Prognostic significance of kynurenine 3-monooxygenase and effects on proliferation, migration, and invasion of human hepatocellular carcinoma

Kynurenine 3-monooxygenase (KMO) is a pivotal enzyme in the kynurenine pathway of tryptophan degradation and plays a critical role in Huntington’s and Alzheimer’s diseases. This study aimed to examine the expression of KMO in human hepatocellular carcinoma (HCC) and investigate the relationship between its expression and prognosis of HCC patients. We first analyzed KMO expression in 120 paired HCC samples (HCC tissues vs matched adjacent non-cancerous liver tissues), and 205 clinical HCC specimens using immunohistochemistry (IHC). Kaplan-Meier survival and Cox regression analyses were executed to evaluate the prognosis of HCC. The results of IHC analysis showed that KMO expression was significantly higher in HCC tissues than that in normal liver tissues (all p < 0.05). Survival and recurrence analyses showed that KMO was an independent prognostic factor for overall survival (OS) and time to recurrence (TTR) (both p<0.01). And in vitro studies revealed that KMO positively regulated proliferation, migration, and invasion of HCC cells. These results suggest that KMO exhibits tumor-promoting effects towards HCC and it may serve as a novel prognostic marker in HCC.

Immunohistochemistry and Scoring. Immunohistochemistry, signal evaluation, and integrated optical density (IOD) analysis were performed as described previously 18 . TDO antibody was purchased from Aviva Systems Biology (1:100) and KMO antibody was purchased from LifeSpan BioSciences (1:100). Photographs of three representative fields were captured under high-power magnification (× 200); identical settings were used for each photograph. IOD was counted and measured using Image-Pro Plus v6.0 software, and mean IOD was calculated from three photographs per specimen.
Immunostaining scores were independently evaluated by two pathologists who were blinded to the clinical outcome. Semiquantitative scores were used to analyze immunostaining of each HCC case in tissue microarray. Intensity of staining was categorized into − , + , + + , or + + + , denoting negative, weak, moderate, or strong staining, respectively. Extent of immunostaining was categorized into 0 (0-5%), 1 (6-25%), 2 (26-50%), or 3 (>51%) on the basis of the percentage of positive cells. Three random microscope fields per tissue were calculated. The sum of intensity and extent of staining was used as final score of expression level, and determined by the formula: final score = intensity score × percentage score. The final score of ≤ 3 was defined as low expression, and > 3 as high expression.
Cell Culture and Transfection. Human normal liver cell lines L02 and SMMC7721 were obtained from Shanghai Institute of Cell Biology, Chinese Academy of Sciences. SK-Hep1 and PLC-PRF5 were purchased from the American Type Culture Collection. MHCC97H, MHCC97L, and HCCLM3 were provided by the Liver Cancer Institute of Zhongshan Hospital of Fudan University (Shanghai, China). Huh7 cells were purchased from Riken Cell Bank (Tsukuba, Japan). All the cell lines were cultured in Dulbecco's modified Eagle's medium (DMEM) (Gibco, Gaithersburg, MD, USA) containing 10% fetal bovine serum (FBS), 100 mg/ml penicillin, and 100 mg/ml streptomycin. All of the cells were incubated in a humidified atmosphere of 5% CO 2  Migration Assay. Cell migration assays were performed using Transwell filter champers (BD Biosciences). 5 × 10 4 cells in 200 μ l serum-free DMEM were seeded in the upper chamber of a transwell and 800 μ l medium supplemented with 15% FBS was added to the lower chamber. After indicated time (12 hours for SK-Hep1, 24 hours for MHCC-97H, 20 hours for SMMC7721 and 20 hours for Huh7) of incubation at 37 °C, migrated cells were fixed and stained with a dye solution containing 0.1% crystal violet and 20% methanol. Cells adhering to the lower side of the inserts were counted and imaged through an IX71 inverted microscope (Olympus). Five random microscopic fields were counted per well for each group, and the experiments were repeated at least three times independently.
Invasion Assay. For in vitro invasion assay, transwell filter champers (BD Biosciences) and transwells coated with Matrigel (BD Biosciences) were utilized according to manufacturer's instructions. 5 × 10 4 cells in 200 μ l serum-free DMEM were seeded in the upper chamber of a transwell and 800 μ l medium supplemented with 15% FBS was added to the lower chamber. After indicated time (20 hours for SK-Hep1, 40 hours for MHCC-97H, 24 hours for SMMC7721 and 30 hours for Huh7) of incubation at 37 °C, invaded cells were fixed and stained with a dye solution containing 0.1% crystal violet and 20% methanol. Cells adhering to the lower side of the inserts were counted and imaged through an IX71 inverted microscope (Olympus). Five random microscopic fields were counted per well for each group, and the experiments were repeated at least three times independently.
Statistic Analysis. Differences among variables were assessed by χ 2 analysis or two-tailed Student t test. Kaplan-Meier analysis was used to assess survival. Log-rank tests were used to compare survival of patients between subgroups. Multivariate analyses were performed by multivariate Cox proportional hazard regression model. The experiments were performed in triplicates and data were presented as mean ± SEM. Differences were considered to be statistically significant for p < 0.05.

Results
Up-regulation of KMO in HCC tissues. KMO expression was first analyzed in 10 matched pairs of HCC and adjacent non-tumorous liver tissue by Western blotting. As shown in Fig. 1A, in most cases, KMO expression in HCC tissue was obviously higher than in adjacent non-tumorous liver tissue of the same HCC patient. We next performed IHC analysis for KMO using a tissue microarray as a test cohort, which contained 50 paired HCC tissue samples. Immunohistochemical results showed the staining density of KMO in HCC group was obviously stronger than that in adjacent non-tumorous liver tissue group ( Fig. 1B,C, p < 0.05). We further analyzed KMO expression in another independent validation cohort of 70 HCC patients by IHC. Similarly, KMO expression was significantly increased in HCC group compared with adjacent non-tumorous liver tissue group (Fig. 1D,E, p < 0.05). We also performed IHC analysis for KMO using a tissue microarray, which contained 43 tumor tissues from HCC patients and 43 liver tissues from liver cirrhosis patients. Our data showed that KMO expression exhibited a progressive increase from liver cirrhosis to HCC ( Fig. 2A   expression. These results indicated that patients with high KMO expression had much shorter OS times (Fig. 3A, p = 0.0005) and a higher tendency of disease recurrence (Fig. 3B, p = 0.0034). Moreover, we also performed IHC analysis for tryptophan 2,3-dioxygenase (TDO), which is the main KP enzyme in liver cells, using a tissue microarray with 182 HCC patients. Kaplan-Meier OS and TTR analyses were conducted and results showed that TDO had no prognostic value for postsurgical HCC patients ( Supplementary Fig. 2).

Correlation of KMO expression with prognosis of HCC patients.
Additionally, prognostic value of KMO was further confirmed by stratified OS and TTR analyses. Results showed that high expression of KMO was closely connected with OS and TTR after surgical resection in subgroups including TNM stage (TNM stage I , Fig. 4A,B; TNM stage II-III, Fig. 4C,D), tumor number (tumor number = 1, Fig. 4E,F), tumor size (tumor size >3 cm, Fig. 4G,H), and AFP concentration (AFP ≤ 20 ng/ml, Fig. 4I,J).
Univariate and multivariate analyses of prognostic variables in HCC patients. We next evaluated prognostic significance of KMO and other clinicopathologic parameters in HCC using univariate analysis. As shown in Table 2, KMO as well as TNM stage, tumor number, tumor differentiation, and microvascular invasion, was responsible for the OS and TTR of HCC patients.

Inhibition of cell proliferation, migration, and invasion by KMO knockdown.
We also assessed protein level of KMO in a normal liver cell line L02 and seven HCC cell lines, including SMMC7721, Huh7, SK-Hep1, PLC-PRF5, MHCC-97L, MHCC-97H, and HCC-LM3. Results showed that KMO expression was much higher in HCC cell lines, compared with the normal liver cell line L02 (Fig. 5A). Then, KMO was knocked down in SK-Hep1 and MHCC-97H cells, which expressed high levels of KMO, by small interfering RNA (siRNA). The mRNA and protein levels of KMO were effectively down-regulated by two KMO siRNAs at 48 hours of posttransfection ( Supplementary Fig. 4 and Fig. 5B). As indicated by results of CCK8 assays (Fig. 5C), the proliferation of SK-Hep1 cells was significantly inhibited after KMO siRNAs transfection on day 2 (p < 0.05) and day 3 (p < 0.01). Likewise, compared to scramble siRNA transfected cells, proliferation of MHCC-97H cells also significantly inhibited after KMO siRNAs transfection on day 3 (p < 0.05). Besides, effects of KMO on cell migration and invasion were also investigated using transwell migration assays and matrigel invasion assays, respectively. Our results showed that KMO knockdown significantly decreased the migration and invasion rates of both SK-Hep1 and MHCC-97H cells in vitro, respectively (Fig. 5D,E, all p < 0.05).

Enhancement of cell proliferation, migration, and invasion by KMO over-expression.
To further confirm the effects of KMO on proliferation, migration, and invasion, SMMC7721 and Huh7 cells were transfected with a vector constitutively expressing KMO (pCMV6-XL5/KMO) and a empty control vector (pCMV6-XL5/Vector), respectively. The mRNA and protein levels of KMO were significantly increased in pCMV6-XL5/KMO cells compared with pCMV6-XL5/Vector (Supplementary Fig. 6 and Fig. 6A). As indicated by results of CCK8 assays (Fig. 6B), the proliferation of SMMC7721 cells was significantly increased after KMO over-expression on day 3 (p < 0.05). Likewise, compared to empty control vector transfected cells, proliferation of Huh7 cells significantly increased after KMO over-expression on day 2 (p < 0.05) and day 3 (p < 0.01). Furthermore, KMO over-expression also significantly increased the migration and invasion rates of both SMMC7721 and Huh7 cells in vitro, respectively (Fig. 6C,D, all p < 0.01).

Discussion
Despite substantial advances in surgery and chemotherapy for HCC in the past time 3,19,20 . therapeutic failure and mortality are still very common. The current pathological TNM (pTNM) stage and histological grading systems are established and can indicate HCC prognosis to a certain extent. However, due to tumor heterogeneity and accumulation of genetic and epigenetic alterations, patients with the same pTNM stage and/or histological grade of HCC often demonstrate considerable variability in tumor recurrence and metastasis 21 . Thus, searching for valuable diagnostic and prognostic predictors that can effectively distinguish between patients with favorable or unfavorable prognoses in the same stage and/ or grade are urgently needed. Although previous studies have suggested several molecular biomarkers of HCC [22][23][24][25][26] , the novel biomarkers that can identify tumor recurrence and aid risk assessment remain substantially limited.
In the present study, we provide the first evidence that increased expression of KMO is correlated with an unfavorable clinical outcome of HCC patients. IHC staining in two independent cohorts (test cohort: 50 paired; validation cohort: 70 paired) indicated that KMO was abnormally elevated in HCC specimen compared with adjacent non-tumorous liver tissue. In a cohort of 205 HCC patients, Kaplan-Meier OS and TTR analyses showed that high KMO expression was associated with short HCC recurrence and poor prognosis after surgical resection. Multiple Cox regression analysis further conformed that KMO was an independent prognostic marker for OS and TTR. In addition, our data also showed a progressive increase of KMO from liver cirrhosis to HCC. These findings implicate that up-regulation of KMO may be a common feature in HCC and it can serve as an independent prognostic marker to identify patients with poor clinical outcome. Identification of novel and specific prognostic biomarker for patients with early stage HCC is remarkably important 27 . Although AFP is most commonly employed and currently available as serological markers of HCC for surveillance, diagnosis and patient outcome prediction [28][29][30] , HCC patients without AFP elevation (AFP < 20 ng/ml) are missed and subsequently progress to late stage HCC before becoming clinically symptomatic and detectable 28,30 . Moreover, the diagnostic and prognostic sensitivity of AFP was poor in the early stage of HCC, especially when used alone. Here, we evaluated the clinical usefulness of KMO as a prognostic factor in HCC patients with AFP < 20 ng/ml, which accounted for 36.1% of HCC patients in our study cohort. In addition, our results also showed that high expression of KMO in HCC patients at TNM I stage was closely associated with the risk of recurrence and shorter survival time (OS, p = 0.0042; TTR, p = 0.0500). Thus, our results have exhibited the potential value of KMO in predicting the risk of recurrence and patient survival in subgroups with normal AFP levels (AFP < 20 ng/ml) or in the early-stage HCC group, which would have been difficult using currently clinically available surrogate biomarkers. Molecular mechanisms underlying recurrence or metastasis of HCC are very complicated. Genetically, chromosomal copy number alterations, such as the loss of alleles on 16q and the amplification of 1q, have been found to associate with HCC metastasis 31,32 . Recently, multiple genes associated with HCC progression and metastasis have also been identified, including PGE2 33 , Keratin 19 34 , FoxQ1 35 , ICAT 36 , and Gankyrin 37 . KMO locates in a pivotal position in KP, which is responsible for >95% of tryptophan degradation in mammals, ultimately leading to the formation of NAD 38,39 . NAD is a key regulator of several energy and signal transduction processes such as transcription, DNA repair, cell cycle progression,   apoptosis and metabolism, which are relevant to tumor cell survival and proliferation. It functions as a coenzyme in metabolic pathways including glycolysis and the pentose phosphate pathway, thereby allowing the efficient production of NADPH, ribose-5-phosphate (Rib-5-P) and biosynthetic compounds used by the tumor for growth and angiogenesis 40 . Moreover, aberrant NAD metabolism is considered a hallmark of cancer: the high rates of aerobic glycolysis (Warburg effect) perturbs NAD metabolism, thereby leading to disrupt the cellular NADH/NAD + redox homeostasis, and promote cancer progression 41,42 .
Our data first exhibited that KMO expression was significantly upregulated in HCC tissues. As a pivotal enzyme in the kynurenine pathway, increased KMO expression might affect NAD concentration in HCC and then involved in HCC progression. Previous studies showed that modulation of KMO activity was mainly involved in several neurodegenerative diseases, including Huntington's disease and Alzheimer's disease 6,43 . However, roles of KMO in tumor, including HCC, remain hitherto unknown. It has been previously reported that selective degradation of tryptophan created an immunosuppressive micromilieu of tumor both by depleting tryptophan and by accumulating immunosuppressive metabolites of the kynurenine pathway: On the one hand, tryptophan shortage inhibited T cells proliferation and causes a lack of accumulation of specific T cells at the tumor site 44,45 . On the other hand, the main tryptophan metabolites, such as Kynurenine, 3-hydroxykynurenine, and 3-hydroxyanthranilic acid, could suppress the T cell response and kill T cells, B cells and natural killer (NK) cells 46,47 . In this study, we found that KMO expression positively regulated proliferation, migration and invasion of HCC cells in vitro. Although underlying mechanism remains to be further investigated, our present study provides a clue for biological function of KMO in HCC. To the best of our knowledge, this is the first study to report that high KMO expression is correlated with aggressive malignant phenotype of HCC. Our data indicate that KMO may serve as a novel prognostic marker for HCC and targeting KMO may provide a promising strategy for HCC treatment.