Acute Toxicity Study of Bioactive Galactomannans From Two Non-Traditional Leguminosae Sources

Galactomannans from Caesalpinia pulcherrima (GM-CP) and Delonix regia (GM-DR) are being pointed as potential therapeutic agents, but systematic evaluations on their acute toxicity are yet to be reported. In order to evaluate the occurrence of systemic toxicity, groups of three female rats received oral GM-CP or GM-DR (300 mg kg -1 ), whereas the control group received vehicle. Since lethality was absent, other groups received doses of 2,000 mg kg -1 , which also did not cause lethality. Collection of organs and blood samples were done at day 14. Dermal toxicity of the galactomannans was also evaluated (2,000 mg kg -1 , n=3 per group), as well both mechanical hypernociception and inammatory cell inux after administration of GM-CP or GM-DR via intra-articular route (200 µg, n = 5 per group). At the routes and doses employed, both galactomannans did not evoke physiological / behavioral changes or skin / joint inammation. Since the LD50 was not inferior to 2,000 mg kg -1, both galactomannans are in the class 5 of the Globally Harmonized System for Classication and Labelling of Chemicals. Hypnea in rats subjected to zymosan-induced arthritis: cGMP signalization


Introduction
Galactomannans are water-soluble neutral polysaccharides found in the seed endosperm of numerous plants, particularly in the Leguminosae family. They are constituted by a (1→4) linked β-mannopyranosyl backbone partially substituted at O-6 with α-D-galactopyranosyl side groups (Meer et al. 1975;Srivastava and Kapoor 2005). Because of their high viscous water solutions, the galactomannans of Ceratonia siliqua, Caesapinia spinosa, Cyamopsis tetragonolobus and Trigonella foenum-graecum are used as emulsion stabilizers in the food, cosmetic and pharmaceutical industries (Prajapati et al. 2013;Johnson et al. 2015). Despite of the galactomannans intake be conventionally assumed as safe, few systematic studies report the acute toxicology of such polysaccharides, especially those obtained from nontraditional sources (Pawar and Lalith 2015;Suryawanshi et al. 2015;Deshpande et al. 2016).
The Leguminosae species Caesalpinia pulcherrima (L.)Sw. and Delonix regia (Bojer ex Hook.)Raf. are native from Central America and Madagascar, respectively, but they are widely used as ornamental plants around the world (Singh and Kumar 2014;Santos et al. 2019). Their easily extractable galactomannans have been exploited as dietary ber, fruit edible coating, preparation of mucoadhesive microspheres, drug delivery systems and injectable regenerative hydrogels (Cerqueira et al. 2009;Tamaki et al. 2010;Lima et al. 2010;Thombre and Gibe 2013;Buriti et al. 2014;Ogunjimi et al. 2017;Lima et al. 2019). Bene cial effects of these galactomannans were recently reported in rodent models of gastritis and osteoarthritis (Marques et al. 2019;Nascimento et al. 2021). Following international protocols preconized by Organization for Economic Co-operation and Development (OECD), this study evaluated the ocurrence of systemic and local acute toxicity for these galactomannans in rats.

Methods
Seeds collection hydrated overnight at 4°C, and homogenized with 500 mL of distilled water. The rst supernatat was collected after an overnight period of decantation. Remaining fragments were resubmitted to homogenization, and a second supernatant was collected. Both supernatants were mixed with two volumes of ethanol 96% (w/v), and the crude galactomannan was ltered under vacuum (glass lter no. 4), and dried at room temperature. Proteolysis was achieved by heating under re ux (85-90°C) of 1 g of crude sample in 200 ml of NaOH 5% (w/v) during 45 minutes under re ux. Alkali and low molar weight compounds were removed by dialysis against distilled water for 72 h, using a semipermeable cellulose membrane (cut off 10,000 g mol -1 ). The dialysate was centrifuged for15 minutes (1000 x g), and ethanol was added to the supernatant. The puri ed galactomannan was ltered under vacuum, and dried at room temperature until constant weight. Removal of proteins was con rmed in both samples through indetectable %N, using a 2400 Series II CHNS/O Elemental Analyzer. Thermogravimetric analysis revealed low levels of inorganic residues in the samples were found (0.24% for GM-CP and 0.56% for GM-DR). As expected for hygroscopic polyssacharides, moisture after desiccation was of 14.76% and 13.78%, respectively.

Animals
Female Wistar rats (170-200 g) were obtained from the Federal University of Ceará, and maintained under padronized conditions (25 °C; cycle of 12 h light/12 h dark, and food and water ad libitum). The procedures were conducted according to the guidelines of the National Council for the Control of Animal Experimentation (CONCEA), under authorization by the ethics committee for animal research of the Federal University of Ceará (No. 52120510/2018, see supplementary material S1).

Evaluation of systemic toxicity
The acute systemic toxicity of the galactomannans was evaluated according to OECD Guideline No.423 (2002). Groups of three animals were subjected to 2-hour fasting, and received a single dose (300 mg kg -1 per os) of the galactomannans from Caesalpinia pulcherrima (GM-CP) or Delonix regia (GM-DR). Fasting was maintained until 2 hours after the administration. Since lethality was absent, other groups received doses of 2,000 mg kg -1 (p.o.) Control group received vehicle (10 mL kg -1 ; sterile saline).
Body weight measurent was recorded during a 14-day period, as well the occurrence of toxicity signals by the Malone's hippocratic screening (Malone and Robichaud 1962;Malone 1977). For evaluation of exploratory activity, animals were acclimated during 1 minute in a 30 x 30 x 15 cm acrylic box with transparent walls and black oor divided into 9 equal quadrants, and the exploratory activity measured by the number of quadrant crossings, self-cleaning (grooming) and rearing during 5 minutes (Archer 1973).
Animals were euthanized under anesthesia (xylazine 10 mg kg -1 + ketamine 100 mg kg -1 i.p.) for blood collection and removal of organs (kidneys, spleen, lungs, heart, liver and brain Counting of blood gurative elements, as well cells volumes and hemoglobin content, were performed using an automated hematology analyzer (SDH-3, Labtest, Brazil). Labtest Diagnostics Kits were used to measure serum levels of urea, creatinine, total cholesterol, triglycerides, bilirubin, total proteins and albumin, and the activity of the enzymes alanine aminotransferase (ALT), aspartate aminotransferase (AST) and alkaline phosphate. Organ samples were embedded in para n, sectioned at 4-μm-thick sections for hematoxylin & eosin staining, and examined under light microscopy by an experienced pathologist without knowledge of the treatments.

Evaluation of dermal toxicity
The dermal toxicity of the galactomannans was evaluated according to OECD Guideline No. 402 (2017) in groups of three animals receiving the highest doses of galactomannans (2,000 mg kg -1 per os).
Animals had their thoracic dorsal area shaved one day before the experiment. Antisepsis was performed with 2% (w/v) chlorhexidine, and GM-CP or GM-DR (2,000 mg/kg, 0.3 mL) was uniformly applied in an area of 3 cm 2 . Control group received vehicle (saline). Inoculated skin was covered with a porous gauze dressing and non-irritating tape throughout 24-hour exposure period. Occurrence of skin reactions (hyperemia and edema) were monitored during 14 days, using the Draize score system (Draize et al. 1944).

Evaluation of intra-articular innocuity
Occurrence of intra-articular pro-in ammatory reactions was veri ed in the right tibiotarsal joints of rat groups receiving GM-CP or GM-DR (200 µg / 25 µL; n=5). Control group received vehicle (saline). For comparison, galactomannan not-submitted to alkaline hydrolysis step were also tested (protein contents of 0.89% and 1.00%, respectively).
After injection, animals were individually placed into Plexiglas boxes with malleable mesh net oors, and joint exion was achieved by pressioning an algesimeter-coupled large probe (4.15 mm 2 ) until the occurrence of paw withdrawal re ex. Reduced mechanical threshold is indicative of hypernociception (Bringel et al. 2020). The test was performed from hour 0 (immediately before the inoculation) to hour 6, in which animals were euthanized. Joint cavity was washed twice with 0.1 ml PBS containing 10 mmol L -1 EDTA, and exudates were collected for total and differential cell counts using Neubauer chambers and stained smears, respectively.

Statistical analysis
Results were expressed as mean ± SEM. Statistical differences between parametric data were determined by one-way / two-way analysis of variance (ANOVA), followed by the Bonferroni's test. Non-parametric data were expressed as median (IQR) and analysed by the Kruskal-Wallis's test, followed by the Dunn's test. P <0.05 was considered signi cant. All data is available as supplentary material.

Results
Acute oral toxicity of the galactomannans No deaths were observed after the oral administration of the galactomannans, which demonstrates an LD50 greater than 2,000 mg kg -1 . Higher-doses of galactomannans had similar increase in body weight, as compared to the vehicle (GM-CP: + 1.0%; GM-DR: + 1.0%; control: + 2.75%, p>0.05) ( Table 1).
Animals did not display alterations related to the central, somatosensory or autonomous nervous systems, as seen using the Malone's Hippocratic triage (Table 2). In all groups, it was observed normal respiration, vibrissae movement, lacrimation and salivation, but no signals of cyanosis, piloerection, analgesia / sedation, ataxia, prostration, catatonia, tail erection, stereotyped movements, tremors nor convulsion. Although exophthalmos seemed to be present at day 14 in the GM-DR group, and in various milestones in the GM-CP group (60 min, day 7 and day 14), it was also found in the vehicle-treated after 60 min.
In all groups, the exploratory activity has decreased along the experiment (Fig. 1). The time of observation signi cantly in uenced the data variability for crossings and rearings (p=0.0241 and 0.0354, respectively), but not for groomings (p=0.2274). The treatments effect was not signi cant for such parameters (p>0.05).
No treatment-related gross pathological change was observed in the brain, liver, kidneys, heart, lungs and spleen (Table 3). Hystopathological alterations were also absent in such organs (Fig. 2), and no changes were detected on serum levels of the creatinine/urea and ALT/AST (Table 4). All groups displayed high blood glucose triglycerides levels, with signi cant hyperglicemia in the GM-DR group (+38.5% and +39.5% in comparison to control group, rspectively). Other serum parameters remained unaltered, as well the hematological parameters (Table 5).

Dermal toxicity of the galactomannans
Topical application of C. pulcherrima or D. regia galactomannans did not evoke signi cant signals of skin irritation at 2,000 mg kg -1 (Table 6). When measured 24 h or 7 days afer induction, the Draize's scores values for hyperemia were slightly higher in the GM-CP and GM-DR groups, but they did not reach statistical signi cance, in comparison to the vehicle-treated group. On the other hand, no signals of edema were observed at any endpoint.

Discussion
Safety of chemical compounds is an important issue for food or biomedical applications, and the conduction of acute toxicity studies has as the main objective the identi cation of side effects. In this study, bioactive and well-characterized galactomannans from Caesalpinia pulcherrima and Delonix had their acute toxicity systematically evaluated in rodents by the use of international preconized protocols, which comprises the administration of single initial doses of 300 mg kg -1 and a 14-day observation period.
As previously reported for others galactomannans, such as those of Senna tora and Trigonella foenumgraecum (Pawar and Lalith 2015;Deshpande et al. 2016), no lethality was observed among the animals treated with GM-CP or GM-DR, even at 2,000 mg kg 1 . Despite Suryawanshi et al. (2015) had already reported similar evidence for the C. pulcherrima galactomannan, such study did not investigate impairments on target organs or biochemical and haematological parameters, as we did (see below). Additionally, animals did not display alterations of spontaneous locomotory activity (Prut and Belzung 2003) or in the nervous systems. Although toxic reactions could alter the body weight (Teo et al. 2002), such parameter was not signi cantly different among the groups along the 14-day period.
Structural and weight changes in organs may re ect toxicity due to several factors, including changes in the enzyme activity, injuries and other physiological disorders, which are dependent on target organs, and make possible the knowledge of the general health status of the animals (Auletta 1995;Michael et al. 2007). No macroscopic lesions or weight organs alterations were induced by the galactomannans on vital organs. Nevertheless, functional events may arise prior to morphological damage, especially in organs such kidneys and liver (Ramaiah 2011;Alimba et al. 2012), but the serum levels of the creatinine/urea and ALT/AST were normal for the animals euthanized 14 days after the acute administration of galactomannans. Moreover, substances affecting bone marrow could inhibit certain steps in the production of hemoglobin, and reduce the ability of the blood to distribute oxygen throughout the body (Jain et al. 2009). Both galactomannans tested did not change the haematological parameters, so that erythrocytes were both normocytic and normochromic. However, animals in all groups displayed high blood glucose levels (>140 mg dL -1 ), even in the control group. Such values of post-brandial capilary glycemia are not uncommon, especially when blood collection is taken after zylaxine/ketamine anesthesia (Saha et al. 2005;Wang et al. 2010), so that such hyperglycemia may not be due to the treatment with the galactomannan.
As evaluated using the Draize's scoring system, cutaneous administration of GM-CP or GM-DR did not cause hyperemia or edema at 2,000 mg kg -1 and these ndings agree with previous reports for innocuity of leaf polysaccharides of D. regia (Shewale et al. 2011;Wang et al. 2016). Regarding the intra-articular innocuity, the protein-free galactomannans did not evoke articular hypernociception or leukocyte in ux.
Similar to reported for the guar galactomannan (Castro et al. 2007), joint in ammatory responses were evoked solely by unpuri ed samples containing remaining proteins.
In conclusion, the lack of systemic and local reactions revealed that the seed galactomannan of C. pulcherrima and D. regia did not cause acute toxicity, and their LD50 values were superior to 2,000 mg kg -1 , being allocated in the class 5 ofthe Globally Harmonised System for Classi cation and Labelling of Chemicals (GHS) (UN, 2019). Further assessments involving different administration schemes, such as repetitive doses, and evaluations in humans are recommended for their safe consumption.

Declarations
Ethical approval This study was performed in accordance with the guidelines of the Brazilian Council of Control in Animal Experimentation (CONCEA). Approval was granted by the local ethics committee (No. 52120510/2018;October 22 nd , 2018). Certi cate is available as supplementary material.

Consent to participate Not applicable
Consent to publish Publication is consented by all authors.
Authors contribution RRC, EMCC and OVOM conceived and designed the research. FCJM, FGSN, DTTN and ICL conducted the experiments. PMGS and MRLM performed the histopathological analysis. AON and GFA perfomed the hematological and biochemical analyses. RRC and AMSA wrote the manuscript. All authors read and approved the manuscript and all data were generated in-house and that no paper mill was used.
Competing interestsThe authors declare no con ict of interest.