Golden berry (Physalis peruviana L.) is a fruit of high commercial importance in some African and Latin American countries, where it is locally consumed and often exported to northern markets, mainly Europe and the US ^[1]. The traditional pharmacopoeia is relatively extensive concerning the potential health benefits of the berry, although it concerns mostly the use of calyces ^[2]. Nonetheless, folk medicine attributes antispasmodic, diuretic, antiseptic, sedative, and analgesic effects to the fruit ^[2]. When considering scientifically based studies, most reviews of the literature conclude that the strongest evidence indicates a hypoglycaemic effect and the improvement of insulin sensitivity ^[3]. Up to four independent studies conducted in India, Colombia, and Peru reported antidiabetic properties. Three interventions with golden berry extracts on streptozotocin-induced diabetic rats [4,5,6]^[4][5][6] and normal mice ^[7] showed positive effects on glycaemic and insulin related metabolites. The fourth study was conducted on 26 young human adults ^[8], and the authors reported that golden berry intake induced a postprandial decrease in glycaemia following a per os glucose challenge. Less documented is an additional potential positive impact of golden berry fruit consumption on oxidative stress and inflammatory processes and status. Nonetheless, most of these studies were conducted in vitro or using animal models, and data on human cohorts are very scarce. Additionally, very little is known about the mechanisms and compounds involved in the observed effects.

The fruit of Physalis peruviana L. contains a wide diversity of biochemical compounds. withanolides and their derivatives are the most emblematic metabolites of Physalis species, and they have been shown to exert a wide range of pharmacological activities in vitro, such as immunomodulatory, angiogenesis inhibitor, anticholinesterase, antioxidant, antibacterial, and antitumoral activities ^[9]. This family of compounds with a steroid backbone has attracted the attention of pharmacologists, as withanolides and derivatives are mostly concentrated in aerial parts of plants, such as the leaves. They were also detected in fruit pulp ^[10] but at low concentrations. Nonetheless, it is not clear whether such compounds could be bioaccessible and bioavailable for humans ^[11]. The fruit also contains an appreciable amount of lipids and carotenoids (essentially β-carotene ^[12]), with some lutein diesters ^[13], phytosterols ^[14], tocopherols [14,15]^[14][15], and flavonoids, essentially rutin ^[16]. Some of these compounds accumulate within diminutive seeds, which are probably not disrupted during passage through gastrointestinal tracts. However, in the case of golden berry, the levels of tocopherols and phytosterols are apparently much higher within the pulp and peel ^[14], which probably makes them more bioavailable. Therefore, compared with other fruit, golden berry is probably an important source of tocopherols (~17 mg/100 g of fruit FW, mainly γ, α, and β in order of importance) with high vitamin E activity and phytosterol content (~10 mg/100 g of fruit FW, mainly 5-avenasterol and campesterol) ^[14]. Additionally, it was recently shown that golden berry fruits could also be a source of trans-resveratrol, which is even richer than red wine ^[17]. On the other hand, specific disaccharide hydroxyesters have also been detected in golden berries [18,19]^[18][19]. These compounds have been associated with the inhibition of alpha-amylase, which could also contribute to the hypoglycaemic effect. Despite the high diversity of compounds present in golden berries, the potential health impacts after fruit consumption cannot yet be attributed to one specific molecule or group of compounds. This could potentially result from synergistic effects of many of these secondary metabolites.

The insulin signalling pathway plays two major roles in cells, both metabolic and mitogenic. First, it regulates metabolic processes such as carbohydrate, lipid, and protein metabolism. Second, it modulates cell division and growth through its mitogenic effects [33]^[20]. In the case of the acute response to golden berry consumption, the effect on the insulin signalling pathway could be due either to the fruit or to the meal that was ingested. Nonetheless, insulin was also at the centre of the medium-term biological network, which compared two fasting plasma metabolomes. Thus, although there may have been interference during the acute intervention, our results seem to reinforce the results already observed in vivo in animal models [5,34]^[5][21]. In mammals, insulin plays a key ubiquitous role in energy homeostasis. It influences the expression and activity of a variety of channels and enzymes involved in the metabolic processes of glucose transport, glycogenesis, glycogenolysis, glycolysis, and inhibition of gluconeogenesis in the liver [33,35,36,37]^{[20][22][23][24]}. Disruption of these mechanisms induces insulin resistance. This poor insulin signalling is a foundational aspect of the pathogenesis of metabolic syndrome, obesity, type 2 diabetes, and most chronic diseases and comorbidities linked with an unhealthy diet, e.g., cardiovascular diseases and cancer [30,31,33]^[25][26][20]. At the molecular level, insulin actions on insulin-sensitive tissues such as liver, muscle, and adipocytes are mediated by its membrane receptors. Activation of insulin receptor (IR) tyrosine kinase reflects insulinaemia and is associated with decreased glycaemia. Upon insulin binding to IR, the receptor is autophosphorylated to subsequently trigger intracellular signalling pathways. These pathways are organized into a complex network of protein interactions and phosphorylation cascades at both the cytosolic and nuclear levels. Two main pathways are mobilized: (1) the PI3K/Akt/mTOR pathway through pleiotropic IRS docking molecules and (2) the Shc/GRB2/SOS/Ras/MAPK pathway. Both pathways control most insulin metabolic actions, such as energy metabolism (carbohydrates, lipids) and gene expression (cell proliferation, differentiation, and growth) [38,39]^[27][28]. These cellular effects correspond to the dual potential of insulin, which behaves as both a hypoglycaemic hormone and an anabolic growth factor. The PI3K pathway also has a key role in the integration of up- and downregulated metabolites for the activation of the insulin signalling pathway [39]^[28], confirming the significant effects on cell-to-cell signalling, cell movement, cell signalling, and protein synthesis predicted by IPA software. A preclinical study already suggested that the consumption of golden berry juice decreased blood glucose and insulin resistance and increased insulin levels in diabetic rats [40]^[29]. Nevertheless, the glucose detected in plasma did not appear as a discriminating metabolite, so we can assume that glucose level was not significantly affected. However, we report for the first time that the relationship in the insulin signalling pathway is supported by integration into the relevant networks of plasma metabolites that are significantly modified after acute ingestion of golden berry fruit. The effect tends to fade after discontinuing consumption of the fruits. Even after medium-term fruit consumption, the effect declined 24 h after the last ingestion (Day 19), remaining significantly altered only at the level of PI3K/Akt/mTOR. Following golden berry ingestion, predictions also mentioned relations with the immune system through modulation of cytokine levels, inflammation, and the NFKB signal transduction channel. Low-grade inflammatory processes are common to diseases associated with energy metabolism (metabolic syndrome, diabetes, obesity, etc.), as well as with tumorigenesis [41]^[30]. Our results suggest a modulatory influence of Physalis peruviana ingestion. Inflammation in metabolic diseases is also detrimental to the cardiovascular system. It is, therefore, interesting to note that the pathway analysis revealed an association with arginase. This enzyme was demonstrated to modulate NO levels in vascular endothelial cells and smooth vascular muscles, thereby impacting arterial blood pressure regulation [42]^[31]. Additionally, our metabolomic results identified norepinephrine variations. This endogenous neurotransmitter from the sympathetic autonomic nervous system is involved in both energy expenditure and blood pressure control through its actions on beta-1 cardiac receptors and alpha-1 vascular receptors [43]^[32]. These results could indicate a potential combination of effects with NO pathways on the cardiovascular system in relation to glucose and lipid metabolism. Research on tyrosine kinase receptors, such as EGFR, revealed the mechanisms of their activation by growth factor ligands [44]^[33]. The EGFR tyrosine kinase intracellular network main channel is the Shc/GRB2/SOS/Ras/MAPK pathway. This signalling cascade controls cell proliferation and differentiation processes. In addition, EGFR is also able to mobilize the PI3K/Akt/mTOR, p53, Ras/MAPK, and NFKB pathways to regulate cell proliferation/growth, amino acid metabolism, cell survival/apoptosis, and cell morphology/motility. In addition, EGFR is known to induce the nuclear factor NFKB via PIK/Akt and MAPK signalling to regulate the immune system and associated inflammation processes, as well as angiogenesis [45]^[34]. In this context, our results showed a significant impact of acute ingestion of golden berry fruit on EGFR. Almost all discriminant metabolites involved within its biological network were significantly changed. These results showed that acute consumption of golden berry may affect the activity of EGFR tyrosine kinase. Furthermore, numerous studies have shown the potential contributions of EGFR-associated signalling pathways in oncogenesis processes, including cell proliferation, angiogenesis, and resistance to apoptosis [46]^[35]. In this context, our results suggest that golden berry fruit consumption as part of a healthy diet could negatively influence EGFR signalling. These combined observations are in accordance with previous reports indicating antitumoral effects observed in a rat model after ingestion of golden berry juice [47]^[36]. Taken together, these data suggest the anti-oncogenic potential of Physalis peruviana. Based on our results, the consumption of the fruits appeared to influence EGFR, which was reported to be involved in a diversity of tumours. These effects might be explained by a loss of EGFR binding affinity, since this receptor is regulated by protein kinase C [44,48]^[33][37]. Alternatively, other more complex mechanisms yet to be discovered could involve modulation of EGFR intracellular signals through crosstalk with insulin signalling. One element that may support this hypothesis is in the link that our results identified between EGFR signalling and insulin-growth-factor binding protein 2 (IGFBP2). This endogenous compound is currently considered a major discriminant metabolites but also regulator of insulin resistance and associated metabolic processes in relation to insulin signalling cascades [48]^[37]. A possible explanation for the results we obtained may lie in the golden berry fruit composition and relative contents of various bioactive compounds. One of these compounds, which is not specific to this fruit, is β-carotene ^[12]. This bioactive molecule can lead to the activation of PI3K/Akt/mTOR [34]^[21] insulin receptor substrate 1 (through IRS-1) phosphorylation, promoting insulin signalling pathway activation and thereby reducing insulin resistance [49]^[38]. In this context, evidence also suggests that a diet enriched in carotenoids with pro-vitamin A functions, such as β-carotene, improves liver function, which is insulin-sensitive tissue [47,50]^[36][39]. The other type of molecule, withanolides, is present in the Physalis sp. composition. Among these compounds, withangulatin-A has previously demonstrated insulin-release stimulatory effects in induced diabetic rats, similar to the reference drug glibenclamide (a potassium channel blocker on endocrine Langerhans insulin-secreting cells), suggesting antidiabetic potential through modulation of insulin levels and glucose homeostasis [51]^[40]. Relatedly, our metabolome analysis also indicated potential impact on potassium levels (Supplemental Figure S1h) after golden berry fruit consumption, but potassium levels in plasma were not measured. Other compounds are much more specific to Physalis peruviana, such as peruviosides, which are sucrose esters that exhibit important inhibition of α-amylase ^[18]. These compounds could also contribute to hypoglycaemic activity, thereby also affecting insulinaemia. These sugars were not detected in the plasma. Nonetheless, a synergistic effect between carotenoids, withanolides, and sucrose esters, along with other unknown compounds, should not be ruled out and could explain our observations. Our findings and mechanistic hypothesis based on our metabolomic approach are supported by a recent report from Pino-de-la Fuente et al. Indeed, the authors recorded positive effects of Physalis peruviana in vivo, i.e., insulin resistance and inflammation improvement in the muscles and liver in a mouse model of diet-induced obesity. These data point in the same direction as ours at the molecular and physiological levels in relation to the compositions of golden berry fruit, as suggested by the authors [52]^[41]. Taking into account up- and downregulated metabolites, as well as data from the literature, three biological networks could be integrated, involving mainly insulin, EGFR, and PI3K/Akt/mTOR. The PI3K pathway is common to insulin and EGFR signalling. These two growth factor molecules also share the MAPK signalling pathway. These signalling cascades control the metabolic and mitogenic effects of insulin and EGFR. Both biological intracellular networks are highly interconnected with the insulin signalling pathway. The insulin network was previously demonstrated to interact with other growth factor signalling pathways, such as EGF/EGFR, through bidirectional crosstalk. More specifically, the EGFR intracellular network, the Shc/GRB2/SOS/Ras/MAPK pathway, was reported to be linked to the insulin signalling cascade, with insulin regulating early signalling events of EGFR [38]^[27]. Therefore, our results suggest that different compounds from Physalis peruviana could be considered good candidates for use as beneficial modulators of pathophysiological processes involving insulin-associated cell dysregulation. Golden berry fruits seem to mobilize the signalling pathways common to the two endogenous anabolic molecules, namely, insulin and EGFR, leading to their direct or indirect modulation of energy and cell cycle homeostasis. These impacts and the technology used in our investigation point out the complex interplay between metabolic and mitogenic processes, wherein cell signalling controls metabolism and reciprocally controls metabolism signalling [53]^[42]. This integrative modulation of cell homeostasis appears to be partially oriented by the molecular contents of Physalis Peruviana, providing some insight explaining the beneficial effects of this fruit on health.