Human bone marrow-derived mesenchymal stem cells promote the growth and drug-resistance of diffuse large B-cell lymphoma by secreting IL-6 and elevating IL-17A levels

Reagents

Cell Counting Kit-8 (CCK-8) was purchased from Dojindo Laboratories (Kumamoto, Japan). Human recombinant IL-6 and IL-17A, and human neutralizing antibodies to IL-6 (aIL-6) and IL-17A (aIL-17A) were obtained from R&D Systems (Minneapolis, MN, USA). The monoclonal mouse anti-human IL-6 antibody (ab9324) was purchased from Abcam (Shanghai, China). Enzyme-linked immunosorbent assay (ELISA) kits for IL-17A, IL-6, IL-10, IL-1β, PGE₂, and TGF-β; fluorescence-activated cell sorting (FACS) human antibodies including anti-CD4-FITC, anti-IL-17A-APC, anti-Foxp3-PE, anti-CD25-APC, and their corresponding anti-mouse IgG1 K-PE/APC; and the Annexin V apoptosis detection kit were all purchased from eBioscience (San Diego, CA, USA). Antibodies against JAK2, p-JAK2, STAT3, p-STAT3, p-AKT, AKT, cyclin D2, P27, and GAPDH were purchased from Santa Cruz Biotech (Santa Cruz, CA, USA). Real-time reverse transcription-polymerase chain reaction (RT-PCR) reagents were obtained from Takara (Beijing, China).Rituximab was purchased from Novartis (Basel, Switzerland). Doxorubicin and Ara-C were obtained from Pfizer (Shanghai, China).

Human samples and cell lines

In vitro cell culture

Four in vitro cell culture experiments were performed. All experiments were performed using SU-DHL-2/4 cells and PBMCs at 1 × 10⁶ cells/well.

Experiment 1: SU-DHL-2/4 cells were cultured with or without hBMSCs at different ratios of 1:1, 1:5, and 1:0.2 in 6-well plates with or without Transwell inserts (CORNING, New York, USA), and then incubated for 72 h. The proliferation of SU-DHL-2/4 cells was analyzed using the CCK-8 assay, and the supernatants were detected by ELISA.

Experiment 2: SU-DHL-2/4 cells were cultured with or without PBMCs and/or hBMSCs at the ratio of 1:1:1 in 6-well plates with Transwell inserts. SU-DHL-2/4 cells were placed in the inserts and the other cells were seeded in the outer wells. All groups were incubated for 72 h and then detected using the CCK-8 assay, colony formation assay, and ELISA.

Experiment 3: Eight groups of cells were cultured with or without Transwell inserts for 72 h as follows: PBMCs alone; PBMCs and hBMSCs; PBMCs and SU-DHL-2 cells; PBMCs, SU-DHL-2 cells, and hBMSCs; PBMCs and SU-DHL-4 cells; PBMCs, SU-DHL-4 cells, and hBMSCs; SU-DHL-2 cells and hBMSCs; SU-DHL-4 cells and hBMSCs. The culture ratio was 1:1:1. After incubation, PBMCs were detected by FACS and qPCR, and the supernatants were detected by ELISA.

Experiment 4: SU-DHL-2/4 cells were co-cultured with or without hBMSCs (1:1), IL-6 (0.5 ng/ml), aIL-6 (50 μg/ml), IL-17A (0.1 ng/ml) and/or aIL-17A (10 μg/ml) for 72 h, and rituximab (10 μg/ml), doxorubicin (2 μM) and Ara-C (2 μM) were added 24 h before detection to induce apoptosis. The SU-DHL-2/4 cells were then collected and analyzed using CCK-8 and apoptosis assays.

Cell viability and colony formation assay

The viability of SU-DHL-2/4 cells was assessed using the CCK-8 assay. Briefly, each sample was allocated in 96-well plates, and CCK-8 was added 4 h before the end of the culture time. The absorbance (optical density, OD) at 450 nm was measured at different time points using a Universal Microplate Spectrophotometer (Thermo Fisher Scientific, Inc., Waltham, MA, USA). The soft agar colony formation assay (CFA) was performed using 24-well plates according to a previous study [22]. The cells were cultured in a cell incubator for 2 to 3 weeks, and then fixed with 10% paraformaldehyde and stained with 1% crystal violet in 70% ethanol. Colonies (with minimal 50 cells) were counted using a stereomicroscope.

Immunohistochemistry

The expression of IL-6 in the 48 human tumor tissues, 18 human benign tissues, and 24 mice tumor tissues was detected by immunohistochemistry (IHC). Tissues were fixed in 10% neutral formaldehyde, embedded in paraffin, sliced and stained with hematoxylin and eosin. Briefly, the paraffin-embedded tissues were serially cut into 4 μm sections, dewaxed, and rehydrated. Sections were then blocked with peroxide and non-immune animal serum and incubated sequentially with primary antibody, biotin-labeled secondary antibody, and streptomycin anti-biotin peroxidase. Finally, the sections were stained with di-n-butyl adipate, counterstained with hematoxylin, dehydrated, cleared in xylene, and fixed. The expression of IL-6 in paraffin sections was quantified by relative integrated optical density (IOD). At least three different images of each paraffin section were acquired at × 40 high power microscopic fields, and the IOD of images and mean IOD of each paraffin section were calculated by experienced pathologists using Image-Pro-Plus 6.0 software (Media Cybernetics, Rockville, MD, USA). The paraffin section with the lowest mean IOD was used as the negative control, and the relative IOD was calculated as the mean IOD/negative control.

qRT-PCR

Western blot analysis

SU-DHL-2/4 cells or tumor tissues were lysed using sodium dodecyl sulfate buffer containing proteinase inhibitors (Roche). Equal amounts of protein (50 μg) were separated by 10% sodium dodecyl sulfate-polyacrylamide gel electrophoresis and transferred onto polyvinylidene difluoride membranes (Bio-Rad, Shanghai, China). The membranes were blocked and incubated with specific antibodies overnight at 4 °C. The antibodies used were JAK2, p-JAK2, STAT3, p-STAT3, p-Akt, Akt, cyclin D2, P27, IL-6, and GAPDH. The membranes were then incubated with horseradish peroxidase-labeled secondary antibody (Santa Cruz Biotechnology). The protein bands were visualized using an enhanced chemiluminescence reagent.

Flow cytometry

The frequencies of Th17 cells and Treg cells were detected by flow cytometry. Cell density was adjusted to 2 × 10⁶/ml. For Th17 cells, cells were stimulated by the addition of 50 ng/ml phorbol myristate acetate, 1 μg/ml ionomycin, and 10 μg/ml Brefeldin A to the medium for 5 h at 37 °C and 5% CO₂. The cells were then stained with anti-CD4-FITC and anti-IL-17A-APC antibodies for 1 h at room temperature. For Treg cells, cells were stained with anti-CD4-FITC, anti-Foxp3-PE, and anti-CD25-APC antibodies for 1 h at room temperature. The cells were analyzed with a flow cytometer (FACS Canto II; BD Bioscience) and the data were analyzed with FlowJo 10 software. CD4 + IL-17A+ cells were defined as Th17 cells, and CD4 + CD25 + Foxp3+ cells were defined as Treg cells.

Elisa

Cell culture supernatants were collected at the indicated times, and then supernatants were assessed for IL-17A, IL-6, IL-10, IL-1β, PGE_2, and TGF-β by ELISA, following the manufacturer’s instructions.

Apoptosis assay

The Annexin V/propidium iodide assay was used to detect DLBCL cell apoptosis. Cells were seeded in 48-well plates and incubated with or without chemotherapy drugs for 72 h. After culture, cells were washed twice with cold PBS and resuspended in binding buffer at a concentration of 1 × 10⁶ cells/ml after which 100 μL of the solution was transferred to a 5-mL tube, and 5 μL of Annexin V–FITC and 5 μL of propidium iodide were added. The tube was gently vortexed and incubated for 15 min at room temperature in the dark. At the end of incubation, 300 μL of binding buffer was added. Flow cytometric analysis was performed immediately with a flow cytometer (FACS Canto II; BD Bioscience).

Human expression microarray analysis and bioinformatics analysis

SU-DHL-2 cells (2 × 10⁶ cells/ml) co-cultured with or without IL-17A (200 pg/ml) for 72 h were harvested, and the total RNA of each sample was extracted using the TRIzol reagent (Takara).The Whole Human Genome Oligo Microarray (4 × 44 K, Agilent Technologies) was generated by KangChen Biotechnology. Briefly, total RNA from each sample was linearly amplified and labeled with Cy3-UTP. The Labeled cRNAs were purified using the RNeasy Mini Kit (Qiagen, Hilden, Germany). The concentration and specific activity of the labeled cRNAs (pmol Cy3/μg cRNA) were measured by NanoDrop ND-1000 (Thermo). Aliquots of 1 μg of each labeled cRNA were fragmented by adding 11 μl of 10 × Blocking Agent and 2.2 μl of 25 × Fragmentation Buffer, then heated at 60 °C for 30 min, and finally 55 μl of 2 × GE Hybridization buffer was added to dilute the labeled cRNA. One-hundred microliters of hybridization solution were dispensed into the gasket slide and assembled to the gene expression microarray slide. The slides were incubated for 17 h at 65 °C in an Agilent Hybridization Oven. The hybridized arrays were washed, fixed, and scanned using the Agilent DNA Microarray Scanner (part number G2505C). Array images were analyzed using Agilent Feature Extraction software (version 11.0.1.1). Quantile normalization and subsequent data processing were performed with the GeneSpring GX v12.1 software package (Agilent Technologies). After quantile normalization of the raw data, genes for which at least 3 out of 6 samples had flags in Detected (“All Targets Value”) were chosen for further data analysis. Differentially expressed genes (DEGs) with statistical significance between the two groups were identified through Volcano Plot filtering. DEGs between the two samples were identified through Fold Change filtering (fold change ≥1.5, P < 0.05). Hierarchical clustering and heat map were performed using the R scripts. The present microarray data is accessible through Gene Expression Omnibus, series accession number GSE115855 (https://www.ncbi.nlm.nih.gov/geo/query/acc.cgi?acc=GSE115855). Gene ontology (GO) analysis and pathway analysis were based on the GO database (http://www.geneontology.org) and KEGG database (http://www.kegg.jp). Geneco-expression networks, in which the core genes that play a vital role in the network were identified, were constructed based on the STRING database (https://string-db.org/) and further analyzed by Cytoscape software (version 3.6.0). In the networks, nodes were mainly genes and edges represented relation types between the nodes. Node size represented the degree of expression of each gene: green and red nodes represented down- and up-regulated genes, and edge size represented the combined score.

Xenograft mouse model

Male nude mice (BALB/c, 20–30 g) (Guangdong Laboratory Animal Center, China) 6–8 weeks of age were used in experiments. Mice were housed and monitored in micro-isolator cages under clean conditions. All experimental procedures and protocols were approved by the Institutional Animal Care and Use Committee at Guangzhou First People’s Hospital. SU-DHL-4 cells (1 × 10⁷cells/100 μL PBS) were subcutaneously inoculated into the right flanks of the nude mice. After the development of 3 mm diameter tumors, mice were randomly assigned to three groups (n = 8 per group). The MSC group was injected with circumtumoral hBMSCs (5 × 10⁵ cells/100 μL PBS per time); the IL-6 group was injected with circumtumoral IL-6 (10 μg/kg per time); and the Control group was injected with circumtumoral PBS (100 μL per time). All groups were injected five times every 3 days (on days 0, 3, 6, 9, and 12). The size of tumors was measured every 3 days, up to 28 days following hBMSC, IL-6, and PBS injection. Tumor volume was calculated according to the following formula: tumor volume (mm³) = (a² × b)/2, where “a” represents the length of the short axis and “b” represents the length of the long axis. The mice were sacrificed 28 days after hBMSC, IL-6, and PBS injection, and the tumors were removed and prepared for IHC, qPCR, and western blot analysis.

Statistical analysis

For microarray analysis, DEGs were confirmed using a p-value threshold and false discovery rate (FDR) analysis. The threshold for truly significant genes was determined at a p-value of < 0.05 and FDR value < 0.05. All analyses were performed using SPSS 17.0. Numerical data were presented as the mean ± standard deviation. Two-tailed independent-sample Student’s t/t′ tests were used for comparisons between two groups. Single-factor analysis of variance (one-way ANOVA) and Student–Newman–Keuls/Dunnett’s T3 tests (between every two groups) were used for comparisons among multiple groups. The Chi-square test was used for enumeration data comparisons between two groups. A p-value of 0.05 was considered significant.

hBMSCs promote the growth of DLBCL cells in vitro, and PBMCs enhance these effects

To investigate whether hBMSCs promote the growth of DLBCL cells in vitro, hBMSCs were co-cultured with SU-DHL-2 or SU-DHL-4 cells at different ratios in a direct or indirect system for 72 h. Then, SU-DHL-2 and SU-DHL-4 cell proliferation was detected using the CCK-8 assay at different time points. HBMSCs significantly increased the proliferation of SU-DHL-4 cells at ratios of SU-DHL-4:hBMSC of 1:1, 1:5, or 1:0.2 in a time-dependent manner in the direct (Fig. 1a–c) or indirect (Fig. 1d–f) co-culture system. Similarly, hBMSCs markedly promoted the proliferation of SU-DHL-2 cells at SU-DHL-2:hBMSC ratios of 1:1 and 1:5 directly (Fig. 1a and b) or indirectly (Fig. 1d and e). However, hBMSCs had no effect at a ratio of 1:0.2 (Fig. 1c and f). These results suggested that hBMSCs promoted the proliferation of DLBCL cells by secreting soluble cytokines instead of cell-to-cell contact.

Open image in new window

In addition to MSCs, many immune cells that infiltrate the TME of DLBCL can affect the growth of tumor cells and the promoting effects of hBMSCs. To simulate the TME of DLBCL, SU-DHL-2 and SU-DHL-4 cells were co-cultured with hBMSCs and PBMCs at a ratio of 1:1:1 for 72 h, and the cells were separated using Transwell inserts. The proliferation and clonogenicity of DLBCL cell lines were tested using the CCK-8 assay and CFA, respectively.. PBMCs significantly increased the effect of hBMSCs on promoting the proliferation of SU-DHL-2 and SU-DHL-4 cells (Fig. 1g and h). Furthermore, hBMSCs accelerated the clonogenicity of SU-DHL-2 and SU-DHL-4 cells, and PBMCs increased these effects (Fig. 1i). According to these data, we hypothesized that hBMSCs might promote the growth of DLBCL cells not only directly through the secretion of paracrine soluble cytokines, but also by affecting PBMC differentiation, as differentiated PBMCs secreted cytokines that enhanced these promoting effects.

hBMSCs secrete IL-6 into the co-culture supernatants, and the IL-6 level is higher in DLBCL tumor tissue than in benign tissue

The results of the present study showed that hBMSCs promoted the growth of DLBCL cells by secreting soluble cytokines; therefore, we attempted to identify these cytokines. The results showed that hBMSCs strongly secreted IL-6 into the co-culture supernatant in a time-dependent manner when co-cultured with or without SU-DHL-2 and SU-DHL-4 cells (Fig. 2a and b).In addition, hBMSCs markedly secreted IL-6 at different ratios in a direct or indirect co-culture system (Fig. 2c and d). The level of PGE2 was decreased in the hBMSC groups at different ratios (Fig. 2e and f). The levels of IL-10, IL-17A, TGF-β, and IL-1-β did not differ between the two groups (Fig. 2g–n). To further verify the IL-6 level in the TME of DLBCL, the relative expression ofIL-6 was examined in tumor tissues and benign tissues by IHC. IL-6 levels were markedly higher in tumor tissues than in benign tissues (Fig. 2o and p). These results indicated that hBMSCs secreted IL-6 into the TME of DLBCL, thereby promoting the growth of DLBCL.

Open image in new window

hBMSCs promote DLBCL growth by secreting IL-6 in vivo

As hBMSCs promoted the growth of DLBCL cells in vitro, we investigated whether hBMSCs have a similar effect in vivo. Because SU-DHL-4 cells showed a high rate of proliferation (Fig. 1c-h), this cell line was used to establish DLBCL xenograft BALB/c nude mouse models. Tumors developed well in nude mice, and tumor tissues were obtained by operation at 28 days after hBMSC, IL-6, or PBS injection (Fig. 3a). The tumor volumes were significantly larger in the MSC and IL-6 groups than in the Control group (Fig. 3b). In addition, IL-6 mRNA and protein levels were markedly higher in the MSC and IL-6 groups than in the Control group (Fig. 3c and d). IHC staining detected higher IL-6 levels in the tumor tissues of the MSC and IL-6 groups than in controls (Fig. 3e). These results suggested that hBMSCs promoted DLBCL growth by secreting IL-6 in vivo.

Open image in new window

hBMSCs induce PBMCs differentiation into Th17 and Treg cells, thereby increasing IL-17A and TGF-β levels in the co-culture supernatants

As PBMCs enhanced the promoting effects of hBMSCs, we next explored the underlying mechanisms. For this purpose, eight groups of co-cultures were generated using hBMSC, PBMC, or DLBCL cell lines for 72 h. The cells were separated using Transwell inserts, and the PBMCs from six groups were detected by FACS and qPCR, and the supernatants from eight groups were detected by ELISA. Comparison of three pairs of groups in the presence or absence of hBMSCs showed that hBMSCs significantly increased the frequencies of Th17 (Fig. 4a and b) and Treg (Fig. 4c and d) cells in PBMCs in each pair of groups. At the transcriptional level, hBMSCs upregulated the relative expressions of RORγt and Foxp3 mRNA in PBMCs when co-cultured with PBMCs or DLBCL cell lines (Fig. 4e–g). In addition, hBMSCs upregulated the relative expression of IL-17A and TGF-β mRNA of PBMCs (Fig. 4e–g). Similarly, hBMSCs increased the levels of IL-17A and TGF-β proteins in the supernatants of co-cultures (Fig. 4h and i). IL-6 levels were significantly higher in the groups that included hBMSCs (Fig. 4k), suggesting that hBMSC-secreted IL-6 in the supernatants induced PBMC differentiation into Th17 cells. In addition, high TGF-β levels induced Treg cell differentiation. However, hBMSCs had no consistent effects on the levels of PGE2, IL-10, and IL-1β (Fig. 4j, l, and m).

Open image in new window

hBMSCs or IL-6 promote the growth of DLBCL cells by protecting them from spontaneous or drug-induced apoptosis, and IL-17A reinforces these effects

Published data indicate that IL-6 or IL-17A acts as a pro-tumor factor in MCL or DLBCL[12, 16]. Based on these data and the present results, we hypothesized that hBMSCs secrete IL-6 into the TME of DLBCL, and IL-6 directly promotes DLBCL growth and induces Th17 cells to secrete IL-17A, which increases this promoting effect. To verify this hypothesis, hBMSCs or exogenous IL-6 and/or IL-17A were co-cultured with SU-DHL-4 cells to induce proliferation, and then exogenous aIL-6 or aIL-17A were used to block these effects. hBMSCs and exogenous IL-6 and IL-17A promoted SU-DHL-4 cell proliferation, and aIL-6 or aIL-17A abolished these effects (Fig. 5a). IL-17A increased the promoting effects of IL-6, whereas aIL-6 combined with aIL-17A abolished these effects (Fig. 5a).

Open image in new window

Next, we examined the effect of exogenous IL-6 and/or IL-17A on spontaneous or drug-induced apoptosis in DLBCL cells. Exogenous IL-6 or IL-17A were added at increasing concentrations into SU-DHL-2 or SU-DHL-4 cell co-cultures for 48 h, followed by rituximab treatment for 24 h before the detection of apoptosis by FACS. The results showed that IL-6 or IL-17A markedly protected SU-DHL-2 or SU-DHL-4 cells from rituximab-induced apoptosis in a concentration-dependent manner (Fig. 5b and c). Based on these data, we determined the most appropriate concentrations of IL-6 and IL-17A for the next anti-apoptosis experiments as 500 pg/ml and 50 pg/ml, respectively. Then, the effects of IL-6 and IL-17A on protecting SU-DHL-2 cells against rituximab-, doxorubicin-, and Ara-C-induced apoptosis were examined. As shown in Fig. 5d, IL-6 significantly decreased the percentage of apoptotic cells in the three groups, whereas IL-17A further enhanced these effects.

Finally, to further confirm the effects of IL-6 and/or IL-17A on DLBCL cell apoptosis, IL-6 and/or IL-17A neutralizing antibodies were used to block IL-6 and IL-17A mediated signaling. The results showed that hBMSCs, IL-6, or IL-17A decreased spontaneous or rituximab-induced apoptosis in SU-DHL-4 cells, and aIL-6 or aIL-17A abolished these effects (Fig. 5e and f). In addition, IL-17A could reinforce the protective effects of IL-6, and aIL-6 with aIL-17A abolished these effects (Fig. 5e and f). Figure 5g shows representative FACS plots of spontaneous or rituximab-induced apoptosis of SU-DHL-4 cells.

IL-6 or hBMSCs promote DLBCL cell growth via the JAK2/STAT3 signaling pathway

To investigate the signaling pathways involved in the effects of hBMSCs or IL-6, JAK2 and STAT3 phosphorylation was assessed in DLBCL cells. As shown in Fig. 6a and b, IL-6 significantly promoted the phosphorylation of JAK2 and STAT3 in both SU-DHL-2 and SU-DHL-4 cells, whereas IL-6-neutralizing antibodies abrogated IL-6-induced phosphorylation of JAK2 and STAT3. Similarly, hBMSCs upregulated p-JAK2 and p-STAT3 in SU-DHL-4 cells, whereas IL-6-neutralizing antibodies abolished these effects (Fig. 6c). Taken together, these results supported that IL-6 or hBMSCs promoted DLBCL cell growth via the JAK2/STAT3 signaling pathway.

Open image in new window

IL-17A promotes DLBCL cell growth by upregulating cyclin D2 via the PI3K/Akt signaling pathway

To further examine the molecular mechanisms underlying IL-17A-mediated DLBCL cell growth, microarray and bioinformatics analyses were performed together with qPCR and western blotting to assess the expression of related genes and proteins. Three samples of SU-DHL-2 cells were co-cultured with IL-17A for 72 h and an additional three samples were cultured without IL-17A as controls. The six samples of SU-DHL-2 cells were subjected to microarray analysis.

As shown in Figs. 7, 288 DEGs were identified (139 upregulated and 149 downregulated), and the heat map is shown in Fig. 7a. The heat map of the top 25 most significant up- and down-regulated DEGs ranked by fold change is shown in Fig. 7b. The top 10 most enriched biological process GO terms of all upregulated DEGs were positive regulation of cyclin-dependent protein serine/threonine kinase activity, positive regulation of cyclin-dependent protein kinase activity, regulation of cyclin-dependent protein serine/threonine kinase activity, and other terms (Fig. 7c). Three of these terms were related to cell cycle regulation, and the involved upregulated DEGs were PKD1, CDKN1B, and CCND2. Based on KEGG database pathway analysis, the most enriched pathways of downregulated DEGs included the Ras signaling pathway and p53 signaling pathway (Fig. 7d). The most enriched pathways of upregulated DEGs included the PI3K/Akt signaling pathway, Rap1 signaling pathway, prostate cancer, and microRNAs in cancer (Fig. 7e). These pathways were all connected to tumor growth, and the PI3K/Akt signaling pathway (eight involved genes: CCND2, CDKN1B, CSF1, EFNA3, FGFR2, FGFR3, IL2RB, and ITGA9) ranked first among upregulated pathways (Enrichment Score = 3.159187). The STRING database and Cytoscape 3.6.0 software were used to generate co-expression networks of all interacting DEGs to identify the genes that may play important roles in IL-17A-mediated SU-DHL-2 cell growth (Fig. 7f and g). Node size represented the degree of expression of each gene; green or red nodes represented down- or up-regulated genes, respectively, and edge size represented the combined score. We found that the eight genes involved in the PI3K/Akt signaling pathway graded high in the gene co-expression networks (Table 3), suggesting that the PI3K/Akt signaling pathway plays an important role in IL-17A-mediated SU-DHL-2 cell growth.

Open image in new window

Table 3

40 highest degree nodes in the DEGs co-expression networks

Nodes	Degree	Up or Down
GNG11	7	down
FGFR3	5	up
FGFR2	5	up
NPY	5	down
ACTN1	5	down
SERPINE1	5	down
ALDH7A1	4	down
TAS2R31	4	down
FCER1G	4	down
CXCR1	4	up
DLG4	4	up
HCAR2	4	up
CLU	4	down
MMRN1	4	down
IL2RB	3	up
CSF1	3	up
GZMA	3	up
CACNG4	3	down
CBLB	3	up
CRHR2	3	down
EFNA2	3	down
KIF19	3	down
CDKN1B	3	up
C1QA	3	down
APCS	2	up
SH3PXD2A	2	up
TXNIP	2	up
EMCN	2	down
ELTD1	2	down
DHRS3	2	down
CTAG2	2	down
GZMK	2	up
GRIN3B	2	up
SOX17	2	down
TAAR8	2	down
HDAC11	2	up
ZNF217	2	up
RPS4Y2	2	down
CCND2	2	up
C1QB	2	up

Finally, the mRNA expression of the eight genes was measured by qPCR, and PI3K/Akt pathway-related protein expression was assessed by western blotting. Consistent with the microarray results, the relative mRNA expressions of CSF1, CCND2, EFNA3, FGFR2, FGFR3, IL2RB, and ITGA9 were significantly higher in IL-17A-treated SU-DHL-2 cells (Fig. 7h). Cyclin D2 is encoded by the CCND2 gene, which is a downstream molecule of the PI3K/Akt pathway and regulates the cell cycle. IL-17A upregulated p-Akt and cyclin D2 expression in SU-DHL-2 cells, whereas aIL-17A abrogated these effects (Fig. 7i). However, IL-17A and aIL-17A had no effect on P27 expression in SU-DHL-2 cells. In summary, the results indicated that IL-17A promoted the growth of SU-DHL-2 cells by upregulating cyclin D2 via the activation of PI3K/Akt signaling.