Correlations between class I glucose transporter expression patterns and clinical outcomes in non‐small cell lung cancer

Abstract

Background

Glucose transporters (GLUTs) are highly expressed in various cancers. However, the implications of these variable expression patterns are unclear. This study aimed to clarify the correlation between class I GLUT expression patterns and clinical outcomes in non‐small cell lung cancer (NSCLC), including their potential role in inflammatory signaling.

Methods

Biopsy tissues from 132 patients with NSCLC (92 adenocarcinomas [ADC] and 40 squamous cell carcinomas [SQCC]) were analyzed. mRNA expression levels of class I GLUTs (solute carrier 2A [SLC2A]1, SLC2A2, SLC2A3, and SLC2A4) and inflammation‐related molecules (toll‐like receptors TLR4, RelA/p65, and interleukins IL8 and IL6) were measured. Cellular localization of GLUT3 and GLUT4 was investigated using immunofluorescence.

Results

Single, combined, and negative GLUT (SLC2A) expression were observed in 27/92 (29.3%), 27/92 (29.3%), and 38/92 (41.3%, p < 0.001) of ADC and 8/40 (20.0%), 29/40 (72.5%, p < 0.001), and 3/40 (7.5%) of SQCC, respectively. In ADC, the single SLC2A3‐expressed group had a significantly poorer prognosis, whereas the single SLC2A4‐expressed group had a significantly better prognosis. The combined expression groups showed no significant difference. SLC2A expression was not correlated with SQCC prognosis. SLC2A4 expression correlated with lower IL8 expression. GLUT3 and GLUT4 expressions were localized in the tumor cytoplasm.

Conclusions

In lung ADC, single SLC2A3 expression correlated with poor prognosis, whereas single SLC2A4 expression correlated with better prognosis and lower IL8 expression. GLUT3 expression, which is increased by IL8 overexpression, may be suppressed by increasing the expression of GLUT4 through decreased IL8 expression.

Class I GLUT expression patterns varied in NSCLC. Single GLUT4 expression, which was a better prognostic factor in lung ADC, was suggested to correlate with IL8 suppression in the cytoplasm. Poor prognostic factor single GLUT3 expression, which is increased by IL8 overexpression, may be suppressed by increasing the expression of GLUT4 through decreased IL8 expression.

INTRODUCTION

Cancer cells require many saccharides for uncontrolled proliferation and highly malignant behaviors such as invasion and metastasis. ¹ In addition to the hallmarks of cancer, “reprogramming energy metabolism” has been added to the six biological capabilities that have been advocated to understand the multistep development of human cancer. ² , ³ Owing to the Warburg effect, cancer cells produce energy through the glucose‐dependent glycolytic pathway instead of oxidative phosphorylation in the mitochondria, regardless of oxygen conditions. ⁴ , ⁵ To absorb large amounts of glucose, cancer cells overexpress glucose transporters (GLUTs), which facilitate the diffusible membrane transport of proteins. ⁶ GLUTs are encoded by the solute carrier 2A (SLC2A) family of genes and are classified into three subclasses, of which 14 members have been identified. ⁶ , ⁷ SLC2A1 (GLUT1), SLC2A2 (GLUT2), SLC2A3 (GLUT3), SLC2A4 (GLUT4), and SLC2A14 (GLUT14) are class I GLUTS; SLC2A5 (GLUT5), SLC2A7 (GLUT7), SLC2A9 (GLUT9), and SLC2A11 (GLUT11) are class II GLUTSs; and SLC2A6 (GLUT6), SLC2A8 (GLUT8), SLC2A10 (GLUT10), SLC2A12 (GLUT12), and SLC2A13 (GLUT13) are class III GLUTs. ⁷ , ⁸ , ⁹ These GLUT family members are expressed in most cancers and cancer cell lines. ⁶ Some studies have shown that GLUT overexpression is not related to cancer malignancy. ⁶ A recent analysis of The Cancer Genome Atlas (TCGA) data reported that among the 14 GLUTs, GLUT1 and GLUT3 are correlated with overall survival and disease‐free survival in colorectal cancer. ¹⁰ Indeed, GLUT1 and GLUT3 are overexpressed in various cancers and involved in their malignancy. ⁹ , ¹¹ , ¹² However, the association between GLUT4 overexpression and malignancy in many cancers remains unclear. ¹² In TCGA, GLUT4 is a favorable prognostic factor for breast cancer. ¹³ In noncancerous tissues, GLUT4 is known to translocate stimulated insulin from the intracellular space to the membrane for glucose uptake. ¹⁴ Otherwise, GLUT4 has shown anti‐inflammatory effects in diabetic and obese rat models. ¹⁵ , ¹⁶ Thus, GLUT expression may have the opposite function against cancers, although, in some cell lines, GLUT1, GLUT3, and GLUT4 are coexpressed. ⁶ The effects of combined GLUT expression on cancer remain unclear. In this study, we aimed to clarify the correlation between class I GLUT expression patterns, including GLUT1, GLUT2, GLUT3, and GLUT4, and clinical outcomes in non‐small cell lung cancer (NSCLC).

METHODS

Patients

Data from patients with NSCLC diagnosed at Nihon University Itabashi Hospital between January 2010 and December 2017 were retrospectively reviewed. We selected 132 formalin‐fixed and paraffin‐embedded (FFPE) tissue specimens that were confirmed to contain sufficient tumor cells among biopsy specimens remaining after diagnosis. Table 1 summarizes the patient characteristics. All the procedures were performed in accordance with the ethical standards of the institutional and national research committees and the Declaration of Helsinki. This study was approved by the institutional review board of Nihon University Itabashi Hospital (RK‐150609‐07), who waived the requirement for written informed consent.

TABLE 1.

Summary of the patient characteristics.

Factors	Number
Age, mean (range)	69.0 (39–89)
Sex
Male	86
Female	46
Smoking status
Former/current	101
Never	27
No information	4
Clinical stage
I	18
II	8
III	37
IV	69
Histological type
Adenocarcinoma	92
EGFR mutant type	26
EGFR wild‐type	66
Squamous cell carcinoma	40
Prognosis
Died of disease	67
Alive	65
Total	132

Total RNA extraction from FFPE biopsy tissue sections and cDNA synthesis

The 8 μm‐thick FFPE sections were mounted on membrane film‐coated slides. After dewaxing with xylene, the sections were lightly stained with toluidine blue. Target tumor cells were then macro‐ and microdissected using a laser‐assisted microdissection system (PALM MBIII‐N; Zeiss). Total RNA was extracted as previously described. ¹⁷ Briefly, each sample was mixed with 200 μL of extraction buffer containing 2% SDS, 0.1 mM EDTA, and 10 mM Tris–HCl and then incubated at 56°C with proteinase K until the sections were completely dissolved. Total RNA was purified with 20 μL of 2 M sodium acetate (pH 4.0), 220 μL of citrate‐saturated phenol (pH 4.3), and 60 μL of chloroform‐isoamyl alcohol, centrifuged, and the upper aqueous layer was transferred to new tubes. Then, 200 μL of isopropanol and 2 μL of glycogen were added and the samples were stored at −80°C for >30 min. They were then washed with 70% ethanol, dried on ice, and dissolved in 5 μL of RNase‐free water. The total RNA samples were stored at −80°C until use. Genomic DNA elimination and cDNA synthesis were performed using the QuantiTect reverse transcription kit (Qiagen) according to the manufacturer's instructions.

Quantitative reverse transcription‐polymerase chain reaction

To investigate the expression status of class I GLUTs and inflammation‐related genes, the mRNA levels of SLC2A1, SLC2A2, SLC2A3, SLC2A4, toll‐like receptor 4 (TLR4), RELA/p65, interleukin 6 (IL‐6), IL‐8, and actin beta (ACTB) were measured as internal controls using quantitative reverse transcription‐polymerase chain reaction (qRT‐PCR). qRT‐PCR was performed using Power SYBR Green Master Mix (ThermoFisher Scientific Inc.) with the respective primers shown in Table 2 on a StepOnePlus real‐time PCR system (ThermoFisher Scientific Inc.). Relative target mRNA values were obtained using the ΔΔCt methods.

TABLE 2.

Primer sequences.

Gene symbol	Gene name	Accession number	Sequences		Product (bp)
SLC2A1	Solute carrier family 2 member 1	NM_006516	Sense	5′‐actcatgaccatcgcgctag‐3′	150
			Antisense	5′‐ggaccctggctgaagagttc‐3′
SLC2A2	Solute carrier family 2 member 2	NM_000340	Sense	5′‐gtggctcagcaattttccgg‐3’	115
			Antisense	5′‐tgtttacagcgccaactcca‐3′
SLC2A3	Solute carrier family 2 member 3	NM_006931	Sense	5′‐gagtgtccagctaccgacag‐3′	145
			Antisense	5′‐ccgatggtggcatagatggg‐3’
SLC2A4	Solute carrier family 2 member 4	NM_001042	Sense	5′‐catccagaatctcgaggggc‐3′	101
			Antisense	5′‐tccgcttctcatccttcagc‐3′
TLR4	Toll‐like receptor 4	NM_138554	Sense	5′‐cgacaacctccccttctcaa‐3′	114
			Antisense	5′‐gcacctgcagttctgggaaa‐3′
RelA	RelA proto‐oncogene, NF‐kB subunit	NM_021975	Sense	5′‐tctgcttccaggtgacagtg‐3′	127
			Antisense	5′‐cggttcactcggcagatctt‐3′
IL‐6	Interleukin 6	NM_000600	Sense	5′‐tgaggagacttgcctggtga‐3′	105
			Antisense	5′‐agtgaggaacaagccagagc‐3′
IL‐8	Interleukin 8	NM_000584	Sense	5′‐ccaaacctttccaccccaaat‐3′	140
			Antisense	5′‐cctctgcacccagttttcct‐3′
ACTB	Actin beta	NM_001101	Sense	5′‐ggcatcctcaccctgaagta‐3′	203
			Antisense	5′‐ggggtgttgaaggtctcaaa‐3′

Immunofluorescence

The 4 μm‐thick sections were cut and dewaxed with xylene and ethanol. For antigen retrieval, the tissue sections were incubated in antigen retrieval buffer (pH 9.0; Nichirei Biosciences Inc.) for 30 min at 95°C and cooled to room temperature. After washing to remove the endogenous peroxidase, the samples were immersed in a 0.3% hydrogen peroxide solution for 10 min. After washing, the sections were incubated with the mixed primary antibodies, including rabbit polyclonal anti‐GLUT3 antibody (1:300, ab41525, Abcam PLC) and mouse monoclonal anti‐GLUT4 antibody (1:300, clone 6H11NB, Novus Biologicals) at 4°C overnight. After washing with phosphate‐buffered saline, the sections were incubated with Alexa Fluor 488‐labeled anti‐rabbit IgG in goat serum (1:500, Thermo Fisher Scientific Inc.) and Alexa Fluor 594‐labeled anti‐mouse IgG with goat serum (1:500, Thermo Fisher Scientific Inc.) for 30 min at room temperature. After washing with phosphate‐buffered saline, the sections were mounted using ProLong Diamond Antifade Mountant with 4′,6‐diamidino‐2‐phenylindole (DAPI) (Thermo Fisher Scientific Inc.). Images were acquired on an Olympus IX71 fluorescence microscope (Olympus Corp.), and color images were obtained using Lumina Vision software (Mitani Co.).

Statistical analysis

The correlations between class I GLUT mRNA expression patterns and histological type in NSCLC and the correlations between clinicopathological features and class I GLUT mRNA expression patterns were analyzed using the chi‐squared test. Survival analysis was performed using the Kaplan–Meier log‐rank method. Inflammation‐related gene expression levels were compared using the Mann–Whitney U test. Statistical analyses were performed using Bell Curve for Excel (Social Survey Research Information Co., Ltd).

RESULTS

Different class I GLUT mRNA expression patterns in NSCLC

The mRNA expression levels of class I GLUTs are shown in Table 3. The mRNA expression statuses of class I GLUTs in the NSCLC samples were categorized as combined (42.4%), single (26.5%), or negative (31.1%). The comparison of class I GLUT (SLC2A, SLC2A2, SLC2A3, and SLC2A4) expression patterns between adenocarcinoma (ADC) and squamous cell carcinomas (SQCC) showed significantly more cases with combined expression in SQCC (72.5%, p < 0.001) and significantly more cases without GLUT expression (negative expression) in ADC (41.3%, p < 0.001). The expression patterns of GLUTs in the NSCLC samples varied. The combined expression patterns of SLC2A1 and SLC2A3 (SLC2A1+/SLC2A3+) and SLC2A1, SLC2A3, and SLC2A4 (SLC2A1+/SLC2A3+/SLC2A4+) were found in significantly more SQCC cases (37.5%, p < 0.001 and 22.5%, p < 0.001, respectively). The combined expression pattern of SLC2A3+ and SLC2A4 (SLC2A3+/SLC2A4+) and single SLC2A3 expression tended to be higher in patients with ADC (18.5% and 15.9%, respectively). No SLC2A2 expression was detected in this study. In ADCs, epidermal growth factor receptor (EGFR) mutations are distinctive genetic variants. However, class I GLUT mRNA expression patterns did not differ between patients with and without pathogenic EGFR mutations (Table S1).

TABLE 3.

Differences in class I GLUT mRNA expression patterns by histology type.

Class I GLUT mRNA expression pattern	Total (n = 132)	ADC (n = 92)	SQCC (n = 40)
	Number (%)	Number (%)	Number (%)
Single expression	35 (26.5)	27 (29.3)	8 (20.0)
SLC2A1	6 (4.54)	3 (3.26)	3 (7.50)
SLC2A2	0 (0.00)	0 (0.00)	0 (0.00)
SLC2A3	21 (15.90)	17 (18.48)	4 (10.00)
SLC2A4	8 (6.06)	7 (7.61)	1 (2.50)
Combined expression	56 (42.4)	27 (29.3)	29 (72.5)**
SLC2A1+/SLC2A3+	24 (18.18)	9 (9.78)	15 (37.50)**
SLC2A1+/SLC2A4+	1 (0.76)	0 (0.00)	1 (2.50)
SLC2A3+/SLC2A4+	21 (15.91)	17 (18.48)	4 (10.00)
SLC2A1+/SLC2A3+/SLC2A4+	10 (7.58)	1 (1.09)	9 (22.50)**
Negative expression	41 (31.06)	38 (41.30)**	3 (7.50)

Abbreviations: ADC, adenocarcinoma; GLUT, glucose transporter; SQCC, squamous cell carcinoma.

^** p < 0.001.

Class I GLUT expression and clinicopathological features in NSCLC

ADC and SQCC showed different mRNA expression patterns for class I GLUTs. Therefore, we examined the correlation between class I GLUT expression patterns and the clinicopathological features of each histological type. The results are shown in Figure 1. Combined SLC2A1+/SLC2A3+ and combined SLC2A1+/SLC2A3+/SLC2A4+ expression patterns were significantly higher in SQCC (p < 0.001; p < 0.001, respectively), and negative class I GLUTs expression was observed in ADC (p < 0.001) (Figure 1a). The combined SLC2A1+/SLC2A3+/SLC2A4+ expression pattern was significantly higher in patients with SQCC >70 years of age (p < 0.001, Figure 1b), female (p < 0.001, Figure 1c), former and/or current smokers (p < 0.001, Figure 1d), with clinical Stage 3 disease or higher (p < 0.001, Figure 1e), male (p < 0.05, Figure 1c), and with less than clinical Stage 3 disease (p < 0.05, Figure 1e). Patients with ADC <70 years of age (p < 0.05, Figure 1b), male (p < 0.05, Figure 1c), former and/or current smokers (p < 0.05, Figure 1d), and clinical Stage 3 disease or higher (p < 0.001, Figure 1e) showed negative class I GLUT mRNA expression patterns. Patients with ADC aged >70 years with less than clinical Stage 3 disease showed significantly higher combined expression of SLC2A3+/SLC2A4+ (p < 0.05, Figure 1b, and p < 0.05, Figure 1e, respectively).

FIGURE 1.

Class I GLUT mRNA expression patterns and clinicopathological features in non‐small cell lung cancer (NSCLC). NSCLC patients were grouped by clinicopathological factors (a) total, (b) age: under or over 70 years, (c) sex: male or female, (d) smoking status: current/former or never smokers, and (e) clinical stage: under or over stage III. The two bars on the left side of the graph show ADCs, and the two bars on the right side show squamous cell carcinomas. Each bar shows the positive ratio (%) of each class I GLUT mRNA expression pattern.

Correlations between class I GLUT mRNA expression patterns and clinical outcome

The correlations between class I GLUT mRNA expression patterns and patient prognosis are shown in Figure 2a–m. In ADC, while a slightly shorter overall survival was observed in the group without any class I GLUT mRNA expression patterns, the difference was not significant (Figure 2a). Hence, the patients were grouped according to their class I GLUT mRNA expression patterns. In ADC, the SLC2A3 single‐expression group had a significantly worse prognosis (p = 0.039, Figure 2d), whereas the SLC2A4 single‐expression group had a significantly better prognosis (p = 0.028, Figure 2e). In contrast, the single SLC2A1‐expressed group (Figure 2c); combined SLC2A1‐expressed group that included SLC2A1+/SLC2A3+, SLC2A1+/SLC2A4+, and SLC2A1+/SLC2A3+/SLC2A4+ patterns (Figure 2f); combined SLC2A3 expressed group that included SLC2A1+/SLC2A3+, SLC2A3+/SLC2A4+, and SLC2A1+/SLC2A3+/SLC2A4+ patterns (Figure 2g); and combined SLC2A4 expressed group that included SLC2A1+/SLC2A4+, SLC2A3+/SLC2A4+, and SLC2A1+/SLC2A3+/SLC2A4+ patterns (Figure 2h) showed no significant difference. In SQCC, no significant correlation was observed between the groups with or without class I GLUT mRNA (SLC2A1, SLC2A3, and SLC2A4) expression and prognosis (Figure 2b). Additionally, the differences in class I GLUT mRNA expression patterns and prognosis showed no correlations (Figure 2i–m). The number of samples with single SLC2A4‐expressed SQCC was too small for analysis.

FIGURE 2.

Survival analysis using the Kaplan–Meier method. The group with or without class I GLUT mRNA expression in (a) adenocarcinoma (ADC), and in (b) squamous cell carcinoma (SQCC). (c) In ADC, the group with or without single SLC2A1, (d) SLC2A3, (e) SLC2A4, (f) combined SLC2A1, (g) combined SLC2A3, and (h) combined SLC2A4. In SQCC, (i) the group with or without single SLC2A1, (j) SLC2A3, (k) combined SLC2A1, (l) combined SLC2A3, and (m) combined SLC2A4.

Inflammation‐related mRNA expression and class I GLUT mRNA expression patterns

In ADC, SLC2A3 expression was associated with poor prognosis, while SLC2A4 was correlated with better prognosis. Because SLC2A4 expression has been reported to suppress the expression of inflammation‐related molecules in noncancerous tissues, we investigated the correlation between class I GLUT expression patterns and inflammation‐related mRNA TLR4, RelA/p65, IL6, and IL8. The results showed that IL8 expression values were significantly lower in the single SLC2A4‐expressed group than in the other groups (p = 0.037) and tended to be lower in the combined SLC2A3+/SLC2A4+ group (p = 0.072) in ADC (Figure 3a). In SQCC, the combined SLC2A3+/SLC2A4+ group showed significantly lower IL8 expression compared to the SLC2A1+/SLC2A3+ and SLC2A3+ groups (p = 0.034). The SQCC groups with SLC2A4 expression (SLC2A1+/SLC2A3+/SLC2A4+ and SLC2A3+/SLC2A4+) and lower IL8 expression compared to the groups without SLC2A4 expression (SLC2A1+/SLC2A3+ and SLC2A3+, p = 0.024) (Figure 3b). TLR4, IL6, and RelA/p65 expression did not differ significantly between ADC and SQCC (Figure S1).

FIGURE 3.

Correlation between inflammation‐related IL8 mRNA expression and class I GLUT mRNA expression patterns. The bars show the median ± standard deviation of the relative gene expression values of inflammation‐related IL8 genes standardized by an internal control ACTB in each group with different SLC2A mRNA expression patterns. (a) Adenocarcinoma and (b) squamous cell carcinoma.

Cellular localization of GLUT3 and GLUT4

GLUTs are expressed from the cytoplasm to the plasma membrane for glucose transport. However, since IL8 is involved in inflammatory signaling in the cytoplasm, we examined the expression of GLUT3 and GLUT4 in ADC. The results are shown in Figure 4. The upper images show normal airway epithelial cells with membranous expression of GLUT4 and without GLUT3. The lower images show ADC cells with cytoplasmic expression of GLUT3 and GLUT4.

FIGURE 4.

Cellar localization of GLUT3 and GLUT4 in adenocarcinoma (ADC). Multiple immunofluorescences showed that red signals were GLUT3, green signals were GLUT4, and DAPI of nucleus. Marge images show colocalization of GLUT3 and GLUT4. The upper photographs show the normal airway epithelium with membranous GLUT4 expression and without GLUT3. The lower photographs show the ADC cells with cytoplasmic GLUT3 and GLUT4 coexpression.

DISCUSSION

This study investigated the expression patterns of class I GLUTs in NSCLC and found that they were characterized by different histological types of lung cancer. In lung ADC, single GLUT3 mRNA (SLC2A3) expression was correlated with poor prognosis, whereas single GLUT4 mRNA (SLC2A4)‐expressing ADC had a better prognosis. Younes et al. reported that in stage I NSCLC, the overexpression of both GLUT1 and GLUT3 was associated with the worst prognosis, followed by either GLUT1 or GLUT3, whereas the best prognosis was associated with no overexpression of either GLUT. ¹⁸ However, their NSCLC cohort included ADC, SQCC, and others, and histological subclass analysis was not conducted. In our study, which was the first to investigate advanced‐stage patients, the prognosis of single GLUT3 (SLC2A3)‐expressed ADC was significantly worse, which was not the case with GLUT1+ and GLUT3+ (SLC2A1+/SLC2A3+) ADC or SQCC (data not shown).

In the present study, there were significantly more SLC2A1+/SLC2A3+ SQCC cases with or without SLC2A4 expression. In particular, SLC2A1+/SLC2A3+ co‐expression was more abundant in clinical Stage 3 or higher SQCC but was not correlated with age, sex, or smoking status. SLC2A expression patterns were restricted in females and never smokers, who were the minority of SQCC cases. Elucidating the relationship between SLC2A1+/SLC2A3+ and the development and progression of non‐smoking related lung SQCC is a topic that should be addressed in the future.

Osumi et al. reported that proinflammatory mediator angiopoietin‐like protein 2 induced both GLUT1 and GLUT3 overexpression in a lung SQCC cell line, but only GLUT3 in a lung ADC cell line, and might have metastatic potential. ¹⁹ GLUT3 expression has been suggested to be associated with aggressive tumor behavior in the early to‐advanced stages of lung ADC. Among the lung ADC cases included in this study, more cases lacked class I GLUT mRNA expression and had malignant clinicopathological features. Although other transporters, such as sodium‐glucose cotransporter (SGLT)‐1 and SGLT‐2, are reportedly expressed in lung cancer, ⁶ , ²⁰ the role of SGLT expression in the prognosis of lung cancer remains a subject for future studies. In contrast, single GLUT4 (SLC2A4)‐expressing lung ADCs showed a significantly better prognosis than ADCs showing combined GLUT4 (SLC2A4) expression. GLUT4 overexpression has been observed in many cancers; however, its function remains unclear. ¹³ One reason why it is difficult to analyze the role of GLUT4 expression in cancer is that few cases express a single GLUT4. However, a recent TCGA analysis showed a correlation between GLUT4 (SLC2A4) expression and a favorable prognosis in breast cancer. ¹² Therefore, GLUT4 expression may suppress tumor cell growth and progression in some cancers.

In normal tissues, GLUT4 is highly expressed in adipose tissues and skeletal muscles, as insulin‐responsive GLUT4 maintains glucose uptake homeostasis. ⁸ , ¹⁴ , ²¹ Recent reports on the effects of GLUT4 on inflammation in mouse models of type 2 diabetes and obesity have shown that metformin treatment increases GLUT4 expression in adipose tissue and may suppress inflammation. ¹⁵ , ¹⁶ Arctigenin, a naturally derived compound, activates GLUT4 and insulin receptor substrate 2 expression and suppresses liver tissue inflammation by suppressing TLR4 expression. ¹⁵

The results of the present study revealed significantly lower IL‐8 mRNA expression in single GLUT4 (SLC2A4)‐expressing lung ADC; SLC2A4 co‐expressed SQCC also exhibited lower IL‐8 expression levels. However, TLR4 was not correlated with any class I GLUT expression patterns. Higher IL‐8 mRNA expression in patients with lung cancer, including ADC and SQCC, is associated with poor prognosis. ²² Shimizu et al. reported that after culturing a human colon and lung ADC cell line with the addition of IL‐8, only GLUT3 was significantly overexpressed among class I GLUTs, and glucose uptake was significantly increased. ²³ GLUT3 expression was decreased by inhibiting IL‐8 expression by shRNA. ²³ Furthermore, IL8 expression in the cytoplasm of some endometrioid carcinomas and colorectal ADCs was shown using immunohistochemistry. ²⁴ Our results showed cytoplasmic expression of GLUT3 and GLUT4 in tumor cells. Therefore, it is suggested that the expression of GLUT4 is correlated with IL8 suppression in the tumor cytoplasm. GLUT3 expression, which is increased by IL8 overexpression, may be suppressed by increasing the expression of GLUT4 through decreased IL8 expression in lung ADC.

In recent years, GLUTs expressed in various cancers have become targets for anti‐cancer therapies. ²⁵ GLUT1 inhibition blocked the growth of RB‐1 positive triple‐negative breast cancer cell lines. ²⁶ In vitro studies showed that GLUT1 expression is inhibited by apigenin, flavonoids, resveratrol, and curcumin ²⁷ , ²⁸ , ²⁹ , ³⁰ , ³¹ , ³² and that GLUT2 expression is inhibited by flavonoids and phloretin. ³³ , ³⁴ However, no agents have been reported to inhibit GLUT3 expression. Our results suggest that the expression of GLUT3, which is a poor prognostic factor in lung ADC, can be suppressed by increasing GLUT4 expression through a decrease in IL8 expression. The enhanced expression of GLUT4 by flavonoids, procyanidin, and quercetin ¹³ , ³⁵ , ³⁶ may inhibit not only inflammation but also cancer progression. Further investigation into the effects of natural ingredients and other substances on the continuous enhancement of GLUT4 expression in cancer is required.

In conclusion, class I GLUT expression patterns were correlated with clinical outcomes in lung ADC but not in SQCC. In ADC, single GLUT3 (SLC2A3) expression was associated with a poor prognosis, whereas single GLUT4 (SLC2A4) expression was associated with a better prognosis. However, combined GLUT expression had a reduced prognostic impact. Furthermore, GLUT4 was suggested to correlate with IL8 suppression in the cytoplasm. GLUT3 expression, which is increased by IL8 overexpression, may be suppressed by increasing the expression of GLUT4 through decreased IL8 expression. Continuously enhanced GLUT4 expression is expected to inhibit not only inflammation but also cancer progression in lung ADC.

AUTHOR CONTRIBUTIONS

All authors had full access to the data in the study and take responsibility for the integrity of the data and the accuracy of the data analysis. Conceptualization, Yoko Nakanishi; Methodology, Yoko Nakanishi, Momoko Iwai, and Yukari Hirotani; Histological analysis, Tomoyuki Tanino, Haruna Nishimaki‐watanabe, Fumi Nozaki, Xiaoyan Tang, and Sumie Ohni; Investigation, Momoko Iwai, Ren Kato and Yukari Hirotani; Writing‐original draft preparation, Yoko Nakanishi and Momoko Iwai; Writing‐review and editing, Shinobu Masuda and Kayoko Sasaki‐fukatsu; Project administration, Yoko Nakanishi and Kayoko Sasaki‐fukatsu; Funding acquisition, Kayoko Sasaki‐fukatsu. All authors reviewed the results and approved the final version of the manuscript.

FUNDING INFORMATION

This work was supported by JSPS KAKENHI (grant no. 21K11601).

CONFLICT OF INTEREST STATEMENT

The authors declare that they have no conflicts of interest to report regarding the present study.

Supporting information

Data S1. Supporting Information

Nakanishi Y, Iwai M, Hirotani Y, Kato R, Tanino T, Nishimaki‐watanabe H, et al. Correlations between class I glucose transporter expression patterns and clinical outcomes in non‐small cell lung cancer. Thorac Cancer. 2023;14(27):2761–2769. 10.1111/1759-7714.15060

DATA AVAILABILITY STATEMENT

The authors confirm that the data supporting the findings of this study are available within the article.