Abstract

   The anti-nerve growth factor antibody class of drugs interrupts
   signaling by blocking NGF binding to TrkA receptors for the treatment
   of pain; however, this target class of drugs has been associated with
   serious adverse effects in the joints during clinical trials. DS002 is
   a novel anti-nerve growth factor antibody drug independently developed
   by Guangdong Dashi Pharmaceuticals. The main purpose of this study is
   to explore the correlation between DS002 and pain as well as cartilage
   and bone metabolism with the help of metabolomics technology and the
   principle of enzyme-linked reaction, and to examine whether DS002 will
   produce serious adverse effects in joints caused by its same target
   class of drugs, in order to provide more scientific basis for the
   safety and efficacy of DS002. Our results showed that DS002 mainly
   affected the metabolism of aromatic amino acids and other metabolites,
   of which six metabolites, l -phenylalanine, 5-hydroxytryptophan,
   5-hydroxytryptamine hydrochloride, 3-indolepropionic acid, kynuric
   acid, and kynurenine, were significantly altered, which may be related
   to the effectiveness of DS002 in treating pain. In addition, there were
   no significant changes in biological indicators related to cartilage
   and bone metabolism in vivo, suggesting that DS002 would not have a
   significant effect on cartilage and bone metabolism, so we hypothesize
   that DS002 may not produce the serious adverse effects in joints caused
   by its fellow target analogs. Therefore, the Anti-NGF analgesic drug
   DS002 has the potential to become a promising drug in the field of
   analgesia, providing pain patients with an efficient treatment option
   without adverse effects.

   Keywords: metabolomics, pain, aromatic amino acids, biomarkers,
   anti-ngf drugs

Graphical Abstract

   [52]graphic file with name FPHAR_fphar-2024-1396790_wc_abs.jpg

1 Introduction

   Chronic pain is an unsolved global healthcare problem that remains a
   major cause of suffering, disability, and substantial economic burden
   worldwide ([53]Cohen et al., 2021). It is estimated that the rate of
   chronic pain in humans is as high as 20% each year ([54]Dahlhamer et
   al., 2018). The International Association for the Study of Pain (IASP)
   defines pain as unpleasant sensory and emotional experiences related to
   or similar to actual or potential tissue damage ([55]Raja et al.,
   20220). Pain can be categorized as acute or chronic based on the
   duration and nature of its occurrence. Modern medical research has
   proven that acute pain is a symptom that goes away as the disease gets
   better; However, chronic pain usually lasts for more than 1 month or
   even 3 months, and this pain is no longer a symptom, but should be
   treated as a disease. The occurrence of pain is caused by many factors.
   In addition to the impact of injury or disease, there are physical
   perception, psychological and emotional changes, genetic factors, etc.
   Due to the complexity of pain, the results of the current treatment of
   chronic pain are unsatisfactory ([56]The, 2021). Currently available
   analgesics do not meet treatment goals. For example, opioids and
   non-steroidal anti-inflammatory drugs (nsaids), two commonly used
   analgesics in clinical practice, have different side effects in chronic
   pain. The main adverse reactions induced by opioids include analgesic
   tolerance, addiction and physical tolerance, nausea and vomiting,
   constipation, and respiratory depression ([57]Paul et al., 2021). The
   long-term use of NSAIDs can produce gastric bleeding and
   cardiovascular-related adverse reactions ([58]Trelle et al., 2011). The
   different side effects greatly limit the widespread use of these two
   classes of drugs. Therefore, there is an unmet need for effective
   analgesic drugs that are safer and more tolerable than currently
   available treatments.

   Nerve growth factor (NGF) was first discovered by Rita Levi-Montalcini
   ([59]Levi-Montalcini, 1952). NGF in the body is both an essential
   protein and an important pain-producing substance ([60]Hirose et al.,
   2016). It has been found that NGF binding can be two types of receptors
   ([61]Landreth and Shooter, 1980), the protomyosin receptor kinase a
   (TrkA), which has a high affinity for NGF, and the p75 neurotrophic
   factor receptor (p75NTR), which has a low affinity for NGF. NGF is
   secreted by inflammatory cells in damaged tissues and by Schwann cells
   in damaged nerves ([62]Thacker et al., 2007; [63]Basbaum et al., 2009).
   When tissue injury occurs, NGFs are abundantly expressed at the site of
   injury and bind to TrkA receptors on the cell membranes of sensory
   nerve endings, which in turn transmits pain signals, ultimately leading
   to the onset of pain. Therefore, in cases where an inflammatory
   response persists with damaging pain, such as low back pain, back pain,
   autoimmune diseases - rheumatism, etc., or in cases of neuropathic
   pain, NGF induces sensitization of the peripheral and central systems,
   which can lead to nociceptive hypersensitivity and anisocoria, and
   analgesics that block NGF/TrKA signaling can be very effective in these
   pathologies ([64]Hirose et al., 2016). In recent years, researchers
   have also identified it as one of the key targets for drugs to treat
   chronic pain. Currently, the main NGF/TrkA inhibitor drugs are anti-NGF
   antibodies and TrkA activity inhibitors ([65]Hirose et al., 2016). This
   study will focus on a new anti-NGF antibody, which have shown promising
   results in clinical studies and are expected to provide new clinical
   treatment strategies for chronic pain in the future.

   Metabolomics is the quantitative and qualitative analysis of all small
   molecules in a biological sample at a certain time and under certain
   conditions. It is a accredited methodology with more than 20 years of
   practical experience. At present, metabolomics has been extended to
   many disciplines ([66]Jeyarajah et al., 2006; [67]Wishart, 2016;
   [68]González-Domínguez et al., 2020). Metabolomics technologies have
   improved the mechanistic understanding, diagnosis, and treatment of
   various diseases by utilizing biological samples from humans and animal
   and cellular models of human diseases. Although the relationship
   between metabolomics and pain may be complex because the occurrence of
   pain is accompanied by multiple physiological symptoms, these issues
   have the potential to manifest through the level of metabolomics
   ([69]Miettinen et al., 2021), which will facilitate the development of
   mechanism-based pain diagnostics to support mechanism-based therapeutic
   approaches. Metabolomics research has the potential to advance these
   goals because it can discover pain biomarkers to improve pain treatment
   options, facilitate drug development progress, and ultimately reduce
   the burden of chronic pain.

   DS002 is an anti-NGF monoclonal antibody developed by Tashi
   Pharmaceutical (Guangdong) Co. Ltd. It interrupts the signal
   transmission by blocking the binding of NGF to the TrkA receptor,
   making it unable to stimulate the corresponding sensory neurons, and
   finally achieving an analgesic effect. Currently, a phase I clinical
   trial of DS002 has been completed and published in our laboratory
   ([70]Ma et al., 2022). This study aims to explore the effects of DS002
   on pain, cartilage and bone metabolism-related markers, and the changes
   in metabolic profile in vivo, to provide more evaluation basis for the
   drug development of DS002.

2 Methods

2.1 Sample collection

   The study samples were obtained from the Phase I clinical Trial Center
   of Nanjing Drum Tower Hospital. Blood samples from 48 healthy subjects
   were enrolled in this study and divided into seven different DS002 dose
   groups (0.5, 1, 2, 4, 7, 12, 20 mg), and 432 blood samples were
   collected. The samples used in this study were PK blood samples at 0 h
   and D4 (72 h), D8 (168 h), D15 (336 h), D29 (672 h), D43 (1008 h), D57
   (1344 h), D85 (2016 h), D113 (2688 h) after administration. The
   collected blood samples were used to determine the changes in serum
   metabonomics before and at different time points after administration,
   and the changes of cartilage and bone metabolism-related indexes before
   and at different time points after administration were determined by
   enzyme-linked immunosorbent assay (ELISA). This study (CTR20210155) was
   approved by the Ethics Committee of Drum Tower Hospital, the Affiliated
   Hospital of Nanjing University Medical School.

2.2 Subjects

   The project included healthy male and female subjects aged 18–45 years,
   with female subjects not pregnant or breastfeeding. Enrolled subjects
   have a body mass index (BMI) of 19.0–26.0 kg/m^2 and weigh ≥50 kg for
   males and ≥45 kg for females. Subjects agree to have no fertility or
   sperm/egg donation plans and to voluntarily use a medically recognised
   and effective form of contraception for the duration of the trial up to
   6 months after dosing. Subjects are sufficiently aware of the content,
   objectives and characteristics of the study to be able to complete the
   study as planned and are willing to serve as subjects and are able to
   sign an informed consent form.

   Subjects were excluded if they met any one of the following exclusion
   criteria: previous history of allergy or hypersensitivity; subjects
   with a history of osteoarthritic disease, a history of peripheral or
   autonomic neuropathy, a history of sensory abnormalities and
   hypoesthesia, a disease or history of unexplained spontaneous bleeding
   and/or post-traumatic hemorrhage with more than one bleed, a history of
   thyroid dysfunction or thyroid hormone abnormality, a history of a
   malignant neoplasm or a history of malignant neoplasms, an individual
   with an inherited immunodeficiency, or a family history of inherited
   immunodeficiencies; smokers, alcoholics or drug users; subjects should
   not take any prescription drugs, over-the-counter medications, or
   natural health supplements for 14 days prior to screening.

2.3 Sample preparation, instrument conditions and data analysis of
metabolomics

   The specific steps of sample processing were as follows: take 100 μL
   serum and add 400 μL methanol solution containing isotope internal
   standard. The samples were shaken for 3 min at 18,000 rpm*10 min and
   centrifuged at 4°C. 400 μL of the supernatant was transferred to an EP
   tube and centrifuged again in the same way. 100 μL of the supernatant
   was taken and 10 μL of the sample was injected. An equal volume of all
   samples was mixed to prepare quality control samples (QC). QC samples
   are part of the system regulation and quality control process. The
   processing and testing of QC samples are similar to those of the
   analyzed samples, which helps to obtain reliable and high-quality
   metabolomics data.

   LC-TOF/MS instrument conditions:the chromatographic column was Waters
   HSS T3 (2.1 mm × 100 mm, 1.8 um), the flow rate was 0.3 mL/min, the
   column temperature was 40°C, and the mobile phase was A: water phase
   (0.1% formic acid-water). B: organic phase (acetonitrile), MS scanning
   range: 50 to 1,200 Da. LC-MS/MS instrument conditions: the
   chromatographic column was Waters HSS T3 (2.1 mm × 100 mm, 1.8 um), the
   flow rate was 3 mL/min, the column temperature was 40°C, the mobile
   phase A was 0.1% formic acid water, and the mobile phase B was 0.1%
   formic acid acetonitrile. Gradient elution: mobile phase B was 10%
   (0.00–1.00 min), 10%–90% (1.00–9.00 min), 90% (9.00–12.00 min), 90%–10%
   (12.00–12.10 min), 10% (12.10–15.00 min), the injection volume was 5 uL
   per injection.

   Based on the metabolomics data obtained by LC-TOF/MS, the raw data were
   converted into mzXML format by MSconvert software. Subsequent analyses
   were performed using the tidymass package installed in the R project
   (version >4.2.2). GraphPad Prism 9.0.0 software was used for
   statistical analysis based on the metabolomics data obtained by
   LC-MS/MS. Data were expressed as mean and standard deviation, and
   differences between groups were analyzed by one-way analysis of
   variance (ANOVA). Differences between all groups were compared first,
   and then between two pairs. p < 0.05 was considered significant in all
   statistical tests.

2.4 ELISA method and data analysis

   The three kits of hyaluronic acid (HA), cartilage oligomeric matrix
   protein (COMP), and collagen type II (COL2α1) used in this study were
   purchased from Signalway Antibody (SAB, United States). The three kits
   of matrix metalloproteinase-3 (MMP-3), human bone-derived alkaline
   phosphatase (BAP), and tartrate-resistant acid phosphatase (TRACP-5)
   were purchased from Shanghai Jining Industrial Co., LTD. Sample testing
   procedures were performed exactly according to the instructions
   included with each kit. The data were analyzed by GraphPad Prism 9.0.0
   software.

3 Results

3.1 Non-targeted metabolomics to screen small molecule metabolites related to
pain

   PLS-DA and OPLS-DA are supervised multivariate data analysis and belong
   to model methods. They use partial least squares regression to
   establish the relationship model between metabolite expression and
   sample type to reduce the data dimension. This supervised mode can
   better establish the relationship between samples and better obtain the
   different information between groups. In this study, PLS-DA and OPLS-DA
   results of each dose group showed that the pre-administration and the
   post-administration could be significantly separated in Electrospray
   ionization positive ion mode (ESI+) and Electrospray ionization
   negative ion mode (ESI-) ([71]Figures 1A–G, [72]2A–G, [73]3A–G,
   [74]4A–G), suggesting that DS002 had A significant effect on the
   metabolism of the body. There was no over-fitting in the models of each
   group, and the prediction performance was good, indicating that the
   results were more accurate. Multivariate analysis was used to analyze
   the control and administration groups of different dose groups, and the
   differential metabolites between the control and administration groups
   were screened by setting the range of change fold and significance. We
   found that the levels of metabolites in vivo increased or decreased
   after administration of DS002. KEGG database was used to perform
   metabolic pathway enrichment analysis of differential metabolites in
   each dose group, and the pathways involved in each dose group were:
   0.5 mg: Phenylalanine, tyrosine and tryptophan biosynthesis,
   Phenylalanine metabolism, Histidine metabolism, Primary bile acid
   biosynthesis, Aminoacyl-trna biosynthesis; 1 mg: Caffeine metabolism,
   Linoleic acid metabolism, α-linoleic acid metabolism, Arachidonic acid
   metabolism, Glycerophospholipid metabolism; 2 mg: Phenylalanine
   metabolism, Phenylalanine, tyrosine and tryptophan biosynthesis,
   Aminoacyl-tRNA biosynthesis, Primary bile acid biosynthesis, Ubiquinone
   and other terpene quinone biosynthesis, Sphingolipid metabolism,
   Arginine and proline metabolism, Tyrosine metabolism; 4 mg:
   Phenylalanine metabolism, Phenylalanine, tyrosine and tryptophan
   biosynthesis, Ubiquinone and other terpene quinone biosynthesis,
   Caffeine metabolism, Sphingolipid metabolism, Primary bile acid
   biosynthesis, Tyrosine metabolism, Purine metabolism; 7 mg:
   Phenylalanine metabolism, Phenylalanine, tyrosine and tryptophan
   biosynthesis, Linoleic acid metabolism, Caffeine metabolism,
   Aminoacyl-tRNA biosynthesis, Ubiquinone and other terpene quinone
   biosynthesis, Pyrimidine metabolism, Tyrosine metabolism; 12 mg:
   Unsaturated fatty acid biosynthesis, Linoleic acid metabolism,
   Phenylalanine, tyrosine and tryptophan biosynthesis, Phenylalanine
   metabolism, Citrate cycle (TCA), Alanine, aspartate and glutamate
   metabolism, Glyoxylate and dicarboxylic acid metabolism,
   Glycerophospholipid metabolism; 20 mg: Aminoacyl-tRNA biosynthesis,
   Phenylalanine, tyrosine and tryptophan biosynthesis, Arginine
   biosynthesis, Arginine and proline metabolism, Glutathione and
   glutamate metabolism, Glutathione metabolism, Ubiquinone and other
   terpene quinone biosynthesis, Phenylalanine metabolism, Alanine,
   aspartate and glutamate metabolism, Tyrosine metabolism ([75]Figure 5).
   It is worth noting that aromatic amino acid metabolic pathways were
   found in the differential metabolic pathways of the six dose groups,
   and aromatic amino acids mainly included tyrosine, phenylalanine and
   tryptophan. Finally, the differential metabolites were summarized in
   the form of Venn diagram through the website Bioinformatics. Tryptophan
   was found in the differential metabolites of the five dose groups under
   ESI+ and ESI- modes ([76]Figure 6). The compounds contained in each
   color region are listed in [77]Supplementary Tables S1, S2. Therefore,
   aromatic amino acids were selected for the next targeted metabolomics
   analysis.

FIGURE 1.

   [78]FIGURE 1
   [79]Open in a new tab

   PLSDA plots of samples in ESI + mode before and after administration in
   different dose groups. The doses were 0.5 mg panel (A), 1 mg panel (B),
   2 mg panel (C), 4 mg panel (D), 7 mg panel (E), 12 mg panel (F), and
   20 mg panel (G).The red color represents the serum sample collected at
   168 h after administration, which is the post-administration. The green
   represents the serum sample collected at 0 h before administration,
   which is the pre-administration.

FIGURE 2.

   [80]FIGURE 2
   [81]Open in a new tab

   PLSDA plots of samples in ESI-mode before and after administration in
   different dose groups. The doses were 0.5 mg panel (A), 1 mg panel (B),
   2 mg panel (C), 4 mg panel (D), 7 mg panel (E), 12 mg panel (F), and
   20 mg panel (G). The red dots represent the serum samples collected at
   168 h after administration, which is the post-administration. The green
   dots represent the serum samples collected at 0 h before administration
   and are the pre-administration.

FIGURE 3.

   [82]FIGURE 3
   [83]Open in a new tab

   OPLS-DA plots of samples in ESI + mode before and after administration
   in different dose groups. The doses were 0.5 mg panel (A), 1 mg panel
   (B), 2 mg panel (C), 4 mg panel (D), 7 mg panel (E), 12 mg panel (F),
   and 20 mg panel (G). The red dots represent the serum samples collected
   at 168 h after administration, which is the post-administration. The
   green dots represent the serum samples collected at 0 h before
   administration and are the pre-administration.

FIGURE 4.

   [84]FIGURE 4
   [85]Open in a new tab

   OPLS-DA plots of samples in ESI-mode before and after administration in
   different dose groups. The doses were 0.5 mg panel (A), 1 mg panel (B),
   2 mg panel (C), 4 mg panel (D), 7 mg panel (E), 12 mg panel (F), and
   20 mg panel (G). The red dots represent the serum samples collected at
   168 h after administration, which is the post-administration. The green
   dots represent the serum samples collected at 0 h before administration
   and are the pre-administration.

FIGURE 5.

   [86]FIGURE 5
   [87]Open in a new tab

   KEGG enrichment analysis of relevant pathways in different
   administration dose groups. Scatter plot of pathways enriched by
   important differential metabolites. The X-axis and size of a node
   indicate the influence of that path in the topological analysis
   pathway, and the Y-axis and color of a node indicate statistical
   significance. The doses were 0.5 mg panel (A), 1 mg panel (B), 2 mg
   panel (C), 4 mg panel (D), 7 mg panel (E), 12 mg panel (F), and 20 mg
   panel (G).

FIGURE 6.

   [88]FIGURE 6
   [89]Open in a new tab

   Common differential metabolites of five drug dose groups (2, 4, 7, 12,
   and 20 mg) in ESI+ and ESI-mode were screened by Venn diagram. Each
   colour area represents a dose group, numbers in individual colour areas
   indicate the number of different metabolites in that dose group, and
   numbers in overlapping colour areas indicate the number of metabolites
   that are common to two or more dose groups. The numbers in the centre
   of the graph indicate the number of metabolites shared among the five
   groups, and the [90]Supplementary Table S1 lists the metabolites for
   each dose group. (A): ESI + mode; (B) ESI-mode.

3.2 Targeted metabolomics for accurate quantification of aromatic amino acids
and their metabolites

   Based on the results of non-targeted metabolomics experiments, we first
   established a targeted determination method for aromatic amino acids by
   liquid chromatography-tandem mass spectrometry. We used targeted
   metabolomics to accurately quantify aromatic amino acids and their
   metabolites. Statistical analysis of the targeted metabolomics data
   revealed that changes in important metabolites were statistically
   significant only in the highest administered dose group. The results
   showed that after administration of 20 mg, the concentrations of
   L-phenylalanine and 5-hydroxytryptophan decreased after administration;
   the concentrations of 3-indolepropionic acid, tryptamine hydrochloride,
   Kynurenic acid, and kynurenine increased first and then decreased
   ([91]Figures 7A–F). The levels of these metabolites were then
   quantified in the placebo group and showed different trends between the
   placebo group and the treated group ([92]Supplementary Figure S1).
   Therefore, we hypothesise that changes in these metabolites may be
   related to the effects of DS002 on pain, and that changes in these
   metabolites may stimulate the antisensory potential of the organism.

FIGURE 7.

   [93]FIGURE 7
   [94]Open in a new tab

   Analysis of concentration trends of important metabolites in vivo. Line
   graphs (A) L-phenylalanine, (B) L-5-hydroxytryptophan, (C)
   Indole-3-propionic acid, (D) 5-Hydroxytryptamine hydrochloride, (E)
   Kynurenic acid and (F) L-kynurenine describe the changing trend of each
   differential metabolite with administration time. * indicates p < 0.05.

3.3 Changes in cartilage and bone metabolism-related indicators

   Identifying the blood levels and distribution of different markers
   related to cartilage and bone metabolism can help to assess the risk of
   osteoarthritis. In our study, the concentrations of serum-related
   markers of cartilage and bone metabolism were determined before and
   after DS002 administration. The results showed that the serum
   concentrations of the six indicators related to cartilage and bone did
   not differ significantly with the changes in dose and time of
   administration ([95]Figures 8A–F), indicating that DS002 had no
   significant effect on the markers related to cartilage and bone
   metabolism. So, we hypothesized that DS002 might not cause the serious
   adverse effects of joint necrosis and rapidly worsening osteoarthritis
   caused by the same target drug class.

FIGURE 8.

   [96]FIGURE 8
   [97]Open in a new tab

   Analysis of the trend changes of markers related to cartilage and bone
   metabolism over time and dosing increments.Scatter plot (A) Hyaluronic
   acid (HA), (B) TypeⅡcollagen (COLII), (C) Cartilage oligomeric matrix
   protein (COMP), (D) Bone-specific alkaline phosphatase (BAP), (E)
   Matrix metalloproteinase 3 (MMP-3), and (F) tartrate-resistant acid
   phosphatase 5b (TRACP-5b) depict the trends of each marker as a
   function of administration time and dose.

4 Discussion

   Pain is a common problem and is the main reason people seek medical
   care. But satisfactory clinical protocols for the treatment of pain
   remain elusive. Anti-NGF monoclonal antibody is a new class of
   analgesic drugs. In the studies that have been done on the treatment of
   pain with this class of drugs, it has shown high efficacy in the
   treatment of pain, far exceeding the efficacy of existing drugs, and is
   non-addictive. In vitro studies showed that DS002 successfully blocked
   the binding of NGF and TrkA receptors, thereby blocking the
   transmission of pain signals. DS002, as a novel analgesic drug
   targeting NGF, has the potential to become a new generation of drugs
   for treating chronic pain, nevertheless, previous studies have shown
   that co-targeted drugs of DS002 cause serious adverse effects on
   joints. In addition, the results of our Phase I clinical trial
   demonstrated that DS002 did not experience dose-limiting toxicity in
   terms of safety and did not meet dose-escalation termination criteria
   in the range of 0.5–20 mg administered ([98]Ma et al., 2022). Five
   subjects developed arthralgia during the course of the trial, four
   subjects treated with DS002 (one in the 4.0 dose group, one in the 12.0
   dose group, and two in the 20.0 mg dose group) and one in the placebo
   group, although the symptoms that developed were medically diagnosed to
   be mild, and all eventually resolved without intervention. No serious
   bone and joint events or peripheral neuropathy/sensory abnormalities
   were noted throughout the trial. Rapidly progressive arthritis (RPOA)
   was not observed even when the drug was administered at high doses. As
   this was the most prominent concern when the FDA shelved the
   development programme for similar anti-NGF drugs. Therefore, in the
   present study, we continued to investigate the effects of DS002 on
   cartilage and bone metabolism-related markers to better clarify whether
   it produces the same target drug-induced adverse effects in the joints
   and to discuss its safety in healthy subjects. Meanwhile, this study
   also explores the changes in metabolite levels in healthy subjects
   after taking DS002 from the perspective of blood metabolomics, and
   links the changes in metabolites to the mechanism of pain occurrence,
   to explore the reasons for the therapeutic effect of DS002, and to
   provide more scientific basis for advancing the marketing of DS002.

   Previous Tanezumab clinical trial data showed joint adverse effects,
   including osteonecrosis and rapidly worsening osteoarthritis, in the
   same class as DS002 ([99]Jayabalan and Schnitzer, 2017). Hyaluronic
   acid (HA) is the main component of the cartilage matrix, which can bind
   to CD44 receptors, thereby inhibiting the expression of interleukin
   (IL) 1β and reducing the production of matrix metalloproteinase 3
   (MMP-3) ([100]Karna et al., 2008). HA can enter the peripheral blood
   circulation through lymphatic circulation, and the expression level of
   HA is positively correlated with the severity of arthritis. Matrix
   metalloproteinases (MMPs) have been reported to play an essential role
   in the pathogenesis of osteoarthritis, as they are involved in the
   degradation of all matrices outside the cell except polysaccharides
   ([101]Grillet et al., 2023; [102]Murphy et al., 2002). Among them,
   MMP-3 is the most important protease involved in cartilage degradation,
   which can be regulated by IL-1 and TNF-α cytokines ([103]Shi et al.,
   2022), followed by activation of other interstitial collagenases to
   degrade type II collagen (COL II), accelerates cartilage destruction.
   COL II is the main extracellular matrix protein in cartilage, and its
   degraded fragments will further accelerate the progression of
   osteoarthritis, promote the degradation of COL II, and accelerate the
   destruction of cartilage ([104]Poole et al., 2002). Cartilage
   oligomeric matrix protein (COMP) is secreted by osteoblasts and
   synovium cells and is related to cartilage metabolism. After articular
   cartilage injury, cartilage matrix degradation produces COMP, which can
   be released into the blood ([105]Maly et al., 2021). Bone alkaline
   phosphatase (BAP) is secreted by osteoblasts during the maturation
   stage and binds to bone matrix after being secreted to the
   extracellular space. It is a special sensitive index reflecting bone
   formation. In osteoporosis, bone calcification is insufficient.
   Osteoblasts proliferate actively and cannot transform into osteocytes.
   Osteoblasts proliferate actively and secrete a large amount of BAP into
   the blood. Tartrate resistant acid phosphatase 5b (TRACP-5b) is an
   enzyme secreted by activated bone calpain and is a good marker of bone
   resorption and osteoclast activity. Its concentration can be measured
   to understand bone metabolism status ([106]Terpos et al., 2005). In
   this study, the concentrations of HA, MMP-3, COLII, COMP, BAP,
   TRACP-5b, which are related to cartilage and bone metabolism, did not
   change significantly over time in healthy subjects after treatment of
   different doses of DS002. These results indicated that the
   administration of monoclonal antibody DS002 did not cause fluctuations
   in the content of metabolites related to cartilage and bone metabolism,
   indicating that DS002 had no significant effect on the relevant
   parameters of cartilage and bone metabolism.

   When pain occurs, a series of changes are reflected in the body’s
   metabolites. Specific blood, urine, and saliva biomarkers can be used
   to determine the degree of risk of chronic pain and can be used as
   prognostic markers of pain progression and treatment response
   ([107]Hagedorn et al., 2021). In the present study, we measured the
   therapeutic effect of DS002 by investigating the blood metabolic
   profile. After DS002 administration, the body’s metabolism was
   significantly affected. The changes in metabolites involved in
   different metabolic pathways. According to previous studies, the
   metabolites in these metabolic pathways have been shown to be related
   to the occurrence and development of different pain. For example,
   researchers have found a link between amino acid metabolism and pain
   ([108]Alexander et al., 2013; [109]Teckchandani et al., 2021), the
   ratio of tryptophan to kynurenine is increased during back pain
   ([110]Staats Pires et al., 2020), glutamate levels increase with the
   onset of fibromyalgia ([111]Clos-Garcia et al., 2019), phosphocholine,
   alanine, and taurine are elevated in neuropathic pain ([112]Finco et
   al., 2016). As the occurrence of pain is accompanied by a series of
   symptoms, different metabolic pathways can also indirectly affect pain.
   Studies have shown that the methionine metabolism pathway is related to
   sleep problems, and the glutathione metabolism pathway is related to
   obesity, and pain, sleep and obesity affect each other ([113]Suzuki et
   al., 2015; [114]Miettinen et al., 2019), so the imbalance of these two
   pathways will also affect pain. So we propose that the treatment of
   pain with DS002 may have a relationship to changes in these pathways.
   In our results, aromatic amino acid metabolic pathways account for the
   most significant proportion, such as the biosynthesis of phenylalanine,
   tyrosine, and tryptophan, as well as phenylalanine and tyrosine
   metabolism. These results may be relevant to the treatment of pain with
   DS002, because the amino acid metabolic pathway is related to the basal
   metabolic recovery of the body, and correct regulation reduces
   inflammatory response and affects nerve conduction. In addition, amino
   acids are the main substrate for producing neurotransmitters, and
   neurotransmitter-mediated pain regulation is the primary way.

   Phenylalanine and tryptophan are essential amino acids for the human
   body and belong to the aromatic amino acids. Most phenylalanine in the
   body is oxidized to tyrosine, an amino acid associated with analgesic
   neurotransmitters, through the catalytic action of phenylalanine
   hydrogenase. The pathophysiological mechanisms underlying the onset and
   resolution of pain are intricate, and many different neurotransmitter
   systems are involved in the transmission, processing, and control of
   pain ([115]Millan, 2002). In previous studies, Amino acid metabolites,
   including L-tryptophan, L-histidine, and L-tyrosine, were significantly
   decreased in the brain of pain model mice, suggesting that neuropathic
   pain may be promoted by the reduction of amino acids related to
   analgesic neurotransmitters. Therefore, we infer that the in vivo
   metabolic conversion of PHE to amino acids related to analgesic
   neurotransmitters would be involved in the therapeutic effect of DS002.
   Tryptophan is an essential amino acid for generating and maintaining
   proteins, muscles, enzymes, and neurotransmitters in the human body
   ([116]Körtési et al., 2022). The tryptophan metabolism pathway includes
   the kynurenine, serotonin, and indole pathways. Changes in tryptophan
   levels can lead to imbalances in the synthesis and metabolism of
   downstream metabolites, which may be related to the pathological and
   physiological mechanisms of neurological and psychiatric diseases
   ([117]Comai et al., 2020). Recent studies have shown that the
   dysregulation of the kynurenine pathway and the changes of metabolites
   in its pathway are associated with neurotoxicity and inflammation, and
   many metabolites are associated with chronic pain ([118]Tanaka et al.,
   2021). Tryptophan enters the kynurenine metabolic pathway through
   indoleamine 2, 3-dioxygenase (IDO) and tryptophan 2, 3-dioxygenase
   (TDO) and generates kynurenine, which is irreversibly converted to
   kynurenic acid (KA) by kynurenine aminotransferase (KATs)
   ([119]Schwarcz et al., 2012). KA is a neuroactive metabolite which
   belongs to the antiexcitotoxin and has neuroprotective effects
   ([120]Fehér et al., 2019). The activity of KA at different targets is
   the primary mechanism by which it achieves specific functions. KA
   exerts its neuroprotective effect by antagonizing NMDA receptors
   ([121]Birch et al., 1988), and previous studies have shown that marked
   reductions in neuroprotective metabolites occur in patients with
   psychiatric disorders ([122]Athnaiel et al., 2022). KA interacts with g
   protein-coupled receptor 35 (GPR35) to indirectly reduce inflammation
   ([123]Wirthgen et al., 2017), which is highly correlated with chronic
   pain. Thus, the neuroprotective and inflammatory response effects of KA
   suggest elevated levels of KA and other metabolites with similar
   properties when treating chronic pain symptoms.

   In recent years, a great deal of research has been conducted on the
   role of 5-HT in chronic pain states.5-HT is metabolised in the body by:
   the small amount of tryptophan in the body is catalyzed by tryptophan
   hydroxylase (TPH) to generate 5-hydroxytryptophan (5-HTP). The aromatic
   L-amino acid decarboxylase interacts with the cofactor Pyridoxal
   5-phosphate to convert 5-HTP to 5-hydroxytryptamine (5-HT)
   ([124]Daubert and Condron, 2010). 5-HT, also known as serotonin, is an
   excitatory neurotransmitter in the human central nervous system, which
   causes the contraction of vascular smooth muscle, thereby potentially
   regulating pain ([125]Taylor and Basbaum, 1995; [126]Xue et al., 2023).
   Endogenous 5-HT was released in response to tissue injury in mast
   cells, platelets and endothelial cells, while multiple neuronal 5-HT
   receptors were detected in the periphery, suggesting that 5-HT can
   interfere with injurious transmission at peripheral sites ([127]Bardin,
   2011). Previous studies have also pointed to the complex relationship
   between 5-HT and pain, for example: intradermal administration of 5-HT
   to rats causes dose-dependent nociceptive hypersensitivity ([128]Taiwo
   and Levine, 1992); plantar injections of 5-HT result in plantar oedema
   and intense flinching and licking behaviour ([129]Sufka et al., 1992);
   In healthy volunteers, intradermal injections of low concentrations of
   5-HT produced burning pain ([130]Lischetzki et al., 2001), while
   injections into the mandibular occlusal muscle produced nociceptive
   hypersensitivity ([131]Ernberg et al., 2000); Furthermore, In a
   carrageenan inflammatory pain model, administration of the 5-HT3
   antagonist ICS-205930 by plantar administration at different times
   resulted in complete suppression of nociceptive hypersensitivity to
   foot inflammation, a process accompanied by 5-HT release
   ([132]Eschalier et al., 1989). Taken together, these studies suggest
   that 5-HT release sensitises sensory nerve fibres in response to
   inflammation or tissue damage and promotes peripheral sensitisation
   through direct or indirect mechanisms. It has been shown that patients
   with chronic pain exhibit weaker endogenous pain inhibitory processes
   compared to pain-free individuals, and that the involvement of downward
   inhibition may represent an endogenous mechanism to prevent the
   chronicisation of pain ([133]Ossipov et al., 2014). In vivo, the
   periaqueductal grey matter (PAG) is a brain region capable of
   activating endogenous pain inhibitory systems, and the PAG influences
   downstream pain modulation primarily through its interconnections with
   the rostral ventromedial medulla (RVM) ([134]Ossipov et al., 2014),
   whereas the activation of downstream projections from the RVM induces
   the release of serotonin ([135]Foo and Mason, 2003; [136]Wei et al.,
   2010; [137]Hossaini et al., 2012). 5-HT can exhibit both pro- and
   antinociception throughout, depending on the subtype of 5-HT receptor
   it activates, with activation of 5-HT1A, 5-HT1B, 5-HT1D, and 5-HT7
   receptors tending to be antinociceptive, and 5-HT2A and 5-HT3 receptors
   tending to be proinjury sensory ([138]Suzuki et al., 2004; [139]Sasaki
   et al., 2006; [140]Bannister et al., 2009; [141]Dogrul et al., 2009);
   there is also a large body of research pointing to the inhibitory and
   facilitatory effects of the downstream 5-HT pathway on spinal neuronal
   excitability and pain-related behaviours, depending on the acute or
   persistent pain state ([142]Bardin, 2011). In conclusion, these results
   suggest that 5-TH and the downstream 5-TH pathway play an important
   role in pain transmission in a complex pathological manner.

   A minimal amount of tryptophan in the body can also be converted into
   indole and its derivatives, such as indole-3-acetic acid,
   indole-3-propionic acid, etc. under the action of the gut microbiota
   ([143]Wei et al., 2021). Indole and its derivatives can contribute to
   the regulation of inflammatory responses ([144]Venkatesh et al., 2014),
   which are highly correlated with the occurrence of pain. Therefore,
   indole and its derivatives can also affect the progression of pain. In
   addition, the gut microbiota is also an important regulator of visceral
   pain. In recent years, it has been found that the gut microbiota also
   plays a crucial role in chronic pain ([145]Guo et al., 2019). In
   summary, these associations potentially suggest a correlation between
   changes in metabolite levels and the efficacy of DS002 in treating
   pain.

   From the above, our findings suggest that DS002 has greater advantages
   in the treatment of pain compared with the existing drugs and the drugs
   in the same target class as DS002. Of course, the study has
   limitations. Our method of administration is a single dose in healthy
   people, which is different from clinical drug administration.
   Therefore, the application of DS002 in the clinical treatment of pain
   needs to be further explored.

5 Conclusion

   Our results showed that DS002 caused the changes in the aromatic amino
   acid metabolic pathway in subjects treated with it. Among them, the
   changes in the concentration of L-phenylalanine, 5-hydroxytryptophan,
   5-hydroxytryptamine, kynurenic acid, kynurenine, and indole-3-propionic
   acid may be related to the therapeutic effect of DS002 for treating
   pain. After administration, there were no significant changes in
   cartilage and bone metabolism indicators, and we preliminarily
   speculate that DS002 will not produce severe adverse reactions as the
   same target class of drugs. Our research provides a specific basis for
   DS002, and aromatic amino acids and their metabolites may serve as
   biomarkers for treating pain. Combined with the clinical results of the
   previous study of DS002 in our laboratory, it can more strongly
   demonstrate the enormous potential of anti-NGF drugs represented by
   DS002 in treating pain.

Acknowledgments