High expression of LAMP2 predicts poor prognosis in patients with esophageal squamous cell carcinoma

Abstract

BACKGROUND:

LAMP2 is one of the major protein components of lysosome. In addition to the expression on the lysosomal membrane, LAMP2 has also been found relocalizing to the cell surface of some highly metastatic tumor cells.

OBJECTIVE:

The aim of this study was to detect the expression levels of LAMP2 and discuss its roles in esophageal squamous cell carcinoma (ESCC).

METHODS:

Six hundred and ten tissue samples of ESCC were collected to construct tissue microarrays, which were stained by immunohistochemistry.

RESULTS:

After immunohistochemical staining, 596 patients including 460 men and 136 women were analyzed. The LAMP2 expression levels were significantly different based on degrees of histological differentiation ( $\chi^{2}=$ 108.906, $P<$ 0.001). The similar results were also observed in TNM stages ( $\chi^{2}=$ 23.835, $P<$ 0.01). LAMP2 expression levels negatively correlated with degrees of histological differentiation ( $P<$ 0.01). Logistic regression analysis showed that the LAMP2 expression levels were correlated with the degrees of histological differentiation (OR ${}_{\textit{differentiation}}=$ 0.452, $P<$ 0.001) and TNM stages (OR ${}_{\textit{TNM}}=$ 1.482, $P=$ 0.42). Besides, Kaplan-Meier survival curves indicated that patients with higher expression of LAMP2 exhibited poor prognosis ( $P<$ 0.05).

CONCLUSIONS:

Our results demonstrated that LAMP2 expression levels correlated with tumor histological differentiation and TNM stages. High expression of LAMP2 predicts poor prognosis in patients with ESCC.

Keywords

LAMP2 differentiation esophageal tumor immunohistochemistry survival analysis

1. Introduction

Esophageal squamous cell carcinoma (ESCC), the fourth most common cause of cancer-related deaths, is one of the most common cancers, especially in China [1, 2]. The fiveyear survival rate of patients with advanced ESCC is still very low [3, 4, 5]. Less than 30% ESCC patients could achieve early diagnosis and treatment. To improve clinical outcome of ESCC, novel molecular biomarkers used for early diagnosis and predicting prognosis have been widely investigated [6, 7].

Lysosomal associated membrane proteins (LAMPs) are major protein components of the lysosome playing an important role in the lysosome-mediated physiological processes [8, 9, 10]. The heavy glycosylation of the LAMPs creates a sugar coat or glycocalyx on the inner side of the lysosomal membrane, which protects the membrane from the hydrolytic enzymes and degradation [11, 12]. Despite having 37% amino acid sequence homology, LAMP1 and LAMP2 are distinct proteins for their separate chromosome localizations. Mice deficient in both of them are embryonic lethal, but viable and fertile upon loss of any one of them, suggesting that LAMPs are, to some degree, functionally redundant in vivo. However, LAMP2 seems to have some specific functions, because LAMP2 single deficiency has more severe consequences than LAMP1loss [13].

In addition to the expression on the lysosomal membrane, LAMP2 has also been found relocalizing to the cell surface of some highly metastatic tumor cells, such as melanoma and colon cancer [14, 15], playing an important role in cell-cell adhesion and cancer cell migration. These observations indicate that LAMP2 may be important for the metastasis of malignant tumors.

In this study, we detected the LAMP2 expression in tissue microarrays (TMAs) [16, 17, 18] by immunohistochemistry (IHC) to analyze the correlation between the LAMP2 expression levels and clinicopathological characteristics of ESCC patients, and evaluated the possibility of LAMP2 as a biomarker for ESCC prognosis.

2. Materials and methods

2.1 ESCC patients and samples

A total of 610 patients with ESCC were enrolled in this study from thoracic surgery department of the Affiliated Hospital of Jining Medical University from 2008 to 2014. All patients did not receive radiotherapy or chemotherapy before surgical resection. Informed consents were obtained from all enrolled patients. The clinical stages were defined according to the 7th edition of TNM classification of the International Union Against Cancer (2010) [19]. Patients’ clinical information was collected from medical records, and follow-up data were collected by telephone interview. This study was reviewed and approved by the Medical Ethics Committee of the Affiliated Hospital of Jining Medical University.

Table 1
LAMP2 Expression and Clinicopathological Characteristics in ESCC

Characteristic Frequency Percent (%) IHC $\chi^{2}$ $r_{s}$

– $+$ $++$ $+++$

No. of Patients 596

Age (Year)-mean (range) 60.9 (34–83)

Age (Year) $<$ 60 248 41.61 45 115 70 18 2.801

$\geqslant$ 60 348 58.39 81 160 85 22

Gender Female 136 22.82 30 62 33 11 0.785

Male 460 77.18 96 213 122 29

TNM Staging I 14 2.35 4 6 3 1 23.835 ${}^{}$ 0.119 ${}^{}$

II 360 60.40 83 167 87 23

III 222 37.25 39 102 65 16

Tumor Location Upper 10 2.24 3 3 3 1 2.170 0.001

Middle 285 63.76 61 134 72 18

Lower 152 34.00 29 77 38 8

Differentiation Well 9 1.51 7 1 1 0 108.906 ${}^{}$ 0.348 ${}^{}$

Well-Moderate 86 14.43 30 44 11 1

Moderate 185 31.04 46 95 39 5

Moderate-Poor 238 39.93 37 116 70 15

Poor 78 13.09 6 19 34 19

Characteristic	Frequency	Percent (%)	IHC	$\chi^{2}$	$r_{s}$
				–	$+$	$++$	$+++$
No. of Patients	596
Age (Year)-mean (range)	60.9 (34–83)
Age (Year)	$<$ 60	248	41.61	45	115	70	18	2.801
	$\geqslant$ 60	348	58.39	81	160	85	22
Gender	Female	136	22.82	30	62	33	11	0.785
	Male	460	77.18	96	213	122	29
TNM Staging	I	14	2.35	4	6	3	1	23.835 ${}^{**}$	0.119 ${}^{**}$
	II	360	60.40	83	167	87	23
	III	222	37.25	39	102	65	16
Tumor Location	Upper	10	2.24	3	3	3	1	2.170	0.001
	Middle	285	63.76	61	134	72	18
	Lower	152	34.00	29	77	38	8
Differentiation	Well	9	1.51	7	1	1	0	108.906 ${}^{**}$	0.348 ${}^{**}$
	Well-Moderate	86	14.43	30	44	11	1
	Moderate	185	31.04	46	95	39	5
	Moderate-Poor	238	39.93	37	116	70	15
	Poor	78	13.09	6	19	34	19

${}^{**}$ , $P<$ 0.01.

2.2 Tissue microarrays (TMAs)

Paraffin-embedded tissue blocks of the ESCC patients were used to construct TMAs. The pathologist microscopically examined hematoxylin and eosin-stained sections and selected representative areas, excluding tumor necrosis. Cores (0.6 mm) of malignant tissues were spotted in duplicates. We constructed 9 TMA blocks containing a total of 610 tissue samples of ESCC. There were up to 70 patient tissue samples on each slice. 5-micrometer sections were cut from completed array blocks and transferred to adhesive slides. Then, the sections were stained with haematoxylin and assessed for adequate tumor representation.

Besides, 60 patients’ paraffin-embedded tissue blo-cks were randomly selected to making routine pathological sections for immunohistochemical staining, which were used to examine the consistency between TMAs and conventional tissue slices.

2.3 Immunohistochemistry (IHC)

TMAs were stained using the rabbit anti-LAMP2 monoclonal antibody (1:50, Proteintech, 10397-1-AP), and the microwave was used to perform the antigen retrieval. The immunostainings, performed according to the manufacturer’s instructions, were completed by the technical staff of the Department of Pathology. Then, the TMAs were counterstained with haematoxylin. LAMP2 antibody omission was used as a negative control. Finally, the TMAs were scanned with CaseViewer 1.4.

The immunohistochemical scores were based on the percentage of LAMP2 positive cells and their staining intensity [20]. Regarding the staining intensity, the score “0” corresponds to no staining, “1” to faint staining, “2” to moderate staining and “3” to intense immunostaining. The necrotic areas were not included. Moreover, the invasion zones were not included. Regarding the percentage of positive tumor cells, the score “0” corresponds to 0–2% positive cells, “1” to 2–25% positive cells, “2” to 25–50% positive cells, “3” to 50–75% positive cells and “4” to 75–100% positive cells. Then intensity and percentage scores were multiplied to conclude a total score (from 0 to 12). The patients were divided into four groups according to the total scores. “ $-$ ” corresponds to 0 $\sim$ 1 scores, “ $+$ ” to 2 $\sim$ 3 scores, “ $++$ ” to 4 $\sim$ 6 scores, and “ $+++$ ” to 8 $\sim$ 12 scores. The scores were judged by two pathologists independently. Cases with different scores were determined by two joint discussions.

2.4 Statistical analysis

Statistical analysis between groups was performed using Pearson chi-squared ( $\chi^{2}$ ) test, and correlation analysis was assessed by Spearman Correlation Coefficient ( ${\bm{r}}_{s}$ ) [16, 21, 22]. Additionally, logistic regression analysis and Kaplan-Meier survival curves were performed. Statistical analysis was performed using SPSS software 22.0. $P<$ 0.05 was considered as statistically significant.

3. Results

3.1 Clinical data

All 610 patients were diagnosed as ulcerative squamous cell carcinoma according to the pathological results. Lymphatic metastasis was present in 281 cases, but no distant metastasis was found. After immunostainings, 14 patients’ tissues were failed to stain due to the tissues removal. The remaining 596 patients were comprised of 460 men and 136 women, ageing from 34 to 83 years (mean age: 60.9 years). The clinicopathological characteristics and the IHC results were listed in Table 1.

3.2 Expression Level of LAMP2 in ESCC

In ESCC tumor cells, the positive immunostaining was mainly located at the cytoplasm (Figs 1, D&E). LAMP2 expression levels in ESCC tissues were significantly different among the different subgroups. The LAMP2 expression levels were significantly different based on degrees of tumor histological differentiation ( $\chi^{2}=$ 108.906, $P<$ 0.001). The similar results were also observed in TNM stages ( $\chi^{2}=$ 23.835, $P<$ 0.01), but not in tumor locations, ages and genders ( $P>$ 0.05).

The correlations between LAMP2 expression levels and clinical data were analyzed by spearman correlation coefficient. LAMP2 expression levels were negatively correlated with degrees of tumor histological differentiation (Fig. 2A, $P<$ 0.01). The worse the degree of histological differentiation, the higher the level of LAMP2 expression. By contrary, positive correlation was observed between LAMP2 expression levels and TNM stages (Fig. 2B, $P<$ 0.01), but no significant correlation was found in tumor locations, ages and genders (Figs 2C, D, E, $P>$ 0.05).

Figure 1.

Immunohistochemistry Stainning of TMAs of the ESCC Patients. A: TMAs of the ESCC patients. B: Negative case of immunohistochemical staining (10 $\times$ ). C: Negative case of immunohistochemical staining (40 $\times$ ). D: Positive case of immunohistochemical staining (10 $\times$ ). E: Positive case of immunohistochemical staining (40 $\times$ ). Scale bars: A, 5000 $\mu$ m; B&D, 200 $\mu$ m; C&E, 50 $\mu$ m.

Figure 2.

Expression Level of LAMP2 in ESCC Tissues. Samples were divided into different subgroups for statistical analysis. Expression levels (95% CI) were correlated with degrees of tumor histological differentiation (A), TNM stages (B), but not with tumor location (C), age (D) and gender (E). *, $P<$ 0.05.

Logistic regression analysis between LAMP2 expression and clinicopathological characteristics was performed. In this analysis, the positive degrees of LAMP2 were redefined. “ $-$ ” and “ $+$ ” were defined as the low expression “ $++$ ” and “ $+++$ ” were defined as high expression. The results showed that the expression levels of LAMP2 were correlated with TNM stages and degrees of differentiation. Further analysis showed that the degrees of differentiation showed medium negative correlation with LAMP2 expression (OR ${}_{\textit{differentiation}}=$ 0.452) with statistical significance ( $P<$ 0.001), while the positive correlation between TNM stages and LAMP2 expression was very weak (OR ${}_{\textit{TNM}}=$ 1.482, $P=$ 0.42).

Kaplan-Meiersurvival curves, used to analyze the correlations between LAMP2 expression levels and patients’ prognosis, indicated that patients with a higher expression of LAMP2 showed worse prognosis (Fig. 3F, $P<$ 0.05). Besides, we also analyzed the relationship between the other clinical data (gender, tumor size, metastasis, and TNM stages) and the prognosis (Fig. 3).

Figure 3.

Kaplan-Meier Survival Curves for ESCC Patients. Kaplan-Meier survival curves for ESCC patients between different subgroups were made. Patients with a higher expression of LAMP2 (F) showed worse prognosis ( $P<$ 0.05). Besides, the prognosis was related to TNM stages (B), organ metastasis (C, follow-up data) tumor size (D) and lymphatic metastasis (E, postoperative pathological diagnosis), but not with genders (A).

Figure 4.

Comparison of Consistency between TMAs and Routine Tissue Slices. 60 patients’ cancer tissues were randomly selected to making routine tissue slices for immunohistochemical staining to examine the consistency between TMAs and conventional slices. Results of 55 cases were consistent with TMAs.

4. Discussion

In the present study, the results showed that the LAMP2 expression levels were significantly different among different TNM stages and degrees of tumor histological differentiation, and that LAMP2 expression levels were negatively correlated with degrees of tumor histological differentiation, but positively with TNM stages (Table 1, Figs 2A & B). The results of correlation analysis and survival curves were consistent with the previous studies which showed that LAMP2 was highly expressed in many cancers, especially in the metastatic cancers [14, 15, 23, 24]. LAMP2 plays an important role in the regulation of autophagy which is involved in cancer cell metabolism [25]. Therefore, the results of this study may imply the enhanced autophagy activity correlated with LAMP2 expression in tumor cells. However, the detailed molecular mechanism needs to be further determined.

LAMP2 expression levels were negatively correlated with degrees of histological differentiation in our study. The worse the degree of histological differentiation, the higher the level of LAMP2 expression. The relationships between LAMP2 and cell differentiation have been reported [26, 27]. This might be related to the glycosylation of LAMP2 [28]. The weak positive correlation between the expression of LAMP2 and TNM stages observed in this study might be associated with the enhancement of autophagy activity in tumor cells [29], because LAMP2 was located on surface of lysosome, the final destination of the autophagy, which involves importantly in cancer cell metabolism. We would continue to study the roles of LAMP2 in ESCC in the following research.

Additionally, we conducted a comparison of LAMP2 expression levels between different genders in consideration of the location of LAMP2 gene on the X chromosome (Xq24) and the gender differences in the incidence of esophageal cancer [1, 2]. According to our results, there was no difference in the expression of LAMP2 between genders (Table 1, $P>$ 0.05). Therefore, LAMP2 might be unrelated to gender differences in incidence of esophageal cancer.

TMAs, arranging many different individual specimens at regular arrays on the same slides [18], were characterized by large scale, high throughput, standardization. The biggest advantage was that all samples were tested under identical conditions, and that the experimental errors were reduced greatly. In this study, we used TMAs to detect the expression levels of LAMP2 in order to reduce the artificial operation errors among samples. But the other problem was that, compared with the traditional tissue sections, the detected area of each sample was greatly reduced, increasing the impact of tumor heterogeneity [30, 31]. With this in mind, we compared the consistency of TMAs and conventional pathological tissue slices. The ratio of consistency was 91.7% (55 of 60, Fig. 4).

Besides, there were other defects in the present study. Firstly, adjacent carcinoma tissues were not involved in this study. The differences between tumor and adjacent tissues could not be analyzed. Secondly, the sample size of some subgroups was small (N ${}_{I}=$ 14, N ${}_{\textit{well differentiation}}=$ 9). This limited number might influence the statistical analysis.

In conclusion, our results demonstrated that LAMP2 might be involved in the histological differentiation, progression and prognosis of ESCC. The molecular mechanism would be discussed in the following research.

Footnotes

Acknowledgments

This study was supported by the following science foundations: Science and Technology Development Project of Jining Science and Technology Bureau (2015-57-35), Scientific Research Project of Jining Medical University (JY2015KJ023), and the Key Project of Scientific Research Program of Affiliated Hospital of Jining Medical University (MP-2015-002), Shandong, People’s Republic of China. All the authors have read the journal’s authorship agreement and policy on disclosure of potential conflicts of interest.

Conflict of interest

The authors declare no conflict of interest.

References

Torre

L.A.

Bray

Siegel

R.L.

Ferlay

Lortet-Tieulent

and Jemal

, Global cancer statistics, 2012, CA Cancer J Clin 65 (2015), 87–108.

Chen

Zheng

Baade

P.D.

Zhang

Zeng

Bray

Jemal

X.Q.

and He

, Cancer statistics in China, 2015, CA Cancer J Clin 66 (2016), 115–132.

Bao

Liu

Zhou

Cai

Anfossi

Liu

Rong

and Liu

, Three-dimensional conformal radiotherapy with concurrent chemotherapy for postoperative recurrence of esophageal squamous cell carcinoma: clinical efficacy and failure pattern, Radiat Oncol 8 (2013), 241.

Zhang

, Epidemiology of esophageal cancer, World J Gastroenterol 19 (2013), 5598–5606.

Rice

T.W.

Rusch

V.W.

Apperson-Hansen

Allen

M.S.

Chen

L.Q.

Hunter

J.G.

Kesler

K.A.

Law

Lerut

T.E.

Reed

C.E.

Salo

J.A.

Scott

W.J.

Swisher

S.G.

Watson

T.J.

and Blackstone

E.H.

, Worldwide esophageal cancer collaboration, Dis Esophagus 22 (2009), 1–8.

Wang

Song

Feng

X.S.

Abnet

C.C.

Zuo

X.B.

Tan

Zhan

Jia

Zhou

Shu

X.O.

Yuan

J.M.

Zheng

Zhao

X.K.

Gao

S.G.

Yuan

Z.Q.

Zhou

F.Y.

Fan

Z.M.

Cui

J.L.

Lin

H.L.

Han

X.N.

Chen

Dawsey

S.M.

Liao

Lee

M.P.

Ding

Qiao

Y.L.

Liu

Chang

Wei

Gao

Y.T.

Koh

W.P.

Xiang

Y.B.

Tang

Z.Z.

Fan

J.H.

Han

J.J.

Zhou

S.L.

Zhang

D.Y.

Yuan

Huang

Liu

Zhai

Qiao

Jin

Guo

Miao

Yang

Wang

Wheeler

W.A.

Gail

Yeager

Yuenger

Guo

E.T.

A.L.

Zhang

X.M.

Sun

L.D.

B.G.

Tang

Peng

X.Q.

Liu

Hutchinson

Jacobs

Giffen

Burdette

Fraumeni

J.F.

, Jr. Shen

Zeng

Kraft

Chung

C.C.

Tucker

M.A.

Hou

Z.C.

Liu

Y.L.

Liu

Wang

Yuan

Chen

L.S.

Liu

Meng

Sun

X.M.

J.W.

Zhou

Y.F.

Yang

L.Q.

Wang

Qige

Yang

W.J.

Lei

G.Y.

Chen

L.Q.

E.M.

Yuan

Yue

W.B.

Wang

L.W.

Fan

X.P.

Zhu

F.H.

Zhao

W.X.

Mao

Y.M.

Zhang

Xing

G.L.

J.L.

Han

Ren

J.L.

Liu

Ren

S.W.

Kong

Q.P.

Sheyhidin

Wei

Zhang

Y.R.

Feng

C.W.

Wang

Yang

Y.H.

Hao

H.Z.

Bao

Q.D.

Liu

B.C.

A.Q.

Xie

Yang

W.C.

Wang

Zhao

X.H.

Chen

S.Q.

Hong

J.Y.

Zhang

X.J.

Freedman

N.D.

Goldstein

A.M.

Lin

Taylor

P.R.

Wang

L.D.

and Chanock

S.J.

, Joint analysis of three genome-wide association studies of esophageal squamous cell carcinoma in Chinese populations, Nat Genet 46 (2014), 1001–1006.

Huang

Luo

Chen

Wang

Jin

Kong

Liu

Rong

Chen

Zhang

Chen

Zhu

Zheng

Wen

Zheng

Tan

Xie

Yang

and Fu

, Identification and Validation of Lymphovascular Invasion as a Prognostic and Staging Factor in Node-Negative Esophageal Squamous Cell Carcinoma, J Thorac Oncol 11 (2016), 583–592.

Schneede

Schmidt

C.K.

Holtta-Vuori

Heeren

Willenborg

Blanz

Domanskyy

Breiden

Brodesser

Landgrebe

Sandhoff

Ikonen

Saftig

and Eskelinen

E.L.

, Role for LAMP-2 in endosomal cholesterol transport, J Cell Mol Med 15 (2011), 280–295.

Kallunki

Olsen

O.D.

and Jaattela

, Cancer-associated lysosomal changes: friends or foes? Oncogene 32 (2013), 1995–2004.

10.

Huang

Pang

Bai

and Yan

, Age-related decrease of the LAMP-2 gene expression in human leukocytes, Clin Biochem 45 (2012), 1229–1232.

11.

Saftig

and Klumperman

, Lysosome biogenesis and lysosomal membrane proteins: trafficking meets function, Nat Rev Mol Cell Biol 10 (2009), 623–635.

12.

Appelqvist

Waster

Kagedal

and Ollinger

, The lysosome: from waste bag to potential therapeutic target, J Mol Cell Biol 5 (2013), 214–226.

13.

Eskelinen

E.L.

, Roles of LAMP-1 and LAMP-2 in lysosome biogenesis and autophagy, Mol Aspects Med 27 (2006), 495–502.

14.

Krishnan

Bane

S.M.

Kawle

P.D.

Naresh

K.N.

and Kalraiya

R.D.

, Altered melanoma cell surface glycosylation mediates organ specific adhesion and metastasis via lectin receptors on the lung vascular endothelium, Clin Exp Metastasis 22 (2005), 11–24.

15.

Saitoh

Wang

W.C.

Lotan

and Fukuda

, Differential glycosylation and cell surface expression of lysosomal membrane glycoproteins in sublines of a human colon cancer exhibiting distinct metastatic potentials, J Biol Chem 267 (1992), 5700–5711.

16.

Leclerc

B.G.

Charlebois

Chouinard

Allard

Pommey

Saad

and Stagg

, CD73 Expression Is an Independent Prognostic Factor in Prostate Cancer, Clin Cancer Res 22 (2016), 158–166.

17.

Kallioniemi

O.P.

Kononen

and Sauter

, Introducing tissue microarrays to molecular pathology, Clin Chem 58 (2012), 1717–1718.

18.

Kononen

Bubendorf

Kallioniemi

Barlund

Schraml

Leighton

Torhorst

Mihatsch

M.J.

Sauter

and Kallioniemi

O.P.

, Tissue microarrays for high-throughput molecular profiling of tumor specimens, Nat Med 4 (1998), 844–847.

19.

Talsma

van Hagen

Grotenhuis

B.A.

Steyerberg

E.W.

Tilanus

H.W.

, van Lanschot and Wijnhoven

B.P.

, Comparison of the 6th and 7th Editions of the UICC-AJCC TNM Classification for Esophageal Cancer, Ann Surg Oncol 19 (2012), 2142–2148.

20.

Azim

H.A.

Kassem

Treilleux

Wang

El Enein

M.A.

Anis

S.E.

and Bachelot

, Analysis of PI3K/mTOR Pathway Biomarkers and Their Prognostic Value in Women with Hormone Receptor-Positive, HER2-Negative Early Breast Cancer, Transl Oncol 9 (2016), 114–123.

21.

Marechal

Demetter

Nagy

Berton

Decaestecker

Polus

Closset

Deviere

Salmon

and Van Laethem

J.L.

, High expression of CXCR4 may predict poor survival in resected pancreatic adenocarcinoma, Br J Cancer 100 (2009), 1444–1451.

22.

Deng

Gao

Cai

Wang

Zhuang

Tan

Liu

Hao

and Lin

, Expression and clinical significance of annexin A2 and human epididymis protein 4 in endometrial carcinoma, J Exp Clin Cancer Res 34 (2015), 96.

23.

Agarwal

A.K.

Gude

R.P.

and Kalraiya

R.D.

, Regulation of melanoma metastasis to lungs by cell surface Lysosome Associated Membrane Protein-1 (LAMP1) via galectin-3, Biochem Biophys Res Commun 449 (2014), 332–337.

24.

Agarwal

A.K.

Srinivasan

Godbole

S.K.

Budnar

Gude

R.P.

and Kalraiya

R.D.

, Role of tumor cell surface lysosome-associated membrane protein-1 (LAMP1) and its associated carbohydrates in lung metastasis, J Cancer Res Clin Oncol 141 (2015), 1563–1574.

25.

Lozy

and Karantza

, Autophagy and cancer cell metabolism, Semin Cell Dev Biol 23 (2012), 395–401.

26.

Yan

Z.W.

Hou

J.K.

Fan

and Huang

, Chloroquine enhances cobalt chloride-induced leukemic cell differentiation via the suppression of autophagy at the late phase, Biochem Biophys Res Commun 430 (2013), 926–932.

27.

Morell

Bort

Vara-Ciruelos

Ramos-Torres

Altamirano-Dimas

Diaz-Laviada

and Rodriguez-Henche

, Up-Regulated Expression of LAMP2 and Autophagy Activity during Neuroendocrine Differentiation of Prostate Cancer LNCaP Cells, PLoS One 11 (2016), e0162977.

28.

Tyleckova

Valekova

Zizkova

Rakocyova

Marsala

Gadher

S.J.

and Kovarova

, Surface N-glycoproteome patterns reveal key proteins of neuronal differentiation, J Proteomics 132 (2016), 13–20.

29.

Cicchini

Karantza

and Xia

, Molecular pathways: autophagy in cancer–a matter of timing and context, Clin Cancer Res 21 (2015), 498–504.

30.

Qian

Wang

D.C.

Chen

and Cheng

, Detection of single cell heterogeneity in cancer, Semin Cell Dev Biol (2016).

31.

Zhang

Guan

Sun

and Guo

, Tumor heterogeneity and circulating tumor cells, Cancer Lett 374 (2016), 216–223.