Nigella sativa (black seed) safety: an overview

Abstract Nigella sativa (commonly known as black seed or black cumin), from the family Ranunculaceae, is a plant that grows in countries bordering the Mediterranean Sea. This narrative review discusses the toxicological profile reported by short- to long-term studies that examined different extracts and oils of N. sativa seeds. Scientific databases including Web of Science, PubMed, Scopus, and Google Scholar were searched using appropriate keywords. LD50 for administered N. sativa seed fixed oil varied from 28.8 mL/kg to 3,371 mg/kg in mice, while 21 g/kg of aqueous, methanol, and chloroform extracts of N. sativa did not lead to any mortality. Subacute toxicity evaluations indicated that aqueous, methanol, and chloroform extracts of N. sativa at doses as high as 6 g/kg do not produce toxicity. Investigation of chronic toxicity found that 2 mL/kg of N. sativa fixed oil is slightly toxic. Cytotoxicity studies indicated that N. sativa chloroform and petroleum ether extracts are more cytotoxic than its other extracts. Although studies that assessed N. sativa toxicity generally introduced it as a safe medicinal herb, to draw a more definitive conclusion on its safety, more detailed studies must be conducted.

The present narrative review discusses the toxicity and adverse effects of different extracts and oils obtained from N. sativa reported in animals, cell lines, and humans.

Methods
We conducted a literature search of scientific databases including Web of Science, PubMed, Scopus, and Google Scholar. In this search, the following string of keywords was used: "Nigella sativa" or "sativa" or "Nigella" or "black seed" or "black cumin" or "black cumins" or "thymoquinone" or "carvacrol" or "4-cymene" and "safety" or "toxicity" or "cytotoxicity" or "genotoxicity" and "acute" or "subacute" or "subchronic" or "chronic". The wild-card asterisk (*) was used to increase the scope of the search. The reference lists of discovered articles were also reviewed to identify reports that evaluated N. sativa toxicity published until March 2020. All studies that assessed the toxicity or reported adverse reactions of N. sativa following topical, parenteral, and oral treatment of experimental animals and humans and data from experiments done on cell lines, were included. All included studies were written in English. Review articles were not included.

Acute toxicity of N. sativa in animal studies
The median lethal dose (50%) (LD 50 ) values of N. sativa seed volatile oil, fixed oil (prepared by hexane extraction), and aqueous extract in male Swiss albino (SWR) mice were calculated using a Litchfield and Wilcoxon method. Intraperitoneal administration of volatile oil, aqueous extract, and fixed oil of N. sativa has LD 50 values of 1,853, 3,020, and 3,371 mg/kg, respectively. Based on these data, authors attributed N. sativa toxicity to its volatile oil rather than aqueous extract and fixed oil [51].
A more recent study reported zero mortality (LD 50 could not be calculated) following administration of N. sativa essential oil at the doses of 0.2, 0.4, 0.4, 0.8, 1 mg/kg to either sex of white Wistar rats and albino mice (NMRI); however, this study lacked a full characterization of the essential oil and only total flavonoid content in terms of quercetin was reported [62]. by sedation. Furthermore, 12 h after administration of N. sativa (50 mL/kg oral or 4 mL/kg intraperitoneal), all animals died. In surviving mice, activity and growth were rapidly retrieved within 4-8 days after oil administration. This study indicated LD 50 values of 28.8 mL/kg body weight (bw) and 2.06 mL/kg bw for oral and intraperitoneal FO, respectively, and also indicated that N. sativa fixed oil is of low toxicity and possesses a wide margin of safety [49].
LD 50 values for aqueous, methanol, and chloroform extracts of N. sativa seeds following a single oral administration of 4 different doses (6,9,14, and 21 g/kg) of the extracts to mice, was evaluated, but zero mortality was observed. The A summary of studies [4,49,51,62] that examined N. sativa acute toxicity is presented in Table 1.

Subacute toxicity of N. sativa in animal studies
To study subacute toxicity of cumin seeds, 10 mL/kg aqueous extract of N. sativa seeds was administered orally to male Sprague Dawley rats for 14 days. In this study, the aqueous extract was not characterized. The results demonstrated a notable increase in gamma-glutamyl transferase (g-GT) amounts, though serum levels of alkaline phosphatase (ALP) remained unchanged and histopathological examinations did not show any damage; possibly, the findings were affected by an ether anesthesia effect on the release of ALP [63]. Oral administration of 10 mL/kg N. sativa seed oil to rats and mice for 48 h caused neither death nor obvious toxicity [53].
Oral administration of 6 g/kg/day of aqueous extract, and methanol and chloroform extracts of N. sativa seeds to male mice, for 14 consecutive days, did not affect plasma concentrations of ALP, alanine transaminase (ALT), or aspartate aminotransferase (AST). The pathological findings outlined only minor alterations in mice livers following treatment with these extracts. This study showed that even the 14-day administration of N. sativa seeds extracts has a wide margin of safety in mice [4].
Furthermore, 90, 180, 360, and 540 mg/kg of N. sativa seed mixed in a powder dough, was given orally for 1, 2, and 4 weeks to normal adult male albino rats. In this study, the authors did not present any information on the chemical composition of the mixture. N. sativa at the tested doses transiently altered both anticoagulant and coagulant profiles of rats. In this context, N. sativa markedly prolonged prothrombin time caused hyperfibrinogenemia and reduced activated partial thromboplastin time and thrombin time. Nevertheless, antithrombin III (ATIII) and AST levels remained unchanged while ALT and albumin levels significantly increased [54].
Powdered N. sativa seeds (0.01, 0.1, and 1 g/kg/day) administered for 28 days to male Sprague Dawley rats, did not affect the bodyweight of the rats, or serum AST and ALT levels compared with a control group. Histological examinations highlighted that 28-days treatment with these doses of N. sativa had no harmful impacts on liver tissue in the rats. However, no information on the composition of the seed powder was provided and the doses administered to rats were chosen based on human N. sativa consumption (viz., 2 g/d) [64].
Daily administration of N. sativa seeds aqueous extract (2, 6.4, 21, 33, and 60 g/kg, orally) to mice for 6 weeks produced hepatotoxic effects (based on histopathological examinations) at doses ≥21 g/kg, but did not influence kidney tissue (in terms of urea and albumin levels). A high mortality rate was found following the administration of 60 g/kg of the extract. The findings highlighted the safety of N. sativa at doses lower than 21 g/kg; however, the aqueous extract of the seeds was not characterized [65].
Although for centuries, a black seed and honey mixture (BSH) has used in traditional folk medicine, especially in Islamic countries because of its mention in the Hadith, its toxic effects are unclear. For this reason, a study was conducted to evaluate the effect of 14-day administration of 100, 500, 1,000, and 2,000 mg/kg doses of BSH to male Sprague Dawley rats and found an LD 50 >2,000 mg/kg. The findings showed no notable alteration in body weight, and differential leukocyte count, or abnormalities in liver and kidney histopathology; however, limitations of this study included a lack of mixture characterization and histopathological assessment of only 2 organs [66]. Table 2 presents investigations [4,53,[63][64][65][66] conducted to evaluate the subacute toxicity of N. sativa in animals.

Subchronic toxicity of N. sativa in animal studies
Powdered N. sativa seeds (20 and 100 g/kg) were added to the diet of 7-day-old Hibro broiler chicks for 7 weeks. Administration of these doses of N. sativa seeds damaged the liver as reflected by increased AST and ALT levels and decreased albumin concentration and cholesterol levels in serum. Furthermore, 100 g/kg of N. sativa decreased red blood cells (RBCs), hemoglobin level, mean corpuscular hemoglobin concentration, and packed cell volume despite increases in mean corpuscular volume levels. Together, the results suggested that a 7-week treatment with N. sativa seeds did not have a negative influence on growth, pathological features of the main organs, or biochemical/hematological profile of the animals [55].
N. sativa oil obtained from Cairo Chemical Industries Co. (15 and 25 mg/kg) was administered to adult male albino rats for 1 month. However, no information on the chemical composition of the oil was noted. At the end of the experiment, histological changes were found in the cortex of the kidney and to a lesser extent, in hepatocytes in a dose-dependent manner. This study indicated that N. sativa oil (oil type not specified) must be used at appropriate doses [48].
Experiments [48,55] that examined the subchronic toxicity of N. sativa are summarized in Table 3.

Chronic toxicity of N. sativa in animal studies
N. sativa fixed oil (prepared by hexane extraction, 2 mL/kg bw chosen based on the LD 50 value obtained in an acute study) was orally given to Wistar-Kyoto rats for 12 weeks. Hematological  examinations showed significant decreases in leukocyte and platelet counts despite significant increases in hematocrit, hemoglobin, mean globular volume, mean corpuscular hemoglobin, and mean corpuscular hemoglobin concentration. No significant alterations were found in AST, ALT, ALP, and GGT levels and no histopathological damages were observed in the heart, kidneys, liver, and pancreas. The researchers suggested that N. sativa fixed oil has a wide margin of safety at therapeutic doses, but hematological impacts should be studied further [49]. Oral daily administration of 1 mL/kg bw N. sativa seed fixed oil (prepared by hexane extraction) for 12 weeks to Wistar-Kyoto rats led to notable decreases in leukocyte and platelet counts. By contrast, hemoglobin and hematocrit levels were raised significantly. Nevertheless, serum levels of key hepatic enzymes (namely, AST, ALT, ALP, and GGT) did not change significantly relative to controls. Together, these results indicated mild toxicity for N. sativa seed FO. In this report, the authors did not specify the sex of the rats or the chemical composition of the fixed oil [56].
Studies [49,56] that evaluated the chronic toxicity of N. sativa in animal models are presented in Table 4.

Cytotoxic and genotoxic properties of N. sativa
Various concentrations (0.25, 0.5, and 1 mg/mL) of N. sativa oil were tested for possible cytotoxicity in cultures of gingival fibroblasts, but no cytotoxicity of the oil against the fibroblasts was found [67]. This study was limited because it did not indicate which type of oil was examined.
The influences of various doses of N. sativa ethanolic extract (100, 500, and 1,000 mg/mL) on viability and proliferation of nonstimulated and concanavalin A or phytohemagglutinin-stimulated isolated rat splenocytes were examined. Based on the results, 1,000 mg/mL of the ethanolic extract resulted in a marked decrease in cell viability and proliferation in both nonstimulated and stimulated cells [57].
Cytotoxic properties of the essential oil and different extracts (methanol, ether, petroleum ether, chloroform, and water) of N. sativa seeds were tested using a brine shrimp lethality assay; results revealed that chloroform and petroleum ether extracts were more toxic than the other extracts [69].
Cytotoxicity of petroleum ether and aqueous extracts of N. sativa for the human acute myeloid leukemia cell line (HL60) was tested, and it was found that petroleum ether extract is more toxic towards these cells having a lower halfmaximal inhibitory concentration (IC 50 ) [71]. Treatment of Huh7 human cells with N. sativa essential oil (100 mg/mL) did not affect cell viability [72].
N. sativa seed oil and its active component thymoquinone caused a decrease in the size and volume of glioblastoma multiforme (GBM) tumors in a xenograft mouse model. In cell lines including U-1242, U251, U-87, U-373, A172, and SNB19, N. sativa seed oil (50-500 mg/mL) and thymoquinone (1-20 mM) inhibited, in a dose-dependent manner cell growth and colony formation in soft agar with an IC 50 of 260 mg/mL and 6 mM, respectively, activated the caspase-9/3 pathway and induced the cleavage of the death substrate poly(ADP-ribose) polymerase (PARP) [58]. Cytotoxicity, genotoxicity, and antigenotoxicity properties of aqueous extract of N. sativa seeds (0.5, 1, 4, 8, 9, and 18 mg/mL) from 3 diverse regions in Morocco were assessed in human C3A cells; the location where the herbs were collected, test system used, and other experimental parameters were important factors influencing the outcomes [73].
Apoptogenic effects of methanolic, n-hexane, and chloroform extracts of N. sativa seed were examined in HeLa cancer cells. The findings suggested that all of these extracts can induce apoptosis in cells as confirmed by western blotting, DNA fragmentation tests, and terminal transferase-mediated dUTP-digoxigenin-end labeling (TUNEL) assay [74].
Potential antigenotoxic properties of aqueous extract of N. sativa 79.5 mg/mL were tested in F344 hepatocytes treated with N-methyl-N-nitro-N-nitrosoguanidine (MNNG); results showed that N. sativa could not protect hepatocytes from the clastogenic effect of MNNG and produced a remarkable increase of chromosomal aberrations when used as a pretreatment. The findings revealed that this extract displayed a small, but significant, genotoxic potential [59].
Based on the cytotoxic studies, we concluded that the ethanolic extract and aqueous extract of N. sativa are safer than other extracts in different cell lines. Essential oil and volatile oil of N. sativa are relatively safe. Table 5 summarizes studies in vitro [57-59, 67-70, 72-74] that evaluated cytotoxicity and genotoxicity of N. sativa.

N. sativa toxicity in humans
Administration of N. sativa oil to human volunteers at 5 mL/day for 26 days produced no significant hepatic, renal, or gastrointestinal adverse effects [75,76]. N. sativa seeds (3 g/day for 3 months) consumed by 39 centrally obese subjects did not lead to marked side effects [77]. Similarly, diabetic patients who took N. sativa seeds (1, 2, and 3 g/day for 3 months) showed no significant alterations in renal or hepatic function [18]. However, epigastric pain and hypoglycemia were observed in hepatitis C virus patients treated with N. sativa seed oil capsules [38]. The use of total oil was associated with a marked increase of blood levels of AST and ALT and both the oil and the crushed seeds resulted in significantly increased g-GT and the ALP activities [78]. N. sativa seeds (5 g/day) had an inhibitory effect on CYP2D6 and CYP3A4 in human liver microsomes and healthy human volunteers [79]. A 28-year-old man who had used pure oil of black cumin topically on his neck for 3 months as a treatment of sore throat, presented with a 2-day history of maculopapular eczema, first on the neck and spreading to his arms and back. To our knowledge, this was the first documented case of allergic contact dermatitis caused by topical black cumin oil [60].
A 31-year-old woman with an 8-month history of eczema on both hands and exacerbation of the skin lesions repeatedly applied an ointment containing essential oil extracted from the seeds of black cumin to her palms as a skin-care product. A biopsy from the left hypothenar showed signs of subacute dermatitis. Patch tests showed reactions to the ointment of black cumin in the form of allergic contact dermatitis [61].
A 56-year-old woman presented a 2-day history of severe bullous target-like lesions, compatible with erythema multiform. Histopathology of the lesions indicated lymphocytic infiltration at the dermal-epidermal junction, dermal edema, basal vacuolization, and keratinocyte necrosis.
During the last 15 days before the eruption, she had been taking 2 capsules of black cumin essential oil, containing 500 mg of organic N. sativa oil (and 7.5 mg of vitamin E) per day. These reactions (presented in Table 6), based on previous reports [60,61,80], should alert physicians to be informed of potentially intense adverse effects of N. sativa essential oils [80].
Clinical trials that examined N. sativa and its active constituent, thymoquinone, found these agents safe, but notably several adverse effects such as bloating, nausea, and burning sensation were observed in functionally dyspeptic patients treated with N. sativa oil. Moreover, after using N. sativa oil and crushed seeds, a slight increase in kidney and liver enzymes was observed [81].
Our literature review to retrieve studies conducted on the short-and long-term toxic effects of N. sativa oils and extracts showed that a considerable number of studies were conducted without characterizing the plant material or its extracts; moreover, studies simultaneously in both sexes, in animals other than rats or mice, or performed to provide a full toxicological screening of all parameters of importance, are lacking. Dose ranges vary among studies making it difficult to draw conclusions.

Conclusions
The essential oil obtained from N. sativa was found to be safer than the volatile oil against different cell lines. Human case reports indicated allergic contact dermatitis following the use of some preparations containing N. sativa. However, clinical trials did not report any severe adverse effects following consumption N. sativa. Studies that assessed N. sativa safety generally introduced this plant as a safe medicinal herb. Nevertheless, more detailed investigations are still required to have a clearer insight into the toxicological profile of N. sativa. EO, essential oil.