Association between XPO5 rs11077 polymorphism and cancer susceptibility: a meta-analysis of 7284 cases and 8511 controls

Moazeni-Roodi A.1, Taheri M.2, Hashemi M.*3

Summary. Aim: Several studies evaluated the association between rs11077 polymorphism located in the 3’UTR of the XPO5 gene and cancer susceptibility. We conducted a meta-analysis to assess the impact of XPO5 rs11077 polymorphism on cancer risk. Materials and Methods: The online databases were searched for relevant case-control studies published up to July 2018. 15 articles of 16 studies, with totally 7284 cancer cases and 8511 healthy controls, were eligible for inclusion in the meta-analysis. The data were extracted from the eligible studies and were processed using Stata 14.1 and Revman 5.3 software. Pooled estimates of odds ratio with 95% confidence intervals were used to evaluate the strength of association between XPO5 rs11077 and cancer risk. Results: Overall, our finding showed no significant association between XPO5 rs11077 variant and overall cancer risk, either performed subgroup analysis by cancer types and ethnic groups in all genetic model. Conclusion: The findings did not support an association between rs11077 variant and cancer risk. Due to small sample sizes particularly in stratified analysis, further large-scale well designed studies between this polymorphism and cancer risk are warranted.

DOI: 10.32471/exp-oncology.2312-8852.vol-41-no-4.13811

Submitted: July 29, 2018.
*Correspondence: E-mail: mhd.hashemi@gmail.com;
hashemim@zaums.ac.ir
Abbreviation used: miRNAs — microRNAs; SNP — single nucleotide polymorphisms; XPO5 — exportin-5.

Cancer is a leading cause of mortality worldwide [1, 2]. There were about 4 292 000 newly-diagnosed cancer cases and 2 814 000 cancer-related deaths in United States in 2017. Although the etiology of cancer is still not clearly disclosed, genetic background and environmental factors are believed to be involved in cancer development [3, 4].

MicroRNAs (miRNAs), as regulators of gene expression, are small single-stranded RNA molecules of about 21–23 nucleotides [5, 6]. The biosynthesis of a functional miRNA involves several miRNA bioge­nesis genes and occurs in multiple steps [7]. The process of miRNA synthesis begins within the nucleus where RNA polymerase II produces large primary miRNA transcripts (about 500 to 3000 nucleotides) known as pri-miRNA. The pri-miRNA is then processed by multiprotein complex that includes DROSHA into pre-miRNA (about 60 to 100 nucleotides). Next, RAN GTPase and exportin-5 (XPO5) complex transfers pre-miRNA to the cytoplasm, and pre-miRNA is then cut into miRNA duplexes by DICER [6, 8] finally forming 18–24 nucleotide single-stranded, mature miRNA [8, 9].

In general, polymorphisms in miRNA processing genes as well as miRNA genes (pri-miRNAs, pre-miRNAs and mature miRNAs) could influence cancer risk by affecting miRNA function [10].

Preceding studies examining the relationship between XPO5 rs11077 gene polymorphism and cancer designated inconclusive findings [11–25]. So, this meta-analysis was performed to evaluate the impact of XPO5 rs11077 polymorphism on cancer risk.

MATERIALS AND METHODS

Literature search. A systemic literature searches in the PubMed, Web of Science, Scopus, and Google Scholar databases was done for all articles focused on association between XPO5 polymorphism and cancer risk published up to June 2018. The search term was “cancer or carcinoma or tumor or neoplasm” and “XPO5 or exportin-5 or miRNA biogenesis” and “polymorphism or mutation or variation or rs11077”.

Inclusion and exclusion criteria. Studies were comprised in the meta-analysis by meeting the following criteria: 1) original case-control studies of the association between the XPO5 rs11077 polymorphism and cancer; 2) studies providing sufficient data of the genotype frequencies of XPO5 rs11077 polymorphism in both cases and controls; 3) the studies have not repeated reports in the same population. The following studies were excluded: 1) conference abstracts, letters, case reports, reviews, overlapped data, animal or mechanism studies for XPO5 rs11077 polymorphism and cancer; 2) studies with insufficient information on genotype frequency. Finally, 15 articles were considered for meta-analysis.

Data extraction. The authors independently extracted data that met the inclusion and exclusion criteria. The following information was collected from each study including the name of first author, year of publication, country, ethnicity, number of cases and controls, and the genotype and allele frequencies of cases and controls.

Statistical analysis. Hardy-Weinberg equilibrium (HWE) for the controls of each study was determined by the chi-square test. We used Revman 5.3 software (Version 5.3. Copenhagen: The Nordic Cochrane Centre, the Cochrane Collaboration, 2014) and STATA 14.1 software (Stata Corporation, College Station, TX, USA) for all statistical analyses and to produce the plots. The strength of the association between XPO5 rs11077 polymorphism and cancer risk was evaluated through calculating pooled odds ratios (ORs) with the corresponding 95% confidence intervals (95% CIs) using following genetic models: codominant, dominant, recessive, overdominant and allele model. The significance of the pooled OR was determined with the Z-test, and p-values less than 0.05 were considered statistically significant.

Heterogeneity between selected studies was inspected using the I2 statistic and the χ2-based Q test. A p < 0.10 representing the presence of significant hete­rogeneity. When significant heterogeneity values were returned, the random-effects model was used to estimate pooled ORs. Otherwise, the fixed-effects model was employed.

Publication bias across enrolled studies was estimated by Begg’s funnel plot. The degree of asymmetry was assessed using Egger’s linear regression test and p < 0.05 was considered significant publication bias.

Sensitivity analysis was conducted through sequential deleting each of included studies so as to ve­rify the stability of overall estimates.

RESULTS

Fifteen articles [11–25] of 16 studies, with totally 7284 cancer cases and 8511 controls, were eligible for meta-analysis. The main detailed characteristics of the eligible studies are listed in Table 1.

Table 1. Characteristics of the studies eligible for meta-analysis
Author Year Country Ethnicity Cancer type Source of control Genotyping method Case/
control
Cases Controls HWE
AA AC CC A C AA AC CC A C
Buas 2015 Europe Caucasian Esophageal cancer HB TaqMan 2495/3206 2879 2111 3751 2661
Cho 2015 Korea Asian Colorectal cancer HB PCR-RFLP 408/400 333 74 1 740 76 337 61 2 735 65 0.667
Ding 2013 China Asian Non-small cell lung cancer PCR-LDR 112/80 94 18 0 206 18 65 14 1 144 16 0.803
Horikawa 2008 USA Caucasians Renal cell carcinoma HB SNPlex 276/277 88 134 54 310 242 89 150 38 328 226 0.044
Kim 2010 Korea Asian Lung cancer HB Sequencing 100/99 88 12 0 188 12 87 9 3 183 15 < 0.001
Kim 2016 China Asian Hepatocellular carcinoma HB PCR-RFLP 147/209 128 19 0 275 19 170 38 1 378 40 0.465
Osuch-Wojcikiewicz 2015 Poland European Larynx cancer HB TaqMan 124/160 36 62 26 134 114 34 44 82 112 208 < 0.001
Sung 2011 Korea Asian Breast cancer HB TaqMan 559/567 473 82 4 1028 90 501 64 2 1066 68 0.977
Thakkar 2018 India Asian Hodgkin Lymphoma PB TaqMan 101/200 39 41 21 119 83 76 92 32 244 156 0.638
Wen 2017 China Asian Thyroid cancer HB TaqMan 1134/1228 907 210 17 2024 244 1023 194 11 2240 216 0.593
Xie 2015 China Asian Gastric cancer HB PCR-LDR 137/142 119 17 1 255 19 123 18 1 264 20 0.705
Yang 2008 American Caucasian Bladder cancer HB SNPlex 746/746 248 356 114 852 584 241 363 122 845 607 0.456
Yao 2013 USA African American Breast cancer PB Illumina GoldenGate 242/411 39 203 45 214 152 304 518 0.018
Yao 2013 USA European American Breast cancer PB Illumina GoldenGate 200/310 76 124 127 130 53 384 236 0.052
Ye 2008 American Caucasian Esophageal cancer HB SNPlex 340/334 129 150 61 408 272 113 175 46 401 267 0.093
Zhao 2015 China Asian Colorectal cancer HB PCR-LDR 163/142 143 19 1 305 21 123 18 1 264 20 0.705

Quantitative synthesis. All eligible studies were pooled into the analysis and the results showed that XPO5 rs11077 polymorphism was not associated with the overall cancer risk in codominant, dominant, recessive, overdominant, and allele genetics models (Fig. 1 and Table 2).

Table 2. The pooled ORs and 95% CIs for the association between XPO5 polymorphism and cancer susceptibility
Polymorphism No Association test Heterogeneity Egger’s testp-value Begg’s testp-value
OR (95% CI) Z p χ2 I2 (%) p    
Overall cancer
AC vs AA 13 1.04 (0.93–1.15) 0.64 0.52 13.72 13 0.32 0.515 1.00
CC vs AA 13 0.96 (0.68–1.36) 0.23 0.82 22.49 47 0.03 0.916 0.929
AC+CC vs AA 13 1.02 (0.92–1.12) 0.33 0.75 18.82 26 0.17 0.101 0.347
CC vs AC+AA 13 0.95 (0.62–1.46) 0.24 0.81 38.13 69 0.0001 0.940 0.531
AC vs CC+AA 13 1.04 (0.87–1.25) 0.45 0.65 30.80 61 0.002 0.983 0.542
C vs A 14 0.99 (0.88–1.12) 0.14 0.89 34.61 62 0.001 0.423 0.208
GI cancer
AC vs AA 5 0.90 (0.73–1.10) 1.03 0.30 4.87 18 0.30
CC vs AA 5 1.10 (0.71–1.70) 0.43 0.67 0.80 0 0.94
AC+CC vs AA 5 9.92 (0.76–1.13) 0.77 0.44 3.80 0 0.43
CC vs AC+AA 5 1.29 (0.87–1.92) 1.25 0.21 1.18 0 0.88
AC vs CC+AA 5 0.88 (0.68–1.14) 0.98 0.33 5.79 31 0.22
C vs A 6 1.03 (0.96–1.10) 0.74 0.46 3.11 0 0.68
Breast cancer
C vs A 3 1.05 (0.87–1.25) 0.49 0.63 3.96 50 0.14
Lung cancer
AC vs AA 2 1.05 (0.58–1.88) 0.16 0.88 0.42 0 0.52
CC vs AA 2 0.18 (0.02–1.58) 1.55 0.12 0.05 0 0.82
AC+CC vs AA 2 0.90 (0.51–1.58) 0.38 0.70 0.09 0 0.76
CC vs AC+AA 2 0.18 (0.02–1.56) 1.56 0.12 0.06 0 0.81
AC vs CC+AA 2 0.75 (0.47–1.20) 1.21 0.23 0.37 0 0.54
C vs A 2 0.78 (0.46–1.32) 0.91 0.36 0.00 0 0.99
Asian
AC vs AA 9 1.14 (0.99–1.31) 1.80 0.07 6.75 0 0.56
CC vs AA 9 1.21 (0.78–1.86) 0.86 0.39 5.29 0 0.73
AC+CC vs AA 9 1.14 (1.0–1.31) 1.93 0.05 7.30 0 0.50
CC vs AC+AA 9 1.24 (0.82–1.87) 1.01 0.31 5.20 0 0.74
AC vs CC+AA 9 1.12 (0.98–1.29) 1.63 0.10 7.55 0 0.48
C vs A 9 1.06 (1.00–1.13) 1.87 0.06 9.55 16 0.30
Caucasian
AC vs AA 3 0.89 (0.75–1.05) 1.39 0.16 1.34 0 0.51
CC vs AA 3 1.08 (0.83–1.40) 0.57 0.57 2.48 19 0.29
AC+CC vs AA 3 0.93 (0.79–1.09) 0.93 0.35 0.66 0 0.72
CC vs AC+AA 3 1.20 (0.87–1.65) 1.13 0.26 4.31 0.54 0.12
AC vs CC+AA 3 0.85 (0.70–1.04) 1.59 0.11 3.16 37 0.21
C vs A 4 1.02 (0.96–1.09) 0.72 0.47 1.66 0 0.65
 Association between <i>XPO5</i> rs11077 polymorphism and cancer susceptibility: a meta analysis of 7284 cases and 8511 controls
Fig. 1. Flow chart of articles selection for this meta-analysis

We also performed stratified analysis by cancer type and ethnicity (see Table 2). The findings proposed that XPO5 rs11077 was not associated with gastrointestinal cancer, breast cancer and lung cancer. Besides, the variant was not associated with cancer risk in Asian as well as Caucasian population.

Heterogeneity. Heterogeneity among the studies included in the meta-analysis is shown in Table 2. The results showed that heterogeneity exists between the studies in homozygous codominant, recessive, overdominant and allele genetic models. So, random-effects model was used to determine pooled ORs.

Publication bias. A funnel plot was created as a visual aid to detect risk of publication bias (Fig. 2). Egger’s linear regression test and Begg’s test proposed no publication bias in all genetic model tested (see Table 2).

214235235 Association between <i>XPO5</i> rs11077 polymorphism and cancer susceptibility: a meta analysis of 7284 cases and 8511 controls
Fig. 2. Forest plots of the association between XPO5 rs11077 A>C polymorphism and cancer risk in the overall study population under the following models: a — AC vs AA, b — CC vs AA, c — AC+CC vs AA, d — CC vs AC+AA, — AC vs AA+CC, and — C vs A

Sensitivity analysis. Sensitivity analysis was done and the findings revealed that our data are stable and reliable in all inheritance genetic models tested (Fig. 3).

 Association between <i>XPO5</i> rs11077 polymorphism and cancer susceptibility: a meta analysis of 7284 cases and 8511 controls
Fig. 3. Funnel plots of the association between cancer risk and XPO5 rs11077 A>C polymorphism in the overall study population under the following models: a — AC vs AA, b — CC vs AA, c — AC+CC vs AA, d — CC vs AC+AA, e — AC vs AA+CC, and f — C vs A
 Association between <i>XPO5</i> rs11077 polymorphism and cancer susceptibility: a meta analysis of 7284 cases and 8511 controls
Fig. 4. Sensitivity analyses for studies on XPO5 rs11077 A>C using different genetic models; а — AC vs AA, в — CC vs AA, с — AC+CC vs AA, d — CC vs AC+AA, e — AC vs AA+CC, and f — C vs A

DISCUSSION

The etiology of cancer is multifactorial in which both host genetic factors and environmental factors play a role [26, 27]. Accumulating evidence proposed that genetic variation is associated with cancer susceptibility [4, 28]. In this study, we conducted a meta-analysis to evaluate the association between XPO5 rs11077 gene polymorphism and cancer risk based on 16 eligible case-control studies with a total of 7284 cancer cases and 8511 healthy controls. Overall, pooled risk estimates proposed that this polymorphism is not associated with cancer risk. Stratified analyses by cancer types and ethnicities did not support an association between rs11077 polymorphism and cancer susceptibility.

Preceding studies examining the association between XPO5 rs11077 gene polymorphism and cancer indicated inconclusive results [11–25]. A genome-wide association study conducted by Buas et al. [11] on miRNA biogenesis genes (157 single nucleotide polymorphisms (SNPs), 21 genes); miRNA gene loci (234 SNPs, 210 genes); and miRNA-targeted mRNAs (177 SNPs, 158 genes) showed no significant association between XPO5 rs11077 A>C polymorphism and risk of esophageal adenocarcinoma. Cho et al. [12] revealed no significant association between XPO5 rs11077 and colorectal cancer risk in Korean population. Horikawa et al. [14] have found no significant correlation between rs11077 variant and risk of renal cell carcinoma. No significant association between rs11077 variant and risk of lung cancer, hepatocellular carcinoma, non-small cell lung cancer were found [13–16]. Osuch-Wojcikiewicz et al. [17] have found that rs11077 variant is associated with the risk of laryngeal cancer in Polish population. Sung et al. [18] have found no significant association between rs11077 variant and risk of breast cancer in Korean population. The rs11077 variant was found to be associated with increased risk of thyroid cancer [19]. No significant association between rs11077 variant and risk of gastric cancer was observed in Chinese population [20]. Yang et al. [21] findings revealed no significant association between rs11077 polymorphism and bladder cancer in American population. Ye et al. [22] reported that rs11077 variant significantly increased the risk of esophageal cancer.

XPO5 gene is mapped to a short arm of chromosome 6 (6p21.1) and encodes XPO5 protein which is involved in export of pre-miRNA from nucleus into the cytoplasm. Hoti et al. [29] reported that a XPO5 knockdown resulted in downregulation of 20 mature miRNAs and overexpression of six miRNAs.

Several studies evaluated the expression levels of XPO5 in various cancers and the findings were controversial. The expression levels of XPO5 were found to be higher in several tumors including breast, ovary, prostate, bladder, and melanoma compared to the normal adjacent tissues, while the lower expression level of XPO5 in kidney, adrenal gland, and hepatocellular carcinoma tumors proposing oncogenic or tumor-suppressor features in different cancer types [29–32].

There are some limitations in our meta-analysis needed to be addressed. First, heterogeneity was observed among the studies possibly resulting from the differences of ethnicity, source of control, and cancer type. Second, this study focused on the effect of rs11077 polymorphism and cancer risk. Gene-gene and gene-environment interactions might also impact in cancer risk. Third, the characteristics of included studies such as age and sex which might affect the results of meta-analysis were not evaluated due to the lack of relevant data across the included studies. Fourth, the majority of the individuals studied were Asian, further studies on other ethnicity groups are needed. Finally, the sample size of our meta-analysis is relatively small especially in subgroup analyses by cancer types (5 studies for gastrointestinal cancer, 3 studies for breast cancer, and 2 studies for lung cancer) and ethnicities (9 studies for Asian and 3 studies for Caucasian). Accordingly, the statistical power of the study is limited and the results should be interpreted with caution.

In conclusion, the results of our meta-analysis based on 16 case-control studies suggested that there is no significant association between the XPO5 rs11077 polymorphism and cancer risk. Statistical power can be improved by pooling analysis from more studies. Considering the limitations mentioned above, further well-designed multicenter studies with large sample sizes, more diverse ethnic groups and cancer types are warranted to verify the findings.

CONFLICT OF INTEREST

The authors have declared that no competing interests exist.

REFERENCES

  • 1. Ferlay J, Soerjomataram I, Dikshit R, et al. Cancer incidence and mortality worldwide: sources, methods and major patterns in GLOBOCAN 2012. Int J Cancer 2015; 136: E359–86.
  • 2. Siegel RL, Miller KD, Jemal A. Cancer Statistics 2017. CA Cancer J Clin 2017; 67: 7–30.
  • 3. Foulkes WD. Inherited susceptibility to common cancers. N Engl J Med 2008; 359: 2143–53.
  • 4. Hassanzarei S, Hashemi M, Sattarifard H, et al. Genetic polymorphisms of HOTAIR gene are associated with the risk of breast cancer in a sample of southeast Iranian population. Tumour Biol 2017; 39: 1010428317727539.
  • 5. Ryan BM. microRNAs in cancer susceptibility. Adv Cancer Res 2017; 135: 151–71.
  • 6. Creugny A, Fender A, Pfeffer S. Regulation of primary microRNA processing. FEBS Lett 2018; 592: 1980–96.
  • 7. Esquela-Kerscher A, Slack FJ. OncomiRs — micro­RNAs with a role in cancer. Nat Rev Cancer 2006; 6: 259–69.
  • 8. Lund E, Guttinger S, Calado A, et al. Nuclear export of microRNA precursors. Science 2004; 303: 95–8.
  • 9. Kim JO, Bae J, Kim J, et al. Association of microRNA biogenesis genes polymorphisms with ischemic stroke susceptibility and post-stroke mortality. J Stroke 2018; 20: 110–21.
  • 10. Georges M, Coppieters W, Charlier C. Polymorphic miRNA-mediated gene regulation: contribution to phenotypic variation and disease. Curr Opin Genet Dev 2007; 17: 166–76.
  • 11. Buas MF, Onstad L, Levine DM, et al. MiRNA-related SNPs and risk of esophageal adenocarcinoma and Barrett’s esophagus: post genome-wide association analysis in the BEACON consortium. PLoS One 2015; 10: e0128617.
  • 12. Cho SH, Ko JJ, Kim JO, et al. 3’-UTR polymorphisms in the MiRNA machinery genes DROSHA, DICER1, RAN, and XPO5 are associated with colorectal cancer risk in a Korean population. PLoS One 2015; 10: e0131125.
  • 13. Ding C, Li C, Wang H, et al. A miR-SNP of the XPO5 gene is associated with advanced non-small-cell lung cancer. Onco Targets Ther 2013; 6: 877–81.
  • 14. Horikawa Y, Wood CG, Yang H, et al. Single nucleotide polymorphisms of microRNA machinery genes modify the risk of renal cell carcinoma. Clin Cancer Res 2008; 14: 7956–62.
  • 15. Kim JS, Choi YY, Jin G, et al. Association of a common AGO1 variant with lung cancer risk: a two-stage case-control study. Mol Carcinog 2010; 49: 913–21.
  • 16. Kim MN, Kim JO, Lee SM, et al. Variation in the Dicer and RAN genes are associated with survival in patients with hepatocellular carcinoma. PLoS One 2016; 11: e0162279.
  • 17. Osuch-Wojcikiewicz E, Bruzgielewicz A, Niemczyk K, et al. Association of polymorphic variants of miRNA processing genes with larynx cancer risk in a Polish population. Biomed Res Int 2015; 2015: 298378.
  • 18. Sung H, Lee KM, Choi JY, et al. Common genetic polymorphisms of microRNA biogenesis pathway genes and risk of breast cancer: a case-control study in Korea. Breast Cancer Res Treat 2011; 130: 939–51.
  • 19. Wen J, Gao Q, Wang N, et al. Association of micro­RNA-related gene XPO5 rs11077 polymorphism with susceptibility to thyroid cancer. Medicine (Baltimore) 2017; 96: e6351.
  • 20. Xie Y, Wang Y, Zhao Y, et al. Single-nucleotide polymorphisms of microRNA processing machinery genes are associated with risk for gastric cancer. Onco Targets Ther 2015; 8: 567–71.
  • 21. Yang H, Dinney CP, Ye Y, et al. Evaluation of genetic variants in microRNA-related genes and risk of bladder cancer. Cancer Res 2008; 68: 2530–7.
  • 22. Ye Y, Wang KK, Gu J, et al. Genetic variations in micro­RNA-related genes are novel susceptibility loci for esophageal cancer risk. Cancer Prev Res (Phila) 2008; 1: 460–9.
  • 23. Zhao Y, Du Y, Zhao S, et al. Single-nucleotide polymorphisms of microRNA processing machinery genes and risk of colorectal cancer. Onco Targets Ther 2015; 8: 421–5.
  • 24. Yao S, Graham K, Shen J, et al. Genetic variants in microRNAs and breast cancer risk in African American and European American women. Breast Cancer Res Treat 2013; 141: 447–59.
  • 25. Thakkar DN, Palugulla S, Selvarajan S, et al. Frequency distribution of BLMH, XPO5 and HFE gene polymorphisms in the South Indian population and their association with Hodgkin Lymphoma. Int J Biol Markers 2018; 33: 514–19.
  • 26. Alizadeh J, Zeki AA, Mirzaei N, et al. Mevalonate cascade inhibition by simvastatin induces the intrinsic apoptosis pathway via depletion of isoprenoids in tumor cells. Sci Rep 2017; 7: 44841.
  • 27. Ghavami S, Shojaei S, Yeganeh B, et al. Autophagy and apoptosis dysfunction in neurodegenerative disorders. Prog Neurobiol 2014; 112: 24–49.
  • 28. Risch N, Merikangas K. The future of genetic studies of complex human diseases. Science 1996; 273: 1516–7.
  • 29. Ott CA, Linck L, Kremmer E, et al. Induction of exportin-5 expression during melanoma development supports the cellular behavior of human malignant melanoma cells. Oncotarget 2016; 7: 62292–304.
  • 30. Khan M, Khan Z, Uddin Y, et al. Evaluating the oncogenic and tumor suppressor role of XPO5 in different tissue tumor types. Asian Pac J Cancer Prev 2018; 19: 1119–25.
  • 31. Hoti N, Yang S, Aiyetan P, et al. Overexpression of exportin-5 overrides the inhibitory effect of miRNAs regulation control and stabilize proteins via posttranslation modifications in prostate cancer. Neoplasia 2017; 19: 817–29.
  • 32. Li Y, Wang X, He B, et al. Downregulation and tumor-suppressive role of XPO5 in hepatocellular carcinoma. Mol Cell Biochem 2016; 415: 197–205.
No Comments » Add comments
Leave a comment

ERROR: si-captcha.php plugin says GD image support not detected in PHP!

Contact your web host and ask them why GD image support is not enabled for PHP.

ERROR: si-captcha.php plugin says imagepng function not detected in PHP!

Contact your web host and ask them why imagepng function is not enabled for PHP.