The Traf2 and Nck-Interacting Kinase Is a Conserved Regulator of Self-Renewal in Hematopoietic and Leukemia Stem Cells

The Traf2 and Nck-interacting kinase (TNIK) is a member of the germinal center kinase (GCK) family. We have recently found that signaling through the tumor necrosis factor superfamily receptor CD27 activates Wnt target genes through Traf2-TNIK-β-catenin signaling, promoting leukemia development and progression in acute and chronic myeloid leukemia (AML and CML).

To explore how TNIK signaling regulates hematopoietic and leukemia stem cells (HSCs and LSCs), we generated a constitutive TNIK^-/- knockout mouse.

Under steady-state conditions, total bone marrow (BM) cellularity of naïve TNIK^-/- mice was slightly reduced (mean BM count (x10⁸): TNIK^-/- = 1.42; TNIK^wt/wt = 2.15, p < 0.0132). In contrast, no differences in absolute number of lineage negative (Lin^-) cells, as well as the more primitive subset Lin^-; Sca1⁺; c-kit⁺ (LSK) cells could be observed. Similarly, proportions of long-term HSCs (LT-HSCs), short-term HSCs (ST-HSCs) and multipotent progenitors (MPPs) did not differ between TNIK^-/- and TNIK^wt/wt mice. Functionally, naïve TNIK^-/- and TNIK^wt/wt HSCs displayed in vitro a comparable serial colony formation capacity in methylcellulose. Altogether these data suggest that TNIK does not affect the maintenance and self-renewal of HSCs under steady-state conditions.

Next the impact of TNIK on hematopoietic recovery after genotoxic stress was analyzed in vivo. TNIK^-/- and TNIK^wt/wt mice were injected i.p. with 5-Fluorouracil (5-FU) (150^mg/_g bodyweight) and BM cellularity as well as HSC phenotype and function were assessed 9 days post injection. The total BM cellularity was significantly higher in TNIK^-/- mice compared to controls (mean BM count (x10⁶/_ml): TNIK^-/- =10.25; TNIK^wt/wt =4.55, p< 0.0382). Furthermore, LSKs were significantly reduced in frequency in TNIK^-/- mice (% LSK of whole BM: TNIK^-/- = 2.93; TNIK^wt/wt = 7.073, p < 0.0009) but not in absolute numbers. Interestingly, however, the total number LT-HSCs was significantly reduced in the BM of TNIK^-/- mice (mean LT-HSC count (x10³): TNIK^-/- = 5.628; TNIK^wt/wt = 16.725, p < 0.0101). These results were functionally confirmed in vivo by competitive BM transplantation (BMTx) of whole BM from 5-FU treated mice into lethally irradiated secondary recipients. Mice receiving BM from 5-FU-treated TNIK^-/- mice showed a significantly reduced peripheral engraftment capacity 6 weeks post BMTx compared to controls (mean peripheral blood chimerism: TNIK^-/- = 64.6%; TNIK^wt/wt = 35.4%, p < 0.0001), suggesting that TNIK signaling mediates HSC maintenance during genotoxic stress conditions in vivo.

In parallel the role of TNIK signaling was analyzed in myeloid LSCs. CML-like disease was induced in mice by transplanting pMSCV-BCR-ABL1-IRES-GFP transduced TNIK^-/- or TNIK^wt/wt LSKs into BL/6 mice. Even though CML developed similarly as indicated by measuring BCR-ABL1-GFP⁺ granulocytes in the peripheral blood, primary TNIK^-/- CML mice survived significantly longer compared to TNIK^wt/wt CML mice (Median survival (days): TNIK^-/- = 33, TNIK^wt/wt = 23.5, p < 0.0216) with 2 out of 7 mice surviving long-term. Next, we determined the effect of TNIK signaling on the functionality of FACS sorted Lin^-; GFP⁺; Sca1⁺; c-kit⁺ LSC in vitro in serial colony forming assays and in vivo in secondary transplantations. Serial colony formation capacity of TNIK^-/- CML LSCs was significantly reduced after re-plating in vitro (CFUs fold change (TNIK^-/- vs TNIK^wt/wt): 1^st Plating = 0.89, P<0.5149, 2^nd Plating = 0.42, 9 < 0.0002). In line with these findings, only 1 out of 9 mice secondarily transplanted with LSCs from primary TNIK^-/- CML mice developed CML within 90 days, while all 9 recipients receiving TNIK^wt/wt LSCs succumbed to the disease within 23 days after transplantation (p < 0.0001). These results suggest that TNIK signaling also promotes self-renewal of CML LSCs.

To summarize, we have described for the first time TNIK signaling as a commonly conserved regulator in stress-induced hematopoiesis and CML, which promotes stem cell self-renewal and maintenance. As TNIK inhibitors are currently developed for the treatment of different kinds of solid tumors, TNIK may be an attractive therapeutic target for LSCs in myeloid leukemia.