Introduction
Chronic low-grade or metabolic inflammation is considered a major hallmark of obesity and its associated comorbidities(Reference Olefsky and Glass1–Reference Kivimäki, Kuosma and Ferrie3), which are themselves risk factors for atherosclerosis and cardiovascular diseases (CVD)(Reference Kivimäki, Kuosma and Ferrie3,Reference Esser, Paquot and Scheen4) . By extension, the term metabolic inflammation will refer in the present review to the inflammation linked to metabolic disorders.
Adipose tissue is one of the organs strongly involved in the initiation of the metabolic inflammation process. Its expansion leads to the expression and endocrine secretion of several pro-inflammatory cytokines, such as tumour necrosis factor alpha (TNF-α), interleukin (IL) 6, IL1β(Reference Olefsky and Glass1,Reference Gregor and Hotamisligil2) and chemokines, including monocyte chemoattractant protein 1 (MCP1/CCL2) or regulated on activation, normal T cell expressed and secreted (RANTES/CCL5)(Reference Tourniaire, Romier-Crouzet and Lee5). Recently, several micro-RNA (miRNA) emerged as new actors in adipose tissue dysfunction during obesity(Reference Karkeni, Astier and Tourniaire6,Reference Landrier, Derghal and Mounien7) , and participate in the aetiology of insulin resistance(Reference Ying, Riopel and Bandyopadhyay8). This pro-inflammatory cascade plays a major role in the prevalence of cardiometabolic diseases. In particular, it initiates insulin resistance, which is linked to the pathophysiology of type 2 diabetes, as well as non-alcoholic fatty liver disease NAFLD(Reference Olefsky and Glass1,Reference Gregor and Hotamisligil2) . In turn, the liver produces acute phase proteins, such as C-reactive protein (CRP), serum amyloid A (SAA) and adhesion molecules, including soluble intercellular adhesion molecule type 1 (sICAM1) or vascular cell adhesion molecule (VCAM), which are markers of cardiovascular disease risk(Reference Pearson, Mensah and Alexander9,Reference Ridker and Morrow10) .
Although the mechanisms of inflammation remain complex, its role in promoting metabolic perturbations is widely accepted. Thus, it is largely accepted that targeting inflammation can be highly attractive in the prevention and/or treatment of metabolic diseases, which have reached epidemic proportions worldwide. Thus, the identification of nutritional strategies, based on food or bioactive compounds in foods, that blunt inflammation constitutes a sensible approach to limit the prevalence and severity of metabolic diseases. In agreement, fruit and vegetable consumption is strongly associated with the promotion of cardiometabolic health(Reference Shi, Morrison and Wiecha11,Reference Micha, Peñalvo and Cudhea12) , and may be related to limiting inflammation(Reference Calder, Ahluwalia and Brouns13–Reference Li, Zhang and Lu15). Among these fruits and vegetables, tomato is one of the most widely consumed worldwide. It is an excellent source of bioactive compounds with anti-inflammatory properties, including various carotenoids. Lycopene is one of the main carotenoids of tomato, and its concentration is particularly increased in tomato-derived products due to the microstructural changes induced by the industrial processes applied, which promote its bioavailability. Lycopene is found in different forms; all-trans form, mainly found in food, and cis-isomeric form, which is the dominant form in biological samples. Indeed, lycopene(Reference Rao, Ray and Rao16) and its metabolites, which have been described in various tissues and biological fluids(Reference Böhm, Lietz and Olmedilla-Alonso17), display antioxidant properties, at least in vitro (Reference Palozza, Parrone and Catalano18) and anti-inflammatory effects in several tissues and in several pathophysiological disorders, which are of significant importance in the context of cardiometabolic disorders. Thus, these anti-inflammatory effects are acknowledged to participate in the numerous lycopene-health effects(Reference Wang19–Reference Imran, Ghorat and Ul-Haq21), related to liver steatosis(Reference Wang19), cardiovascular diseases(Reference Muller, Caris-Veyrat and Lowe22–Reference Tierney, Rumble and Billings24), adiposity, obesity and diabetes(Reference Bonet, Canas and Ribot25–Reference Zhu, Chen and Bai28). Such anti-inflammatory effects are strongly related to its ability to interfere with several signalling pathways including NF-κB, Nrf2(Reference Bohn, Bonet and Borel29), but also nuclear receptors such as RAR(Reference Aydemir, Carlsen and Blomhoff30–Reference Bohn, de Lera and Landrier32). Tomatoes are also a source of antioxidant vitamins A, B, C and E, and polyphenols known for their powerful anti-inflammatory and antioxidant actions(Reference Wang19). Although the beneficial effect of tomato or tomato-derived product supplementation is generally attributed to lycopene, other tomato compounds also play an essential synergistic role in anti-inflammatory processes. Studies evaluating the effect of lycopene isolated from its matrix provide the advantage of considering lycopene’s specific mechanisms of action, independently of other compounds. Studies investigating the effect of tomato or tomato-derived product supplementation present the advantage of experimenting with a matrix that contains other bioactive compounds on human health, and is accessible to a large proportion of the population. However, in such studies, the clinical effects reported cannot be attributed to the effect of lycopene exclusively, but to the synergy of all the phytomicronutrients.
The aim of the present review is therefore to summarise findings regarding the anti- inflammatory potential of tomato (and tomato-derived products) and lycopene. The review focused on the relationship between tomato consumption and/or tomato-derived products and/or pure lycopene and inflammatory markers in studies involving healthy participants (without age restriction) or patients with cardiometabolic diseases (obesity, type-2 diabetes, metabolic syndrome, cardiovascular diseases). We excluded research articles that studied the relationship between lycopene and the inflammatory process related to pathological situations such as cancer, chronic obstructive pulmonary disease, asthma, kidney diseases and neurological disorders, to focus on relevant studies in the context of cardiometabolic diseases.
The review was performed using PubMed and Web of Science scientific databases, using tomato, tomato products, lycopene and inflammation as keywords. Publications that were not relevant or focused on metabolic inflammation were excluded. Original publications related to cell, animal or clinical studies, in English and prior to December 2022 were included.
Cross-sectional and prospective studies depicting the relationship between lycopene and metabolic inflammation in human
The role of raw tomato or tomato-derived products has not been investigated in cross-sectional and observational studies. Nonetheless, numerous observational studies have consistently shown a robust depicted a strong inverse association between circulating concentrations of lycopene and CRP(Reference Riccioni, D’Orazio and Palumbo33–Reference Kim, Yoe and Kim35), especially in extensive investigations like the National Health and Nutrition Examination Survey (NHANES)(Reference Ford, Liu and Mannino36–Reference Mazidi, Kengne and Katsiki38) (Table 1). In an Aboriginal population (n = 171), a reverse correlation was observed between plasma lycopene concentration and plasma CRP levels, and this relationship remained statistically significant even after adjustment for classical confounding factors(Reference Rowley, Walker and Cohen39). On the other hand, in a general Swedish population (n = 285), elevated plasma lycopene concentration was associated with lower plasma CRP and IL6 concentrations, but these associations became non-significant after adjustments for age, sex, alcohol intake, BMI, systolic blood pressure and total cholesterol(Reference Ryden, Garvin and Kristenson40). Lycopene concentration was also found to be inversely associated with IL6 concentration(Reference Walston, Xue and Semba41) and other endothelial markers, such as sICAM 1(Reference van Herpen-Broekmans, Klopping-Ketelaars and Bots42). In the Coronary Artery Risk Development in Young Adults study (CARDIA) prospective study, the plasma concentration of lycopene at baseline was inversely linked to sICAM1 protein level 15 years later(Reference Hozawa, Jacobs and Steffes43). Conversely, no relationship between plasma lycopene concentration and CRP level was observed in a population of middle-aged and older females (n = 2895)(Reference Wang, Gaziano and Norkus44) and in a population of patients with coronary artery diseases (CAD) and healthy controls (n = 113)(Reference Jonasson, Wikby and Olsson45).
Table 1. Cross-sectional studies investigating the relationship between plasma lycopene concentration and plasma inflammatory markers
M, male; F, female; OR, odds ratio; CI, confidence interval.
Altogether, these studies partly support the inverse association between plasma lycopene level and pro-inflammatory marker concentrations; however, it is mandatory to mention that such association does not demonstrate any causal role of lycopene in the control of inflammation, especially since lycopene is a marker of tomato-products consumption. Moreover, these associations could also result of several confounding factors since high consumers of tomato or lycopene are high consumers of other fruits and vegetables, which could provide several phytochemicals with anti-inflammatory effect.
Intervention studies in human depicting the effect of tomato-based products or lycopene on metabolic inflammation
Intervention involving tomato or tomato-based product supplementation
Some studies have been conducted in healthy participants. Most studies reported a reduction of pro-inflammatory marker plasma levels (Table 2). Indeed, tomato juice consumption for 2 weeks (daily dose of lycopene: 20·6 mg) decreased plasma CRP level(Reference Jacob, Periago and Bohm46) and MCP1 concentration decreased in twelve healthy participants consuming a vegetable soup constituted mainly of tomato (500 ml/d for 14 d)(Reference Sanchez-Moreno, Cano and de Ancos47), but did not modulate TNFα and IL1β or TNFα, IL6 and IL1β, respectively. Tomato juice consumption (280 ml/d, containing 32·5 mg lycopene) for 8 weeks also reduced MCP1 concentration and increased adiponectin concentration in young females (n = 30)(Reference Li, Chang and Huang48). Recently, the consumption of sofrito (Mediterranean tomato-based sauce; 240 g/70 kg body weight) in twenty-two healthy males was associated with a decrease in plasma TNFα and CRP concentration (no modification regarding IL6 concentration), 24 h after consumption(Reference Hurtado-Barroso, Martínez-Huélamo and Rinaldi de Alvarenga49). However, one study reported no impact on plasma concentrations of ICAM and CRP in 103 healthy participants submitted to a tomato-rich diet (300 g/d for 1 month)(Reference Blum, Monir and Khazim50), while another one depicted negative result (i.e. increase of plasma TNFα and lack of modulation of IL4, after 330 ml/d tomato juice supplementation for 2 weeks in 22 healthy males)(Reference Watzl, Bub and Briviba51).
Table 2. Clinical trials that studied the effect of supplementation with tomato or tomato-derived products on inflammation biomarkers
↘, decrease of the measured parameter ; ↗, increase of the measured parameter; NA, data not available.
The anti-inflammatory effect of tomato-based products has been investigated during post-prandial metabolic inflammation. In this particular context, consumption of tomato products reduced IL6 plasma level in healthy females and males (n = 25) during the post-prandial phase(Reference Burton-Freeman, Talbot and Park52). Similarly, different types of tomato-rich diets (raw tomatoes, tomato sauce or tomato sauce with olive oil) were evaluated regarding post-prandial plasma inflammatory response in healthy participants (n = 40). Such regimens led to a decrease in MCP1 concentration and to an increase in IL10 concentration (anti-inflammatory interleukin) following consumption of the three diets, a decrease in IL18 concentration following tomato sauce consumption and a reduction of IL6 and VCAM concentration following consumption of tomato sauce plus olive oil(Reference Valderas-Martinez, Chiva-Blanch and Casas53).
In heathy participants with overweight, a cocktail of dietary anti-inflammatory products, including tomato extracts, increased plasma adiponectin concentration (known to display anti-inflammatory activity(Reference Ruhl and Landrier54)) but did not modify CRP levels(Reference Bakker, van Erk and Pellis55).
Several studies have also been conducted in patients with metabolic disorders (obesity, type-2 diabetes or metabolic syndrome). One study conducted in patients with type-2 diabetes (n = 15) did not report reduction of CRP, ICAM or VCAM plasma concentrations after tomato juice supplementation for 4 weeks (300 ml/d) compared with placebo(Reference Upritchard, Sutherland and Mann56). Conversely, 20 d of supplementation with 330 ml/d of tomato juice for significantly decreased plasma IL8 and TNFα concentrations in participants with overweight and IL6 levels in participants with obesity(Reference Ghavipour, Saedisomeolia and Djalali57). Patients with metabolic syndrome who consumed tomato juice four times a week for a duration of 2 months experienced a comparable decrease in their plasma concentrations of IL6 and TNFα(Reference Tsitsimpikou, Tsarouhas and Kioukia-Fougia58). In a population of twenty-seven participants at high cardiovascular risk, a supplementation for 4 weeks with 200 ml or 400 ml of tomato juice resulted in a decrease of ICAM1 and VCAM1 concentrations, without modification of CRP or IL8(Reference Colmán-Martínez, Martínez-Huélamo and Valderas-Martínez59). Finally, in fifty-two obese children with fatty liver, the assignment to tomato sauce for 60 d increased adiponectin plasma concentration and decreased leptin plasma concentration(Reference Negri, Trinchese and Carbone60).
Altogether, these studies provide arguments in favour of a limitation of metabolic inflammation under tomato-based product supplementation, especially in participants with metabolic inflammation at baseline of the study. In agreement, a recent meta-analysis found a reduction in IL6 levels after tomato supplementation (standardised mean difference −0·25; p = 0·03)(Reference Cheng, Koutsidis and Lodge61). Nevertheless, we have to keep in mind that such results have been obtained following the consumption of tomato sauces or juices, which include several other ingredients (vegetable extracts, olive oil, oregano, basil extracts, etc.). Thus, the anti-inflammatory effect depicted in these studies may not be entirely attributed to tomato-derived products. Indeed, the presence of many other phytochemical-rich extracts could be involved in the anti-inflammatory response. In addition, if the anti-inflammatory effect has been characterised at the systemic level in these trials, it is not clear if such response is only due to circulating immune cells or may be related to other cells, tissues or organs. To our knowledge, no data are available regarding the specific role of tomato on the inflammation process in human organ or tissue; however, the fact that adiponectin and leptin plasma concentrations were modulated in some studies suggests that adipose tissue inflammatory status may be directly or indirectly impacted by tomato supplementation.
Another important issue relies on the identification of bioactive compounds mediating the tomato and tomato-derived product anti-inflammatory effects. We cannot exclude that the anti-inflammatory effect is mainly due to a cocktail effect of phytochemicals (including carotenoids, phenolic compounds, vitamins, etc.). Nevertheless, lycopene is classically considered as the major bioactive compound of tomato. Even if the correlations between the anti-inflammatory effects highlighted in clinical trials and plasma lycopene concentrations are sometimes conflicting(Reference Hurtado-Barroso, Martínez-Huélamo and Rinaldi de Alvarenga49,Reference Colmán-Martínez, Martínez-Huélamo and Valderas-Martínez59) , the specific effect of lycopene is traditionally suggested, based on its anti-inflammatory and antioxidant activity shown in in vitro and in animal studies.
Intervention studies involving lycopene supplementation
Lycopene supplementation for 1 week (Lyc-O-Mato; 80 mg/d lycopene) did not reduce post-prandial inflammation markers (CRP, ICAM or VCAM) in young healthy participants (n = 27)(Reference Denniss, Haffner and Kroetsch62) (Table 3). Among participants with obesity (n = 16), lycopene supplementation (4 weeks, Lyc-O-Mato, 30 mg/d) did not bring about any changes in the plasma levels of inflammatory markers (TNFα, IL6, CRP)(Reference Markovits, Ben Amotz and Levy63). Similarly, no impact of 1-month lycopene supplementation (7 mg/d) was reported regarding CRP of patients with hypertension(Reference Petyaev, Dovgalevsky and Klochkov64). In smoking (n = 12) and non-smoking (n = 15) participants, 2 weeks of supplementation (Lyc-O-Mato, 14·64 mg of lycopene/d) had no effect on circulating TNFα and IL2 levels compared with placebo(Reference Briviba, Kulling and Moseneder65). The consumption of 7 mg/d lycopene for 2 months in patients with CVD and healthy participants (n = 36 in each group) did not modulate CRP(Reference Gajendragadkar, Hubsch and Maki-Petaja66). In 224 middle-aged participants with moderate overweight, 12 weeks of supplementation with 10 mg/d lycopene or tomato-rich diet did not modify CRP, IL6 and ICAM plasma concentrations(Reference Thies, Masson and Rudd23). In patients with coronary vascular disease, 7 mg/d lycopene provided by microencapsulated lycopene for 30 d did not modify CRP plasma concentrations but reduced non-classical inflammatory parameters such as Chlamydia pneumonia IgG and oxidative stress as revealed by oxidised LDL(Reference Petyaev, Dovgalevsky and Klochkov67).
Table 3. Clinical trials that assessed the effect of supplementation with lycopene on blood biomarkers of inflammation
↘, decrease of the measured parameter ; ↗, increase of the measured parameter.
In contrast, a study indicated that in overweight middle-aged population (n = 54), consuming a lycopene-rich diet (providing 224–350 mg lycopene/week) for 12 weeks resulted in a substantial 36% decrease in SAA concentration within the HDL3 fraction. In the same study group, a 12-week lycopene supplementation regimen (using capsules, 70 mg/week) led to a reduction in SAA levels in both the plasma and HDL3 fraction(Reference McEneny, Wade and Young68). Similar results, that is, a decrease of serum and HDL3 under lycopene or high tomato diet has been recently published in a larger panel of volunteers (n = 225; 40–65 years and BMI between 18·5 and 35 kg/m2)(Reference McEneny, Henry and Woodside69). In healthy men (n = 126), the intake of lycopene (6 or 15 mg/d for 8 weeks) resulted in decreased plasma concentrations of ICAM and VCAM in both groups. Furthermore, the group receiving a daily dosage of 15 mg also showed a reduction in CRP levels as well(Reference Kim, Paik and Kim70). Among twenty-six healthy volunteers, the administration of Lyco-O-Mato, containing 5·7 mg of lycopene for a period of 26 d led to a 34·4% reduction in TNFα production from whole blood following an in vitro lipopolysaccharide (LPS) challenge(Reference Riso, Visioli and Grande71).
Altogether, these intervention studies using lycopene mostly failed to demonstrate a beneficial anti-inflammatory impact at the systemic level. As previously evoked, we cannot exclude that lycopene supplementation could reduce organ or tissue inflammatory status, but to our knowledge, no data are presently available. Nevertheless, current data are not in favour of an anti-inflammatory effect of pure lycopene in human, whereas observational studies clearly support such assumption. It is noteworthy that, in observational studies, plasma lycopene level is a good marker of tomato and tomato-derived product consumption. Thus, the association between high lycopene and low inflammatory status is more in favour of an effect of tomato and derived product consumption on inflammation, via a combination of phytochemicals, rather than solely a direct effect of lycopene.
Animal studies depicting the effect of tomato product and/or lycopene consumption on metabolic inflammation
Effect on inflammation in obesity and associated comorbidities context
The anti-inflammatory effects of the consumption of tomato-based products, especially tomato powder, or lycopene have been extensively studied in the context of obesity and its associated comorbidities (Table 4). Most studies were based on diet-induced obesity (DIO), using different types of regimens, supplemented or not with lycopene or tomato powder.
Table 4. Animal trials that assessed the effect of supplementation with lycopene and/or tomato products on biomarkers of inflammation
↘, decrease of the measured parameter; ↗, increase of the measured parameter.
The high fat/high cholesterol (HF/HChol) diet has been regularly used in DIO animal models. It is reported that lycopene, isolated from algae or tomato, significantly reduced the liver steatosis and blunted atherosclerotic depot in the aorta inflammation in rats (decreased ceruloplasmine, CRP, myeloperoxidase, decreased PBMC 15-lipoxygenase and cyclo-oxygenase activities)(Reference Renju, Kurup and Saritha Kumari72). This study points out that lycopene from algae gave more favourable results than lycopene isolated from tomatoes, implying a potential synergistic effect with other compounds of algae. A reduction of inflammatory status (TNFα, IL1β, ICAM and p65) in kidneys and heart of male rats fed a high-cholesterol diet supplemented with lycopene (50 mg/kg body weight (BW)/d) was also reported(Reference Albrahim73). Nevertheless, other studies using HF/HChol diet did not observe local and/or systemic anti-inflammatory effects following tomato product/lycopene consumption in rats(Reference Bernal, Martin-Pozuelo and Lozano74,Reference Martin-Pozuelo, Navarro-Gonzalez and Gonzalez-Barrio75) . It is noteworthy that, in these studies, the HF/HChol diet does not induce any pro-inflammatory response (at the plasma and liver level), making the demonstration of a lycopene-mediated anti-inflammatory effect unlikely. A similar observation was made in HF/HChol-fed mice supplemented for 12 weeks with 9% or 17% of dried tomato peel (Reference Zidani, Benakmoum and Ammouche76), where the dose–response effect was not obvious and only insulin sensitivity was restored.
Others model of DIO, mixing high-fat diet plus sucrose and supplemented or not with lycopene, have been used in rats. Under these conditions, lycopene supplementation reduced insulin resistance-related parameters (plasma insulin, glycaemia) and blunted inflammation at the systemic level (decreased plasma concentrations of leptin, resistin and IL6 but not TNFα), in the adipose tissue (decreased mRNA expression of Leptin, Resistin, Mcp1 and Il6 but not Tnfa)(Reference Luvizotto Rde, Nascimento and Imaizumi77) and in the brain(Reference Yin, Ma and Hong78). Pierine et al. investigated in DIO rats (rats were already obese at the beginning of the supplementation), the impact of a supplementation with lycopene. Although no modulation of plasma TNFα level was observed, in kidney its production decreased(Reference Pierine, Navarro and Minatel79), suggesting a beneficial impact on kidney oxidative stress and inflammation associated with obesity.
Several studies have been implemented in DIO models. For example, Bahcecioglu et al. supplemented rats with 2 and 4 mg/kg BW/d of lycopene for 6 weeks and observed an improvement of insulin sensitivity, liver steatosis and inflammation with the lower dose of lycopene (2 mg/kg BW/d)(Reference Bahcecioglu, Kuzu and Metin80). The dose of 4 mg/kg BW/d of lycopene led to a lower reduction of those parameters, supporting the idea of an optimal dosage of lycopene to blunt metabolic disorders. Similar hepatic limitation of inflammation (TNFα and IL1β) was observed in male rats subjected to HF diet and supplemented with lycopene at 25 or 50 mg/kg BW/d)(Reference Albrahim and Alonazi81). In female rats, 10 weeks of lycopene supplementation with 20 or 40 mg/kg BW/d reduced proinflammatory markers (Il1b, p65 and Il6 mRNA) in the heart of obese rats and induced the expression of Il10 (Reference Ugwor, Ugbaja and James82). Similar reduction of inflammation (TNFα, MCP1 and IL6) in the heart of obese male rats with cardiac dysfunction was observed after a supplementation of 10 weeks with tomato oleoresin (equivalent to 10 mg/kg BW/d)(Reference Ferron, Francisqueti-Ferron and Silva83). In DIO mice, Singh et al. in a very elegant study showed the anti-inflammatory effects of lycopene (10 mg/kg of diet) but also the limitation of adiposity, hyperinsulinaemia and insulin resistance. Authors also reported a limitation of liver inflammation and steatosis (with decreased concentrations of IL6, TNFα, NF-κB and TLR4)(Reference Singh, Khare and Zhu84) and a reduction of plasma inflammatory markers (TNFα, IL1β) in lycopene-supplemented group(Reference Singh, Khare and Zhu84). Similarly, we reported that supplementation for 12 weeks with lycopene but also tomato powder (providing the same amount of lycopene) reduced hepatic steatosis, reduced inflammation and improved hepatic lipid metabolism in a DIO mouse model(Reference Fenni, Hammou and Astier85). The hepatic anti-inflammatory effect of lycopene supplementation has been further confirmed(Reference Wang, Zou and Suo86,Reference Ni, Zhuge and Nagashimada87) . In this last study, authors demonstrate that the effects of lycopene are associated with a reduction of Kupffer and T cells (both helper and cytotoxic), a reduction of pro-inflammatory cytokine expression (Il6, Tnfa, Il1β) and a reduction of NF-κB, JNK and p38 signalling pathway activation (Fig. 1)(Reference Ni, Zhuge and Nagashimada87). In addition, the hepatic anti-inflammatory effect of tomato powder was confirmed in Bco1−/−/Bco2−/− mice, possibly through a Sirt1/AMPK pathway-dependent mechanism(Reference Li, Liu and Fu88).
Fig. 1. Transcription factors activated or inhibited by lycopene or metabolites, leading to anti-inflammatory effects in different organs or cell types. NF-kB, nuclear factor kappa B; JNK, c-Jun N-terminal kinase; NRF2, nuclear factor (erythroid-derived 2)-like 2; PI3K, phosphatidylinositol-3-kinase; ERK, extracellular signal-regulated kinase; RAR, retinoic acid receptor; PPARγ, peroxisome proliferator-activated receptor γ; Sirt1, Sirtuin 1.
The study by Singh et al. mentioned earlier also presents a novel finding, suggesting that lycopene may influence gut microbiota and dysbiosis associated with obesity, as indicated by changes in caecal bacterial abundance and the production of short-chain fatty acids(Reference Singh, Khare and Zhu84). Therefore, the potential attenuation of dysbiosis and the preservation of gut and colon epithelial integrity, both linked to obesity and insulin resistance, might contribute to the broader systemic anti-inflammatory effects of lycopene. Such observation of an impact of lycopene on intestinal permeability has been further confirmed(Reference Wang, Suo and Zhang89), as well as the effect of tomato powder on microbial richness(Reference Li, Liu and Fu88).
Effect on adipose tissue inflammatory status
Adipose tissue and liver constitute the main lycopene storage tissues, and respectively contain 60% and 30% of the total lycopene body stores(Reference Böhm, Lietz and Olmedilla-Alonso17,Reference Landrier, Marcotorchino and Tourniaire26,Reference Chung, Ferreira and Epstein90,Reference Moran, Erdman and Clinton91) . In the liver, lycopene uptake may be mediated via chylomicron-remnant receptors (LDL-receptor (LDLR), LDL-receptor related protein 1 (LRP1) and heparan sulphate proteoglycans (HSPGs))(Reference Dallinga-Thie, Franssen and Mooij92). In adipose tissue and adipocytes, the lycopene uptake is mediated by cluster of differentiation 36 (CD36)(Reference Moussa, Gouranton and Gleize93) and seems to be independent of its physico-chemical properties(Reference Sy, Gleize and Dangles94). Lycopene is then stored in lipid droplets and membranes of adipocytes(Reference Gouranton, Yazidi and Cardinault95).
The effect of lycopene and tomato powder supplementation was evaluated on adipose tissue inflammation. A down-regulation of genes encoding pro-inflammatory markers, including cytokines, adipokines, acute phase proteins, chemokines and metalloproteinases (Mmp3 and Mmp9) and an up-regulation of genes encoding anti-inflammatory proteins were reported(Reference Fenni, Hammou and Astier85). Some of these regulations were confirmed at the protein level (IL6, TNFα, MCP1, CCL5). We also observed that the phosphorylation levels of p65 and IκB were reduced under lycopene and tomato powder supplementation. Similar anti-inflammatory response under lycopene effect in adipose tissue was observed in other studies(Reference Wang, Suo and Zhang89,Reference Chen, Ni and Nagata96) . In addition, Chen et al. reported a reduction of M1 macrophages (pro-inflammatory macrophages) together with an increase of M2 macrophages (anti-inflammatory macrophages), and a decrease of p38 phosphorylation(Reference Chen, Ni and Nagata96), whereas Wang et al. reported an induction of autophagy in adipose tissue(Reference Wang, Suo and Zhang89). Altogether, these data suggest that the anti-inflammatory effect of lycopene and tomato powder on adipose tissue rely on their ability to inhibit NF-κB, p38 and JNK signalling and/or to restore autophagy in adipose tissue. Nevertheless, we cannot exclude that this anti-inflammatory effect was strictly due to the reduced adiposity.
Since apo-10′-lycopenoids were proposed to be major BCO2-mediated metabolites of lycopene(Reference Harrison and Kopec97), the effect of apo-10′-lycopenoic acid on liver steatosis was studied in the Ob/Ob mouse model(Reference Chung, Koo and Lian98). Apo-10′-lycopenoic acid supplementation diminished liver steatosis via a SIRT1-dependent mechanism leading to the inhibition of NF-κB inflammatory signalling(Reference Schug and Li99). The effect of apo-10′-lycopenoic acid was also reported in a high-fat diet mouse model, where this molecule blunted pro-inflammatory response in the liver through the modulation of SIRT1 activity(Reference Ip, Hu and Liu100). The study on the efficacy of lycopene compared with apo-10′-lycopenoic acid in reducing liver steatosis and inflammation in DIO Bco2−/− mice has been documented. Ip et al. observed that these two compounds operated through distinct mechanisms, with apo-10′-lycopenoic acid influencing SIRT1 activity and expression in the liver, whereas lycopene exerted its effects at the mesenteric fat level by modulating the actions of PPARα and PPARγ(Reference Ip, Liu and Lichtenstein101).
The anti-inflammatory effect of apo-10′-lycopenoic acid has been investigated in adipose tissue(Reference Gouranton, Aydemir and Reynaud102). We reported that apo-10′-lycopenoic acid transactivated RAR in vivo (Reference Gouranton, Aydemir and Reynaud102). Such activation of RAR has been associated to a decrease of inflammatory markers in adipose tissue, via a NF-κB deactivation(Reference Karkeni, Bonnet and Astier103), suggesting that putative metabolites of lycopene may also display efficiency regarding inflammatory process in adipose tissue.
Altogether, data generated in animal model support the anti-inflammatory effect. It is somehow important to keep in mind that some studies (especially those with lycopenoids) have been conducted with huge amount of lycopene or tomato-derived products, which sometimes makes the physiological relevance questionable. It is noteworthy that the dose of 10 mg of lycopene per kilogram of animal body weight could be of physiological interest in animal studies. Such dose is equivalent to 0·81 mg/kg body weight in human (taking classical conversion factor used to translate doses from mice to human). Thus, for a 70 kg human, it represents 56 mg lycopene per day. Based on a mean lycopene content in tomato paste approximately 1500 mg/kg(Reference Rao, Ray and Rao16), it represents approximatively 300 g of tomato paste to obtain 56 mg of lycopene. Such quantity corresponds to doses classically used in human intervention studies (Table 2) and can thus be considered as relevant for human health.
In vitro studies
The cell-autonomous anti-inflammatory effects of lycopene and/or tomato-based products have been evaluated in several cell types linked to metabolic disorders including endothelial cells, immune cells, hepatocytes or adipocytes (Table 5).
Table 5. In vitro experiments that assessed the effect of supplementation with lycopene and/or tomato products on biomarkers of inflammation
↘ : decrease of the measured parameter ; ↗ increase of the measured parameter.
Although the ability of lycopene to reduce pro-inflammatory response in hepatocytes has been recently demonstrated to occur via mechanisms involving indirect activation of Sirt1/AMPK and inactivation of NF-kB signalling (Fig. 1)(Reference Zhu, Liu and Shen104), data remain scarce in hepatocytes. Conversely, the anti-inflammatory effect of lycopene towards the gene expression of cytokines and chemokines has been demonstrated in adipocytes (both human and murine)(Reference Gouranton, Thabuis and Riollet105), and further confirmed in human adipocytes(Reference Warnke, Jocken and Schoop106). More recently, we compared the effect of the two main isomers of lycopene (i.e. all-E and 5-Z-lycopene) and reported similar effect in terms of limitation of cytokine, chemokine and acute phase protein expression as well as p65 dephosphorylation(Reference Fenni, Astier and Bonnet107). Such impact on the phosphorylation level of p65 agreed with the initial report of the deactivation of NF-κB signalling pathway(Reference Gouranton, Thabuis and Riollet105). Similar inhibition of cytokine and chemokine expression was reported with apo-10′-lycopenoic acid,(Reference Gouranton, Aydemir and Reynaud102). Finally, we reported that lycopene limited the induction of TNFα mediated by LPS in macrophages through an NF-κB and JNK deactivation and subsequent macrophage migration in vitro (Reference Marcotorchino, Romier and Gouranton108). As a result, lycopene reduced the mRNA levels of cytokines, acute-phase proteins and chemokines induced by macrophages in adipocytes. Likewise, tomato extracts blunted inflammation and the production of pro-inflammatory cytokines during the adipocyte and macrophage crosstalk(Reference Kim, Mohri and Hirai109).
In macrophages, lycopene (Reference Zou, Feng and Ling110–Reference Feng, Ling and Duan112) and tomato aqueous extracts (Reference Schwager, Richard and Mussler113,Reference Navarrete, Alarcon and Palomo114) suppressed the LPS-mediated induction of pro-inflammatory cytokines and interleukins (including TNFα, IL1β, IL6) in RAW264·7 cells, possibly via an NF-κB pathway inhibition(Reference Hadad and Levy111). In THP-1 cells, lycopene exhibited a preventive effect against the activation of pro-inflammatory pathways mediated by oxysterols, either through NF-κB inhibition and an elevation in PPARγ expression(Reference Palozza, Simone and Catalano115), or by PMA(Reference Makon-Sébastien, Francis and Eric116). Apo-10′-lycopenoic acid and apo-14′-lycopenoic acid, two lycopene metabolites, also suppressed the activation of NF-κB mediated by cigarette smoke extract in THP-1 macrophage cells(Reference Catalano, Simone and Cittadini117) and the expression of MMP9(Reference Palozza, Simone and Catalano118).
The impacts of lycopene and tomato-based products have been investigated in various vascular endothelial cell models, including HUVEC cells. In these cell lines, tomato oleoresin and pure lycopene notably inhibited the TNFα-mediated expression of multiple adhesion molecules, such as ICAM1 and VCAM1, by inhibiting NF-κB signalling, marked by a decrease in IκB protein levels and in p65 phosphorylation levels (Reference Armoza, Haim and Bashiri119–Reference Ucci, Di Tomo and Tritschler121). Tomato extracts also reduced pro-inflammatory cytokines and chemokines (Tnfa and Il8) gene expression and induced anti-inflammatory cytokines (Il10) gene expression, mediated by TNFα stimulation, probably through an inhibition of NF-κB signalling (Reference Hazewindus, Haenen and Weseler122,Reference Di Tomo, Canali and Ciavardelli123) . Interestingly this effect was linked to a restraint in monocyte migration, probably related to a decrease of chemokine expression. Similarly, tomato aqueous extracts dampened the production of a wide array of interleukins and chemokines(Reference Schwager, Richard and Mussler113). Beside its impact on NF-kB signalling, lycopene was able to mediate anti-inflammatory action through deactivation of several inflammatory-linked cell signalling, such as ERK, p38(Reference Chen, Lin and Yang124), PI3K/AKT or NRF2(Reference Sung, Chao and Chen125). If TNFα has been largely used as pro-inflammatory signal, other pro-inflammatory stimuli including HMGB1 (high mobility group box 1(Reference Lee, Ku and Bae126)) and LPS(Reference Wang, Gao and Yu127) have been tested regarding the ability of lycopene to blunt the pro-inflammatory responses.
Altogether, data generated in vitro also support the putative anti-inflammatory effect of lycopene, even if it is important to remind that some studies have been conducted with extremely high concentrations of lycopene, which presents a very low solubility, a high hydrophobicity and low stability due to its high oxidability. In addition, the physiological relevance of some studies that incubate non-intestinal cells with tomato extracts remains questionable.
Conclusions and perspectives
Based on the reported studies, the anti-inflammatory effect of tomato-derived products and/or lycopene is supported by at least in vitro and animal studies, with several limitations evoked throughout the text. Indeed, tomato-derived products or lycopene exert inhibition of the expression and secretion of cytokines and interleukins in several cell models related to cardiometabolic health and impact several signalling pathways that could participate in the overall effect including NF-κB. Animal studies also support these data, notably in the liver, in adipose tissue and in some cases at the plasma level, where a reduction of the inflammatory status has been reported in vivo. Most of the data generated in the context of obesity and its comorbidities indicate that lycopene, and in certain instances, tomato-based products, exert a positive impact on inflammation in crucial organs and are also linked to enhancements in metabolic disorders. Nevertheless, it is mandatory in future studies to implement physiologically relevant studies in terms of quantity of tomato-derived products or lycopene administrated to animals or cell cultures.
Notably, this anti-inflammatory property of tomato-based products and/or lycopene has been partially corroborated in specific clinical investigations at the systemic level, particularly using tomato products. It is noteworthy that the effect of isolated lycopene on inflammatory status is still not clear. However, when tomato-based products are provided to participants with elevated level of inflammation (overweight/obesity or during post-prandial inflammation), it becomes more evident that this dietary approach may offer advantages in managing inflammation. In line with this, it is important to specify that the health benefits of tomato consumption are generally attributed to lycopene(Reference Canene-Adams, Campbell and Zaripheh128). Nonetheless, these findings indicate that the alignment between lycopene supplementation and tomato-based product consumption is not flawless in terms of their anti-inflammatory properties, in agreement with prior suggestions(Reference Burton-Freeman and Sesso129).
Many significant questions remain unanswered. It remains uncertain whether it is lycopene itself or its metabolites, and which specific metabolites, that are responsible for the effects observed. Indeed, the activity of lycopene metabolites to control inflammation has been poorly investigated so far. It is also noteworthy that in the case of tomato product consumption it cannot be excluded that the reported effects are not only due to lycopene but could be related to other molecules found in tomato products. Indeed, tomato contains several other bioactive phytochemicals, including β-carotene, vitamin C and vitamin E, but also phenolic compounds such as hydroxycinnamic acid derivates and flavonoids and glycoalkaloids such as tomatine that have been demonstrated to mediate biological, including anti-inflammatory effects(Reference Chaudhary, Sharma and Singh130). Hence, we could speculate that all these phytochemicals display additive or synergic effect together with lycopene.
Additionally, lycopene bioavailability, and probably that of its metabolites, displays a large interindividual variability, due to many host-related factors (Reference Bohn, Desmarchelier and Dragsted131,Reference Desmarchelier, Landrier and Borel132) . This can act as a confounding variable in human studies, thereby blurring the relationship between lycopene consumption, from food or supplements, and metabolic inflammation. All these questions require further investigations in vitro, in animal models, but also in dedicated and adapted clinical trials.
So far, it is important to note that only ‘classical’ inflammatory biomarkers (cytokines, chemokines and interleukins) have been investigated. Studying other mediators of inflammation, including miRNA, could also be of particular interest. Indeed, these inflammatory-linked miRNAs are produced by several organs, including adipose tissue during inflammation(Reference Karkeni, Astier and Tourniaire6), and they actively participate in metabolic disorders(Reference Landrier, Derghal and Mounien7) and mediate insulin resistance(Reference Ying, Riopel and Bandyopadhyay8). Moreover, we previously reported that phytochemicals or vitamins were able to modulate miRNA expression (Reference Milenkovic, Deval and Gouranton133–Reference Karkeni, Payet and Astier135). Hence, it is tempting to hypothesise that lycopene and/or tomato-derived product phytochemicals could also regulate inflammatory-linked miRNA, thus representing a new anti-inflammatory mechanism of action of lycopene and/or phytochemicals present in tomato products.
In the field of clinical studies, it is crucial to note that the anti-inflammatory impacts of lycopene and/or tomato-based products have, until now, been assessed only in plasma. Therefore, it will be of particular interest to validate some of these findings using tissue biopsies, especially from adipose tissue, to assess the role of lycopene or tomato-based products in local tissue inflammation. Indeed, such impact at local level may profoundly modify organ biology, without any sign of inflammation modification at plasma level.
Finally, although there is a strong suspicion that the NF-κB signalling pathway is inactivated by lycopene or its metabolites, the exact molecular mechanism underlying this anti-inflammatory effect has yet to be revealed. Several potential mechanisms could be implicated, including an influence on the expression of phosphatases involved in NF-κB proteins dephosphorylation, a direct interaction with NF-kB signalling proteins, or an effect on the production of lipid mediators such as resolvins.
In conclusion, the impact of lycopene and tomato-derived products on the control of inflammation associated with cardiometabolic disorders is suggested from in vitro and in animal studies and in clinical trial only with tomato-derived products but not lycopene. Additional research is essential to establish the interest of this nutritional approach in meaningful clinical studies. Should these findings be validated, this type of supplementation potentially offers a novel strategy to limit or prevent inflammation and its related consequences.
Acknowledgements
The authors thank all team members for their support.
Authorship
J.F.L. designed the review, wrote the paper and had primary responsibility for the final content. T.B., L.S., C.D., L.M. and P.B. participated in the writing and correction of the review. All authors have read and approved the final manuscript.
Financial support
This work has been funded by grants from INRAE, INSERM and AMU and by the TomHealth ANR project (ANR-20-CE21-0010).
Conflict of interest
Authors have no conflict of interest to declare.
