Mild and reversible nephrotoxicity following repeated administration of damnacanthal in nude mice.

1
Introduction
Damnacanthal (DAM), an anthraquinone compound extracted from the root of Morinda citrifolia (Noni), possesses promising pharmacological properties [1], [2], [3], [4]. Oral administration of 20 mg/kg DAM (DAM20) has demonstrated superior efficacy to 5-FU in suppressing colorectal cancer xenograft growth in nude mice [4], [5]. Acute oral toxicity testing, following OECD Guideline 423, revealed low toxicity, with an estimated LD₅₀ cut-off of 2500 mg/kg. At 300 mg/kg (DAM300), mild reversible effects were observed, while 2000 mg/kg (DAM2000) caused moderate diarrhoea, dehydration, and one mortality, along with evidence of renal tubular damage in one animal [4], [6]. Although DAM exhibits low acute toxicity in ICR mice, its safety at therapeutic doses, particularly with repeated DAM20 administration in xenografted nude mice, remains unclear. In addition, it was previously reported that long-term treatment of anthraquinone was nephrotoxic to rats and patients [7], [8]. As DAM20 efficacy depends on multiple dose administration, further investigation into its long-term cumulative toxicity, in particular, nephrotoxicity, is essential before advancing toward clinical application.
Caspase-3 plays an essential role in the activation of the apoptosome or the activation of phospholipase A2 (iPLA2) for preventing renal injury and regeneration of normal renal tubular cells (Fig. 1) [9], [10], [11], [12], [13]. The release of TNF-α and DAMPs from injured proximal tubule epithelial cells could induce apoptosis, necrosis and recruitment of macrophages [14], [15], [16]. Classical M1 subpopulation of macrophages promotes further release of TNF-α and gradually induces apoptosis of injured proximal tubule epithelial cells, resulting in either regeneration or remodelling of proximal tubule epithelial cells [15], [17]. If cellular injuries are unresolved or prolonged to sustain the inflammation, the polarisation of M1 to M2 macrophages as well as M2 macrophage subtypes is destined, resulting in the kidney profibrotic stage. Release of fibrotic factors such as TGF-β, TNF-α and Osteopontin at the injury sites acts as EMT-activators to stimulate the differentiation of fibroblasts and the development of fibrosis (Fig. 1) [17], [18], [19], [20], [21], [22], [23]. Kidney Injury Molecule-1 (KIM-1) is a sensitive early biomarker of proximal tubular injury, markedly upregulated in response to epithelial stress but not directly involved in apoptotic signaling [24]. Although KIM-1 does not induce cell death, its expression often coincides with early apoptosis markers, indicating sublethal injury or the initiation of programmed cell death [25], [26]. Co-detection of KIM-1 and caspase-3 activation in murine kidneys may reflect an early, potentially reversible stage of tubular damage, where stress responses and apoptotic signaling pathways overlap [11], [25], [26].
T-cell-deficient nude mice were commonly used in colorectal cancer research due to their compromised adaptive immune system, which allowed the successful engraftment and sustained growth of human colorectal cancer xenografts. This enabled in vivo investigation of tumour growth, invasiveness, and metastasis, and played a critical role in the evaluation of anti-cancer therapies [27]. Assessing nephrotoxicity in xenografted nude represented a critical facet of preclinical drug development, as numerous chemotherapeutic compounds, while demonstrating their tumour-suppressing effects, could induce considerable renal damage in these models [28], [29]. Cisplatin induces acute tubular necrosis primarily in the proximal tubules through ROS generation, mitochondrial dysfunction, and activation of p53 and caspase pathways, leading to reduced GFR, electrolyte imbalances, and early elevation of renal injury biomarkers such as KIM-1 and neutrophil gelatinase-associated lipocalin (NGAL) before histological damage becomes apparent [30]. Carboplatin, though less nephrotoxic, exerted similar effects [30]. Ifosfamide caused proximal tubular injury via its metabolite chloroacetaldehyde, which disrupts mitochondrial respiration, leading to tubular injury and acute kidney injury (AKI) [31]. Targeted anti-cancer therapies such as EGFR and VEGF inhibitors could also impair tubular repair and endothelial integrity, respectively, and may result in thrombotic microangiopathy [32]. Immunotherapies, including checkpoint inhibitors and CAR-T cells, could trigger immune-mediated nephropathies via cytokine release and T-cell activation, presenting as interstitial nephritis or glomerulonephritis [32.33]. These findings highlight the importance of early nephrotoxicity detection using sensitive biomarkers like KIM-1 and NAGL to monitor subclinical renal injury in conjunction with apoptotic activity during preclinical drug development.
This study evaluated blood and urine biochemistry, renal ultrasonography, and histopathology in nude mice following repeated DAM20 administration. Renal injury was assessed using histological staining, immunohistochemistry, and detection of KIM-1 expression and DNA fragmentation to elucidate the underlying mechanisms of DAM-induced nephrotoxicity

2
Materials and methods
2.1
Damnacanthal (DAM) preparation
Purified DAM was kindly provided by the Drug Discovery and Development Center (DDC), Thammasat University. DAM stock solution was dissolved in Dichloromethane, dimethyl- sulfoxide (DMSO) (ACI Labscan, Bangkok, Thailand), and Tween-80 (Esteem Industries, PVT. LTD, Mumbai, India) and mixed using an ultrasonic bath at 50 °C. The 2000 (DAM2000), 300 (DAM300), and 20 (DAM20) mg damnacanthal per kilogram of mouse body weight was prepared from stock solution by dilution with 10 % DMSO and Tween80 in Dulbecco's phosphate-buffered saline (DPBS) (Gibco, Carlsbad, CA, USA) [4].

2.2
Animal housing and experiment
Animal protocols were approved by the Institutional Animal Care and Use Committee (IACUC) of Thammasat University (Protocol no. 003/2022). Six to eight-week-old male athymic BALB/cALcl-nu/nu nude mice were purchased from Nomura Siam International Company Limited (Thailand) and acclimatised for seven days before initiating the experiment. Mice were housed in individually ventilated cages (IVC) at 21 ± 1 °C, 30–70 % humidity, and with a 12-hour light-dark cycle. Sterile pellet feed and water were provided ad libitum.
Sample size determination was performed using G*Power software version 3.1 [33], yielding a total of 16 nude mice. The calculation was based on an estimated effect size (Cohen’s f) of 0.875, derived from preclinical studies involving rodent models of cancer and stroke, as reported by Holman et al. (2016) [34]. The statistical parameters included an α-error probability of 0.10 and a power (1 −β) of 0.80, assuming a one-way ANOVA design with four independent experimental groups. Nude mice were randomly assigned into four mice per group. Three experimental groups received repeated administrations of damnacanthal (DAM20; 20 mg/kg) via oral gavage every other day for three, seven, or fourteen doses, respectively, as illustrated in Fig. 2
[4]. Corresponding vehicle control groups were administered 10 % DMSO/Tween 80 in DPBS (Veh) following the same dosing schedule and duration.
Following administration, feed and water consumption were monitored daily. Mice grimace scale (MGS) consisted of orbital tightening, nose bulge, cheek bulge, and ear and whisker positions were investigated daily for mice behaviour presentation. MGS was presented as an Action Unit of 0–2 where 0, 1 and 2 were criterion is absent, moderately present and prominently present, respectively. Physical abnormality signs, time of onset, length of recovery period, and mortality were observed after treatment for the first four hours and then daily. Physical signs included changes in skin and fur, eyes, mucous membranes, respiratory, tremors, convulsions, salivation, diarrhoea, lethargy, sleep, coma, and death, according to OECD Guideline 423 [6]. At the end of each experiment, mice were euthanised by intraperitoneal injection of 50 mg/kg Thiopental followed by cervical dislocation [35]. Blood samples were collected from cardiac puncture. Kidneys were excised and immediately fixed in 10 % neutral buffer formalin. (Surgipath, Leica, USA) (Fig. 2)

2.3
Urinalysis and Fecal occult blood (FOB)
The fresh urine sample was collected from DAM20 treated and Veh control mice according to the schedule in Fig. 2. Urine creatinine and microalbumin were measured using the URIT-500B automated urine analyser, which operates on the principle of reflectance photometry. The intensity of the photometric signal change was directly proportional to the concentration of microalbumin, expressed in mg/dL. Microalbumin/creatinine ratios were also calculated [36]. FOB was determined to indicate gastrointestinal tract bleeding, which could be a sign of end-stage of chronic kidney disease (CKD) [37]. Fresh faeces were collected according to the schedule in Fig. 2 to determine haemoglobin (HB) concentration using FOB test kit (CTK Biotech, USA). The FOB test kit lowest limit HB detection was 25 ng/ml.

2.4
Kidney ultrasonography
Mice were anaesthetised with isoflurane inhalation, and the right kidney was subjected to ultrasound imaging by using Vevo3100 Visual Sonics (FUJIFILM, Japan). Kidney volume, percentage of kidney vascularity and mean velocity of the renal artery were assessed [38], [39].

2.5
Hematology and blood biochemistry determination
Blood samples were collected, and a complete blood count (CBC) was determined by using ABX Micros (ESV60, HORIBA, USA). Kidney function tests for DAM20 treated mice, blood urea nitrogen (BUN) and creatinine were evaluated using a DRI-CHEM NX500 automated analyser (FUJIFILM) [40], [41].

2.6
Kidney histopathology
2.6.1
Preliminary study
Positive and negative renal tubular fibrosis tissues were from ICR mice administered with a single dose of DAM300, DAM2000, and veh control from the previous study as a paraffin-embedded kidney tissue [4]. Embedded renal tissues were sectioned to three micrometres thickness and stained with Hematoxylin and Eosin (H&E) (C.V. Laboratories, Thailand) followed by dehydration and mounted [42], [43]. The level of extracellular matrix deposition in the renal cortex was determined by staining for fibronectin and collagen deposition with Periodic Acid-Schiff (PAS; ITW reagents, VWR Chemicals, Germany) and Masson's trichrome staining (Abcam, UK), respectively [44], [45]. Thirty non-overlapping snapshot micrographs of H&E stained renal cortex area at a magnification of 400x were randomly captured under the automated laser microscope (BioTek Lionheart FX, Agilent) [43], [44], [45]. The renal tubule abnormalities, including desquamated structure, vacuolated and pyknotic nuclei cells, were visualised and scored into five scores where 0 was an absence of abnormality, 1 was one to two, 2 was three to six, 3 was seven to ten abnormality foci and 4 was coalescing to diffuse lesion or seven to ten abnormality foci, respectively [43], [45], [46], [47].
Thirty glomeruli were randomly selected and graded for shrunken (atrophy) or necrotic glomeruli into five scores where 0 was absent, 1 was one to five, 2 was six to ten, 3 was 11–15 and 4 were more than 16 atrophy or necrotic glomeruli [47], [48].
Thirty of PAS and Masson's trichrome-stained renal cortex images were randomly selected and individually captured under the automated laser microscope (BioTek Lionheart FX) at a magnification of 400x. PAS stained fibronection in fuchsia pink, and Masson's trichrome stained collagen fibres in blue. Fuchsia pink or blue staining areas in each micrograph were then marked and calculated for the percentage of positive staining relative to the entire area of capture frames using ImageJ software (http://imagej.en.softonic.com). The single colour threshold for fibronectin and collagen fibre were individually calibrated and set. The average area percentage of fibronectin/collagen fibre deposition was calculated from randomly selected 30 cortex fields of each section and compared to that of the section from the Veh control mouse to indicate the presence of fibrosis.
Images that were captured from randomly selected fields within the renal cortex for each sample were further visualised and observed in a blinded manner by three independent observers who were unaware of the treatment groups, to minimise bias during characterisation and quantification.

2.6.2
Kidney histopathology of nude mice treated with repeated doses of DAM20
Kidneys were collected from mice treated with three, seven and 14 repeated doses of DAM20 and their control groups and fixed in 4 % paraformaldehyde buffer. Fixed kidney tissues were processed and paraffin-embedded using an automated tissue processor (HistoCore PEARL, Leica, USA). Subsequently, three micrometre-thickness sections were cut and stained with H&E, PAS or Masson's trichrome staining. Pathological lesions, as well as fibronectin and collagen fibre depositions were visualised and then graded or calculated as previously described [49].

2.7
Immunohistochemistry
2.7.1
Detection of caspase 3 activation and macrophage infiltration
Five micrometre-thick sections of paraffin-embedded kidney tissues from mice treated with three, seven and 14 repeated doses of DAM20 and their respective control groups were subjected to deparaffinisation, antigen retrieval and blocking.
In order to detect whether renal epithelial cell injury and apoptosis were initiated, kidney sections were incubated with 1:200 caspase 3 recombinant rabbit monoclonal anti-body (9H19L2) (Thermo Fisher Scientific, USA) [49], [50], [51]. Macrophage infiltration and M2 macrophage polarisation in kidney sections were separately detected by incubation with 1:200 dilution of F4/80 BM8 and CD206 MR5D3 rat monoclonal antibodies, respectively [52], [53]. Subsequently, the sections were washed and incubated with 1:1000 secondary antibodies (HRP-conjugated goat anti-rabbit IgG [H+L] or HRP goat anti-rat IgG [H+L]) (Thermo Fisher Scientific) for one hour. After washing four times, Diaminobenzidine (DAB) substrate was added, followed by counterstained with hematoxylin, dehydrated and mounted [53], [54], [55]. Section images were visualised and captured at magnifications 400x (to detect activated caspase-3) and 200x (to detect F4/80 and CD206 macrophage markers). The brownish immunostaining indicated a positive area of activated caspase-3 and F4/80 and CD206 positive macrophages

2.7.2
Detection of DNA fragmentation in kidney tissue using TUNEL assay
In order to detect DNA Fragmentation, a functional consequence of apoptosis, a TUNEL Assay Kit-HRP-DAB (Abcam; ab206386) were used. Briefly, 10 μm-thick sections of paraffin-embedded kidney tissue from mice treated with three, seven, and 14 repeated doses of DAM20 and their respective control groups were subjected to deparaffinisation, rehydration, and treated with Proteinase K to permeabilise the kidney cells. Endogenous peroxidase activity was quenched using 3 % hydrogen peroxide (H₂O₂). The exposed 3’-OH ends of fragmented DNA generated during apoptosis in the tissue sections were then labelled by incubating with Terminal deoxynucleotidyl Transferase (TdT). Next, sections were blocked with blocking buffer and incubated with an HRP-conjugated antibody. The signal was visualised by the addition of DAB substrate and counterstaining with methyl green. The section slides were dehydrated and mounted for imaging at 200x magnification. Twenty micrographs of the renal cortex were captured and analysed using Gen5 Image Prime 3.14 software as described previously. The presence of brown-stained cells indicated the apoptotic renal cells with the presence of DNA fragmentation. Negative control was an untreated mouse kidney tissue section. A positive control was prepared by treatment of mouse kidney tissue section with DNase I (1 μg/μL in TBS/1 mM MgSO₄; Thermo Scientific, EN0523) prior to Proteinase K incubation.

2.7.3
Computer-assisted microscopical analysis
For the quantitative evaluation an area percentage calculation, each selected-area was photographed using automated laser microscope (BioTek Lionheart FX) at a specified magnification. The numbers of labelled cell or areas were determined using Gen5 Image Prime 3.1.4 in a blinded manner by three independent observers as described previously.

2.8
Determination of KIM-1 expression by relative q-RT PCR
To assess KIM-1 gene expression, total RNA was extracted from the right kidney tissues preserved in RNAlater solution (Thermo Fisher Scientific) at −20 °C, using the RNeasy Mini Kit (Qiagen, Hilden, Germany). A total of 100 ng RNA was used in a quantitative reverse transcription PCR (qRT-PCR) reaction. qRT-PCR was performed using one-step SYBR-based real-time RT-PCR with the Brilliant II SYBR Green qRT-PCR Master Mix Kit (Agilent Technologies, USA) on a CFX Opus 96 Dx Real-Time PCR System (Bio-Rad, Ho Chi Minh City, Vietnam). The primers used for quantitative real-time PCR were 5’- TCA GAA GAG CAG TCG GTA CAA C-3’ (forward) and 5’- TGT AGC TGT GGG CCT TGT AGT-3’ (reverse) [56]
β-actin served as the internal control for calculation of relative gene expression. A cycle threshold (Ct value of 37 was set as the cut-off negative threshold value. Total RNA extracted from A549 cell line (40 ng) was used as a positive control [57], [58].

2.9
Statistical analysis
The results are present as the mean ± standard deviation. Comparisons between the Veh and DAM20 groups were performed using ANOVA or Dunnett's test as appropriate, p < 0.05 was statistically significant difference.

3
Results
3.1
Monitoring of DAM20 and Veh-treated nude mice during the study
Mice orally administered with three, seven and 14 repeated doses of DAM20 and vehicle control (Veh) showed no clinical signs of any toxicity effects such as pale membrane, tearing or diarrhea or behaviour changes. All mice also showed an MGS action unit of zero (Table S1). Feed and water intake in the DAM20 and Veh groups fluctuated slightly with no statistically significant differences throughout the experiments (Fig. S1). Moreover, none of the mice from repeated doses DAM20 administration groups demonstrated the 10 % decrease in body weight compared to Veh control suggested that multiple repeated DAM20 administrations in nude mice did not cause any physical toxicity (Fig. 3) [6].

3.2
Renal Ultrasonography of DAM 20 treated nude mice
Ultrasonography of the kidney indicated that vascularity percentage and mean blood velocity from all mice in DAM20 multiple treatment had no significant difference compared to Veh groups.
Interestingly, the average kidney volumes of mice from 14 repeated DAM20 (28-day duration) treatment were significantly larger than that of the respective Veh control and the other DAM20 treatment groups (Fig. 4 and Table S2). Moreover, among 14 repeated DAM20 groups, only one mouse (number DL04) had the standard size of normal kidneys.

3.3
Urinalysis of DAM20 and Veh treated nude mice
Although no significant difference in urine protein and microalbumin/creatinine ratio was found in all DAM20 treated mice compared to Veh control, urine microalbumin of mice from the seven repeated doses of the DAM20 treatment group (19.59 ± 4.48 mg/dl) were significantly elevated from its Veh control group (12.48 ± 2.25 mg/dl) (Fig. 5 and Table S3). However, the microalbumin levels were reversal to be insignificant from the control after the 14 repeated doses of DAM20 (Fig. 5C). Interestingly, mice treated with 14 repeated doses of DAM20 exhibited significantly elevated urine creatinine levels (53.19 ± 5.55 mg/dL) compared to the vehicle-treated control group (42.75 ± 7.27 mg/dL) (Fig. 5B and C). This elevation might be the consequence of various factors and was generally indicative of increased urinary concentration or reduced urine output. However, to accurately evaluate renal function, urine creatinine levels must be interpreted alongside other renal biomarkers. Future studies assessing glomerular filtration rate (GFR) and creatinine clearance using specialised metabolic cages for precise urine volume collection or measuring the elimination kinetics of fluorescein isothiocyanate-labeled sinistrin (FITC-sinistrin) via transcutaneous sensors are necessary for a more comprehensive analysis of renal function.

3.4
Hematology and Blood chemistry evaluation of DAM20 and Veh-treated nude mice
CBC information, including white blood and red blood cell counts, hematocrit, haemoglobin concentration, and platelet count from all mice in DAM20 and Veh groups, were within the normal range of male nude mice (Table S4) [41]. The absence of anaemias, which were the expected manifestation of chronic kidney disease, suggested that the presence of chronic kidney malfunction in repeated dose DAM20-treated mice, if any, was either in an early stage or mild. Serum BUN and creatinine measurements, as well as urine Creatinine/serum creatinine (UCr/SCr) ratio, revealed that following the three or 14 repeated doses of DAM20, the kidney function remained in a normal range (Table 1) [41]. However, treatment with seven repeated doses of DAM20 had significantly lower BUN compared to the Veh control. Whereas UCr/SCr ratios of mice in this group were also significantly increased, suggesting the possibility of either early or mild renal tubular injury following exposure to nephrotoxicant or enhancing in glomerular filtration of creatinine. However, this elevated ratio could also be the consequence from various physiological or pathological conditions such as reduction of plasma volume (e.g., due to dehydration), increased muscle mass, high-protein intake, mild liver dysfunction, or malnutrition.

3.5
Histopathology examination of mice treated with high doses of damnacanthal
3.5.1
Histopathology examination of mice treated with DAM300 and DAM2000
The H&E staining of renal tissues from nude mice treated with DAM300, DAM2000 and Veh control showed no sign of shrunken glomerulus, renal tubule abnormalities at proximal and distal tubules [47], [48]. These data further confirm that damnacanthal had a relatively low acute oral toxicity (Fig. 6).
The PAS and Masson's trichrome staining were used to detect the deposition of fibronectin and collagen fibre, respectively. The increase in the density of both dyes in D300- and D2000-treated mice demonstrated the thickening of the interstitial space and tubular basement membrane containing fibronectin and collagen fibre, which is the hallmark of renal tubular fibrosis (Fig. 7) [44], [46], [49]. Furthermore, fibronectin and collagen fibre deposition densities in renal tubular areas were further subjected to quantitative measurement by calculating the mean density of the PAS- and Masson's trichrome-positive structures relative to the entire visualisation microscopic field using ImageJ Software. The deposition of fibronectin and collagen fibre in kidney tissues of mice treated with DAM300 and DAM2000 was significantly higher compared to the control group (p < 0.05) (Table 2).

3.5.2
Histopathology examination of mice treated with repeated doses of DAM20
Similar to H&E staining of renal tissue from mice treated with DAM300 and DAM2000 treated mice, renal tissue from mice treated with the repeated doses of DAM20 and Veh control also showed no sign of any pathological lesion in the glomerulus, proximal and distal tubules (Fig. S2). Interestingly, the area percentages of extracellular matrixes (fibronectin and collagen fibre) deposition in the renal tissue of mice treated with repeated doses of DAM20 had no significant difference from that of the Veh control group (Fig. 8) (Table 3)

3.6
Assessment of apoptosis and macrophage subtype infiltration in renal tissue of mice treated with repeated dosed of DAM20 by immunohistochemistry
3.6.1
Caspase-3 activation in renal tissue of mice
Immunohistochemical staining of the renal cortex of mice treated with three and seven repeated doses of DAM20 revealed that none were positive for caspase-3 activation, suggesting that none of the renal cortex tubular cells from those groups of mice underwent apoptosis (Fig. 9) Surprisingly, renal cortex tissue from one mouse in 14 repeated doses DAM20 (mouse no. DL01) showed an average 0.14 % positive area for caspase-3 activation, suggesting that the 14 repeated doses of DAM20 caused renal tubular cell apoptosis (Fig. 9 and Table 4).

3.6.2
Macrophage subtype infiltration renal tissue of mice
Macrophage infiltration and macrophage subtypes play a crucial role in renal injury, repair and tissue remodeling to cause renal fibrosis. The antibodies against surface markers F4/80 and CD206 were used in immunohistochemical reactions to detect pan and M2 macrophage populations specifically. It was found that the renal cortex tissues from all tested mice showed no brown staining of DAB substrate, indicating no macrophage as well as M2 macrophage infiltrations (Fig. 10 and Table 4).

3.6.3
TUNEL assay for detection of DNA fragmentation in renal tissue of mice
TUNEL assay of the renal cortex in mice treated with three, seven, and 14 repeated doses of DAM20 showed no positive staining for DNA fragmentation, indicating the absence of apoptosis in renal tubular cells [Fig. S3]. However, a weak caspase-3 positivity in one mouse treated with 14 repeated doses of DAM20 suggested that, despite initiating the apoptotic pathway, it did not progress to the terminal stage. This partial activation might reflect the non-apoptotic roles of caspase-3, such as promoting tissue repair through activation of phospholipase A2 and induction of vascular endothelial growth factor (VEGF) production [10], [11] [Fig. 1].

3.7
Expression of KIM-1 biomarker for early kidney injury
No significant differences in KIM-1 expression relative to β-actin were observed in the kidney tissues of nude mice treated with three or seven repeated doses of DAM20 compared to the vehicle control (p = 0.277 and 0.070, respectively) [Fig. 11]. Notably, KIM-1 expression was significantly elevated in the group treated with 14 repeated doses of DAM20 compared to its vehicle control (p = 0.012) [Fig. 11]. This finding suggests early signs of proximal tubular epithelial cell injury, despite only one mouse in this group showing caspase-3 activation (Fig. 9). Furthermore, the absence of TUNEL-positive staining in this group indicates that the kidney tissue may not have progressed to the stage of DNA fragmentation.

4
Discussion
DAM was previously demonstrated as an effective anticancer candidate for the treatment of colorectal cancer with a relatively low acute oral toxicity. However, the efficacy of the therapeutic regimen required multiple repeated administrations. This study, therefore, aims to assess the renal toxicity of therapeutic DAM20 following three, seven and 14 repeated administrations in nude mice, recipients of colorectal cancer xenograft.
The increase in renal volume following 14 repeated doses of DAM20 (28-day duration of treatment) suggested the signs of high fluid retention in the kidney. Generally, exposure to neurotoxicants stimulates caspase-3 activation in the renal epithelial cells. Apart from apoptosis initiation, the activation of iPLA2 and prostaglandin E2 by activated caspase-3 could lead to cycle repair in renal cells (Fig. 1). During this process, the renal arteriole could undergo vasoconstriction and increased reabsorption, resulting in renal fluid retention and oliguria, as demonstrated by the increase in renal volume seen in 14 repeated doses of DAM20 [38], [39]. After injury, the renal tubule might successfully repair, recover and restore its function. The polyurea to eliminate the accumulated metabolic waste frequently occurs prior to gaining its normal function. However, in this study, gross observation of the wetness of the bedding of mice receiving 14 repeated doses of DAM20 showed no sign of increasing wetness or polyurea (data not shown). Nevertheless, the measurement of 24-hour urine production requires future studies to confirm the details of polyuria condition. If acute renal injury persisted, could not be resolved, or partially resolved, stage one asymptomatic chronic kidney disease (CKD) would be the outcome. The possibility that asymptomatic CKD might occur in nude mice treated with multiple treatments of DAM20 was further confirmed by the finding that seven repeated DAM20 administrations in mice resulted in low BUN, urine microalbumin leakage and increased UCr/SCr ratios. Although DAM20-treated nude mice exhibited no obvious clinical manifestations or behavioural indicators of nephrotoxicity, biochemical profiling revealed early subclinical alterations suggestive of renal involvement. Specifically, a reduction in blood urea nitrogen (BUN) levels, coupled with an elevated urine-to-serum creatinine (UCr/SCr) ratio, was observed. The decrease in BUN might not reflect improved renal function but could instead be the consequence of extrarenal influences such as suppressed hepatic urea synthesis, reduced dietary protein intake, or DAM-induced modulation of nitrogen metabolism [59], [60]. Additionally, altered tubular reabsorption of urea secondary to proximal tubular dysfunction could not be excluded. The increased UCr/SCr ratio, while sometimes indicative of enhanced renal excretory function or recovery from injury, may alternatively signal early compensatory responses to subtle tubular damage or segment-specific dysfunction not yet evident on histological examination [61], [62]. These findings implied that glomerular and tubular functions might have been partially compromised, potentially due to low-grade inflammation or microvascular injury involving renal endothelial cells [63]. Moreover, the dissociation between biochemical abnormalities and the absence of histopathological or gross phenotypic changes emphasized the limitations of conventional toxicity endpoints in detecting early renal impairment. This highlights the critical need to incorporate sensitive renal injury biomarkers such as KIM-1, NGAL, and DNA fragmentation assays (TUNEL) to improve the determination of nephrotoxicity assessment in preclinical models [64], [65].
Although the BUN and microalbumin leakage levels were reversible after the 14 repeated doses of DAM20, the high urine creatinine in this period was the manifestation of partially resolved renal injury or asymptomatic CKD. However, the low BUN compared to the Veh control could also suggest the possibility of either mild liver impairment or malnutrition. However, the body weight and food intake in this group were similar to those of other control and treatment groups (Figure S1). Furthermore, the low BUN in the seven repeated doses of the DAM20 group might be due to overhydration from the interference of kidney function in the respective urine microalbumin leakage. Moreover, the interference of liver and renal activities, as well as undetectable loss of appetite, could not be excluded because Woradulayapinij and colleagues previously reported that ICR mice treated with single DAM2000 caused mid-zonal hepatocellular necrosis and mild infiltration of inflammatory cells and a low degree of single-cell necrosis in the renal tubules [4]. In addition, the absence of faecal occult blood in all mice treated with repeated doses of DAM20 indicating that this asymptomatic CKD did not lead to severe kidney injury or symptomatic end stage CKD [22], [23].
Furthermore, for a better understanding of the interference of kidney function by DAM20, the blood and urine chemistries should be interpreted in combination with renal histopathologic data. H&E, PAS and Masson’s trichrome staining elucidated that the glomerulus atrophy, renal cell damage and necrosis were absent from all mice treated with repeated doses of DAM20 and Veh control, indicating the absence of glomerular cell atrophy and renal cell damage or necrosis [47], [49]. However, the previous finding reported that treating mice with a single DAM2000 resulted in a few single-tubular cell necroses in some mice. Moreover, immunohistochemical staining of the renal cortex treated with three and seven repeated doses of DAM20 revealed that none were positive for caspase-3 activation, suggesting that none of the renal cortex tubular cells from these mice underwent apoptosis initiation [9], [15], [16], [23]. However, the finding that the caspase-3 activation was found in kidney tubular cells of one out of four mice treated with 14 repeated doses of DAM20 indicated that renal injury did occur as apoptotic pathway was initiated in this mouse. Moreover, the renal tubules of mice in all DAM20 repeated treatment groups showed negative results in the TUNEL assay, suggesting that although apoptotic signaling might had been triggered, it did not progress to the stage of DNA fragmentation or cell death. This observation correlated with the previous finding that caspase-3 activation could occur in non-lethal cellular processes such as differentiation and tissue repair. Furthermore, the induction of KIM-1, an early biomarker for renal tubular injury without any histopathological lesion following 14 repeated DAM20 treatments clearly demonstrated that the nephrotoxicity of DAM20 repeated treatment in nude mice was at a very early stage, mild, or potentially reversible.
In an unresolved renal tubule injury, inflammation persists and macrophages infiltrate. Macrophage subtypes play a crucial role in the decision to repair or remodel renal tubules to cause renal fibrosis (Fig. 1). The immunohistochemical assay indicated that none of the mouse renal cortex tissue sections of mice treated with the repeated doses of DAM 20 and Veh had pan macrophage (surface markers F4/80 positive) and M2 (CD206 positive) M2 macrophage populations infiltration. Consequently, this data indicated that the caspase-3 activation found in mouse tubular cells treated with 14 repeated doses of DAM20 might lead to cycle repair by activation of phospholipase A2 and induced the production of vascular endothelial growth factor in the absence of macrophage infiltration. In addition, the absence of macrophage infiltration, apoptotic cells and the deposition of extracellular matrix in the renal cortex of all DAM 20 treated mice indicated that the injury to renal tubular cells underwent a repair cycle and restored their normal morphology and most of their function without a sign of any tissue remodelling. However, the enlargement of the kidney and increased urine creatinine without any histopathological lesion following 14 repeated DAM20 treatments clearly demonstrated that the nephrotoxicity of DAM20 repeated treatment in nude mice was either mild or reversible, and the signs of chronic kidney disease were mild. Nevertheless, currently, it was not known whether apoptosis continues to activate the other factors further downstream of the apoptotic pathway beyond 14 repeated doses of DAM20. Future studies are warranted to clarify the progression and potential reversibility of DAM20-induced renal effects using a larger-scale longitudinal design with extended observation periods. This should include kinetic assessment of renal function, early kidney injury and inflammatory and endothelial biomarkers, and long-term apoptotic and histopathological outcomes. Real-time monitoring of DAM-induced nephrotoxicity may be achieved using KIM-1-GFP transgenic mice or 3D human kidney organoids, while integrated transcriptomic, proteomic, and metabolomic analyses can be used to elucidate key cellular pathways involved.

5
Conclusions
DAM effectively inhibits colorectal tumour xenografts in nude mice, and the previous study showed that the administration of the therapeutic dose of DAM20 can be generally safe. However, mice administered with a single dose of DAM2000 (100 times higher amount than the therapeutic dose) showed signs of toxicity to the liver, kidney, and gastrointestinal tract. The repeated doses of damnacanthal were necessary for anti-cancer activity. The mild impairment of renal and hepatic function was detected following seven repeated doses of DAM20 treatment without any histopathological lesion of the renal tubule. Upon 14 repeated doses of DAM20, significant kidney enlargement without the sign of histopathological lesion of the renal tubule and some impairment of kidney function were demonstrated. Although one mouse showed positivity for caspase-3 activation, none of the apoptotic and inflammatory activities were detected, indicating nephrotoxicity was mild and the repair and regeneration processes of renal injury restoring to normal tissue without further tissue remodelling and fibrosis. Mild CKD renal caused by the repeated administration of DAM20 might be reversible. However, this finding warrants further investigation using comprehensive molecular, functional, and longitudinal analyses to accurately characterize DAM renal toxicity and safety profile.

Authors statement
We would like to express our sincere appreciation to you and the reviewers for the continued time, effort, and insightful feedback provided throughout the review process. In response to the most recent comments, we have carefully revised the manuscript once again to address all remaining concerns.
A detailed, point-by-point response to each comment has been included in the accompanying response letter. The revised manuscript incorporates additional clarifications, further improvements in data presentation, and refinements to the methodology and discussion to enhance scientific rigor and clarity. All revisions have been clearly highlighted in the manuscript for your convenience.
We are grateful for the opportunity to submit this second amended version and respectfully request its reconsideration for publication in Toxicology Reports.

CRediT authorship contribution statement
Jeeraphong Thanongsaksrikul: Writing – review & editing, Supervision, Funding acquisition, Formal analysis. Potjanee Srimanote: Writing – review & editing, Supervision, Project administration, Funding acquisition, Conceptualization. Thunyatorn Yimsoo: Writing – review & editing, Validation, Resources, Methodology, Investigation. Noppanan Kotsaouppara: Writing – original draft, Methodology, Investigation, Formal analysis. Worapapar Treesuppharat: Writing – review & editing, Resources, Methodology.

Institutional review board statement
Study design and experimental procedures were approved by the Institutional Animal Care and Use Committee (IACUC) of Thammasat University (Protocol no. 003/2022).

Informed consent statement
Not applicable.

Supplementary materials
Supplementary data associated with this article can be found in the online version, Table S1: Action Unit (AU) of Mice grimace scale (MGS) following 14 repeated doses of DAM 20 (28-day duration); Table S2: Ultrasonography analysis of renal physiology following DAM20 and Veh treatment; Table S3: Urinalysis data of mice following DAM20 and Veh treatment; Table S4: CBC parameters of mice following DAM20 and Veh treatment; Table S5: KIM-1 expression relative to β-actin in mice kidney following DAM20 and Veh treatment Fig. S1: Feed and water intake of nude mice treated with DAM 20 and Veh control; Fig. S2: H&E staining of renal tissues of mice treated with Veh and DAM20; Fig. S3: TUNEL assay to detect DNA fragmentation;

Funding
This work was supported by the Thailand Science Research and Innovation Fundamental Fund fiscal year 2023, grant number 180178, and the Thammasat University Research Unit in Molecular Pathogenesis and Immunology of Infectious Disease, grant number 2567 to J.T.

Declaration of Competing Interest
The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.