between groups (Gao et al. 2018). In another study, 100 patients with type 2 diabetes randomized to receive 1000 mg/day of Iranian propolis or placebo for 90 days, in which serum TNF-α significantly decreased in the propolis group compared with the placebo group (Zakerkish et al. 2019).

Ginger
Ginger, a member of the Zingiberaceae family, consists of several components, including gingerol, zingerone, shogaol, paradols, and β-bisabolene. This plant is used as a spice in both foods and beverages in Asian countries for thousands of years (Sahebkar 2011;Singletary 2010). This herb exhibits several unique beneficial effects on human health, such as its antioxidant, anti-inflammation, antimicrobial, antihypertensive, antidiabetic, cardioprotective, anticancer, antiemetic, chemopreventive, and gastroprotective (Ali et al. 2008;Baliga et al. 2011;Panahi et al. 2012). It is declared that most of these beneficial effects attributed to the ginger inflammatory responses, which might be mediated through inhibition of the activity of cyclooxygenase-1 (COX-1) and cyclooxygenase-2 (COX-2) resulted in prostaglandin synthesis suppression. In addition, the interleukins and TNF-α production in activated macrophages could be blocked by ginger (Grzanna et al. 2005;Liguori et al. 2018;Thomson et al. 2002). In a recent metaanalysis, which evaluated the effects of ginger on inflammatory markers based on clinical trials, it has been shown that in comparison to the control group, ginger consumption had a significant effect on the reduction of serum CRP, IL-6 and TNF-α (Jalali et al. 2020a, b). In a recent RCT, results indicated that supplementation with 1.5-g ginger/day for 12 weeks in overweight and obese patients with active rheumatoid arthritis had no significant effects on TNF-α level compared with placebo (Aryaeian et al. 2019). In another recent RCT, patients with type 2 diabetes with chronic periodontitis were asked to consume 2 g of ginger/day or a placebo for 8 weeks.
There was a significant reduction of TNF-α in the intervention group compared with placebo (Javid et al. 2019). In another Fig. 3 The potential mechanisms of the effectiveness of herbal bioactive compounds on inflammatory markers and cytokines and its relation to the obesity related diseases. Obesity, excess fat and adipose tissue increase the activity of LOX and COX-2. LOX increases the levels of LTE-4 and COX-2 increases the level of PGE2, which both are metabolites of poly unsaturated fatty acids which play a significant role in increasing inflammation. Obesity, and excess fat also increase the amounts of NF-Kb, which increases inflammatory cytokines. NF-Kb also increases the level of NO and increase the level of COX-2. These inflammatory factors significantly increase several obesity-related diseases. Herbal medicine suppresses the activity of LOX, COX-2 and NF-Kb, which resulted in a significant reduction in inflammatory markers. PG prostaglandin, COX cyclooxygenase, LOX lipoxygenase, TNF-α tumor necrosis factor-α, LT leukotriene, NO nitric oxide, iNOS inducible NO synthase, NF-Kb nuclear factor-KB, IL interleukin study, NAFLD patients supplemented with daily 1.5 g of ginger after 12 weeks; serum TNF-α did not change significantly compared with control patients (Rafie et al. 2020). In a previous clinical trial study, 46 obese women with diabetes were divided into 4 groups, including (ginger, aerobic exercise training with ginger, aerobic exercise training) to receive 1000 mg/day of ginger extract for 10 weeks. At the end of the study, no significant differences were observed on the TNF-α level in the four groups (Asadi et al. 2017). In an earlier study, patients with osteoarthritis received 1 g of ginger for 3 months, in which TNF-α was significantly reduced compared with the control group (Mozaffari-Khosravi et al. 2016).

Cinnamon
Cinnamon has been used since time immemorial as a food spice and flavoring with known medicinal properties in both traditional and modern science. Promising health benefits of cinnamon such as antioxidant, anti-inflammatory and insulin-sensitizing characteristics made it one the most popular herbs for prevention and treatment of the various pathologies (Bagherniya et al. 2018). It is asserted that cinnamon has antimicrobial, antioxidant, anti-inflammatory, antidiabetic and cardiovascular disease lowering properties (Gruenwald et al. 2010;Rao and Gan 2014;Sadeghi et al. 2019). In addition, its beneficial effects on some neurological diseases (Rao and Gan 2014;Zareie et al. 2020). It shows that cinnamon with several flavonoids might have an inhibitory role in nitric oxide production through suppression of NF-Κb activation, which mediated its anti-inflammatory effects on cinnamon (Lee et al. 2005;Rao and Gan 2014). In addition, it may reduce the activation of Src/spleen-tyrosine-kinase-(Src/Syk-) mediated NF-κB (Rao and Gan 2014;Youn et al. 2008;Yu et al. 2012). Suppression of the production of inducible nitric oxide synthesis (iNOS), COX-2, and nitric oxide in the central nervous system (CNS) might be another pathway of anti-inflammatory effects of cinnamon (Hwang et al. 2009;Rao and Gan 2014). A recent study using 3 g (3 × 1 g) of cinnamon for 8 weeks among patients with type 2 diabetes mellitus had no substantial effects on serum TNF-α compared with placebo (Davari et al. 2020). In another RCT, women with rheumatoid arthritis were asked to consume 2 g (4 × 500 mg) of cinnamon/day for 8 weeks.
There was no significant reduction in serum TNF-α with cinnamon consumption compared with placebo (Shishehbor et al. 2018).

Quercetin
Quercetin, a plant flavonoid that is found in a wide range of plants, fruits and vegetables. It has been shown that quercetin has beneficial effects on cardiovascular health, cancers, diabetes, eye disorders, allergic diseases, arthritis, neurological diseases, and many other diseases (Elumalai and Lakshmi 2016;Lakhanpal and Rai 2007;Patel et al. 2018;Serban et al. 2016). It is revealed that quercetin has substantial antioxidant properties (Alrawaiq and Abdullah 2014; Zhang et al. 2011). Preclinical studies have shown that the production of COX and lipoxygenase, which regularly cause inflammation induction, was inhibited by quercetin (Chen et al. 2016;Lee et al. 2010). In addition, quercetin dramatically inhibited proinflammatory cytokines in cultured fibroblasts (Chen et al. 2016;Yoon et al. 2012). Likewise, COX-2 and the nitric oxide and NF-κB production were suppressed by quercetin (Ramyaa and Padma 2014). Moreover, it is shown that quercetin suppresses the secretion of TNF-α and IL-6 in LPS-stimulated macrophages (Mueller et al. 2010). A recent meta-analysis showed that in comparison to the control group, quercetin consumption had no significant effect on CRP, IL-6, and TNF-α, which included 6, 5 and 4 clinical trials, respectively (Ou et al. 2020). In a previous crossover trial, 93 overweight or obese subjects with metabolic syndrome traits were asked to consume 150 mg quercetin/d for 6 weeks, in which serum TNF-α did not significantly change (Egert et al. 2009). In another study, 500 mg quercetin/da, and 1000 mg quercetin/day for 12 weeks had no effects on TNF-α concentration on healthy female subjects (Heinz et al. 2010). In another study, supplementation of 500 mg/ day of quercetin on post-myocardial infarction patients for 8 weeks showed a significant reduction of TNF-α level in the treatment group but not significantly different from the control group (F. Dehghani et al. 2020).

Curcumin
Curcumin is a bioactive compound of turmeric. Besides curcumin's various health benefits, widespread molecular targets have been shown to interact with curcumin (Esatbeyoglu et al. 2012; Mohammadi et al. 2013). The anti-inflammatory effects of curcumin are well-recognized attributable to its effects on the activity of COX-2, lipoxygenase, and inducible nitric oxide synthase iNOS enzymes and inhibition of inflammatory cytokines (Goel et al. 2008). In one study, 37 obese patients were recruited, and each of them received curcuminoids (1 g/day) or placebo and then crossed over. Each treatment period was 30 days with a 2-week wash-out interval. They found that serum levels TNFα did not alter (Ganjali et al. 2014). In another double-blind, randomized placebo-controlled trial, 84 overweight/obese patients with NAFLD diagnosed were randomized into 2 equal groups to receive either 2 40-mg capsules of nano-curcumin (NC) per day or placebo for 3 months. Results showed that in both NC and placebo groups, the mean difference of TNF-α decreased (Jazayeri-Tehrani et al. 2019). Also, Saadati and colleagues investigated the effects of either 1500 mg curcumin or placebo on 50 overweight/obese patients with NAFLD for 12 weeks in a randomized, double-blind trial. They found that TNF-α significantly reduced in both groups at the end of the study, but there was no significant difference between them (Saadati et al. 2019). Abdolahi et al. conducted a double-blind, randomized placebo-controlled trial on 74 overweight patients with episodic migraine. They were divided into four groups; (a) the group treated with 2500 mg ω-3 fatty acids and 80 mg nano-curcumin supplementation, (b) the group treated with 2500 mg ω-3 fatty acid supplementation, (c) the group treated with 80 mg nano-curcumin supplementation, and (d) the control group. After 2 months of intervention, they observed a significant reduction in the serum levels of TNF-α in the combination group, but no significant changes in other groups were observed (Abdolahi et al. 2017). In a randomized, double-blind placebo-control clinical trial,l 40 overweight/obese subjects with knee Osteoarthritis were allocated randomly to receive either 1500 mg/ day pure curcuminoids or a placebo for 6 weeks. At the end of the study, there were no significant differences in TNF-α levels between the two groups (Rahimnia et al. 2015).

Garlic
For several thousand years BC, Garlic (Allium sativum) has been used for medicinal purposes. The beneficial effects of garlic are due to its more than 2000 active ingredients that work synergistically. The most important compounds are different enzymes, including alliinase, sulphur-containing substances, such as diallyl sulfide and alliin, and enzyme products, such as ajoenes and allicin (Majewski 2014). The fructose-containing carbohydrates, sulfur compounds, free amino acids, protein and fibre, constitute the bulk of garlic's dry weight, and water is the major part of garlic (65%) (Rola et al. 2007). It also comprises high phosphorus levels, saponins, zinc, potassium, sulfur, moderate levels of selenium and Vitamins A and C, and low levels of sodium, calcium, manganese, magnesium, iron, and B-complex vitamins. The phenolic content is also high in garlic. A high concentration of garlic inhibits cytokine production in endothelial cells (Sharifi et al. 2010). The role of garlic in reducing the risk of cardiovascular disease and other inflammatory disorders have been confirmed in several studies. Evidence for the efficacy of the active components of garlic (e.g., diallyl sulfide) in lessening inflammatory factors, include IL-1β induced COX-2 upregulation, have been provided in cell culture studies (Dehghani et al. 2018). In a randomized, placebocontrolled clinical trial, 80 post-menopausal overweight and obese women with mild to moderate knee osteoarthritis were given either garlic (1000 mg tab twice daily), or placebo tabs, for 12 weeks. After 12 weeks, no significant changes in TNF-α concentration were observed within or between the two groups (S. Dehghani et al. 2018). In another randomized placebo-controlled trial conducted by Sharifi et al.,50 women, including 40 cases with metabolic syndrome and 10 normal female controls, were recruited. They were randomly allocated to two parallel treatment groups and received either garlic tablets (1.8 g/day; two 300 mg garlic tablets three times per day), or a placebo for 6 weeks. At the end of the study, serum TNF α did not change in the study groups (Sharifi et al. 2010). Van Doorn and colleagues investigated the effects of garlic powder on 90 overweight and smoker subjects. Patients were randomized into three parallel groups to receive 2.1 g/d garlic powder, 40 mg/d atorvastatin or placebo. After 3 months of intervention, none of the variables, including TNF-α, showed significant differences between the garlic-treated and placebo groups (van Doorn et al. 2006). Also, Xu et al. conducted another study on 51 healthy obese individuals to investigate the effect of 3.6 g/d doses of aged garlic extract (AGE) or placebo on inflammation biomarkers and immune function after 6 weeks of supplementation. At post-intervention serum TNF-α in participants supplemented with AGE were significantly lower than those with placebo capsules (Xu et al. 2018).

Flaxseed
The properties of flaxseed as a functional food is attributed to its components α-linolenic acid (ALA), soluble and insoluble fibres such as lignans with antioxidant and estrogen-like properties (Bloedon et al. 2008;Goyal et al. 2014). Because flaxseed is a rich source of ALA, it is known for its beneficial effects on cardiovascular diseases (e.g., atherosclerosis), diabetes, metabolic syndrome, hypertension, and dyslipidemia (Brant et al. 2012;Fukumitsu et al. 2010;Goyal et al. 2014;Hutchins et al. 2013;Pan et al. 2009;Ursoniu et al. 2015). Flaxseed oil, flaxseed lignan, or flaxseed supplementation markedly reduced serum TNF-α, CRP, IL-6, IL-1 β, glycosylated haemoglobin concentrations as well as increased insulin sensitivity in humans (Caughey et al. 1996;Hallund et al. 2008;Paschos et al. 2005;Zhang et al. 2008). In a prospective, single-blinded 42 days study, 27 overweight/obese men with cardiovascular risk factors were assigned to 2 groups with either low carbohydrates intake and 60 g of flaxseed powder per day or low carbohydrates intake and 60 g of raw rice powder per day. At the end of the study, levels of TNF-α were reduced only in the flaxseed intake group (Cassani et al. 2015). In a randomized, doubleblind, placebo-controlled trial with 60 overweight patients with diabetes and CHD, the participants were randomized into 2 groups to intake either 1000 mg flaxseed oil supplement or placebo twice a day for 12 weeks. Results showed that flaxseed oil supplementation down-regulated TNF-α in patients with diabetes and CHD (Hashemzadeh et al. 2017). Paschos et al. conducted a single-blind, parallel design intervention on 35 overweight, nondiabetic, dyslipidemic men. Participants were allocated into two groups to receive either 15 ml of flaxseed oil rich in ALA or 15 ml of safflower oil per day for 12 weeks. They found no changes in plasma TNF-α in the flaxseed oil versus the control group (Paschos et al. 2007). In another study, 44 patients with coronary artery disease were randomized to 12 weeks consumption of flaxseed (30 g/day) or usual care as control. A significant reduction in plasma TNF-α was observed in the flaxseed consumption group at post-intervention compared with the control (Khandouzi et al. 2019). Rhee and colleagues investigated the effects of 40 g/day ground flaxseed compared with wheat bran as a placebo on inflammatory biomarkers, using a randomized crossover design for 12 weeks with a 4-week washout period on nine obese glucose intolerant participants. They found no changes in plasma TNF-α at the end of the study (Rhee and Brunt 2011). In a single-blind clinical study, 75 overweight adolescents were divided into 3 groups to take 28 g/day of brown flaxseed, golden flaxseed or control in different preparations at school from Monday to Friday for 11 weeks. Although all groups showed increased levels of TNF-a, groups did not differ significantly on the values of TNF-α (Machado et al. 2015). In another randomized controlled study, 50 overweight and obese adults randomized to take lifestyle advice or lifestyle advice plus 30-g milled flaxseed every day for 12 weeks. They investigated that in both groups (flaxseed and control), TNF-α decreased significantly but decreasing was higher in the flaxseed group (Yari et al. 2019).

Ginseng
Panax ginseng Meyer root has been regularly utilized in East Asia for about 2000 years (Heo et al. 2016;Hong et al. 2016). Ginsenoside Rb1 is the most plentiful ginsenoside in ginseng that inhibits gene expression, encoding lipogenesis-inducing enzymes in rats diagnosed with fatty liver disease. Numerous studies have shown that ginseng might positively affect diabetes, stress, inflammatory, or hyperlipidemia (Bang et al. 2014;Heo et al. 2016;Hong et al. 2016;Jeong et al. 2018). Several ingredients in ginseng have anti-inflammatory effects (Im 2020;Ratan et al. 2020). In M1-polarized macrophages and microglia, pro-inflammatory cytokine and enzyme expressions were inhibited, demonstrating the anti-inflammatory mechanism of ginsenosides (Im 2020;Xue et al. 2020). Ginsenoside inhibited the expression of TNF-α, IL-1 and overcame TANK-binding kinase 1/IκB kinase ε/interferon regulative factor-3 and p38/ ATF-2 signaling (Huang et al. 2021;Im 2020). A previous meta-analysis, which analyzed the findings of seven clinical trials, indicated that ginseng consumption resulted in a significant reduction in serum TNF-α (Mohammadi et al. 2019). In 1 clinical trial study, 35 patients with NAFLD were asked to take Korean red ginseng (KRG) (ginsenosides Rg1 + Rb1 6.0 mg/g; 3,000 mg/day) (intervention group), and 31 patients received placebo, and both groups were on healthy eating with regular exercise for 3 weeks. There was a significant reduction of TNF-α in the intervention group compared to the control group . In another study, 72 diabetic patients were randomized to receive a placebo or 1500, 2000, or 3000 mg of ginseng for 8 weeks. At post-intervention, TNF-α significantly decreased in 1500 and 3000 groups compared with the control group (Yoon et al. 2012a, b).

Resveratrol
Resveratrol, a polyphenolic compound, has beneficial effects in several diseases, including CVD, diabetes, and other metabolic disorders (Saiko et al. 2008;Shakibaei et al. 2009;Shankar et al. 2007) despite some controversies (Chudzińska et al. 2021;Sahebkar et al. 2015). Some reports imply that resveratrol plays an important role in inhibiting inflammation, oxidative stress, carcinogenesis, and ROS generation (Xing et al. 2020). Resveratrol was reported to increase the mRNA expression of cytokine genes such as COX-2, TNFα, IL-1β, IL-8, and improve the levels of TNF-α and NF-κB mRNA. Reseveratrol reduced the activation of NF-κB and TNF-α (Xiao et al. 2021). In a double-blind, randomized controlled study, 58 patients with non-alcoholic fatty liver disease were assigned into 2 groups to receive a placebo or 300 mg daily resveratrol for 12 weeks. At post-intervention, resveratrol capsules significantly reduced TNF-α to a greater extent than placebo . In a previous clinical trial, 44 overweight or obese women were randomly assigned to receive 15% of their energy from grape seed oil (GSO) as an intervention group or 15% of their energy from sunflower oil as a control group 8 weeks. At the end of the study, the TNF-α level significantly reduced in the intervention group (Irandoost et al. 2013). In another study, 24 obese men were randomized into 2 groups to receive a placebo or 500 mg daily resveratrol for 4 weeks. There were no demonstrable changes in TNF-α levels in both groups (Poulsen et al. 2013). In another study, 40 polycystic ovary syndrome (PCOS) patients were randomized into 2 groups to receive 800 mg resveratrol/day or a placebo for 40 days. Serum TNF-α was significantly reduced in the intervention group, and this reduction was marginally significant compared with the control group (p = 0.056) (Brenjian et al. 2020). In another clinical trial, using 1 g resveratrol/day for 3 months was associated with a significant reduction in serum TNF-α in the rheumatoid arthritis patients compared with placebo (Khojah et al. 2018).

Green tea
Green tea polyphenols include anthocyanins, flavonoids, catechins, and phenolic compounds Sabu et al. 2002;Xia et al. 2019). Some studies have shown green tea's remedial effects in metabolic syndromes, type 2 diabetes, and repressing lipogenesis in hepatocytes Xia et al. 2019). Evidence has revealed polyphenols of green tea, such as epigallocatechin gallate (EGCG), to restrain matrix-metalloproteinase-2 and matrix-metalloproteinase-9 (Demeule et al. 2000;Hagiu et al. 2020). Green tea has anti-inflammatory activity by repressing the synthesis and inhibiting various proinflammatory mediators, nitric oxide synthase, peroxynitrite, reactive oxygen/nitrogen species, and COX-2 (Hagiu et al. 2020;Paquay et al. 2000;Zhong et al. 2012). A recent meta-analysis has shown that green tea consumption significantly reduces TNF-α according to six clinical trials (Haghighatdoost and Hariri 2019). In a clinical trial, 46 obese diabetic women were randomly recruited to 4 groups: (1) 1500 mg/d green tea, (2) aerobic exercise training with green tea, or (3) aerobic exercise training, or (4) control. After 10 weeks, three times a day, there were no significant effects on TNF-α levels between the four groups (Banitalebi et al. 2016). In another study, overweight middle-aged males who consumed 500 mg/day of green tea extract + endurance training had no effects on TNF-α compared with placebo and endurance training + placebo groups (Bagheri et al. 2020).

Anthocyanin
Anthocyanins are bioactive compounds belonging to polyphenols obtained in berry fruits, including blackberries, blueberries, and strawberries (Ding et al. 2020). Anthocyanins act as reactive oxygen species, free radical scavenger, and have several health benefits on obesity and diabetes (Ding et al. 2020). Studies published that foods containing anthocyanin appeared to lower inflammatory and oxidative stress biomarkers (Zhang et al. 2020). Anthocyanins have a vital role in reducing the electron-transfer reaction pathways, and these beneficial effects are because of their antioxidant capacity (Zhang et al. 2020). A recent meta-analysis study showed that dietary intervention with anthocyanins significantly reduced serum TNF-α based on the 32 randomized controlled trials (Fallah et al. 2020).
In a previous clinical trial study, 11 obese and overweight women were divided into 2 groups to receive 500 ml/day red orange juice. After 12 weeks, there was no significant effect on TNF-α level in baseline and after 12 weeks with consumption of red-orange juice (Azzini et al. 2017

Soy
Soybean is a good source for all essential amino acids discovered in animal proteins without cholesterol and limited saturated fat (Chatterjee et al. 2018 Most of the health benefits of soy proteins are their associated phytochemicals, principally isoflavones (Chatterjee et al. 2018). Some anticancer properties of soy protein with or without isoflavones show anti-inflammatory and antioxidant outcomes by inhibiting NF-κB and blocking the proinflammatory cytokines in an oxidative stress-inducible rat model (Chatterjee et al. 2018;Matemu et al. 2021). Soy milk digested with pepsin and pancreatin inhibited the production of IL-1β, nitric oxide, nitric oxide synthase, and COX-2 (Chatterjee et al. 2018;González-Montoya et al. 2018). A meta-analysis, which analyzed findings of 18 clinical trials, showed that soy isoflavones and soy isoflavones plus soy protein had no significant effects on serum TNF-α (Hariri et al. 2021). In a doubleblind, randomized controlled study, 63 overweight or obese individuals were supplemented with 2500 mg soybeans daily or 2500 mg starch in the control group for 8 weeks. Results showed that the level of TNF-α was significantly reduced in both groups with no statistical difference between them . In a recent double-blind, randomized controlled study, 100 hypercholesterolaemic healthy individuals were asked to consume soy supplements or allocated a placebo group. After 24 weeks of intervention, there was no significant effect on TNF-a between the two groups (Hermansen et al. 2005). In another study, 31 women with post-menopausal were randomized to receive three servings of soy [vanilla soymilk (244 mL) with 6 g protein] or dairy milk for 28 days matched by a single bout of downhill running. Results showed that soy and dairy milk consumption had no significant effects on the inhibition of muscle inflammation and proteolysis and did not attenuate up-regulation of exercise-induced changes (Serra et al. 2012). In a previous randomized longitudinal prospective cohort study, 87 healthy women with postmenopausal were divided into 2 groups to receive diet and exercise for the control group or diet, exercise, and intake of a soy isoflavones extract for the intervention group. After 6 months of intervention, the combination of diet, exercise, and intake of soy isoflavones had a significant decrease in TNF-a levels in both groups (Llaneza et al. 2011). In another study, 80 postmenopausal women were randomly recruited to 2 groups: Mediterranean diet and exercise for the control group, or this intervention with soy isoflavone extract for 24 months. After these interventions, results showed that the level of TNF-α significantly declined in both groups (Llaneza et al. 2012).
In a previous study, 41 men with hypercholesterolemia and women with postmenopausal were assigned to get a lowfat dairy food for the control group, and intervention group including foods including 2 soy protein phases, one high and the other low in isoflavones with the amount of 50 g and 52 g soy protein daily for high-isoflavones phase, and 73 mg and 10 mg isoflavone daily for low-isoflavones phase. After three 1-month phases, there were no significant changes in TNF-α in men and women (Jenkins et al. 2002).

Conclusion and future perspective
This review attempts to clarify the effectiveness of phytochemicals and herbal bio-active compounds on serum TNF-α in overweight and obese subjects. Evidence indicates that herbal medicine has favorable effects on the reduction of TNF-α concentration. However, the results were not uniform. Among the plants, ginger, ginseng, resveratrol, and flaxseed have more promising effects than other phytochemicals and herbal bio-active compounds. Nevertheless, in almost all of the reviewed studies, TNF-α was evaluated as the primary outcome. Moreover, it is not clear that reduction in TNF-α has led to disease treatment or not. More importantly, studies are too heterogeneous regarding the low sample size and the study population, dose of the herbs, and duration of the treatment, making it difficult to draw a definitive conclusion.
To accurately assess the efficacy of the phytochemicals and natural bioactive compounds on TNF-α in obese patients, larger clinical trials are warranted to determine the optimal dose and recognize the correct dosing regimen (dosing frequency and duration) to reap their full therapeutic potential.
Funding None.

Conflict of interest
The authors declare that they have no competing interest.