Open Access

Severe hypoxia induces chemo-resistance in clinical cervical tumors through MVP over-expression

  • Pedro C Lara1, 2Email author,
  • Marta Lloret1, 2,
  • Bernardino Clavo1, 2,
  • Rosa M Apolinario2,
  • Luis Alberto Henríquez-Hernández2, 3,
  • Elisa Bordón2,
  • Fausto Fontes2 and
  • Agustín Rey2, 4
Radiation Oncology20094:29

DOI: 10.1186/1748-717X-4-29

Received: 3 June 2009

Accepted: 6 August 2009

Published: 6 August 2009

Abstract

Oxygen molecule modulates tumour response to radiotherapy. Higher radiation doses are required under hypoxic conditions to induce cell death. Hypoxia may inhibit the non-homologous end-joining DNA repair through down regulating Ku70/80 expression. Hypoxia induces drug resistance in clinical tumours, although the mechanism is not clearly elucidated. Vaults are ribonucleoprotein particles with a hollow barrel-like structure composed of three proteins: major vault protein (MVP), vault poly(ADP-ribose) polymerase, and telomerase associated protein-1 and small untranslated RNA. Over-expression of MVP has been associated with chemotherapy resistance. Also, it has been related to poor outcome in patients treated with radiotherapy alone. The aim of the present study was to assess the relation of Major Vault Protein expression and tumor hypoxia in clinical cervical tumors. MVP, p53 and angiogenesis, together with tumor oxygenation, were determined in forty-three consecutive patients suffering from localized cervix carcinoma. High MVP expression was related to severe hypoxia compared to low MVP expressing tumors (p = 0.022). Tumors over-expressing MVP also showed increased angiogenesis (p = 0.003). Besides it, in this study we show for the first time that severe tumor hypoxia is associated with high MVP expression in clinical cervical tumors. Up-regulation of MVP by hypoxia is of critical relevance as chemotherapy is currently a standard treatment for those patients. From our results it could be suggested that hypoxia not only induces increased genetic instability, oncogenic properties and metastatization, but through the correlation observed with MVP expression, another pathway of chemo and radiation resistance could be developed.

Introduction

Growing cancers often acquire an increasing number of genetic alterations. Such genetic changes, including chromosomal translocation, gene amplification, intragenic mutation, and gene silencing, are responsible for the activation of oncogenes and the inactivation of tumour-suppressor genes [1]. How cancer cells acquire genetic instability remains unclear. Exposure of cells to adverse conditions like hypoxia can lead to genome alterations, enhancing the progression potential of tumor cells and resistance to oncological treatments [1]. Hypoxia may lead to conditions that causes increased spontaneous damage to DNA or inhibit DNA repair processes, impair DNA repair and cause tumor progression by altered p53 expression and increased angiogenesis [2, 3]. Deregulation of DNA repair pathways can contribute to the phenomenon of hypoxia-induced genetic instability within the tumor [4]. Hypoxia is measured in clinical tumors by several techniques, including the Eppendorf polarographic method [2, 5]. In cervical cancer patients, hypoxia is commonly associated to a lesser response to treatment and lower survival rates [6, 7]. Hypoxic tumors have a significant higher probability of relapse and death [7] and they are resistant to chemotherapy [8]. Chemo-resistance would be mediated by up-regulation of Major Vault Protein (MVP) through the Hypoxia-inducible factor 1 (HIF-1) as shown in previously studies performed in vitro [9]. Hypoxia inhibits the non-homologous end joining (NHEJ) DNA repair through down-regulating Ku70/80 expression, combined with increased angiogenesis and altered p53 expression [2]. Cervical tumors over-expressing MVP also showed down-regulation of Ku70/80 and BAX [10]. MVP over-expression has been associated with a suppression of NHEJ repair, and subsequent genomic instability [10]. These mechanisms would be responsible for tumor progression in cervical carcinoma. Moreover, MVP over-expression was associated to reduced long-term local control in patients who achieved clinical complete response to radio-chemotherapy [11]. The aim of the present study was to assess the relation between the expression of the Major Vault Protein and tumor hypoxia in clinical cervical tumors.

Methods

Forty-three consecutive patients suffering from localized cervix carcinoma were prospectively included in this study from July 1997 to September 2001 [2]. Patients were diagnosed and treated by definitive radiation at the Hospital Universitario Materno-Infantil, at the Hospital Universitario Dr. Negrín and at the Hospital Universitario Insular in Las Palmas de Gran Canaria (Spain). Written informed consent was given previously. The study was approved by the Research and Ethics Committee of our institution. The mean age of the patients was 49.48 ± 12.79 years (median 48, range 29–81 years). Fourteen patients had stage I disease, 22 stage II and 7 stage III-IVA. MVP expression was studied by immunohistochemistry in paraffin-embedded 4 μm sections incubated for the specific primary antibody (MVP, Neomarkers CA, USA). A secondary biotinated antibody (Dako Detection Kit, LSBA) was incubated for 30 minutes, and peroxidase-streptavidin-biotin complex (Dako) was used afterward. Staining was revealed by using diaminobenzidine tetra-hydrochloride substrate (DAB Chromogen; Dako), followed by light counterstaining with Harris hematoxylin as previously described [10]. Data of p53 and angiogenesis, estimated by CD-31 staining, were obtained from our files [2]. Paraffin-embedded tissues from tumor biopsies were available from all patients, and the most representative tumor block was used for immunohistochemical analysis. Blocks were handled as previously described and then incubated for the specific secondary antibody (p53, Clon:DO-7, Novocastra Laboratories Ltd., Newcastle upon Tyne, UK; CD-31 Clon:JC/70A, Dako, Carpintería, CA, USA) [2]. The primary antibody was omitted in one section as a negative control in each set of slides. As a positive control, a strong positive tumor for the oncoprotein was used. Tumor oxygenation was measured by an Eppendorf device following standard criteria as previously described [2, 12] using a polarographic probe system "pO2 Histograph" (Eppendorf AG, Hamburg, Germany). For each set of measurements obtained from tumor, 200 single pO2 values were recorded using at least 6 different electrode tracks. Tumor hypoxia data were reanalyzed for detecting cases of severe hypoxia and the percentage of pO2 values < 2.5 mmHg were obtained from the pooled data and for each individual. Assessment of immunostaining or tumor oxygenation result was blinded to knowledge of the clinical outcome of the patient. Statistical analysis was performed by SPSS 15.0 software.

Results

All immunohistochemical markers and hypoxia values were known in all 43 cases (Figure 1). MVP expression was considered low (negative/slightly positive) in 23 cases and high (strongly positive) in 20 cases. Data of mean vascular density (MVD) and p53 expression were obtained from our files [2] (Table 1). MVD was 49.62 ± 33.98% (median 41%, range 0–160). P53 expression showed a mean value of 39.15 ± 27.62% (median 35%, range 0–92%). Tumor hypoxia was also known in all patients. Mean tumor hypoxic fraction <2.5 mmHg (HF 2.5) values were 35.89 ± 26.80 (median 35.20%, range 0–91.30%). MVP expression was independent of clinical and histological variables, except for adenocarcinoma tumors. In fact adenocarcinoma tumors (5 cases) included in the present study over-expressed MVP versus 15 out of 38 squamous cancers (p = 0.011). Besides, high MVP expression was related to severe hypoxia as determined by higher hypoxic fractions HF (2.5) (45.82 ± 28.00%) compared to low MVP expressing tumors (27.26 ± 22.96%) (p = 0.022) (Figure 2a). Tumors over-expressing MVP also showed increased angiogenesis (65.41 ± 38.38) compared to low expressing cases (35.89 ± 22.55) (p = 0.003) (Figure 2b). MVP expression was independent of p53 protein expression.
Table 1

Characteristics of the patients in the study

Characteristics

All patients

(n = 43)

MVP low

(n = 23)

MVP high

(n = 20)

P value

Age

49.48 ± 12.79

49.47 ± 13.68

49.50 ± 12.04

 
 

(29–81)

(29–81)

(32–72)

0.325

Stage

    

   I

14

5

9

 

   II

22

13

9

 

   III

7

5

2

0.228

Histology

    

   Epidermoid

38

23

15

 

   Adenocarcinoma

5

0

5

0.011

Grade

    

   I

5

3

2

 

   II

19

10

9

 

   III

19

10

9

0.952

p53

39.15 ± 27.62

37.53 ± 28.04

41.02 ± 27.74

 
 

(0–92)

(0–92)

(0–81)

0.685

Vascular density

49.62 ± 33.98

35.89 ± 22.55

65.41 ± 38.38

 
 

(0–160)

(0–113)

(12–160)

0.003

Hypoxic fraction

35.89 ± 26.80

27.26 ± 22.96

45.82 ± 28.00

 
 

(0–91.30)

(0–66.30)

(0–91.30)

0.022

Median pO2

7.61 ± 8.98

7.84 ± 7.85

7.36 ± 10.34

 
 

(0–41.90)

(0–24.30)

(0–41.90)

0.863

Mean ± standard deviation and range are included as well as p53, vascular density, hypoxic fraction and median of pO2

https://static-content.springer.com/image/art%3A10.1186%2F1748-717X-4-29/MediaObjects/13014_2009_Article_166_Fig1_HTML.jpg
Figure 1

Representative immunostaining of MVP (a), p53 (b) and micro-vessels (c).

https://static-content.springer.com/image/art%3A10.1186%2F1748-717X-4-29/MediaObjects/13014_2009_Article_166_Fig2_HTML.jpg
Figure 2

Relationship between (a) MVP and hypoxic fraction (HF 2.5) and (b) mean vascular density.

Discussion

In this study we show for the first time that severe tumor hypoxia is related to high MVP expression in clinical cervical tumors. MVP is ubiquitously expressed and, besides chemotherapy resistance, it has been implicated in the regulation of several cellular processes including transport mechanisms, signal transmissions and immune responses [13]. Previous studies have demonstrated that vaults are up-regulated in different multidrug resistant cancer cell lines [14] and resistance models [15, 16]. Increased levels of MVP have been reported in numerous cell lines after selection with a wide panel of cytostatic drugs (e.g. doxorubicin, methotrexate, vincristine or cisplatin) [17]. By contrast, tumour necrosis factor-either applied externally or after gene transduction, led to down-regulation of MVP transcription [18]. There are several publications concerning to the relationship between MVP expression and drug resistance in clinical oncology [1922]. The role of MVP in clinical outcome after radio-chemotherapy in cervical carcinoma [11] and other cancers [23] has been reported. MVP seems to down-regulate the pro-apoptotic gene BAX through its relation with Ku70/80. Ku70/80 are key genes in the NHEJ repair pathway for radiation-induced DNA double strand breaks. Expression of Ku70/80 has been related to survival in patients treated with x-rays [24, 25]. Ku70/80 is a central regulator of apoptosis by interacting with BAX [26] and BCL-2, which in turn has been shown to suppress Ku, thus inhibiting NHEJ repair [27]. In the clinical setting, up-regulation of MVP by hypoxia is of critical relevance because chemotherapy is currently a standard treatment for those patients. In the other hand, hypoxia inhibits the NHEJ DNA repair through down-regulating Ku70/80 expression [2]. Preclinical studies about the role of hypoxia in cancer cells showed that reduction of pO2 is a favoring factor to increase chemo-resistance [8, 28]. In cancer, hypoxia is an adverse prognostic indicator associated with tumor progression and resistance to therapy [29]. Cellular drug delivery and uptake in hypoxic areas are affected by hypoxia. Some chemotherapeutic drugs require oxygen to generate free radicals that contribute to cytotoxicity. Hypoxia induces cellular adaptations that compromise the effectiveness of chemotherapy. Moreover, the expression of several genes controlling tumor cell survival is regulated by hypoxia (e.g., growth factors governing the formation of new blood vessels and hypoxia-responsive transcription factors modulating the expression of genes). The transcription factor Hypoxia-inducible factor 1 (HIF-1) is one of the principal mediators of homeostasis in human tissues exposed to hypoxia. It is implicated in virtually every process of rapid gene expression in response to low oxygen levels [30]. HIF-1alpha is over-expressed in the majority of common human cancers and their metastases, due to the presence of intratumoral hypoxia and as a result of mutations in genes encoding oncoproteins and tumor suppressor genes [31, 32]. Whether in clinical tumors this chemo-resistance can be reverted by HIF-1 inhibitors deserves to be studied [9]. Pharmacologic manipulation of HIF-1 levels may provide a novel therapeutic approach to diseases like cancer, especially in combination with anti-angiogenic agents [33] that would further reduce tumour oxygenation. Our previously clinical results showed a close relation of clinical hypoxia to increased angiogenesis and in a lesser extent to p53 oncoprotein alteration [2]. Clinical outcome in patients suffering different types of tumours mainly treated by radiation (i.e., cervical and head & neck cancers) depends, at least in part, of those parameters. An increased genetic instability, oncogenic properties, resistance to treatment and increased ability to metastization are expected.

From our results it could be suggested that hypoxia not only induces increased genetic instability, oncogenic properties and metastatization, but through the correlation observed with MVP expression, another pathway of chemo-resistance could be developed.

Conflict of interests

The authors declare that they have no competing interests.

Abbreviations

HIF-1: 

Hypoxia-inducible factor 1

MVD: 

Mean vascular density

MVP: 

Major Vault Protein

NHEJ: 

non-homologous end joining.

Declarations

Acknowledgements

This work was subsidized by grants: FIS 1035/98, 0855/01. Henríquez-Hernández LA, Bordón E and Fontes F were supported by an educational grant from the Instituto Canario de Investigación del Cáncer (ICIC).

Authors’ Affiliations

(1)
Radiation Oncology Department, Hospital Universitario de Gran Canaria Dr. Negrín
(2)
Canary Institute for Cancer Research (ICIC)
(3)
Clinic Sciences Department of Las Palmas de Gran Canaria University (ULPGC)
(4)
Pathology Department, Hospital Universitario de Gran Canaria Dr. Negrín

References

  1. Huang LE, Bindra RS, Glazer PM, Harris AL: Hypoxia-induced genetic instability – a calculated mechanism underlying tumor progression. J Mol Med 2007, 85: 139-148. 10.1007/s00109-006-0133-6View ArticlePubMedGoogle Scholar
  2. Lara PC, Lloret M, Clavo B, Apolinario RM, Bordon E, Rey A, Falcon O, Alonso AR, Belka C: Hypoxia downregulates Ku70/80 expression in cervical carcinoma tumors. Radiother Oncol 2008, 89: 222-226. 10.1016/j.radonc.2008.07.018View ArticlePubMedGoogle Scholar
  3. Wood RD: DNA repair in eukaryotes. Annu Rev Biochem 1996, 65: 135-167. 10.1146/annurev.bi.65.070196.001031View ArticlePubMedGoogle Scholar
  4. Bindra RS, Crosby ME, Glazer PM: Regulation of DNA repair in hypoxic cancer cells. Cancer Metastasis Rev 2007, 26: 249-260. 10.1007/s10555-007-9061-3View ArticlePubMedGoogle Scholar
  5. Tatum JL, Kelloff GJ, Gillies RJ, Arbeit JM, Brown JM, Chao KS, Chapman JD, Eckelman WC, Fyles AW, Giaccia AJ, Hill RP, Koch CJ, Krishna MC, Krohn KA, Lewis JS, Mason RP, Melillo G, Padhani AR, Powis G, Rajendran JG, Reba R, Robinson SP, Semenza GL, Swartz HM, Vaupel P, Yang D, Croft B, Hoffman J, Liu G, Stone H, Sullivan D: Hypoxia: importance in tumor biology, noninvasive measurement by imaging, and value of its measurement in the management of cancer therapy. Int J Radiat Biol 2006, 82: 699-757. 10.1080/09553000601002324View ArticlePubMedGoogle Scholar
  6. Hockel M, Schlenger K, Aral B, Mitze M, Schaffer U, Vaupel P: Association between tumor hypoxia and malignant progression in advanced cancer of the uterine cervix. Cancer Res 1996, 56: 4509-4515.PubMedGoogle Scholar
  7. Fyles AW, Milosevic M, Wong R, Kavanagh MC, Pintilie M, Sun A, Chapman W, Levin W, Manchul L, Keane TJ, Hill RP: Oxygenation predicts radiation response and survival in patients with cervix cancer. Radiother Oncol 1998, 48: 149-156. 10.1016/S0167-8140(98)00044-9View ArticlePubMedGoogle Scholar
  8. Sasabe E, Zhou X, Li D, Oku N, Yamamoto T, Osaki T: The involvement of hypoxia-inducible factor-1alpha in the susceptibility to gamma-rays and chemotherapeutic drugs of oral squamous cell carcinoma cells. Int J Cancer 2007, 120: 268-277. 10.1002/ijc.22294View ArticlePubMedGoogle Scholar
  9. Lelong-Rebel I, Brisson C, Fabre M, Bergerat JP, Rebel G: Effect of pO2 on antitumor drug cytotoxicity on MDR and non-MDR variants selected from the LoVo metastatic colon carcinoma cell line. Anticancer Res 2008, 28: 55-68.PubMedGoogle Scholar
  10. Lloret M, Lara PC, Bordon E, Fontes F, Rey A, Pinar B, Falcon O: Major vault protein may affect nonhomologous end-joining repair and apoptosis through Ku70/80 and bax downregulation in cervical carcinoma tumors. Int J Radiat Oncol Biol Phys 2009, 73: 976-979.View ArticlePubMedGoogle Scholar
  11. Lloret M, Lara PC, Bordon E, Rey A, Falcon O, Apolinario RM, Clavo B, Ruiz A: MVP expression is related to IGF1-R in cervical carcinoma patients treated by radiochemotherapy. Gynecol Oncol 2008, 110: 304-307. 10.1016/j.ygyno.2008.04.034View ArticlePubMedGoogle Scholar
  12. Clavo B, Perez JL, Lopez L, Suarez G, Lloret M, Morera J, Macias D, Martinez JC, Santana M, Hernandez MA, Robaina F, Gunderoth M: Influence of haemoglobin concentration and peripheral muscle pO2 on tumour oxygenation in advanced head and neck tumours. Radiother Oncol 2003, 66: 71-74. 10.1016/S0167-8140(02)00391-2View ArticlePubMedGoogle Scholar
  13. Berger W, Steiner E, Grusch M, Elbling L, Micksche M: Vaults and the major vault protein: novel roles in signal pathway regulation and immunity. Cell Mol Life Sci 2009, 66: 43-61. 10.1007/s00018-008-8364-zView ArticlePubMedGoogle Scholar
  14. Izquierdo MA, Shoemaker RH, Flens MJ, Scheffer GL, Wu L, Prather TR, Scheper RJ: Overlapping phenotypes of multidrug resistance among panels of human cancer-cell lines. Int J Cancer 1996, 65: 230-237. http://creativecommons.org/licenses/by/2.0 10.1002/(SICI)1097-0215(19960117)65:2<230::AID-IJC17>3.0.CO;2-HView ArticlePubMedGoogle Scholar
  15. Izquierdo MA, Scheffer GL, Flens MJ, Shoemaker RH, Rome LH, Scheper RJ: Relationship of LRP-human major vault protein to in vitro and clinical resistance to anticancer drugs. Cytotechnology 1996, 19: 191-197. 10.1007/BF00744212View ArticlePubMedGoogle Scholar
  16. Siva AC, Raval-Fernandes S, Stephen AG, LaFemina MJ, Scheper RJ, Kickhoefer VA, Rome LH: Up-regulation of vaults may be necessary but not sufficient for multidrug resistance. Int J Cancer 2001, 92: 195-202. 10.1002/1097-0215(200102)9999:9999<::AID-IJC1168>3.0.CO;2-7View ArticlePubMedGoogle Scholar
  17. Lange C, Walther W, Schwabe H, Stein U: Cloning and initial analysis of the human multidrug resistance-related MVP/LRP gene promoter. Biochem Biophys Res Commun 2000, 278: 125-133. 10.1006/bbrc.2000.3782View ArticlePubMedGoogle Scholar
  18. Stein U, Walther W, Laurencot CM, Scheffer GL, Scheper RJ, Shoemaker RH: Tumor necrosis factor-alpha and expression of the multidrug resistance-associated genes LRP and MRP. J Natl Cancer Inst 1997, 89: 807-813. 10.1093/jnci/89.11.807View ArticlePubMedGoogle Scholar
  19. Mossink MH, van Zon A, Scheper RJ, Sonneveld P, Wiemer EA: Vaults: a ribonucleoprotein particle involved in drug resistance? Oncogene 2003, 22: 7458-7467. 10.1038/sj.onc.1206947View ArticlePubMedGoogle Scholar
  20. Suprenant KA: Vault ribonucleoprotein particles: sarcophagi, gondolas, or safety deposit boxes? Biochemistry 2002, 41: 14447-14454. 10.1021/bi026747eView ArticlePubMedGoogle Scholar
  21. Steiner E, Holzmann K, Elbling L, Micksche M, Berger W: Cellular functions of vaults and their involvement in multidrug resistance. Curr Drug Targets 2006, 7: 923-934. 10.2174/138945006778019345View ArticlePubMedGoogle Scholar
  22. Izquierdo MA, Scheffer GL, Schroeijers AB, de Jong MC, Scheper RJ: Vault-related resistance to anticancer drugs determined by the expression of the major vault protein LRP. Cytotechnology 1998, 27: 137-148. 10.1023/A:1008004502861PubMed CentralView ArticlePubMedGoogle Scholar
  23. Silva P, West CM, Slevin N, Valentine H, Ryder WD, Hampson L, Bibi R, Sloan P, Thakker N, Homer J, Hampson I: Tumor expression of major vault protein is an adverse prognostic factor for radiotherapy outcome in oropharyngeal carcinoma. Int J Radiat Oncol Biol Phys 2007, 69: 133-140.View ArticlePubMedGoogle Scholar
  24. Beskow C, Kanter L, Holgersson A, Nilsson B, Frankendal B, Avall-Lundqvist E, Lewensohn R: Expression of DNA damage response proteins and complete remission after radiotherapy of stage IB-IIA of cervical cancer. Br J Cancer 2006, 94: 1683-1689.PubMed CentralPubMedGoogle Scholar
  25. Wilson CR, Davidson SE, Margison GP, Jackson SP, Hendry JH, West CM: Expression of Ku70 correlates with survival in carcinoma of the cervix. Br J Cancer 2000, 83: 1702-1706. 10.1054/bjoc.2000.1510PubMed CentralView ArticlePubMedGoogle Scholar
  26. Amsel AD, Rathaus M, Kronman N, Cohen HY: Regulation of the proapoptotic factor Bax by Ku70-dependent deubiquitylation. Proc Natl Acad Sci USA 2008, 105: 5117-5122. 10.1073/pnas.0706700105PubMed CentralView ArticlePubMedGoogle Scholar
  27. Wang Q, Gao F, May WS, Zhang Y, Flagg T, Deng X: Bcl2 negatively regulates DNA double-strand-break repair through a nonhomologous end-joining pathway. Mol Cell 2008, 29: 488-498. 10.1016/j.molcel.2007.12.029PubMed CentralView ArticlePubMedGoogle Scholar
  28. Cosse JP, Michiels C: Tumour hypoxia affects the responsiveness of cancer cells to chemotherapy and promotes cancer progression. Anticancer Agents Med Chem 2008, 8: 790-797.View ArticlePubMedGoogle Scholar
  29. Shannon AM, Bouchier-Hayes DJ, Condron CM, Toomey D: Tumour hypoxia, chemotherapeutic resistance and hypoxia-related therapies. Cancer Treat Rev 2003, 29: 297-307. 10.1016/S0305-7372(03)00003-3View ArticlePubMedGoogle Scholar
  30. Adams JM, Difazio LT, Rolandelli RH, Lujan JJ, Hasko G, Csoka B, Selmeczy Z, Nemeth ZH: HIF-1: a key mediator in hypoxia. Acta Physiol Hung 2009, 96: 19-28. 10.1556/APhysiol.96.2009.1.2View ArticlePubMedGoogle Scholar
  31. Semenza GL: Expression of hypoxia-inducible factor 1: mechanisms and consequences. Biochem Pharmacol 2000, 59: 47-53. 10.1016/S0006-2952(99)00292-0View ArticlePubMedGoogle Scholar
  32. Zhong H, De Marzo AM, Laughner E, Lim M, Hilton DA, Zagzag D, Buechler P, Isaacs WB, Semenza GL, Simons JW: Overexpression of hypoxia-inducible factor 1alpha in common human cancers and their metastases. Cancer Res 1999, 59: 5830-5835.PubMedGoogle Scholar
  33. Bergers G, Javaherian K, Lo KM, Folkman J, Hanahan D: Effects of angiogenesis inhibitors on multistage carcinogenesis in mice. Science 1999, 284: 808-812. 10.1126/science.284.5415.808View ArticlePubMedGoogle Scholar

Copyright

© Lara et al; licensee BioMed Central Ltd. 2009

This article is published under license to BioMed Central Ltd. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.