In Vivo Genotoxicity and Cytotoxicity Kinetics of Trimethoprim Sulfamethoxazole in Well-nourished and Undernourished Young Rats

Abstract

Background/Aim

Undernutrition is a serious health problem prevalent in poor countries, affecting millions of people worldwide, especially young children, pregnant women, and sick elderly individuals. This condition increases vulnerability to infections, leading to widespread use of antibiotic treatments in undernourished populations. The objective of the present study was to determine the in vivo genotoxic and cytotoxic effects of trimethoprim-sulfamethoxazole (TMP-SMX) treatment according to nutritional conditions.

Materials and Methods

The effects of TMP-SMX treatment were measured by analyzing the kinetics of micronucleated reticulocytes (MN-RET) induced in the peripheral blood of young, well-nourished (WN) and undernourished (UN) rats.

Results

In the WN group, two distinct peaks of MN-RET were observed, while the UN group had a significantly higher basal frequency of MN-RET compared to the WN group and only a later peak. Reticulocyte (RET) frequency slightly decreased in WN, indicating a poor cytotoxic effect. In contrast, in the UN, the treatment caused a significant increase in RET frequency. The results indicate that SMX’s aromaticity index decreases when formed with TMP, suggesting potentially fewer toxic effects.

Conclusion

In vivo TMP-SMX produces two MN-RET induction peaks in WN animals, indicating two DNA damage induction mechanisms and consequent micronucleus production. The UN rats did not display the two peaks, indicating that the first MN induction mechanism did not occur in UN, possibly due to pharmacokinetic effects, decreased metabolism or effects on cell proliferation. TMP-SMX has a slight cytotoxic effect on WN. In contrast, in the UN, the antibiotic treatment seems to favor early erythropoiesis.

Undernutrition, commonly referred to as malnutrition (protein-calorie malnutrition), is caused by deficiencies in micronutrients and macronutrients, such as proteins, carbohydrates, and fat. Illness is caused by an imbalance in the amount and caliber of macro- and micronutrients, which are essential for an organism’s healthy growth and development (1). Undernutrition accounts for approximately 45% of child mortality in individuals under the age of five (2).

Malnutrition is a cause of increased infant mortality and is often associated with infectious diseases in countries where poverty and poverty-related factors are prevalent (3). Diarrhea and acute respiratory infections are the main causes of death in children under five years of age (4). The impact of undernutrition is often underestimated, and its sequelae are diverse and include those already mentioned, deterioration of child development and increased susceptibility to obesity at future ages.

Since undernutrition is frequently associated with several infections, antibiotic treatments are extensively used for undernourished patients. The prescribed drug Bactrim consists of trimethoprim (TMP) in combination with sulfamethoxazole (SMX).

Trimethoprim and sulfamethoxazole (TMP-SMX). Sulfamethoxazole and trimethoprim are bacteriostatic and obstruct the synthesis of folate. The trimethoprim-sulfamethoxazole (TMP-SMX) combination has a synergistic antifolate effect. Therefore, this mixture may have a bactericidal effect (5,6). Prophylactic antibiotic administration for severe malnutrition has been shown to improve the rate of recovery, promote weight gain, and decrease mortality in children (7).

Since antibiotics are considered essential in various protocols for the management of undernourished patients, studies on their effects are required, and alternatives for their proper use are needed. Optimal TMP-SMX dosing strategies for patients with pneumonia have been proposed (8). Care should also be taken when prescribing TMP-SMX, and bacterial resistance should be considered (9).

Several investigations have been conducted to ascertain the genotoxic potential of TMP. TMP treatment increased the frequency of micronuclei and sister chromatid exchanges (SCE) in human cells in vitro (10). Additionally, an increase in micronuclei frequency was observed using the cytokinesis-block micronucleus (CBMN) assay (11,12). By using the same technique, minor increases in micronuclei and SCE were found following SMX therapy. These findings revealed drug genotoxicity (10).

Several investigations have been conducted on the genotoxicity of the combination of TMP-SMX and undernutrition, and the authors found that when therapeutic doses were given to undernourished (UN) rats, the frequency of micronuclei (MN) increased (13).

Additionally, in a prior study, the frequency of mutations was measured using the Pig-a assay, which is a useful method for assessing in vivo gene alterations. The frequency of Pig-a mutations increased in both WN and UN rats treated with TMP-SMX, indicating the mutagenic potential of the drug (14).

Numerous investigations on the cellular and genetic consequences of starvation have been conducted on both experimental animals and humans. These investigations demonstrated that chromosomal abnormalities (15), sister chromatid exchange (16), and micronuclei (17,18) all indicate that nutritional undernutrition increases susceptibility to chromosomal damage.

Micronucleus (MN). A commonly used technique for assessing the genotoxicity of various chemical, physical, and biological factors is the MN assay (19). The in vivo application of the MN assay in rodents has made possible hazard identification and risk assessment (20).

MN can arise due to mitotic spindle apparatus dysfunction, typically caused by aneuploidogens, or double-stranded DNA breaks induced by clastogens (20). By integrating the use of flow cytometry into the analysis of MN in erythrocytes, it was possible to increase the accuracy and speed of the assay (21-23).

The kinetics of MN induction in polychromatic erythrocytes or reticulocytes demonstrated the relationship between the production of micronuclei and chromosomal abnormalities (24). Since only a few blood samples are required for these investigations, it is possible to examine the kinetics of MN-RET induction over time for each animal. Furthermore, by tracking the increase in micronuclei in immature erythrocytes (reticulocytes) with time after treatment, the in vivo genotoxic profiles of various drugs can be determined. Additionally, cytotoxic effects were assessed by measuring the decrease in the frequency of reticulocytes (25).

Morales-Ramírez et al. (26) studied the genotoxic action mechanism of bifunctional alkylating agents on MN-RET induction and demonstrated that the kinetics of micronucleated reticulocyte induction allow responses to be assayed more accurately than single-point measurements.

In this work, we used food competition to induce undernutrition in rats during the lactation stage. The kinetics of MN and cytotoxicity induction in reticulocytes following treatment with TMP-SMX were subsequently evaluated. In addition, as a positive control group, the kinetics and cytotoxic activity as well as MN induction by the potent mutagen N-ethyl-N-nitrosourea (ENU) were examined.

In recent decades, computational chemistry has been an important tool for evaluating the physicochemical properties of molecules with pharmacological activities, such as TMP-SMX. Since toxicity is a critical parameter, quantum chemical calculations were performed to evaluate the electrophilicity index (ω) and aromaticity index (Δ), which have been linked to the carcinogenicity and mutagenicity of organic compounds. According to Roy et al. (27,28), there is a correlation between high electrophilicity indices and a greater probability of covalent bond formation with DNA, which may result in mutagenicity. In addition, Barone et al. (29,30) identified a possible relationship between carcinogenicity and aromaticity indices greater than 0.36 eV (Δ); therefore, carcinogenicity can be used as an indirect measure of toxicity according to Barone’s rules.

The aim of the present study was to assess the impact of TMP-SMX treatment on the kinetics of MN-RET induction and its cytotoxic action, as determined by the decrease in reticulocyte frequency in peripheral blood of young (23-day-old) WN and UN rats.

Materials and Methods

Animals. The National Institutes of Health (NIH), the Universidad Autonoma Metropolitana (UAM), and the Official Mexican Guidelines (Norma Oficial Mexicana NOM-062-ZOO-1999) were followed in the execution of all the experiments. Han-Wistar rats were housed at the Bioterio of the Universidad Autonoma Metropolitana Campus Iztapalapa, Mexico City. The animal conditions and nursing mothers’ feeding were previously described by our laboratory (14).

Experimental malnutrition. Food competition during lactation caused undernutrition. Rats with severe starvation or third-degree malnutrition were selected (31).

Animal treatment and blood collection. The following four groups of six three-week-old rats were analyzed: WN and UN rats exposed to ENU or TMP-SMX. ENU dosage preparation and administration were performed as previously described for the gene mutation study (32). TMP-SMX was obtained as Bactrim, which was prepared in Mexico by Roche (Parque Industrial Lerma, Toluca, Estado de México, México).

The dose of TMP-SMX was 80 mg/kg (TMP)-400 mg/kg (SMX) body weight (14). The control samples were obtained prior to treatment. Blood samples (25 μl) were obtained from the tail vein every 8 h until 72 h after ENU or TMP-SMX administration. Sample preparation, cell staining and flow cytometry analysis, were performed as previously described (33,34) and presented below.

Sample preparation. Blood was collected in tubes with sodium heparin, diluted in bicarbonate-buffered saline solution, fixed in ultracold methanol, and stored at –70˚C for more than 24 h before staining (33,34).

Cell staining. Two tubes were utilized, one containing RNase for autofluorescence measurements and the other RNase and anti-CD71-FITC antibody for reticulocytes identification. The samples were incubated in the dark at 4˚C and subsequently at 25˚C at the indicated times. To identify the DNA of the micronuclei, we used propidium iodide (PI). Fifteen min before flow cytometry analysis, the tubes containing anti-CD71-FITC were stained with 2 μl of PI to detect DNA in reticulocytes (19).

Flow cytometry analysis. To differentiate reticulocytes (RETs) from erythrocytes, we used anti-CD71-FITC, while propidium iodide was utilized to identify micronuclei in both cell groups. An argon laser (488 nm excitation) fitted to a FACSCalibur (Becton Dickinson, Immunocytometry Systems, San Jose, CA, USA) was used to perform the flow cytometry. Using Cell Quest Software, 1×10⁶ events (total cells) were obtained. The data were then analyzed using WinMDI 2.9. Propidium iodide and CD71-FITC were observed in channels FL1 and FL2, respectively. The analysis was performed according to Dertinger et al. 2003 (35). At least 4,000 RETs were scored in accordance with the OECD guidelines (22).

Genotoxicity. The area beneath the curve (ABC) of the MN-RET frequency vs. time was used to determine genotoxicity, and the ABC value per dose, expressed in μmol/kg of bd wt, was used to determine genotoxic efficiency (26).

Cytotoxicity. By computing the decrease in the ABC between the RET number and time, cytotoxicity was defined as the decrease in the frequency of RETs over time. Assuming that the basal RET frequency remained relatively constant throughout time, the total ABC was estimated (34).

Computational model. The chemical descriptors were determined using the Gaussian 09 software package (36). The geometric structures were optimized with the M06-2X functional, which is a hybrid meta-exchange-correlation functional that well describes noncovalent interactions, such as aromatic groups (37-39), using the 6-311+G(d,p) basis set (40). Harmonic vibrational frequencies of the optimized geometries verify that the structures obtained are minima on the potential energy surface. At the same theoretical level, single-point calculations were carried out on optimized structures using a 6-311++G(2d,2p) basis set for both the complex and the individual molecules. All calculations were performed in an aqueous solution; the solvent effect was described by the SMD model (41).

Statistics. The Shapiro test was used to ascertain whether the data were parametric. Paired student’s t-test was used to determine whether there were significant differences in the MN-RET between the various time points. To determine statistically significant differences across groups, student’s t-tests were employed. The threshold of significance was established at p<0.05 (42).

Results

Body weight. Rats were weighed from 1 to 23 days of age during the nursing period; animals were classified as severely malnourished upon weaning if their weight was 40% lower than that of WN controls. The average body weight was 35.1±2.0 g for UN rats and 65.8±3.4 g for WN rats. Experimental animals undernourished during lactation exhibit other indicators of undernutrition, including low levels of total serum proteins, frail bones, sparse hair, and low activity levels.

Kinetics of MN induction with ENU. N-ethyl-N-nitrosourea (ENU) is a monofunctional alkylating agent that was used as a positive control. Its primary action is the formation of DNA mono adducts. The mechanism of micronucleus formation involves DNA breaks during repair (43). Figure 1 demonstrates the kinetics of MN-RET induction following ENU administration.

Figure 1. Kinetics of MN-RET/2000 RET induction. From time zero and every 8 h until 72 h, well-nourished and undernourished rats (A and B) were treated with ENU. The frequency was recorded as MN-RET/2000 RET, which is the method it is usually used in microscopy studies. The asterisks indicate significant differences from the baseline (student’s ttest, paired p<0.05). The baseline frequency of % MN-RET was significantly greater in the UN group (0.51±0.12) than in the WN group (0.1±0.04) (p<0.05, student’s t-test).

The time at which the MN-RET induction peaks occurred was determined with the Origin 8.0 software package from OriginLab (Northampton, MA, USA). As shown in the graph, a very evident main peak was observed at 40 h in the WN rats. For the UN group, it was observed that ENU treatment induced two kinetic peaks: the main (and first) peak also appeared at approximately 40 h, and a late peak was observed at 56 h. The late response associated with the second peak could be related to decreased proliferation/differentiation.

In the WN, there was a significant difference with respect to the baseline at all time points (p<0.05). In the UN group, a significant difference was obtained between 8 and 40 h.

The total induction of MN-RET was quantified according to the ABC; the results indicated that UN induced MN-RET (104.75) slightly more strongly than did WN (96.5).

RET percentages in WN and UN rats treated with ENU. The percentage of RET vs. time for the UN and WN groups is shown in Figure 2. The baseline frequency of %RET was significantly lower in the UN group (6.52±2.6) than in the WN group (9.14±3.1) (p<0.05, student’s t-test).

Figure 2. Percentage of RET at different times in well-nourished (WN) and undernourished (UN) rats treated with ENU (A and B). The data were collected beginning at time zero and every 8 h until 72 h. The asterisks indicate significant differences from the baseline (student’s t-test, paired p<0.05).

WN rats showed a decrease in the frequency of RET at 16 h that continues and peaks at 48 h and then begins to increase in the following times (56 to 72 h); however, these rats exhibited a statistically significant decrease in relation to the basal level at all times points. In the UN group, a significant decrease was observed with respect to the baseline only at 44 and 48 h and later at 64 h.

Kinetics of MN induction with TMP-SMX. Figure 3 displays the kinetics of TMP-SMX treatment-induced MN-RET induction in WN and UN rats. The data indicate that the main difference between the WN group and the UN group is the high baseline value (approximately three times greater in the UN group than in the WN group).

Figure 3. Kinetics of MN-RET/2000 RET induction. At time zero and every 8 h until 72 h, well-nourished and undernourished rats (A and B) were treated with TMP-SMX. The asterisks indicate significant differences from the baseline (paired student’s t-test, p<0.05).

In addition, the induction curve of MN-RET was different between the UN group and the WN group. MN-RET from the WN presented two very evident peaks, one at approximately 16 h and one at approximately 48 h. In the UN, there was a reduction at approximately 32 h and a late induction at approximately 48 h; this point was the only point at which a significant difference was observed in relation to the baseline.

The total ABC in the curve of WN was 56.2, the first peak, ABC_1, was at 21.8 h, and the second peak, ABC_2, was at 34.4 h. In the UN group, there were not clearly two peaks, but in the period’s equivalents, ABC₁ was at 34.7 h and ABC₂ was at 55.4 h. These findings indicate that the number of MN was nearly double in UN compared with that in WN.

RET percentages in WN and UN rats treated with TMP-SMX. Figure 4 displays the RET frequency curves as a percentage vs. time for the WN and UN groups. This shows that the WN curve has a slight reduction in %RET, with a significant difference at 40 and 48 h. Regarding the UN group, there was an early increase at 8 h, with a statistically significant difference, then a decrease and a late induction with a significant difference at 72 h.

Figure 4. Percentage of RET at different times in WN and UN rats treated with TMP-SMX (A and B). At time zero and every 8 h until 72 h, the asterisks indicate significant differences from the baseline (paired student’s t-test, p<0.05).

Aromatic index (∆) and electrophilic index (ω). The indexes for SMX, TMP, SMX-TMP, 5-azacytidine (aza-C) and gemcitabine (Gem) are shown in Table I. Aza-C and Gem were included as positive controls. Likewise, TMP and SMX alone were tested and analyzed to determine their effects alone and in combination in the TMP-SMX complex.

Table I. The aromatic index (Δ) and electrophilic index (ω) at the M06-2X/6-311++G(2d,2p)//M06-2X/6-311+G(d,p) levels of theory.

The electrophilic index values shown in Table I indicate that Gem, aza-C and SMX have a greater predisposition to acquire an additional electron, thus, according to Roy et al. (27,28), they could be more mutagenic molecules than their analogs. In contrast, the values for TMP and the SMX-TMP complex are the smallest, i.e., the electrophilic index decreases for SMX when it forms a complex with TMP (Figure 5); therefore, TMP and its SMX-TMP complex could be molecules with fewer toxic effects.

Figure 5. Optimized molecular structures of SMX (a), TMP (b), the SMX-TMP complex (c), gemcitabine (d) and 5-azacytidine (e). The C, H, N, F, O and S atoms are represented as gray, white, blue, cyan, red, and yellow balls, respectively.

Table I also displays the aromatic indices, which show that SMX, aza-C and Gem have the highest aromatic indices (Δ), while TMP and the SMX-TMP complex have the lowest values; therefore, according to the rules of action of Barone, TMP and its complex with SMX could have fewer carcinogenic effects than their analogs.

Discussion

Malnutrition, which generally refers to undernutrition caused by a diet lacking certain nutrients, primarily proteins, is still a major public health issue, particularly in poor nations. It is necessary to insist on the importance of related studies in many different fields. Therefore, hospitals must evaluate each patient’s nutritional risk at admission and monitor any at-risk patients. Additionally, there are continuing studies on infections since they are associated with an increase in morbidity and mortality in undernourished individuals. It is also necessary to analyze the protection from infections through vaccination. Immunodeficiency results from undernutrition, and the immunological responses of undernourished children to vaccination are concerning (44).

Morales-Ramírez et al. (2008) reported that 5-aza induces MN-RET through multiple alternatives (45), including DNA demethylation (46) and repair of adducts (47), which causes persistent chromosome fragility in the G-C rich sites caused by DNA demethylation and chromatin decondensation, producing late induction of MN (48). These results indicate that the kinetics of MN-RET can reveal DNA breaks indirectly caused by agents and also DNA breaks generated during the processing of some lesions caused by the agents (25).

It has been demonstrated that assessing antibiotic resistance and preventing the overuse of antibiotics are crucial for providing effective treatment for malnutrition (49). Furthermore, several researchers have documented modifications in the effectiveness of different medications in cases of protein undernutrition (50).

The effects of undernutrition have been extensively studied using experimental animal models. Lares-Asseff et al. (51) analyzed the kinetics of TMP-SMX penetration in tissues of severely malnourished rats. According to their findings, undernutrition affects the biotransformation capacity of trimethoprim more than it does for sulfamethoxazole. Infected individuals with severe undernutrition should be treated with special care when considering a TMP-SMX treatment plan.

In general, undernourished organisms have increased susceptibility to infections; therefore, antibiotic treatments are needed. Treatment for bacterial infections of the urinary, respiratory, and gastrointestinal systems can be achieved with the combination of TMP-SMX. The FDA-approved use of TMP-SMXs include dosage schedules, significant side effects, allergies, toxicity, and monitoring (52). Additionally, TMP-SMX has the potential to be bactericidal (these medications can have bacteriostatic effects only when taken alone). Furthermore, the current study’s data showed that the aromaticity index decreases for SMX when it forms a complex with TMP (Table I). Therefore, the use of this complex could have fewer toxic effects.

The analysis of the kinetics of the genotoxic activity of TMP-SMX showed that, in WN rats, there were two very well-defined peaks, one at approximately 16 h and another late, at 48 h; these findings suggest, from experiences with other chemical agents, such as bis-chloroethyl-nitrosourea (BCNU) and aza-cytidine (azaC), that these peaks are the products of different mechanisms of DNA damage induction and therefore of micronucleus production (Figure 3) (25).

In contrast, the curve of the UN for the kinetics of MN-RET induction by TMP-SMX showed a tendency to decrease from the high basal frequency until approximately 40 h and late increase at approximately 48 h, which coincided with that of the WN. This is the only point at which a significant difference is observed in relation to the baseline in the UN curve. Notably, the response observed in the UN curve showed substantially greater variability. It is possible that the first peak in the UN curve is masked by the high basal frequency and with the increase in variability since there seems to be a shoulder at 16 h. In addition, the absence of this first peak could be related to the alterations in the pharmacokinetics of the drug observed in malnourished organisms (51).

Regarding the kinetics of the cytotoxicity cause by TMP-SMX, WN had a slight reduction in %RET, with significant differences at 40 and 48 h. Conversely, in the UN group, there was an early increase in the %RET at 8 and 16 h, with a statistically significant difference, followed by a decrease and a late increase with a significant difference at 72 h. The initial increase of %RET may have been the antibiotic’s positive reaction to the animals’ chronic low basal frequency of RET; undernutrition is known to cause this kind of low frequency. In this instance, the data dispersion at the UN was likewise noticeably greater.

Conclusion

Undernutrition modifies the body’s response to TMP-SMX, at least in terms of the response over time and the degree of variability observed between organisms. An important role is played by the presence of a remarkably high basal frequency of MN-RET in the undernourished group, which was 5 to 7 times greater.

TMP-SMX produces two MN-RET induction peaks in WN animals, suggesting that this event is the product of two mechanisms of DNA damage induction and therefore micronucleus production. In the UN group, the two peaks were not observed, which suggested that the first mechanism of MN induction did not occur in the malnourished group, perhaps due to pharmacokinetic effects, decreased metabolism, or alterations in cell proliferation.

In conclusion, ENU caused a similar maximal response to MN-RET in both WN and UN animals, with greater dispersion in the UN group. Cytotoxicity is more significant in WN due to the higher basal frequency.

The results of quantum mechanical calculations at the atomic level indicate that the aromaticity index of the TMP-SMX complex decreases, so it has fewer toxic effects than its individual molecules. Therefore, the intermolecular interactions formed in the complex have a significant impact on the chemical structure and, by extension, the biological activity.

Conflicts of Interest

The Authors declare no conflicts of interest in relation to this study.

Authors’ Contributions

PM-R and RO-M conceived the project; EC-R conducted the experiments and analyzed the data; L R-C and EC-B contributed to the manuscript editing and experiments; CS-C and LC-F coordinated the computational work and wrote part of the manuscript; PM-R analyzed the results, prepared the figures and coordinated the kinetics analysis; PM-R and RO-M wrote the manuscript; and RO-M coordinated the laboratory work. All the Authors participated in reading and reviewing the manuscript.

Acknowledgements

The Authors thank Monserrat Pacheco-Martínez, a doctoral student, for her invaluable support with the experiments. We thank the Dirección General de Cómputo y de Tecnologías de Información y Comunicación (DGCTIC) at the Universidad Nacional Autónoma de México (UNAM) for allocating computer time to the supercomputer (Miztli, LANCAD-UNAM-DGTIC-203).