Studies have shown that immune dysfunction (Liu et al., 2023), Glu excitotoxicity (Wang and Reddy, 2017) and tau phosphorylation (Rawat et al., 2022) are closely related to the abnormal development of AD, while the immune system, glutamatergic system and protein kinase/phosphatase system involved in the above factors are cross-regulated through KP. KP, which is hyperactivated by pro-inflammatory factors, tends to neurotoxic branches to generate QUIN and 3-HK, among others, and plays an important role in the pathogenesis of AD through oxidative stress and/or inflammatory responses.

The academic community has long shown that KP is abnormally activated in the brain tissue of AD patients. Baran et al. reported that KYN and 3-HK contents were significantly decreased in the frontal cortex, caudate nucleus, putamen, hippocampus, and cerebellum, but significantly increased in the putamen and caudate nucleus in AD patients (Baran et al., 1999); Some scholars attribute the abnormal increase of KYNA content to the compensatory physiological response of the body and believe that it is the compensatory mechanism by which KYNA antagonizes QUIN-mediated excitotoxicity (González-Sánchez et al., 2020; Jacobs et al., 2019). Subsequently, numerous studies have shown that KP metabolites and enzymes are associated with AD. Guillemin et al. found that IDO was overexpressed and QUIN content was significantly increased in microglia, astrocytes, and neurons in the medial temporal lobe, frontal lobe, and cingulate cortex of the brain in AD patients by immunohistochemical techniques, with IDO and QUIN in microglia and astrocytes being most significantly expressed and diffusely distributed in the periphery of senile plaques; QUIN colocalized with hyperphosphorylated tau in cortical neurons and also existed evenly in NFT in the form of granular deposits (Guillemin et al., 2005) providing conclusive evidence for KP involvement in AD pathogenesis. However, the sample size of the above studies is small and needs to be further verified in combination with large-sample studies. Van der Velpen et al. used untargeted metabolomics and targeted quantification analysis to confirm that cerebrospinal fluid QUIN levels were elevated in AD patients, and microtubule-associated protein tau and threonine-181 phosphorylated tau (pTau-181) were significantly negatively correlated with cerebrospinal fluid QUIN content. B-Amyloid-42 (Aβ1-42) was significantly positively correlated with KYNA content in cerebrospinal fluid (van der Velpen et al., 2019). Bakker et al. found that plasma TRP and KYN contents were significantly negatively correlated with total tau and phosphorylated tau, cerebrospinal fluid TRP and KYN contents were significantly negatively correlated with phosphorylated tau, and cerebrospinal fluid KYN, KYNA, and KYN/TRP contents were significantly positively correlated with Aβ1-42 in AD patients using ELISA (Bakker et al., 2023). There is now evidence that Aβ1-42 induces IDO1 expression and enhances QUIN production, and significantly elevates pro-inflammatory cytokine responses, thereby upregulating IDO1, TDO2, and KMO expression (Guillemin et al., 2003; Pocivavsek et al., 2024). However, a large study reported that neuroprotective metabolites KYNA and Pic were increased in cerebrospinal fluid of AD patients and were also associated with slow progression of AD, while toxic metabolites 3-HK and QUIN were not significantly increased (Knapskog et al., 2023). The reasons for this finding may be methodological differences, or influenced by disease stage vs. population cohort characteristics. In addition, age as a factor associated with KP may also influence the study results, as controls were younger than AD patients in van der Velpen's study, which suggests that future studies need to strictly match age variables (Giil et al., 2017; Knapskog et al., 2023). It has also been demonstrated that patients with AD have decreased KYNA/QUIN ratio in cerebrospinal fluid and decreased XA/3-HK ratio in serum. Age was strongly associated with increased KYN, KYNA, and QUIN contents in serum and cerebrospinal fluid, and KYNA concentrations were significantly lower in AD patients than in age-matched controls. Serum KYN and QUIN levels are strongly positively correlated with corresponding levels in cerebrospinal fluid (Sorgdrager et al., 2019). This suggests that KP metabolite levels are altered in both the periphery and CNS of AD patients and are positively correlated with the severity of clinical symptoms, in good agreement with what has been reported in previous studies (Jacobs et al., 2019). In addition, there was a significant association between changes in peripheral levels of KP metabolites and cognitive dysfunction in AD patients. Gulaj et al. used HPLC to analyze plasma TRP and KP metabolites in AD patients and age-matched controls and confirmed that plasma TRP and KYNA levels and AA/KYN, KYNA/KYN, and 3-HK/KYN ratios were significantly decreased, and QUIN levels and KYN/TRP and QUIN/3-HK ratios were significantly increased in AD patients. Plasma TRP levels were positively correlated with instrumental activities of daily living (IADL) and physical activities of daily living scores (PADL) assessing self-care ability and independent living standards in individuals with cognitive impairment groups. AA levels were positively correlated with IADL scores. KYNA levels and KYNA/KYN ratio were positively correlated with Mini-Mental State Examination (MMSE) scores reflecting cognitive function in AD. QUIN levels were negatively correlated with clock drawing test (CDT) scores reflecting the severity of cognitive impairment in AD patients (Gulaj et al., 2010). Another study showed that age was significantly positively correlated with KYN, 3-HK, QUIN content and KYN/TRP ratio in plasma and significantly negatively correlated with XA content in AD patients. QUIN levels were significantly negatively correlated with Cambridge Cognitive Examination (CAMCOG) scores, which reflect the degree of dementia and assess the degree of cognitive impairment (Giil et al., 2017).

In summary, the levels of heavy KP metabolites in the peripheral and central nervous systems of AD change, and are closely related to the development of AD. However, research regarding KP abnormalities and aspects of AD remains problematic. (1) The results of some current studies on KP metabolite levels in AD are conflicting and lack sufficient experimental data to support them; small sample sizes in some studies and differences in detection methods between different studies may affect the generalizability of the results, (2) existing studies have not been able to determine whether KP imbalance is a causative factor or a secondary phenomenon in the development of AD. It is thought that KP imbalance may precede clinical symptoms (Knapskog et al., 2023), (3) there is also a lack of a strong explanation for the differences in KP metabolism that appear in different brain regions. In response to the above issues, further studies are still needed to elucidate the potential mechanism of action of KP in AD.

The association between AD symptoms and levels of specific KP metabolites suggests that reversing KP imbalance would be beneficial in alleviating AD progression. One way to achieve this goal could be to exert a direct impact on KPs. KYNA has the limitation of low permeability through blood-brain barrier and insignificant administration effect. In animal models of AD, it has been shown that KYNA analogs or KYNA prodrugs are effective methods to prevent and/or delay AD progression. Majerova et al. used the KYNA analog kynurenoquinolinic acid N-(2-N, N-dimethylaminoethyl)−4-oxo-1 H-quinoline-2-carboxamide (KYNA-1), which has similar biological activity but is more brain permeable, to overcome BBB restriction and found that KYNA-1 reduced hyperphosphorylation of insoluble tau in the brain and plasma total tau and inhibited neuroinflammation to alleviate AD progression in SHR-24 transgenic rats when administered chronically (Majerova et al., 2022). Deora et al. developed multi-target KYNA series analogs 5b and 5c with structural modification of canine urinary quinolinic acid (KYNA) mother nucleus. In AD transgenic C. elegans strain GMC101, 5c administration significantly suppressed Aβ42 profibrotic levels and prevented Aβ42-induced cytotoxicity. 5b inhibited NMDAR and had moderate potency during DPPH radical scavenging. 5b and 5c are highly permeable in tests based on the blood-brain barrier (BBB) model of MDR1-MDCKII cells (Deora et al., 2017). It was shown that systemic administration of 4-chloro-KYN (KYNA analog), a prodrug of 7-chloro-KYNA (synthetic KYNA prodrug), protected rat hippocampus from QUIN-induced excitotoxicity (Wu et al., 2000). 3-HAA is an essential precursor for the synthesis of QUIN and has dual oxidative stress and antioxidant properties, which are associated with a variety of physiological and pathological processes. Studies have confirmed that transgenic C. elegans AD models expressing amyloid-β in body wall muscles effectively prevent Aβ toxicity after 3-HAA supplementation, and their efficacy is equivalent to direct knockdown of HAAO-1 (Hull et al., 2024).

IDO and TDO are responsible for regulating the rate of KYN production, and their activity determines the potential role of KP neurotoxic metabolites in neurodegenerative diseases, which are now confirmed as one of the key targets for AD drug therapy. Souza et al. showed that 1-MT treatment with IDO inhibitor decreased KYN content and KYN/TPR ratio in prefrontal cortex and hippocampus, and improved cognitive memory ability and non-cognitive dysfunction in Aβ1-42-induced AD model in mice. It is characterized by increased cognitive indices in novel object recognition experiments and opening and closing activity times in elevated plus maze as well as decreased immobility times in tail suspension experiments (Souza et al., 2016). Breda et al. showed that peripheral blood KYNA, 3-HK, XA, 3-HAA, and QUIN levels were lower in β-amyloid precursor protein (APP233) transgenic mice than in control wild-type mice with AD. After 6 weeks of oral administration of TDO inhibitor 680C91, peripheral blood KYN, QUIN, and PIA levels were significantly reduced, and memory recognition and spatial learning ability and anxiety-related behaviors were improved. It is characterized by a decrease in cognitive index in the novel object recognition test, a decrease in escape latency and second quadrant activity time in the Morris water maze; and an increase in open-arm activity time and a decrease in closed-arm activity time in the elevated plus maze (Breda et al., 2016). Minhas et al. showed that 4-week treatment with the IDO1 inhibitor PF068 (15 mg/kg) decreased KYN content in the hippocampus of APP/PS1, 5XFAD, and PS19 (P301S) tau transgenic mouse AD models, significantly reduced latency to reach the target hole in the Barnes maze test and significantly increased the identification index in the novel object recognition test. Aβ peptide accumulation in thioflavin S (ThioS) -positive dense core plaques and 6E10-positive diffuse plaques was significantly reduced in the hippocampus of the 5XFAD mouse model, and Thr231 phosphorylated tau in the soluble and insoluble tau fractions and Thr181 phosphorylated tau in the insoluble fraction of the hippocampus were significantly reduced in the PS19 mouse model (Minhas et al., 2024). It has also been shown that coptisine, an IDO-1 inhibitor, decreased IDO concentrations in serum and mRNA levels of IDO1, KYNU, KMO, and 3-HAAO in the hippocampus of AβPP/PS1 transgenic AD mouse models, and eliminated neuroinflammatory responses and neurotrophic defects in the hippocampus (Yu et al., 2015). KMO is a key enzyme in the toxic branch of KP and is not only directly involved in the synthesis of 3-HK, but also plays a central role in the formation and function of downstream metabolites 3-HAA and QUIN (Hughes et al., 2022). Zwilling et al. showed that KYNA levels in the brain of β-amyloid precursor protein (APP) transgenic mouse AD model were lower than those in littermate controls, and after long-term oral administration of JM6 (weak KMO inhibitor), KYNA levels in brain tissue and peripheral blood were significantly increased, and extracellular Glu in neurons in the brain was continuously reduced and synaptic loss in the hippocampus and cortex was prevented, and spatial memory deficits in the Morris water maze test and anxiety-like behavior in the elevated plus maze were improved, as shown by increased activity time in the platform quadrant and closed/open arm activity time ratio, respectively (Zwilling et al., 2011).

In summary, targeting the KP major enzyme system to regulate KP homeostasis could provide new therapeutic directions for AD. Modulators of the major KP enzymes, such as agonists or antagonists, and analogs and precursors of their major neuroprotective metabolites can be potential therapeutic targets for AD and one of the strategies to achieve effective neuroprotection in AD. However, current evidence supporting these strategies stems almost exclusively from animal model studies and no relevant human clinical studies have been identified. In addition, targeted regulation of the KP system is highly complex, and some KP metabolites play different mechanistic roles according to concentration gradient and microenvironment specificity (such as the dual characteristics of 3-HAA), so supplementation of KP metabolite analogs may trigger unpredictable pathway compensation; while long-term intervention against specific KP enzymes will likely lead to disturbance of the degradation cascade downstream of KP and affect the production of downstream products resulting in impaired KP function, and comprehensive studies are needed to evaluate the strategy of KP enzyme inhibitors and agonists and KP metabolite analogs to target KP for the treatment of AD in the future.

Among various types of exercise programs, aerobic exercise is considered the most potential and cost-effective intervention (Lee et al., 2023). Aerobic exercise such as outdoor aerobic walking (Venturelli et al., 2011), power cycling (Yu et al., 2021), and treadmill training (Vidoni et al., 2019) are widely used as adjuncts to AD drug therapy and have been shown to have direct beneficial effects in improving cognitive function, motor function, and quality of life in AD patients. Aerobic pedal rehabilitation training, based on the fixation device ReckMOTOmed, improves multidimensional cognitive function (including executive control, verbal fluency, response speed, and attention allocation) in AD patients and greatly relieves stress on care in AD patients; follow-up studies have shown that its improvement remains sustained for at least 3 months after termination of the intervention (Holthoff et al., 2015). Comparative studies have found that different aerobic exercise training modes enhance aerobic fitness and physical performance in AD patients, with intermittent aerobic training (IAT) also significantly improving self-care ability in AD patients (Enette et al., 2020). It has also been shown that even a single acute work rate cycling training session significantly improves cognitive dysfunction in patients with moderate AD, and this improvement is enhanced when combined with cognitive play (Ben Ayed et al., 2021). Subsequent studies have shown that there is also no significant gender difference in the benefit of a single acute aerobic exercise in AD patients (Ben Ayed et al., 2024) (as shown in Table 1).

In conclusion, aerobic exercise, as a mainstream intervention for cognitive rehabilitation, shows immediate benefits in symptom management in AD patients and is regarded as one of the potential non-pharmacological intervention strategies. However, large-scale randomized controlled trials with long-term follow-up are needed to confirm the current findings, and the optimal intensity-frequency combinations and optimal implementation protocols for different training modalities (e.g., continuous training vs. intermittent training) remain to be further explored and evaluated.

Resistance training is a periodic form of physical exercise that induces neuromuscular adaptive changes by gradually overload of skeletal muscle using external weights and is considered an important intervention to improve most motor performance (Sepúlveda-Lara et al., 2024). It has been suggested that diminished muscle strength may be an early indicator or contributory factor in the development of Alzheimer's disease (Buchman et al., 2007). Lower muscle strength is associated with an increased risk of cognitive impairment, including AD (Boyle et al., 2009). Whereas, for every unit increase in muscle strength, the prevalence of AD decreases by 43% at the beginning of cognitive impairment (Li et al., 2024).

Resistance training has been shown to positively impact agility, lower limb strength, balance, and flexibility in AD (Garuffi et al., 2013). Papatsimpas et al. demonstrated that resistance training was effective in slowing global cognitive function, executive function, and working memory decline and improving instrumental daily activity in patients with mild AD; the effect was more significant if combined with aerobic exercise (Papatsimpas et al., 2023). Xiao et al. showed that isokinetic muscle strength training can not only significantly promote the recovery of cognitive function and activities of daily living in AD patients, but also significantly improve motor function and balance function, and advocated that isokinetic muscle strength training should be used as an adjuvant therapy for the treatment of AD patients (Xiao et al., 2017). Chang et al. showed that resistance training targeting the trunk and lower limbs resulted in significant improvements in depression, muscle mass, and muscle function in patients with mild AD-sarcopenia comorbidity (Chang et al., 2020). Resistance training of the upper and lower extremities based on elastic bands has also been shown to significantly improve muscle strength and endurance, cardiopulmonary function, and gait speed in patients with mild AD, and resistance training is considered an effective rehabilitation program for patients with AD (Ahn and Kim, 2015). Other studies have also shown that resistance training does not significantly improve cognitive function in AD patients. The reason for the lack of expected results may be that the low-volume, low-intensity exercise program adopted under the principle of safety first failed to fully activate the neural adaptive mechanism (Vital et al., 2012). Based on the multifaceted improvement of AD patients by resistance training with different intensity and cycle parameters mentioned above (including cognitive function, motor function, muscle strength, functional flexibility, and quality of life), it should be considered as an important component of AD public health promotion programs (as shown in Table 2).

Physical and mental exercise is a non-pharmacological adjunctive strategy to improve symptoms in AD patients. Traditional Chinese physical and mental exercises such as Tai Chi and Baduanjin Qigong are mostly based on Traditional Chinese Medicine (TCM) theory, combined with respiratory control, stretching and relaxation of skeletal muscles and concentration of mental ideation, which aim to enhance strength, flexibility, balance and proprioception, while increasing attention to reduce anxiety and stress, and have a positive effect on improving cognitive impairment, physical function and quality of life in AD patients (Jiang et al., 2022). Yao et al. found that Tai Chi significantly improved functional activity performance associated with fall risk and effectively reduced fall risk in AD patients (Yao et al., 2013). Li et al. showed that Tai Chi exercise combined with Naoling decoction improved cognitive function and anxiety status and quality of life in AD patients (Liu et al., 2013). It has also been confirmed that Tai Chi combined with traditional Chinese art calligraphy and painting exercises that have a stabilizing physical and mental effect can significantly improve cognitive function and quality of life in AD patients (Tai et al., 2016). In addition, Baduanjin Qigong has been shown to positively improve cognitive function and activities of daily living in patients with mild-to-moderate AD (Wei et al., 2024). In recent years, dance and yoga have also been used in the rehabilitation of AD patients and have been shown to show positive benefits in improving cognitive function, neuromental status, balance ability, functional flexibility, and lower limb strength in AD patients (Tao et al., 2023). Chiesi et al. showed that Biodanza intervention significantly improved neuropsychiatric symptoms in AD patients and relieved aggressive behavior and verbal agitation symptoms in agitated states, with a positive promoting effect on mental health status (Chiesi et al., 2021). Salsa dance has also demonstrated significant improvements in range of motion, strength, balance, functional flexibility, gait distance, and speed in AD (Abreu and Hartley, 2013). In addition, yoga significantly improves cognitive function and quality of life in AD patients, while effectively relieving depressive symptoms and enhancing their social engagement. This therapy not only has a positive impact on mental health and overall quality of life in AD patients, but also significantly reduces caregiver stress (Kaushik et al., 2025) (as shown in Table 3).

In summary, various types of exercise therapy have potential clinical benefits in the overall management of AD. Although there is no consensus on improving the optimal type and intensity of exercise in AD patients to date, it has been comprehensively documented that regular moderate-intensity exercise or combined exercise (e.g., aerobic exercise plus resistance training), and maintaining a high frequency (e.g., >30 min per day, ≥2 days per week for ≥4 weeks), can bring significant clinical benefits to delay the progression of AD. Key implementation principles include the development and implementation of personalized exercise prescriptions under the guidance of professionals, whose content, intensity, and supervision level need to be adjusted in a timely manner according to individual health status and needs, especially considering the staging characteristics of AD, and early patients are recommended to pursue the benefits of maximizing cognitive and physiological function improvement and delaying decline using moderate-intensity exercise (e.g., brisk walking, power cycling) or combined training (e.g., aerobic resistance + resistance) under the premise of assessing good tolerability. For patients in the middle and advanced stages, in view of their significant cognitive and motor dysfunction, it is necessary to give priority to ensuring safety and compliance. Physical and mental movements (such as yoga, tai chi, qigong) emphasize physical and mental integration, and have low fall risk characteristics and adaptability to residual executive function, which can be used as a core intervention for the management of behavioral psychiatric symptoms (BPSD) and maintenance of ADL in patients with advanced AD. Exercise protocols must also be developed to integrate patients' current physical condition exercise type preferences, and previous medication use to ensure compliance and long-term sustainability. Although the current AD exercise intervention strategy has certain applicability, there are few randomized controlled studies on the effect of different types of exercise on improving the symptoms of AD patients, and it is necessary to compare the improvement effect of different types of exercise on the symptoms of AD patients through a larger sample size. This requires that future studies should focus on the following: (1) consider combined exercise and focus on optimizing the dose, and investigate whether specific characteristics of AD patients are associated with different types of exercise; (2) clinical intervention programs for AD patients need to fully consider specific factors such as gender, disease severity, specific drug use, and intervention cycles to effectively control heterogeneous factors and make clinical exercise intervention programs objective, scientific, and effective for AD patients; (3) evidence-based recommendations on which exercise is most suitable for an individual and the optimal length and intensity of intervention required to produce clinically relevant positive effects remain lacking, and it is necessary to systematically determine the most effective treatment for specific signs and symptoms from all available exercise types and provide individual evidence-based recommendations for AD patients.

In summary, KP is closely related to the occurrence and development of AD, regular exercise may regulate the balance of KP metabolism in the central and/or peripheral, so as to achieve AD exercise prevention and treatment, and this still needs to be confirmed by further studies.

Scientific exercise/physical activity is an important component of the public health promotion program for AD (AD health management). However, the neuroprotective effect of exercise intervention in AD and the molecular basis of its pathology and symptom improvement remain to be deeply elucidated. Recent studies have mainly focused on autophagy, oxidative stress, Glu excitotoxicity, mitochondrial dysfunction, cholinergic deletion and other perspectives to explore the possible neurobiological mechanism of exercise in the prevention and treatment of AD. Combined with the above exercise and KP, KP and AD occurrence and development and the role of prevention and treatment and exercise in AD prevention and treatment, it is speculated that KP key metabolic enzymes (IDO, KMO and KAT) may be a new target to mediate AD exercise prevention and treatment. In subsequent studies, we will employ integrated pharmacological modulation and genetic modification strategies. This will include utilizing gene editing technologies (e.g., Cre-LoxP and CRISPR-Cas9) alongside advanced cellular and molecular imaging techniques to achieve cell- and tissue-specific knockdown or knockout of key KP metabolic enzymes. These approaches will elucidate the critical role of KP enzymes in exercise-mediated prevention and mitigation of AD, thereby validating their feasibility as novel therapeutic targets for exercise-based AD interventions. This work will provide essential theoretical foundations and innovative perspectives for research on the neurobiological mechanisms underlying exercise-induced alleviation of AD-related symptoms, as well as for targeted therapeutic strategies.