Human exposure to chitosan occurs primarily through consumption of dietary supplements, as chitosan is marketed as a fiber-like supplement to increase satiation and promote weight loss through inhibition of fat absorption.¹⁶ The acute toxicity of chitosan has previously been examined in human studies (12 days or up to 8 weeks) evaluating the effectiveness of chitosan as a weight-loss supplement, and the results from these studies demonstrated no observable toxicity following oral administration of chitosan.^32-34 However, there is indication of serum vitamin and bone mineral depletion following consumption of chitosan in rats.²⁶ Therefore, NTP conducted 6-month feed studies to evaluate the effects of dietary chitosan on bone metabolism, fat-soluble vitamin levels, and dietary fat and calcium absorption, as well as general toxicity in Charles River Sprague Dawley rats.

Feed concentrations of 1%, 3%, and 9% chitosan, which resulted in average daily doses of approximately 450, 1,500, and 5,200 mg chitosan/kg body weight per day to males and 650, 1,800, and 6,000 mg/kg per day to females, were selected based on existing data from animal studies.^24;26 The 9% concentration is higher than the typical 5% NTP concentration limit, but the 9% diet was considered to be nutritionally adequate. The AIN-93M feed was selected for this study over the NTP-2000 feed based on the high levels of fat-soluble vitamins and higher total fat content found in the NTP-2000 feed. The NTP-2000 feed contains almost double the amount of required fat-soluble vitamins and has a higher fat content (7% to 8%) than the AIN-93M feed (4%).^40,41 One of the primary rationales for this chitosan study was the potential for decreases in fat-soluble vitamin concentrations, and therefore utilizing a diet with lower levels of preexisting vitamins and a lower fat content was ideal to avoid confounding potential results.

The animals used in this study were split into three groups, the core group, Group A, and two special study groups, Groups B and C. Different parameters were evaluated in each group, which, while allowing for the collection of extensive endpoints, meant that only 10 animals were examined per endpoint instead of 30, as there was no crossover of analyses between the groups.

Multiple endpoints were evaluated at multiple time points (6, 12, 18, and 24 weeks) in Group C rats to determine effects on fat absorption. Treatment-related decreases in percentage fat digestion of 20% to 33% in males and 5% to 14% in females relative to control, were consistently observed in the 9% group with effects also noted in males in the 3% group (decreases of 2% to 8%). Stronger responses were observed in males relative to females. Additionally, fecal weight was significantly increased in 1% females at weeks 12, 18, and 24 (19%, 18%, and 29%, respectively), and in 3% (35% to 56%) and 9% (96% to 170%) males and females relative to controls at all time points. These data suggest that consumption of chitosan reduced the absorption of fat in the feed, resulting in increased fecal weight due to fat being excreted. Similar results have been observed in other studies. Deuchi et al.⁵⁵ reported that rats fed deacetylated chitosan had decreased fat digestion; as the degree of deacetylation increased, fat digestibility decreased. The chitosan used by Deuchi et al.⁵⁵ was 70% to 90% deacetylated, which is a level very similar to the chitosan (86.5% deacetylated) used in the current study. Gallaher et al.¹² demonstrated that male Wistar rats exposed to 10% chitosan in AIN-93 feed had increased fecal fat excretion and dry fecal weight and decreased cholesterol absorption relative to control rats, similar to what was observed in the current study.

Due to the high percentage of chitosan in the feed of the 9% group, it is possible that the observed decreases in percentage fat digested were due to bulk chitosan in the feces confounding the amount of fat actually being excreted. Misrepresented fecal weights would alter the calculated amount of fat excreted in the feces, which would subsequently affect the calculation of percentage fat digested. The observed increases in fecal weight could also be attributed to an increase in the percentage fecal moisture, which was significantly increased in both males and females in the 3% and 9% groups. In Group A, there were decreases, albeit not significant, in mean body weights of 9% males and females (decreases of 9% and 11%, respectively), but overall there were no significant changes in the body weights of rats exposed to chitosan; the mean body weights of exposed animals were similar to those of control animals. Considering the large decrease in percent fat digested, combined with the significant increase in fecal weight observed in 9% males and females, it would be expected that mean body weights would significantly decrease due to more fat being excreted than digested. The slight mean body weight decrease observed in this study could be due in part to excretion of bulk chitosan, but regardless, the magnitude of increase in fecal fat excretion as well as the decrease in hepatic periportal fatty change still indicates a treatment-related response.

Consistent significant decreases in cholesterol levels were observed in 9% male and female rats; triglycerides levels were also affected but not as consistently as cholesterol. Decreases in cholesterol were consistent with many other studies and not an unexpected finding, as chitosan is well known to have a cholesterol lowering effect in rats.^{14; 56-59} The mechanism by which chitosan lowers cholesterol is still controversial, but recent studies indicate that chitosan, acting as a weak anion exchange resin, reduces cholesterol by causing a decrease in its absorption in the small intestine and by inducing increases in bile acid excretion.^11-13 With bile acid excretion, plasma or liver cholesterol is utilized to maintain the bile acid pool.¹² Alternatively, the cholesterol lowering effects of chitosan may be related to an increase in viscosity of intestinal contents, which entrap fat and prevent lipolysis, or this mechanism may be in addition to chitosan’s ability to bind bile acids.^13-15

Along with an inhibition in dietary fat absorption and decreases in serum lipids there were also treatment-related decreases in the levels of fat-soluble vitamins A and E. Serum and liver vitamin E levels were substantially affected, being 62% to 87% lower in the 9% males and females. These findings are similar to those of Deuchi et al.²⁶ where decreases in serum and liver vitamin E levels were observed after 14 days of consuming a 5% chitosan feed. In this same study, liver vitamin A levels were decreased, but vitamin A serum levels were unchanged. Bile and lipids are needed for the absorption of dietary vitamins A and E, as both must be incorporated into intestinal micelles for their absorption.⁶⁰ Thus, it is highly plausible that the decrease in dietary fat absorption, including cholesterol, led to the decreases in serum and liver concentrations of these vitamins. It is also possible that, by some unknown mechanism, chitosan may enhance vitamin A or E requirements in the peripheral tissues.

There were no histologic changes associated with the observed decreases in vitamin levels; however, the decreases were significant enough to suggest nutritional inadequacies. The long-term effects of vitamin A and vitamin E deficiencies are well-known,^60-63 and it is unknown what deficiency-related effects would have been observed had these decreased levels been maintained for a longer period of time. When circulating levels of vitamin E, specifically α-tocopherol, are depleted, tissue damage can occur. Vitamin E depletion in humans has subsequently been correlated with anemia, disruption of normal growth, decreased responses to infection, and pregnancy concerns.⁶² Vitamin A is essential in numerous biological processes and pathways, including growth, vision development, immune function, and metabolism. Severe vitamin A deficiency (VAD) results in disruption of normal tissue function and is associated with childhood blindness, anemia, and depressed responses to infection; VAD during a severe infection may result in death.^61-63 While the long-term effects of vitamin deficiency in rodents are not as well understood, the available literature on human deficiencies suggests that the decreases in vitamin A and E observed in this study may be detrimental over time.

In contrast to decreases in vitamins A and E, 1,25 (OH)₂ vitamin D (bioactive vitamin D) levels were significantly elevated in 9% male and female rats. Vitamin D’s main function is to help maintain normal calcium and phosphorus levels by regulating the intestinal absorption of these minerals from the diet. In addition to the increased 1,25 (OH)₂ vitamin D levels, significant decreases in serum phosphorus were also seen in male and female rats. Although intestinal absorption of phosphorus was not measured in this study, chitosan has been observed by others to cause a significant reduction in intestinal phosphorus absorption.⁶⁴ Low phosphorus concentrations stimulate 1,25 (OH)₂ vitamin D production by the kidney, therefore the increased levels of 1,25 (OH)₂ vitamin D observed in this study may be the result of the low phosphorus levels. Increased levels of 1,25 (OH)₂ vitamin D can cause an increase in intestinal absorption of calcium regardless of serum calcium levels. Significant elevation in intestinal absorption of calcium was observed sporadically in the female rats, but serum calcium levels were relatively stable. This effect is most likely due to a loss of calcium through the urine, which has been observed in other chitosan feed studies^64;65 and is known to occur in cases of hypophosphatemia-induced elevations in 1,25 (OH)₂ vitamin D due to Fanconi’s syndrome.⁶⁶ The reported urinary calcium loss in chitosan feed studies may be compensatory or directly induced by the chitosan.

Significant decreases in urine volume were observed in various male and female groups, but most consistently in the 9% group where decreases of 40% to 58% of the control group volume were observed. As the urine volumes decreased, urine creatinine concentrations were seen to increase significantly. This is consistent with proper renal function. The most likely cause of the decrease in urine volume is decreased consumption of water, although water consumption was not measured, so this cannot be certain. However, the mild increases in urea nitrogen in the 9% male and female rats at 25 weeks (the only time point measured) supports decreased water consumption (i.e., mild dehydration). Water retention in the intestine may have contributed to the decreases in urine volume, as fecal moisture was mildly increased in some of the treatment groups, although it is highly unlikely this would be the primary cause and no diarrhea was observed.

There was a significant decrease in the occurrence of periportal fatty change, or lipid accumulation, in the livers of 9% females relative to the controls, and this negative trend was maintained in both 1% and 3% females, although not significantly. In male rats, the incidences of periportal fatty change were decreased in both 1% and 9% groups and the severities were decreased in both the 3% and 9% groups. The decrease in lipid accumulation was inconsistent between male and female rats in the 9% exposure groups, as a more severe decrease was observed in the 9% female rats (100% lower) compared to the 9% male rats (50% lower) relative to the respective controls. The morphologic features observed during this study (periportal hepatocytes with large, single, well-defined intracytoplasmic vacuoles displacing the nucleus), were consistent with the intracytoplasmic lipid accumulation that is associated with fatty change.⁶⁷ During normal function, fatty acids circulate between the liver and adipose tissue, which maintains a balance of triglycerides between the two locations. When this balance becomes skewed, hepatic fatty acids can accumulate as small vacuoles in the hepatocytes and progress over time into larger globules.^67;68

Lipid accumulation in the liver can occur via multiple mechanisms, including 1) increased synthesis of fatty acids, 2) increased uptake of fatty acids from adipose tissue and/or the diet, 3) improper removal of fatty acids from the liver, or 4) decreased oxidation of fatty acids.⁶⁹ Diet and nutritional status can also influence lipid accumulation.^68,70 Singh et al.⁷¹ demonstrated that albino rats administered vitamin A orally for 2 days had increased hepatic lipid accumulation. In the present study, there were treatment-related decreases in hepatic vitamin A and E in both male and female rats, which could have contributed to the loss of periportal lipid accumulation observed in the animals fed 9% chitosan. Lipid accumulation in the liver can also occur due to imbalanced uptake of lipids from the blood and secretion of lipoproteins from the hepatocytes.⁷² In this chitosan study, the fatty change (lipid accumulation) observed was periportal, or in Zone 1. Zone 1 is closest to the incoming vasculature and receives the majority of oxygenated blood, and Zone 1 hepatocytes are generally resistant to the effects of nutritional deficiencies.⁷³ Therefore, the decrease in fatty change observed in rats fed 9% chitosan could be an adaptive response to the vitamin and mineral depletion noted in this study.

The incidences and severities of fatty change in both male and female control animals was particularly high (6/10, males; 7/10, females; average severity 1.7 and 1.1, respectively), suggesting that the Charles River Sprague Dawley rats used in this study may have a normally high level of hepatic periportal lipid accumulation. Figure 3, Figure 4, and Figure 5, included in this report, are well representative of the observations made in this study, as the increased severity of periportal fatty change in control animals was a strong response.

Absolute and relative liver weights of male and female rats were significantly decreased in animals fed 9% chitosan relative to control animals. As described above, there were decreases in the incidence of periportal fatty change in all exposed animals, particularly in the female rats fed 9% chitosan. The decrease in liver weights observed in the 9% animals could be due to the loss of fat accumulation in the livers, which would alter the weight of the organs.

The absolute and relative thymus weights of 3% and 9% males and 9% females were also significantly decreased relative to those of control groups. The thymus is extremely sensitive to toxic compounds and similar stressors, and alterations in thymus weight can be an indicator of apoptosis and organ atrophy in response to a toxic insult. Nutritional status can cause a decrease in thymus weight, in particular vitamin, mineral, and fatty acid deficiencies.⁷⁴ In the current study, male and female rats fed 9% chitosan had depleted levels of serum vitamin A and E, liver vitamin E, and serum cholesterol and triglycerides, indicating nutritional inadequacies. The observations from this chitosan study, combined with what is known about the thymus, suggest that exposure to chitosan may have induced reductions in thymus weight secondary to nutritional deficiencies.

Results of this study did not support chitosan as a cause of bone resorption. Significant elevation of parathyroid hormone levels occurred occasionally and inconsistently, while calcium levels were relatively stable. Calcium was mildly, but significantly, decreased at only two time points in male groups by no more than 4%. Additionally, serum total osteocalcin and urinary deoxypyridinoline level, both biomarkers of bone turnover, while occasionally significantly elevated, lacked any consistent increases over time or between sexes. In fact, deoxypyridinoline was significantly decreased at some time points. Lastly, bone calcium, bone length, and the histology findings of this study did not support calcium loss from the bone.

Although bone parameters were unaffected by chitosan exposure, a limitation of this study may be that the time frame of the study was not extensive enough to adequately evaluate bone loss. Rats are generally not considered skeletally mature until 10 months of age, and the long bones in rats can continue to grow until 30 months of age, making it difficult to observe any loss of bone before that point.⁷⁵ In a study of female Charles River Sprague Dawley rats, Wronski et al.⁷⁶ observed closed growth plates in the tibias of 15-month-old animals. In a separate study, Fukuda and Iida⁷⁷ noted that natural decreases in bone mineral density did not begin until 15 months of age in female Wistar rats. Also, standard osteoporosis studies using rat models commonly utilize ovariectomized animals, which mimics the conditions of menopause and generally increases rates of bone remodeling and bone loss. Ovariectomized SHRSP rats fed 10% chitosan alongside a low calcium diet exhibited decreased bone mineral density and increased femur stiffness.⁶⁴ Following ovariectomy, bone loss in the femurs, specifically the femoral neck, is still not observed until a minimum of 30 days postprocedure.⁷⁵ Therefore, given the time frame of the study there was reduced likelihood of observing any osteologic changes possibly induced by chitosan exposure.

There were no treatment-related clinical findings in the core, Group A animals, but there were instances of seizures in Groups B and C animals. Thirteen animals from Groups B and C (two 1%, one 3%, and ten 9%) were observed with seizures either during or after the 18-week blood collection. Seizures were not noted at any other time point. Similarly, there was no treatment-related mortality in the Group A animals, but five animals from Groups B and C died, often after seizures, near the time of blood collection. Cause of death was undetermined for these animals. While there was no clear connection between chitosan treatment and the incidence of seizures, there was an exposure concentration-related increase in the occurrence of seizures. Therefore, it is possible that chitosan exposure may have induced the increased rate of seizures observed in this study.

Under the conditions of the 6-month feed study of chitosan, male and female rats fed 3% and 9% chitosan in the diet had significantly decreased levels of serum vitamin A and serum and hepatic vitamin E and increased levels of serum 1,25 (OH)₂ vitamin D. Consumption of high levels of chitosan decreased percentage fat digestion and increased fecal weight and moisture, as well as reduced levels of phosphorous, cholesterol, and triglycerides. Female rats exposed to 9% chitosan also had significant liver weight and histologic changes. Based on the above results, the lowest-observed-effect level for chitosan exposure was 1% (approximately equivalent to 450 mg/kg) in male and 9% (approximately equivalent to 6,000 mg/kg) in female rats.