A total of 28 patients with DLBCL (18 men and 10 women; age range, 53–93 years; mean age, 72) and 34 healthy adult volunteers serving as controls were included in the serum analyses. The Ethics Committees of Fujita Health University (Toyoake, Japan; approval no. HM20-268) and Gifu University (Gifu, Japan; approval no. 2018-25) approved all procedures involving human subjects, and the study was performed in accordance with the principles of the Declaration of Helsinki. All patients and healthy volunteers signed informed consent before study participation. The present study investigated healthy volunteers and 28 patients, including 7 patients for Fig. S2 (3 men and 4 women; age range, 53-90 years), who were histologically diagnosed with DLBCL according to the World Health Organization classification of hematopoietic tumors (24) between April 2008 and November 2016. Patients with DLBCL metastasis or without complete clinical information were excluded from the present study. Serum samples from patients with DLBCL were obtained before the initiation of therapy. All serum samples were separated by centrifugation (1,500 × g; 15 min; 24°C) and stored at −80°C until analysis. Patients <70 years of age were assigned to receive eight cycles of R-CHOP or rituximab, pirarubicin, cyclophosphamide, vincristine and prednisone (R-THP-COP) therapy in Gifu University Hospital (Gifu, Japan) (25–27). Patients ≥70 years old received six cycles of R-CHOP or R-THP-COP therapy in Gifu University Hospital, which is recommended for elderly patients with DLBCL (28). The clinical characteristics of patients with DLBCL at the time of diagnosis are summarized in Table I. Concerning age as a prognostic factor, a patient age cut-off of 60 was determined according to the age cut-off of the International Prognostic Index (IPI) and Revised International Prognostic Index (R-IPI) (29,30). Healthy controls, without renal dysfunction, hepatic dysfunction, pregnancy or breastfeeding, immune-mediated inflammatory disease or medications, and blood dyscrasia or anemia, were sex-matched (20 male patients; 14 female patients) and age-matched (<60 years, 7 patients; ≥60 years, 27 patients; age range, 47–80 years; mean age, 66) to the patients with DLBCL. Blood samples from newly diagnosed patients and age- and sex-matched controls were collected at the inpatient and outpatient department (Gifu University Hospital, Gifu, Japan), respectively. All follow-up dates were based on the last entries on December 1, 2020. The sample size used in the present study was determined by the number of samples collected during the study period. Therefore, due to the small sample size, statistical analysis was performed using non-parametric analysis.

3-HK and KYN measurements were performed as previously described (19,31). For KYN measurement, serum was diluted (4:1, v/v) in 10% perchloric acid. After thorough mixing, the precipitated proteins were removed by centrifugation (7,000 × g; 10 min; 4°C). A total of 50 µl of the resulting supernatant was subjected to high-performance liquid chromatography (HPLC Prominence; Shimadzu) analysis. KYN was isocratically eluted from a reverse phase column [TSKgel ODS-100V; 3 µm, 4.6 mm (ID) × 150 mm (L); Tosoh] using a mobile phase containing 10 mM sodium acetate and 1% acetonitrile (pH adjusted to 4.5 with acetic acid) at a flow rate of 0.9 ml/min. KYN was detected using an ultraviolet and visible spectrophotometric apparatus (SPD-20A; Shimadzu) (UV wavelength, 365 nm).

For 3-HK measurement, serum was diluted (1:4, v/v) in 10% perchloric acid. After thorough mixing, the precipitated proteins were removed by centrifugation (7,000 × g; 10 min; 4°C). A total of 20 µl of the supernatant was applied to a 3-µm HPLC column (HR-80; 80×4.6 mm; ESA), using a mobile phase consisting of 1.5% acetonitrile, 0.9% triethylamine, 0.59% phosphoric acid, 0.27 mM EDTA and 8.9 mM sodium heptane sulfonic acid, at a flow rate of 0.5 ml/min. 3-HK was detected electrochemically using an ECD 300 detector (oxidation potential: +0.55 V; Eicom). KMO activity was calculated from the 3-HK/KYN ratio as previously described (32,33). Furthermore, the present study investigated the association between changes in 3-HK levels and KMO activity and sex, age, performance status (PS), serum lactate dehydrogenase (LDH) levels, soluble interleukin-2 receptor (sIL-2R) levels, extranodal lesions, clinical stage (CS), B symptom, IPI and R-IPI to establish a link between KMO activity and characteristics of patients with DLBCL. Extranodal lesions, CS and B symptoms were assessed according to the Ann Arbor criteria, and classification was performed for all patients (34).

Human B-cell non-Hodgkin lymphoma cell lines (KML-1: JCRB1347) and human DLBCL cell lines (STR-428: JCRB1384) were purchased from the Japanese Collection of Research Bioresources Cell Bank. KML-1 and STR-428 cells are positive for the surface markers CD10, CD19, CD20 and human leukocyte antigen DR (HLA-DR) (35,36). Therefore, both of these cell lines are categorized as germinal center B-cell-like (GCB) types in the Hans classification (37). All tumor cells were cultured in RPMI-1640 medium (FUJIFILM Wako Pure Chemical Corporation) supplemented with 10% FBS (HyClone; Cytiva), 50 U/ml penicillin (Sigma-Aldrich; Merck KGaA) and 50 µg/ml streptomycin (Sigma-Aldrich; Merck KGaA) at 37°C with 5% CO₂.

Total RNA was isolated from KML-1 and STR-428 cells using an Isogen RNA Isolation kit (Nippon Gene Co., Ltd.). The RNA (250 ng) was then used for first-strand synthesis of cDNA with a high-capacity cDNA reverse transcription kit (Applied Biosystems; Thermo Fisher Scientific, Inc.). Semi-quantitative RT-PCR in KML-1 and STR-428 cells was performed to detect KMO, IDO1 and β-actin cDNA using the KAPA Taq Extra HotStart Readymix PCR Kit (Nippon Gene Co., Ltd.) according to the manufacturer's protocol. The PCR products were separated on a 2% agarose gel and stained with ethidium bromide to visualize the amplified nucleic acid fragments. The mRNA expression levels of kynureninase (KYNU), 3-hydroxyanthranilate-3,4-dioxygenase (3-HAO), quinolinate phosphoribosyltransferase (QPRT) and β-actin KML-1 and STR-428 cells were quantified by qPCR on a 7900HT Fast Real-Time system (Applied Biosystems; Thermo Fisher Scientific, Inc.). KYNU-, 3-HAO-, QPRT- and β-actin-targeted reactions were performed using Sso Advanced SYBR-Green Supermix (Bio-Rad Laboratories, Inc.) according to the manufacturer's protocol (38), and data were analyzed using 7900HT software (version 2.3; Applied Biosystems; Thermo Fisher Scientific, Inc.). The PCR primers are listed in Table SI. Thermocycling conditions were as follows: Initial denaturation, 95°C for 30 sec; 40 cycles of amplification (denaturation, 95°C for 5 sec; annealing/extension, 58°C for 60 sec). Gene expression levels were normalized to β-actin expression levels using a standard curve.

KML-1 and STR-428 cells were collected, 100 µl RIPA extraction buffer (FUJIFILM Wako Pure Chemical Corporation) was added per 1×10⁶ cells, and the samples were sonicated and centrifuged at 10,000 × g for 5 min at 4°C. The protein concentration in the supernatant was measured using a BCA protein assay. Subsequently, 10 μg of protein was loaded onto a 10% Mini-PROTEAN TGX gel (Bio-Rad Laboratories, Inc.), separated by SDS-PAGE and transferred to PVDF membranes. After blocking non-specific reactions with 5% skimmed milk for 1 h at 24°C, the membranes were first incubated with the primary antibodies for 12 h at 4°C, including anti-KMO (dilution, 1:1,000; cat. no. 10698-1-AP; ProteinTech Group, Inc.) and anti-β-actin (dilution, 1:1,000; cat. no. A5441; Sigma-Aldrich; Merck KGaA). The membrane was then incubated with horseradish peroxidase-conjugated secondary antibody for 1 h at 24°C (dilution, 1:10,000; cat. nos. 115-035-144 and 115-006-020; Jackson ImmunoResearch Laboratories, Inc.). Detection was performed using the Immunostar LD reagent (FUJIFILM Wako Pure Chemical Corporation) and a WSE-6100 Lumino Graph I instrument (ATTO Corporation).

The lymph node of a single patient with DLBCL was fixed in 10% formalin in PBS overnight at 24°C and then embedded in paraffin. Sections (thickness, 3-µm) were used for H&E staining and KMO immunostaining. The primary antibody used was a mouse anti-KMO antibody (dilution, 1:2,000; cat. no. 60029-1-Ig; ProteinTech Group, Inc.). After deparaffinization and rehydration, sections were heated at 121°C for 20 min in Histofine antigen retrieval solution (pH 9.0; Nichirei Biosciences, Inc.). The sections were soaked in 3% hydrogen peroxide in methanol for 30 min to eliminate endogenous peroxidase activity. After nonspecific binding was blocked for 15 min at 24°C with G-block (cat. no. GB-01; GenoStaff Co., Ltd.), the sections were incubated with or without (negative control) primary antibody overnight at 4°C. Secondary antibody, conjugated with a peroxidase polymer (ready to use; cat. no. MP-7500; ImmPRESS Reagent anti-rabbit IgG; Vector Laboratories, Inc.) was added for 30 min at 24°C according to the manufacturer's protocol, followed by the addition of the substrate 3,3′-diaminobenzidine tetrahydrochloride (Dako; Agilent Technologies, Inc.). The sections were then counterstained with hematoxylin for 10 sec at 24°C.

KML-1 and STR-428 cells (2×10⁵ cells/200 µl) were seeded on 6-well coverslips and fixed with 4% paraformaldehyde at 24°C for 30 min. Fixed cells were permeabilized in 0.1% Triton X-100 in 1X PBS (FUJIFILM Wako Pure Chemical Corporation) for 30 min at 24°C, and blocked with 5% horse serum (Vector Laboratories, Inc.) for 1 h at 24°C. Anti-KMO antibody (dilution, 1:100; cat. no. 10698-1-AP; ProteinTech Group, Inc.) was used as the primary antibody, which was hybridized with the samples for 1 h at 24°C. The cells were then stained with Northern Lights anti-rabbit IgG-NL557 as the secondary antibody (dilution, 1:1,000; cat. no. NL004; R&D Systems, Inc.) for 1 h at 24°C in the dark. Nuclei were stained with DAPI (dilution, 1:1,000; cat. no. BS04; Dojindo Molecular Technologies, Inc.) for 5 min at 24°C. All immunostained slides were observed under a BX51 fluorescence microscope equipped with a DP74 digital camera (Olympus Corporation).

KML-1 and STR-428 cells were seeded in 96-well plates at a density of 1×10⁴ cells/well, and 10 µl 3-HK (1 and 10 µM; Sigma-Aldrich; Merck KGaA) or KMO inhibitor Ro61-8048 (10 and 100 nM, Sigma-Aldrich; Merck KGaA) was added. As a control, 10 µl PBS was added to the cells. After 24 h at 37°C with 5% CO₂, 10 µl water-soluble tetrazolium salt (WST)-1 reagent (Premix WST-1 Cell Proliferation Assay System; Takara Bio Inc.) was added and the cells were incubated for 1.5 h at 37°C with 5% CO₂. Absorbance measurements were performed on a microplate reader (main wavelength, 450 nm; auxiliary wavelength, 620 nm). The number of viable cells was quantified based on the absorbance.

NAD⁺ levels in KML-1 and STR-428 cells were measured using the NAD⁺/NADH assay kit-WST (cat. no. N509; Dojindo Laboratories Inc.), which allows for the determination of intracellular amounts of total NAD⁺/NADH and NADH alone. Intracellular NAD⁺ levels were determined by subtracting the levels of NADH from total NAD⁺/NADH levels.

GraphPad Prism software version 6 (GraphPad Software, Inc.) was used for all statistical analyses. Experiments were repeated at least three times. All data are presented as the mean ± SD. The Mann-Whitney U test (Fig. 1) and two-sided paired t-test (Fig. S2) was used to test for significant differences between two groups for patient data. Correlations between serum KYN and 3-HK levels in patients with DLBCL were analyzed using Pearson's product-moment correlation coefficient (r). Categorical variables in Tables II and III were presented as numbers, and the groups were compared using the χ² or Fisher's exact test, as appropriate. For in vitro experiments, unpaired Student's t-test was used to identify significant differences between two groups. One-way analysis of variance was used for significance testing of three groups. Tukey's multiple comparison test was used for post-hoc pairwise analyses of multiple comparisons. P<0.05 was considered to indicate a statistically significant difference.

The levels of serum 3-HK and the ratio of 3-HK/KYN, as an indicator of KMO activity (32,33), were significantly higher in patients with DLBCL before treatment than in healthy controls (Fig. 1A-C). Similarly, it has been reported that the levels of KYN in patients with DLBCL are increased compared with those in healthy controls (4). However, there was no correlation between KYN and 3-HK levels in patients with DLBCL (Fig. 1D). Therefore, the levels of serum 3-HK in patients with DLBCL were altered independently of KYN levels.

To investigate associations between the serum 3-HK levels or KMO activity and clinical characteristics of patients with DLBCL, patients with DLBCL were classified into low or high groups based on the median serum 3-HK levels (cut-off value, 21.9 nM) or KMO activity (cut-off value, 9.2). No significant associations between 3-HK levels or KMO activity and sex, age, PS, LDH, sIL-2R, molecular subtypes, extranodal lesions or B symptoms were observed. Importantly, 3-HK levels were significantly associated with CS, IPI and R-IPI. KMO activity was significantly associated with CS, but not IPI and R-IPI (Tables II and III). Furthermore, the levels of serum 3-HK and KMO activity in patients after treatment tended to be decreased compared with those measured before treatment (Fig. S2). These results suggested that serum 3-HK levels and KMO activity may reflect tumor progression.

To confirm KMO expression and its roles in DLBCL cells, the present study investigated KMO expression in KML-1 and STR-428 cells, and lymph nodes of patients with DLBCL. KML-1 (B-cell non-Hodgkin lymphoma) and STR-428 (DLBCL) cell lines exhibited a difference in KMO expression despite being derived from patients with the same disease and pathological classification (GCB type). Notably, STR-428 cells exhibited high levels of KMO expression (KMO^high), whereas KML-1 cells exhibited low levels of KMO expression (KMO^low; Fig. 2A-C). Both cell lines were negative for IDO1 mRNA using RT-PCR (data not shown). High levels of KMO expression were also observed in the lymph nodes of patients with DLBCL compared with those in the negative controls (Fig. 2D).

To investigate the role of KMO expression in DLBCL cells, the present study examined cell viability upon KMO inhibition or 3-HK addition. Spontaneous cell viability of KMO^high STR-428 cells, which have the ability to produce higher 3-HK levels, was much higher than that of KMO^low KML-1 cells, which have the ability to produce lower 3-HK levels (untreated ‘0’ in Fig. 2). Notably, although the addition of 3-HK to KMO^low KML-1 cells significantly improved cell viability, KMO^high STR-428 cells were not affected (Fig. 2E). By contrast, KMO inhibition in KMO^high STR-428 cells using Ro61-8048 reduced cell viability, but KMO^low KML-1 cells were not affected (Fig. 2F). These results suggested that the viability of DLBCL cells may be regulated by the production of 3-HK through KMO activity.

NAD⁺ is an essential coenzyme involved in cell redox reactions and is required for cell proliferation as a substrate for NAD⁺-dependent enzymes (39). Therefore, it was hypothesized that the difference in viability of these DLBCL cell lines resulted from an increase in the levels of NAD⁺, the final product of the KYN pathway. To test this hypothesis, the present study investigated the mRNA expression levels of KYNU, 3-HAO and QPRT, which encode metabolic enzymes downstream of KMO. The mRNA expression levels of KYNU and QPRT were significantly higher in KML-1 cells than in STR-428 cells. Furthermore, the mRNA expression levels of 3-HAO were significantly lower in KML-1 cells than in STR-428 cells (Fig. 2G). Notably, NAD⁺ levels in KMO^high STR-428 cells were significantly increased compared with those in KMO^low KML-1 cells (Fig. 2H).