*2.4. MAP1B Promotes the Cell Proliferation, Migration, and Invasion of UC Cell Lines*

To investigate the biological effects of MAP1B, we first characterized endogenous MAP1B expression in eight UC cell lines and noticed RTCC1 and J82 cells had the most abundant MAP1B transcripts and protein expression (Figure 4A). We next successfully knocked down MAP1B in both the RTCC1 (Figure 4B, left) and J82 (Figure 4B, right) cell lines using short hairpin RNA (shRNA). We found significantly attenuated proliferation (viability) in stable MAP1B-silenced RTCC1 (Figure 4C1) and J82 (Figure 4C2) cells. Due to the positive relationship between MAP1B expression and the development of metastasis, we evaluated the effect of MAP1B in UC cell migration and invasion. MAP1B knockdown significantly decreased the migratory and invasive abilities of RTCC1 (Figure 4C3,C5) and J82 (Figure 4C4,C6) cells.

**Figure 4.** *MAP1B* expression promotes the growth of UC cells in vitro. (**A**) As compared with RT4 cells, endogenous *MAP1B* mRNA (upper) and protein (lower) expressions were increased in cells from the J82 and RTCC1 cell lines. (**B**) The two cell lines with high endogenous *MAP1B* expression were stably silenced against *MAP1B* expression by a lentiviral vector bearing one of the two clones of *MAP1B* shRNA with different sequences for both RTCC1 (left panel) and J82 (right panel) cells. Using an ELISA-based colorimetric assay to assess the rate of BrdU uptake, cell proliferation was significantly reduced in stable *MAP1B*-knockdown (**C1**) RTCC1 and (**C2**) J82 cell lines compared with that in the corresponding shLacZ controls. Similar trends were found for cell migration and invasion among cells from the (**C3** and **C5**) RTCC1 and (**C4** and **C6**) J82 cell lines. (\* *p*<0.05). More details of western blot, please view at the supplementary materials.

## *2.5. MAP1B Expression Correlates with Chemoresistance In Vitro and In Vivo*

Flow cytometric analysis of stable *MAP1B* knockdown RTCC1 and J82 cell lines showed stable *MAP1B* knockdown significantly increased the sub-G1 population, indicating induced cell apoptosis (Figures 5 and 6). Further analysis of vinblastine-treated RTCC1 and J82 cell lines also disclosed induced cell apoptosis (Figures 7 and 8). In other words, MAP1B expression might lead to a resistance to anti-mitotic chemotherapeutics. In the independent UBUC patient cohort receiving adjuvant chemotherapy, Kaplan–Meier survival analysis showed high MAP1B expression correlated with inferior DFS (Figure 9), further supporting the role of MAP1B in chemoresistance.

**Figure 5.** Stable *MAP1B* knockdown increases the sub-G1 population with significantly altered cell-cycle progression. Cell-cycle analysis as conducted by flow cytometry identified a remarkable increment of sub-G1 population indicating cell death in *MAP1B*-knockdown RTCC1 (upper panel) and J82 (lower panel) cells.

**Figure 6.** *MAP1B* knockdown induces apoptosis. Flow cytometric analysis of annexin V/propidium iodide-stained RTCC1 (upper panel) and J82 (lower panel) cell lines disclosed *MAP1B* knockdown significantly increased percentage of apoptosis. (\* *p* < 0.05).

**Figure 7.** Stable *MAP1B* knockdown increased vinblastine-induced apoptosis. Flow cytometric analysis of vinblastine-treated RTCC1 (upper panel) and J82 (lower panel) cell lines disclosed that *MAP1B* knockdown significantly increased the sub-G1 population, indicating induced cell apoptosis.

**Figure 8.** Stable *MAP1B* knockdown increase vinblastine-induced apoptosis. Flow cytometric analysis of annexin V/propidium iodide-stained RTCC1 (upper panel) and J82 (lower panel) cell lines demonstrated *MAP1B* knockdown significantly increased percentage of vinblastine-induced apoptosis. (\* *p* < 0.05).

**Figure 9.** Kaplan–Meier survival analysis of *MAP1B* expression for the DFS of the UBUC patient cohort receiving adjuvant chemotherapy. Kaplan–Meier survival analysis showing the prognostic significance of *MAP1B* expression for the DFS of the UBUC patient cohort receiving adjuvant chemotherapy.

#### **3. Discussion**

It is estimated that one-third of patients with UBUC have advanced disease at presentation [16]. A similar poor prognosis was found among patients with advanced UTUC in that the DSS has not changed significantly during the last two decades [17]. Regardless of the high initial response, the therapeutic effects of current treatment were insufficient and resulted in recurrence and death. Currently, there are no effective salvage regimens for treating metastatic UC. Metastasis requires the inherent dynamic instability of microtubules for cell motility, and many changes in the microtubule network have been identified in various cancers [14]. There is accumulating evidence that MAPs are associated with changes in microtubule dynamics, that they can determine the effects of microtubule-targeting agents, and that they play a role in cancer resistance [14]. However, reliable tumor markers that predict the sensitivity to chemotherapy and resistance to tumor metastasis remain elusive.

MAPs contain products of oncogenes, tumor suppressors, and apoptosis regulators thought to be involved in microtubule assembly. On the other hand, vinblastine, listed in the World Health Organization's List of Essential Medicines, binds tubulin and inhibits the assembly of microtubules [18]. It causes M-phase–specific cell-cycle arrest by breaking microtubule assembly and proper formation of the mitotic spindle and the kinetochore, which were essential for the separation of chromosomes during the anaphase of mitosis. Due to the possibility of sharing a common function, the rational microtubule-targeting cancer therapeutic approaches should preferably include proteomic profiling of tumor MAPs before the administration of antimicrotubule agents preferentially in combination with agents that modulate the expression of relevant MAPs [14].

Histologically, MAPs were originally related to the development of the nervous system, based on their very early detection in neurons. However, the aberrant expression of primarily neuronal MAPs has since been detected in non-neural cancer tissues [14]. We also assessed MAP1B expression across various cancer types using Oncomine™ Platform (Thermo Fisher, Ann Arbor, MI). Data revealed a diverse expression of MAP1B in various cancers. Of these, CNS tumor has highest MAP1B expression; bladder tumor has moderate expression. In our present results and using a published transcriptome dataset (GSE31684), we first found that *MAP1B* was significantly upregulated in UC and associated with more advanced pT status and metastatic disease in UBUC. Next, we found using immunohistochemistry that *MAP1B* overexpression markedly correlated with disease status in affected patients. In patients with UTUC, *MAP1B* overexpression was positively associated with synchronous multiple tumors, advanced pathological tumor stage, positive lymph node metastasis, the presence of vascular invasion, and an increased mitotic rate. However, in patients with UBUC, *MAP1B* overexpression was associated with advanced pathological tumor stage, positive lymph node metastasis, high histological tumor grade, the presence of vascular invasion, and an increased mitotic rate. Furthermore, using survival analysis, we demonstrated an association between *MAP1B* and aggressive clinical progression, whereby *MAP1B* overexpression independently predicted poor DSS and MFS rates for all patients with UC. These findings indicate that standard clinical practices may benefit from evaluating the *MAP1B* status to improve the risk stratification of patients with UC.

Different *MAP1B* interactors can be grouped into seven different categories, including signaling, cytoskeleton, transmembrane proteins, RNA-binding proteins, apoptosis, neurodegeneration-linked proteins, and neurotransmitter receptors [19]. *MAP1B* is translated as a precursor polypeptide that undergoes proteolytic processing to cleave into an N-terminal heavy chain (*MAP1B* HC) and a C-terminal light chain (*MAP1B* LC1). *MAP1B* LC1 overexpression, which can generate protein aggregates, has been observed in endoplasmic reticulum-related stress-induced cell apoptosis. This effect is blocked by DJ-1, a Parkinson's disease–related protein that has been proposed to act like a molecular chaperone, and inhibits α-synuclein aggregation [20]. However, in contrast to the proapoptotic effects caused by LC1 overexpression, *MAP1B* overexpression is not related to cell death related to *p53*, a tumor-suppressor gene; in fact, *MAP1B* overexpression reduces *p53* transcriptional activity and inhibits doxorubicin-induced apoptosis [21]. In addition, we found that the percentages of cells in the early and late stages of apoptosis were significantly increased between shLacZ controls and shMAP1B-treated cells. Further in vivo studies are warranted to confirm our findings and to determine whether such results may lead to new therapeutic targets for UC.

Recent studies have found that changes in the expression of MAPs are associated with chemotherapy resistance and cancer progression [14,22]. For example, stathmin plays a role in regulating neuroblastoma cell migration and invasion [22]. Silencing stathmin expression using RNAi gene silencing significantly reduced lung metastasis in neuroblastoma in vivo. Similarly, we demonstrated using UC cell lines with high endogenous *MAP1B* expression that silencing by *MAP1B* shRNA significantly reduced cell proliferation, migration, and invasion ability. Based on these findings, we posit that *MAP1B* may be a clinically valuable diagnostic marker for early cancer detection and a promising prognostic marker.

Further, *MAP1B* interacts with several other proteins associated with cancer. For example, Ras-association domain family 1 isoform A (RASSF1A), a tumor suppressor whose inactivation is implicated in the development of many human cancers, interacts with *MAP1B* to influence microtubule dynamics in the cell cycle and is involved in the inhibition of cancer cell growth [23]. Through distinct bifunctional structural domains, C19ORF5, a sequence homolog of *MAP1B*, mediates the communication between the microtubular cytoskeleton and mitochondria in the control of cell death and defective genome destruction. In addition, it has been proposed that the accumulation of C19ORF5 results in microtubule hyperstability, which may be involved in the tumor suppression activity of RASSF1A [24]. In the mammary cancer susceptibility 1 (*Mcs1*) region in chromosome 2 (a region that expresses centromeric proteins), Laes et al. analyzed candidate genes in the region and found that *MAP1B* was expressed in the mammary glands of rats [25]. Interactions with other proteins not related to its role in stabilizing microtubules suggest that *MAP1B* may be part of a "signaling protein" that regulates molecular pathways [19]. We propose that *MAP1B* has multiple functions, and whether the main function of *MAP1B* is microtubule stabilization or whether it has many cellular functions warrants further investigation.

A recent study that focused on kidney glomerular development and function found that *MAP1B* was specifically expressed in podocytes in human and murine adult kidney tissues [26]. In a mouse model, *MAP1B* was not essential for glomerular filtration function but may play a role in the development and differentiation of the kidney tubular system. The authors hypothesize that *MAP1B* may be related to either stress maintenance or the aging process in the kidney. It is clear that the overall effects of *MAP1B* on UC are complex, with reports of associations between *MAP1B* and survival and metastasis. Research aimed at decoding the functional consequences of *MAP1B* and signaling cross-talk with other proteins in different cancers is needed in the future. However, due to a slight predominance toward females, it is unclear if the results can easily be transferred to the rest of the world.
