Amino acids change liver growth factors gene expression in malnourished rats

Los aminoácidos alteran la expresión genética de factores de crecimiento hepático en ratas macho desnutridas

R. Passos de Jesus¹, L. De Nardi², N. Da Rós³, S. Salaorni³, M.^a A. Nagai³, M.^a Mitzi Brentani³, D. Akamine⁴, and D. L. Waitzberg²

¹Nutrition School of Federal University of Bahia. Nutrition Science Department. Bahia. Brazil.
²University of Sao Paulo. School of Medicine. Department of Gastroenterology. Digestive Surgery Division. LIM 35. Sao Paulo. Brazil.
³University of Sao Paulo. School of Medicine. Laboratory of Experimental Oncology. LIM-24. Radiology Department. Sao Paulo. Brazil. ]]> ⁴Farmoterápica. Manipulação de Nutrição Parenteral Ltda. Sao Paulo. Brazil

This study was funded by a grant from Fundação de Apoio a Pesquisa do Estado de São Paulo (FAPESP), project N^o 98-11856-1. This study was supported by Fresenius-Kabi, who provided Dipeptiven used in this study, Farmoterápica, who provided the solution containing a mixture of alanine plus proline and Abbott Laboratories who supplied infusion pumps. The authors thank Prof. Dr. Neil Ferreira Novo from the Federal University of Sao Paulo for the statistical analysis, and the Central Pharmacy of Hospital das Clinicas de Sao Paulo for pharmaceutical support.

Correspondence

ABSTRACT

Background: Glutamine and proline are metabolized the liver and may collaborate on its regeneration. Parenteral nutrition (PN) containing either glutamine or proline was given to partially hepatectomized rats. The total RNA content and growth factor gene expression in hepatic remnants was measured, to determine the effects of these amino acid supplementation on the expression ofgrowth factors during liver regeneration.
Methods: Wistar rats nourished (HN) and malnourished (HM) were hepatectomized and divided in two groups: 20 receiving PN enriched with Alanyl-Glutamine (HN-Gln and HM-Gln) and 20 PN enriched with proline+alanine (HN-Pro and HM-Pro). The control groups comprised 7 nourished (CN) and 7 malnourished (CM) rats that didn't undergo surgery. Growth factor and thymidine kinase mRNA levels were measured by RT-PCR.
Results: In nourished rats, total hepatic RNA levels were lower in the HN-Gln and HN-Pro groups (0.75 and 0.63 μg/mg tissue, respectively) than in control group (1.67 μg/mg tissue) (P < 0.05). In malnourished rats, total hepatic RNA content was higher in the HM-Pro group than HN-Pro, HM-Gln, and CM (3.18 vs. 0.63, 0.93 and 1.10 μg/mg, respectively; P < 0.05). Hepatocyte growth factor mRNA was more abundant in the HM-Gln group when compared to CM (0.31 vs. 0.23 arbitrary units) and also in HM-Pro in relation to HM-Gln, HN-Pro, and CM(0.46 vs. 0.33 and 0.23, respectively, P < 0.05).
Conclusions: Proline or glutamine supplementation in malnourished rats improves total RNA content in the remnant hepatic tissue. Amino acids administration increased HGF gene expression after partial hepatectomy in malnourished rats, with a greater effect of proline than glutamine.

RESUMEN

Introducción: La glutamina y la prolina de metabolizan en el hígado y pueden contribuir a la regeneración de este. Se administró nutrición parenteral con glutamina o prolina a ratas sometidas a hepatectomía parcial. Se midieron el contenido de ARN total así como la expresión genética del factor de crecimiento en el tejido hepático remanente, con el objetivo de determinar los efectos que provocaban estos aminoácidos en la expresión genética de factores de crecimiento durante el proceso de regeneración del hígado.
Métodos: ratas macho Wistar nutridas (HN) y desnutridas (HM) se sometieron a hepatectomía parcial y se dividieron en dos grupos: 20 recibieron nutrición parenteral enriquecida con Alanil-Glutamina (HN-Gln y HM-Gln) y 20 nutrición parenteral enriquecida con prolina+alanina (HNPro y HM-Pro). Los grupos de control estaban formados por 7 ratas nutridas (CN) y 7 desnutridas (CM) que no se sometieron a la cirugía. Los niveles de factor de crecimiento y timidina quinasa mRNA se midieron por RT-PCR.
Resultados: En las ratas nutridas, lo niveles de ARN hepático total fueron inferiores en los grupos que recibieron HNGln y HN-Pro (0,75 y 0,63 mg/mg tejido, respectivamente) que en el grupo control (1,67 mg/mg tejido) (P < 0,05). En las ratas desnutridas, el contenido total de ARN hepático fue superior en el grupo que recibió HM-Pro que el que recibió HN-Pro, HM-Gln y CM (3,18 vs 0,63, 0,93 y 1,10 mg/mg, respectivamente; P < 0,05). El mRNA del factor de crecimiento de hepatocitos fue más abundante en el grupo que recibió HM-Gln con respecto a CM (0,31 vs 0,23 unidades arbitrarias), así como en HM-Pro con relación a HM-Gln, HN-Pro y CM (0,46 vs 0,33 y 0,23, respectivamente, P < 0,05).
Discusión: La suplementación de prolina o glutamina en ratas macho desnutridas mejora el contenido total de ARN en el tejido hepático remanente. Al administrarse aminoácidos aumentó la expresión genética del HGF tras someter las ratas desnutridas a una hepatectomía parcial, obteniéndose un resultado mejor con prolina que con glutamina.

Palabras clave: Aminoácidos. Nutrición parenteral. Hígado regeneración. Prolina. Glutamina.

Introduction

Advances in surgical technique have increased the frequency of partial hepatectomy (PH) to remove malignant neoplasia.¹ Since the nutritional status of the patient is frequently compromised, a specialized nutrition during the perioperative period may improve the clinical course, support weight gain and reduce infectious morbidity.^2,3 Liver regeneration induced by hepatic resection occurs through hyperplasia and compensatory cellular hypertrophy of the remaining lobes.⁴ It is controlled and mediated by intra and extra hepatic factors such as hormones, neurotransmitters, nutrients, proto-oncogenes, and growth factors.^5,6.

Proline is catabolized primarily in the liver, and when completely metabolized to glucose and urea, ATP is produced for gluconeogenesis and ureogenesis.⁸ Until the present moment, it is not totally clear how glutamine could influence cellular metabolism and hepatic regeneration in nourished animals after partial hepatectomy.^9-11 In this study, we determine the effect of glutamine or proline enriched parenteral nutrition on the content of total RNA, hepatocyte growth factor (HGF), transforming growth factor-alfa (TGF-α, thymidine kinase (TK) and mRNA levels in the hepatic remnants of well-nourished and malnourished rats.

Methods

All experimental procedures were approved by the Ethical Research Committee (CAPPesq), School of Medicine, University of Sao Paulo, (Sao Paulo, Brazil). All animals received appropriate care throughout the experimental procedures.

Experimental protocol

Male Wistar rats, mean body weight 250 g, from the University of São Paulo Medical School animal house (Centro de Bioterismo, FMUSP) were kept at room temperature with 12 hours light cycles. Ten days before the experiment, the rats were transferred to individual metabolic cages, with water ad libitum. Fifty four Wistar rats nourished (HN) and malnourished (HM) were hepatectomized and divided in two groups as previously described in an identical protocol.⁷ Briefly, twenty rats that received parenteral nutrition enriched with Ala-Gln (10-HN-Gln and 10-HM-Gln) and twenty rats enriched with proline plus alanine (10-HN-Pro and 10-HM-Pro). The control groups comprised seven nourished (CN) and seven malnourished (CM) rats that did not undergo surgery.

The nourished groups were fed with normocaloric (100 kcal/day) and normoprotein oral diet (14% of the total caloric value as casein) for ten days.¹² The malnourished group received a low-protein oral diet (4% of the total calorie value as casein) for ten days, until 10-15% of the initial body weight was lost.⁷.

After these periods (nourished or malnourished), all animals were anesthetized with intraperitoneal injection of ketamine chlorhydrate (80 mg/kg of body weight) and submitted to jugular vein catheterization and partial hepatectomy (PH) under sterile conditions. Central vein cannulation was performed according to a standard technique. Rats were submitted to PH through 68-70% total liver tissue extirpation, including the median and left lateral hepatic lobes.¹³ After PH, 3.0 mL of 10% glucose and 1.0 mL of 10% sodium bicarbonate solutions were intravenously injected in the animal.⁷

All experimental animals received isocaloric and isonitrogenic intravenous parenteral nutrition (PN) by a peritaltic pump for 96 h. They were randomly selected to receive non-fat parenteral nutrition regimens containing 188 kcal/kg/day and 1.12 g N/kg/day, with 10% amino acid standard solution, supplemented with 10.4 g of Lalanyl-L-glutamine dipeptide (Alanine = 3.94 g and Glutamine = 6.46 g; Ala-Gln; Dipeptiven^®, Fresenius-Kabi-Germany) or a solution containing a mixture of 11.2g of alanine and proline (Alanine = 4.88 g and Proline = 6.32 g; Ala/Pro, Farmoterápica-Brazil). The experimental animals received an average of 5.14 g/kg/day of glutamine and 4.29 g/kg/day of proline.⁷

The nourished and malnourished rats of the control groups (CN and CM) were not submitted to hepatectomy or catheterization. Four days after the adaptation period (10 days), with standard oral diet, control rats were anesthetized and laparotomized for total liver extirpation.

RNA extraction

Total RNA was extracted from liver remnant as follows. Approximately 0.15 g of hepatic tissue liquid nitrogen powdered tissue was resuspended in 1 ml of lysis buffer (4 M guanidinium isothiocyanate, 25 mM sodium citrate, pH 7, 0.5% sarcosyl, and 100 mM β-mercaptoethanol), and homogenized. Total RNA was extracted with phenol-chloroform and isopropanol precipitation as described.¹⁴ The total RNA concentration was obtained by spectrophotometry at 260 nm.¹⁴ The concentration of total RNA, is reported as μg/μL, or μg/mg of hepatic tissue.

RT-PCR analysis

Total RNA quality and integrity were assessed by electrophoresis on a formaldehyde-agarose gel.¹⁴ All RNA samples were treated with DNAse I (Amersham-Pharmacia) for 30 minutes at 37^oC to eliminate genomic DNA contamination. One mg of DNasetreated RNA was used to measure thymidine kinase, HGF-α, TGF-α and mRNA. by RT-PCR with a Superscript One-step Platinum^® Taq kit (Invitrogen^®).¹⁵ β₂-microglobulin (bMG) amplification served as an internal control for RNA quantization. Amplification in the absence of reverse transcriptase was a negative control. RT-PCR products were separated by electrophoresis on 1.5% agarose gels stained with ethidium bromide¹⁶. The numbers of amplification cycles and the annealing temperatures used in the RT-PCR were as follows: HGF-α (26, 60 ^oC); TGF (22, 55 ^oC) and TK (28, 60 ^oC).^17,18

Primers based on rat sequences were designed for cDNA synthesis and for PCR amplification of the HGF, TGF-α and TK genes. The primer sequences and the predicted product sizes in base pairs (bp) were as follows: HGF sense 5'CCC GGT GCT GCA GCA TGT CCT3' and antisense 5'TCC CCT CGA GGA TTT CGA CAG3', 530 bp; TGF-α sense 5'GAC AAG TTG AAC AAG AAC CTC3' and antisense 5'CGT CAT CCA CCT AAT ACA TAA G3', 250 bp; TK sense 5'GCA GAT CCA GGT GAT TCT CG and antisense 5'ATC GAT GCC AAT GAC AGC CA3', 234 bp.^16,18

Statistical analysis

After densitometry measurement and normalization with βMG, each RT-PCR result was subjected to statistical analysis. Results were analyzed by nonparametric tests appropriate for the distribution and type of variables studied, using SPSS software. P values ≤ 0.05 were considered significant. Data are presented as the mean ± SD, in arbitrary units.

Kruskal-Wallis method was used to compare levels of hepatic gene expression among the different subgroups (Gln, Pro, and control without PH).¹⁹

Results

Total RNA content

Partial hepatectomy (PH) and amino acid parenteral infusion influenced total RNA. The maximum percentage of total RNA among the experimental groups was: HM-Pro = 289% (3.18 μg/mg hepatic tissue), HM-Gln = 84% (0.93 μg/mg), HN-Gln = 45% (0.75 μg/mg) and HN-Pro = 38% (0.63 μg/mg). In relation to control group: HN-Gln and HN-Pro < CN; HM-Gln = CM; HM-Pro > CM and CN > CM (P < 0.05) and when comparing hepatectomized rats HN-Gln > HN-Pro; HN-Pro < HM-Pro; HN-Gln = HM-Gln; HM-Gln < HM-Gln (P < 0.05), as shown in figure 1.

RT-PCR reactions

We performed semi-quantitative RT-PCR to determine the levels of HGF, TGF-α and TK gene expression.

Different effects on HGF expression was observed after PH, when considering the animals' nutritional status. In malnourished rats, a significant increase in HGF expression was demonstrated, when compared to malnourished control group (HM-Gln = 0.31 and HMPro = 0.46 AU > CM, 0.23 AU, with P < 0.05). In nourished animals, the opposite was observed and a significant reduction in HGF expression was demonstrated, when compared to nourished control group (HN-Gln = 0.26 and HN-Pro = 0.33 AU < CN, 0.48 AU, with P <0.05) - figure 2.

Malnourished rats, independently of the parenteral nutrition received, demonstrated higher HGF expression than nourished rats (P < 0.05). Nevertheless, the nourished control group had significant more expression of HGF than malnourished control group (0.48 vs 0.23 AU, respectively, with P < 0.05).

All the groups have TGF-α and TK positive expression, but without statistic difference between then.

Discussion

The clinical results after PH depend on the regenerative liver process, liver resection extension, nutritional status, and per operative nutritional therapy.³ The majority of patients considered candidates to hepatic resection may be malnourished and with protein depletion. Due to the depleted nutritional state, most of these patients are submitted to specialized nutritional therapy. As specific amino acids may have a particular role at the liver regeneration, it is worthwhile to study their effect in an experimental malnourished liver resection model, looking after specific changes in liver regeneration gene expression. We have previously observed that the hepatic regeneration index and hepatic growth percentage of malnourished rats subjected to PH using parenteral nutrition with glutamine or proline were higher than in control groups without hepatectomy.⁷ The subsequent step was to study, in the same experimental model, the effects of parenteral glutamine and proline infusion on liver RNA content and hepatic regeneration gene.

In this study, our results have demonstrated that the nutritional status and the parenteral infusion supplemented with glutamine or proline could influence RNA content and gene expression involved in the hepatic regeneration. Parenteral nutrition supplemented with proline in malnourished rats (HM-Pro) protected against the reduction in content RNA, usually occurring after PH (fig. 1).

TGF-α and HGF are the main growth factors involved in the hepatic regeneration. HGF is considered to be the most powerful stimulator of hepatocyte proliferation, acting through paracrine and endocrine mechanisms to stimulate cellular mitosis. TGF-α probably participates in the later stages and acts through an autocrine mechanism.⁵ TK levels are usually very low in normal hepatic tissue. However, 12 to 24 hours after PH, a significant increase is observed in this enzyme level which remained elevated for 60 hours and declined to reach the baseline about 7 to 8 days. This event coincides with the greater cellular division activity, observed during hepatic regeneration.¹⁶.

Therefore, we performed semi-quantitative RT-PCR to determine the levels of HGF, TGF-α, TK and mRNAs. The TGF-α and TK expression was verified for all groups with no statistical differences between them.

We found higher total RNA values in the malnourished group given parenteral nutrition (PN) supplemented with alanine plus proline, probably because the liver uses proline efficiently, even under conditions of amino acid deprivation.^9,21 Our results suggest that in the malnourished rats, supplementation of PN with proline probably favored the synthesis of total RNA during regeneration, possibly by triggering compensatory mechanisms for better use of the proline, with consequent economy in energy and amino acid use for transcription, protein synthesis, and proliferation.

The higher concentration of total RNA obtained in the HM-Pro group in relation to the HN-Pro group indicates a higher capacity for protein synthesis in the malnourished animals given PN enriched with proline. Proline might be related to the increase in total RNA, that reflects protein synthesis, and the possibility that this amino acid is being used in the production of collagen and not necessarily for hepatocyte proliferation.

Following PH, the genes involved in the expression of acute phase proteins and gluconeogenesis are quickly over expressed in order to restore metabolic homeostasis and to repair the tissue damage to the hepatic remnant.²² An immediate cellular response to surgical trauma is fundamental to induce hepatocytes entering in the cell cycle.²³ Therefore our experimental model was not capable of demonstrating differences in TGF-α expression among the various experimental groups, probably because our samples were collected 96 h after partial hepatectomy, a point at which TGF-α mRNA levels have decreased gradually until they reached basal values, around the 6th postoperative day.¹⁷.

We also found that HGF mRNA levels of HM-Pro group exceeded those of the HN-Pro group, while HGF expression in CM group was lower than in the CN group, suggesting that in these animals, the nutritional state and PH have a significant influence on HGF expression. In addition, proline administration increased the HGF gene expression after partial hepatectomy in malnourished rats, more than glutamine. See figure 2 for more details.

In this study, we used the liver remnant, composed 90% of hepatocytes, for RNA extraction. Nonparenchymatous cells are the main secretors of HGF during hepatic regeneration^24,25 so, the use of total RNA origin from these cells as opposed to total RNA extracted from the residual liver, optimizes the amount of amplified HGF product, thereby improving the sensitivity of the method. This explains the significant differences between our results and those of Masson et al (1999), in which the RT-PCR reactions were carried out with RNA extracted from non-parenchymatous cells isolated from the residual liver.¹⁷.

Since changes in TK gene expression can be related to growth and cell proliferation,^16,26 we measured TK mRNA levels 96 h after PH to determine whether TK over expression preceded a peak in DNA synthesis in this period, in order to identify the influence of parenteral infusion of amino acids on the later phases of hepatic regeneration. However, our results did not demonstrate any alterations in TK gene expression in this period in the various experimental groups, confirming that the increase in enzymatic activity up to 96 hours after partial hepatectomy does not occur in association with an increase in TK mRNA, which does occur earlier.¹⁷

Our results from nourished and malnourished hepatectomized rats submitted for 96 hours to PN supplemented with dipeptide Ala-Gln or Ala/Pro lead to the following conclusions: (i) hepatic total RNA levels were lower in nourished animals submitted to partial hepatectomy, independently of amino acid infusion; (ii) proline or glutamine supplementation in malnourished rats improves total RNA content in the remnant hepatic tissue; (iii) the administration of amino acids probably increased HGF gene expression after partial hepatectomy in malnourished rats, with proline having a greater effect than glutamine; and (iv) parenteral supplementation with glutamine or proline did not affect the levels of TGF-αand TK mRNAs in the hepatic remnant.

Finally, further studies are necessary to clarify the metabolic and molecular functions of amino acids, such as glutamine and proline, during hepatic regeneration under various nutrition conditions. These studies could involve the administration of nutrients, different models of hepatectomy, and the use of sensitive methods (in situ hybridization or microarray) to study the remnant hepatic tissue. The results of these studies could provide information on the safety and efficacy of the administration of amino acids during hepatic regeneration.

1. Livraghi T. Guidelines for treatment of liver cancer. Eur J Ultrasound 2001; 13: 167-76.        [ Links ]

2. Fan ST, Lo CM, Lai EC, Chu KM, Liu CL, Wong J. Perioperative nutritional support in patients undergoing hepatectomy for hepatocellular carcinoma. N Engl J Med 1994; 331: 1547-52.        [ Links ]

3. Ziegler TR. Perioperative nutritional support in patients undergoing hepatectomy for hepatocellular carcinoma. JPEN J Parenter Enteral Nutr 1996; 20: 91-2.        [ Links ]

4. Arias M, Irwin M, Boyer JL, Fausto N. Hepatic regeneration. In: Fausto N, editors. The liver: biology and pathobiology. New York: Raven Press; 1994: 32-53.        [ Links ]

5. Michalopoulos GK, DeFrances MC. Liver regeneration. Science 1997; 276: 60-6.        [ Links ]

6. Jesus RP, Waitzberg DL, Campos FG. Regeneração hepática: papel dos fatores de crescimento e nutrientes. Rev Ass Med Brasil 2000; 46: 242-54.        [ Links ]

7. Passos de Jesus Mazza R, Bertevello PL, Torrinhas RSMM et al. Effect of glutamine dipeptide on hepatic regeneration in partially hepatectomized malnourished rats. Nutrition 2003; 19: 930-5.        [ Links ]

8. Hensgens HESJ, Meijer AJ, Williamson JR, Gimpel JA, Tager JM. Proline metabolism in isolated rat liver cells. Biochem J 1978; 170: 699-707.        [ Links ]

9. Ito A, Higashiguchi T. Effects of glutamine administration on liver regeneration following hepatectomy. Nutrition 1999; 15: 2328.        [ Links ]

10. Yamaguchi T, Minor T, Isselhard W. Effect of glutamine or glucagon-insulin enriched total parenteral nutrition on liver and gut in 70% hepatectomized rats. J Am Coll Surg 1997; 185: 156-62.        [ Links ]

11. Yoshida S, Yunoki T, Aoyagi K et al. Effect of Glutamine supplement and hepatectomy on DNA and protein synthesis in the remnant liver. J Surg Res 1995; 59: 475-81.        [ Links ]

12. Reeves PG, Nielsen FH, Fahey Jr GC. AIN-93: purified diets for laboratory rodents: final report of the American Institute of Nutrition ad hoc writing Committee on the Reformulation of the AIN-76A Rodent Diet. J Nutr 1993; 123: 1939-51.        [ Links ]

13. Higgins GM, Anderson RM. Experimental pathology of the liver. Arch Pathol 1931; 186: 186-202.        [ Links ]

14. Chomczynski P, Sacchi N. Single-step method of RNA isolation by acid guanidium thiocyanate phenol chloroform extraction. Anal Biochem 1987; 162: 159-69.        [ Links ]

15. Drosten C, Seifried E, Roth WK. TaqMan 5'-nuclease human immunodeficiency virus type 1 PCR assay with phagepackaged competitive internal control for high-throughput blood donor screening. J Clin Microbiol 2001; 39: 4302-8.        [ Links ]

16. Masson S, Daveau M, Hiron M et al. Differential regenerative response and expression of growth factors following hepatectomy of variable extent in rats. Liver 1999; 19: 312-7.        [ Links ]

17. Masson S, Scotte M, Francois A et al. Changes in growth factor and cytokine mRNA levels after hepatectomy in rat with CCl(4)-induced cirrhosis. Am J Physiol 1999; 277(4 Pt 1):G838-46.        [ Links ]

18. Demetris AJ, Murase N, Ye Q et al. Analysis of Chronic Rejection and Obliterative Arteriopathy. Possible contributions of Donor antigen-Presenting Cells and Lymphatic Disruption. Am J Pathol 1997; 150: 563-78.        [ Links ]

19. Siegel S, Castellan Jr N J. Nonparametric statistics. 2nd ed. New York: Mac Graw-Hill; 1988.        [ Links ]

20. Carrillo MC, Carnovale C E, Favre C, Monti J A, Scapini C. Hepatic protein synthesis and serum aminoacid levels during liver regeneration in young and old malnourished rats. Mech Ageing Dev 1996; 91: 55-64.        [ Links ]

21. Watanabe M, Sugimura K, Yamanoha B. Effect of acute deficiency of dietary proline on proline balance in the rat small intestine and liver. J Anim Physiol 1999; 82: 294-304.        [ Links ]

22. Leu JI, Crissey MA, Leu JP, Ciliberto G, Taub R. Interleukin-6-induced STAT3 and AP-1 amplify hepatocyte nuclear factor 1-mediated transactivation of hepatic genes, an adaptive response to liver injury. Mol Cell Biol 2001; 21: 414-24.        [ Links ]

23. Xu W, Wang S, Wang G, Wei H, He F, Yang X. Identification and characterization of differentially expressed genes in the early response phase during liver regeneration. Biochem Biophys Res Commun 2000; 278: 318-25.        [ Links ]

24. Matsuda Y, Nakamura T. Molecular biology of hepatocyte growth factor (HGF). Nippon Rinsho 1993; 51: 435-45. [Abstract].        [ Links ]

25. Matsumoto K, Nakamura T. Hepatocyte growth factor: molecular structure, roles in liver regeneration and other biological functions. Crit Rev Oncog 1992; 3: 27-54.        [ Links ]

26. Beyer HS, Sherman R, Zieve L. Aging is associated with reduced liver regeneration and diminished thymidine kinase mRNA content and enzyme activity in the rat. J Lab Clin Med 1991; 117: 101-8.        [ Links ]

Correspondence: ]]> Dan L. Waitzberg.
Nutrition and Metabolic Digestive Surgery Laboratory.
School of Medicine. Gastroenterology Department.
University of Sao Paulo.
Avda. Dr. Arnaldo, 455 sala 2208.
CEP 01246-903 Sao Paulo SP. Brazil.
E-mail: dan@ganep.com.br

Recibido: 8-IV-2009.
Aceptado: 7-IX-2009.