Aspects of gene regulation in the diploid and tetraploid Odontophrynus americanus ( Amphibia , Anura , Leptodactylidae )

Erythropoietic and hemoglobin DNA transcriptional activities were analyzed in the diploid and the tetraploid Odontophrynus americanus. Flow cytometric analyses of DNA, RNA and mitochondrial contents showed increased genic activity in both diploid and tetraploid animals during erythropoiesis in vivo elicited by pretreatment phenylhydrazine. Generally, higher values were seen in immature tetraploid erythroid cells. On the 10th day of recovery from anemia, large amounts of messenger RNA were found in both specimens. Based on the mitochondrial content, the tetraploid cells had more intense energy metabolism than the diploid cells. Diploid O. americanus had about three times more erythroid cells than tetraploid specimens, indicating that there were differences in the regulatory mechanisms of erythroid cells. Hematological parameters showed that tetraploid cells had 30% more hemoglobin than the diploid, suggesting a regulatory mechanism of hemoglobin synthesis at the transcriptional level. Cytoplasmic inclusions resembling Heinz bodies were found in both typ es of cells. In the tetraploid cells they were previously found associated with RNA or RNP, suggesting that other regulatory system which co ntrols the accumulation of nontranslated RNA transcribed in excess must be present. These differences at the physiological and molecul ar levels during erythropoiesis reinforce the hypothesis that speciation is occurring between diploid and tetraploid O. mericanus . Laboratório de Genética, Instituto Butantan, Av. Vital Brazil, 1500, 05503-900 São Paulo, SP, Brasil. Send correspondence to A.M.C. Fax: +55-11-815-1505. E-mail: amcianciarullo@hotmail.com Laboratório de Hemoglobinas e Hemoglobinopatias, Departamento de Biologia, UNESP, São José do Rio Preto, Av. Cristóvão Colombo, 2265, 15054-000 São José do Rio Preto, SP, Brasil. Departamentos de Protozoologia and Ultra-estrutura e Biologia Celular, Instituto Oswaldo Cruz/FIOCRUZ, Av. Brasil, 4365, 21045-900 Rio de Janeiro, RJ, Brasil. 358 Cianciarullo et al. sualize immature forms of erythroid cells. Smears were counterstained with Rosenfeld’s dye (Rosenfeld, 1947), to differentiate the cytoplasmic and nuclear acidophily and basophily. The distinct stages of erythropoiesis, as a function of time, were then determined by light microscopy.


INTRODUCTION
Ecological, biogeographical, and evolutionary processes encompass several categories of individual or populational responses that may affect the survival of a species or the rise of a new species (Endler, 1977).Genomic evolution by polyploidy is observed in plants (Stebbins, 1950) and animals (Beçak et al., 1966).When doubling of the entire chromosomal set occurs simultaneously, adaptive polymorphism will be fixed.This mechanism, which produces great genetic variability in a relatively fast manner, seems to have played an important role in the evolution of the lower vertebrates (Beçak, 1968;Beçak and Beçak, 1998).However, the question remains as to how gene expression is regulated in polyploid systems.Several hypotheses have been proposed but not fully demonstrated experimentally (Beçak and Pueyo, 1970;Cortadas and Ruiz, 1988;Ruiz and Brison, 1989;Álvares et al., 1998).In this context, Odontophrynus americanus, a South American frog, may provide a useful experimental genetic model.This species is a fossorial-terrestrial anuran and was the first reported example of a bisexual natural polyploid species of vertebrate (Beçak et al., 1966).Diploid and tetraploid cryptic specimens occur in allopatric populations of this frog in São Paulo State, Brazil.The aim of the present study was to examine erythrocyte maturation in vivo in a relatively synchronized polyploid system, to quantify some of the genic products synthesized during distinct stages of the erythropoiesis elicited by phenylhydrazine, and to correlate these results with gene regulation and speciation.sualize immature forms of erythroid cells.Smears were counterstained with Rosenfeld's dye (Rosenfeld, 1947), to differentiate the cytoplasmic and nuclear acidophily and basophily.The distinct stages of erythropoiesis, as a function of time, were then determined by light microscopy.

Flow cytometry
Analyses of 15,000 cells for each normal and anemic blood sample from four diploid and four tetraploid frogs were made with an EPICS-751 flow cytometer (Coulter Electronics, USA), using the following parameters: cell size after red cell sphering (Mohandas et al., 1986), DNA and RNA content by acridine orange staining (Traganos et al., 1977) and mitochondrial content by rhodamine 123 staining (Darzynkiewicz et al., 1982).The intensity of red fluorescence emission, F >600 (RNA, mitochondria) , measured within a 600-650 nm band, and the intensity of green fluorescence emission, F >530 (DNA) , measured within a 515-575 nm band, were obtained for cells excited at sharply focused 488 nm by an argon ion laser beam and recorded by separate photomultipliers.Controls were performed by incubating cell samples for 20 min at 37 o C with 0.004% RNase diluted in PBS, pH 7.2, and by omission of the respective fluorochromes in each measurement.
Total RNA extraction Blood of two normal and two anemic (10 days after anemia induction) adult Bufo ictericus toads was used as a control for the RNA synthesis, because of the great volume of blood these animals can provide.Total RNA extraction was performed according to Chomczynski and Sacchi (1987).

Transmission electron microscopy
Erythroid cells were fixed in a gradient concentration of 4 and 2% glutaraldehyde in 0.2 M and 0.1 M phosphate buffer, respectively (Brunner Jr. et al., 1975), postfixed in 1% OsO4 diluted in the same buffer, contrasted en bloc with 2% uranyl acetate in aqueous solution, dehydrated in acetone series and embedded in Epon 812 resin.Thin sections were contrasted with lead citrate and examined in a Zeiss EM 109 transmission electron microscope.

Determination of hematological parameters
Classic clinical hematology procedures (Oliveira Lima et al., 1992) were applied to four normal diploid and tetraploid frogs.Red cells were counted in a Neubauer chamber using Hayen's liquid.Hematocrit was determined in order to establish the erythrocyte concentration (%) in blood and to determine hematimetric values such as the mean corpuscular volume (MCV), mean corpuscular hemoglobin (MCH), and mean corpuscular hemoglobin concentration (MCHC).

Recovery from hemolytic anemia
Flow cytometry of blood samples from diploid and tetraploid O. americanus frogs before and after the induction of anemia revealed a small nonsignificant difference in the size of the erythroid cells in the two populations, with tetraploid cells usually being larger.Furthermore, immature cells were smaller than mature erythrocytes (Figure 1).These results were confirmed by light microscopy (data not shown).
The DNA, RNA and mitochondrial contents were quantified indirectly by measuring the incorporation of fluorescent dye into these structures (Figure 2A-C).DNA levels were usually proportional to the ploidy number.At all time points analyzed, the tetraploid cells had higher amounts of DNA compared to the diploid cells.However, the measurements were not uniform since they included cells in different stages of DNA synthesis and mitosis.Tetraploid cells showed a first DNA peak on the 10th day after the induction of anemia and a new, more pronounced peak at the 20th day, thus representing two distinct waves of erythropoiesis (Figure 2A).Diploid cells had a single peak 15-20 days after the induction of anemia and maintained these levels until the 30th day.After 50 days, the DNA content returned to normal levels in both diploid and tetraploid cells.The controls were negative (zero) for all measurements performed (data not shown).
The tetraploid cells produced two distinct RNA peaks, one on the 10th day and other on the 20th day after the induction of anemia, thus coinciding with the DNA peaks (Figure 2B).The diploid cells produced only one peak on the 10th day, but this was less intense than that of the tetraploid cells.Total RNA extraction of blood collected from the diploid Bufo ictericus obtained on the 10th day after the induction of anemia confirmed the data obtained by flow cytometry.A high RNA concentration in the 28S, 18S and 5S bands was found at this time (Figure 3: lane 2).The control, using blood samples from non-anemic toads, showed no detectable RNA under the same experimental conditions (Figure 3: lane 1).
The relative mitochondrial content showed that the tetraploid cells had a higher number of mitochondria (or larger ones) than the diploid cells at all time points (Figure 2C).A small increase in fluorescence was observed on the 10th day of recovery from anemia in both cell types, and this was maintained up to the 50th day.
The data obtained by flow cytometry and light microscopy were confirmed at the ultrastructural level, where a correlation between the erythrocyte maturation stage, the presence of cytoplasmic organelles and the variations in cell size was possible (Cianciarullo et al., 2000).The cyto-plasmic inclusions previously found in the tetraploid cells (Cianciarullo and Meirelles, 1994) were now detected in both diploid and tetraploid erythroid cells (Figures 4a,b).

Hematological parameters
The approximately two-fold higher hematocrit in diploid cells probably reflects the fact that the blood of diploid O. americanus contains about three times more erythroid cells than that of tetraploid organisms (Table I).The MCV was 50% greater in tetraploid cells, whereas the total hemoglobin concentration was 50% higher in the whole blood of the diploid animals.However, the MCH and MCHC measurements showed that tetraploid cells had 80 and 30% more hemoglobin per cell, respectively.

DISCUSSION
We examined the genic regulatory system that acts on erythropoietic cell production in vivo in diploid and tetraploid O. americanus frogs.A relatively synchronized system obtained by treating the frogs with phenylhydrazine hydrochloride helped to enhance erythropoiesis.This drug has been used to study erythrocyte maturation in amphibians and fishes (Cianciarullo et al., 1989(Cianciarullo et al., , 1999;;Spadacci Morena et al., 1991;Cianciarullo and Meirelles, 1994).
A follow-up of the in vivo erythropoiesis in diploid and tetraploid frogs by flow cytometry showed higher DNA, RNA and mitochondrial contents during this pro-cess, indicating increased genic activity in both diploids and tetraploids.The values were higher and distributed over a wider range in immature tetraploid erythroid cells, suggesting that gene activity in tetraploid O. americanus is more intense than in diploids during recovery from anemia.These data indicate that, because of their tetravalent state, the gene alleles in tetraploid animals may be more active under physiological stress than in diploid animals.
The DNA peak observed on the 20th day agrees with the results obtained by transmission electron microscopy, which showed very immature erythroid cells around the 30th day.Based on their ribosome content, these cells were classified as proerythroblasts and basophilic erythroblasts (Cianciarullo et al., 2000).
The RNA content of erythroid cells consisted predominantly of messenger RNA (mRNA), since the erythroid cells were not treated to analyze the ribosomal RNA (rRNA) or transfer RNA (tRNA), which are inaccessible without adequate previous treatment (Darzynkiewicz et al, 1975;Bauer and Dethlefsen, 1980).Thus, only singlestranded RNA was detected by our procedure.The RNA peaks on the 10th and 20th days of recovery from anemia corresponded to increased amounts of mRNA, most probably for globin synthesis.
The mitochondrial content reflects the state of energy metabolism in these cells.During erythropoiesis, the tetraploids showed more intense energy metabolism than diploid cells at all time points analyzed.Nevertheless, diploid cells also had an increased mitochondrial content, in-  dicating that mitochondria may play an important role in the maturation of erythroid cells in these amphibians.
The DNA/RNA/hemoglobin ratio between diploid and tetraploid O. americanus was approximately 1:2/1:1.3/1:1.3, suggesting that gene expression was being regulated at the DNA transcriptional level.Apart from the diploidtetraploid O. americanus, another example of anuran cryptic species is found in the family Hylidae: the diploid Hyla chrysoscelis and the tetraploid Hyla versicolor tree frogs (Bogart and Wassermann, 1972).Measurements of several cytological characters in these Hyla species did not show 1:2 ratios as expected, further suggesting that the cytological factors analyzed were being regulated at the diploid level (Bachmann and Bogart, 1975).
The erythrocyte measurements summarized by Glomski et al. (1997) revealed ratios of 1:1.2 (length) and 1:1.1 (width) in diploid:tetraploid Hyla species, respectively.Our corresponding measurements were 1:1.1 (length) and 1:1.3 (width), respectively, for the diploid:tetraploid O. americanus (Cianciarullo et al., 2000), showing that erythrocyte size is not correlated with ploidy.The other hematological parameters could not be compared here because they were not described for the Hyla species.In some parameters, the tetraploid O. americanus had similar hematologic values to the diploid Bufo paracnemis toad, such as in the MCHC, which was 32.8% in toads (Naoum et al., 1986).Diploid O. americanus had almost three times more erythroid cells than the tetraploids, but only 50% more total hemoglobin content.This fact appears to be due to the higher DNA content of the tetraploid cells.It also suggests a regulatory mechanism for maintaining the hemoglobin levels in the blood via a balance between the increased production of erythroid cells in diploid animals and lower levels of DNA transcription in tetraploid frogs.
The hematological data showed that MCHC was only 30% higher in the tetraploid cells.These data agree with morphometric analyses performed by transmission electron microscopy (Cianciarullo et al., 2000), which showed that tetraploid erythroid cells produce only 25-30% more ribosomes than diploid cells.These data also support the hypothesis that a regulatory system acts at the transcriptional level.Ruiz and Brison (1989) showed that methylation of ribosomal genes was increased in tetraploid genomes of adult O. americanus.If only 30% more ribosomes and 30% more hemoglobin are synthesized by the tetraploid genome, DNA methylation levels could be the a b mechanism responsible for this genic regulation system, since the reduction of genic activity in these tetraploid amphibians does not result from the loss of ribosomal DNA (Schmidtke et al., 1976).Two other regulatory elements were suggested for the O. americanus model, including the presence of intergenic spacers (IGSs), i.e., enhancers of about 87 bp, and promoters of about 220/390 bp (Álvares et al., 1998).The role of these sequences in the regulation of rDNA activity needs to be confirmed.
The cytoplasmic inclusions detected here at the ultrastructural level, which are similar to the cytoplasmic inclusions associated with RNA or RNP and rich in the element phosphorus detected in the tetraploid cells in previous experiments described in detail by Cianciarullo and Meirelles (1994), have not been observed in other anuran species analyzed so far, or in other vertebrate classes under the same experimental conditions.In addition, nuclear structures similar to the cytoplasmic inclusions were equally intensely labeled by RNase-gold complexes and phosphorus detection.Rifkind and Danon (1965) described Heinz bodies (denatured hemoglobin) in rabbit erythroid cells, induced by phenylhydrazine treatment.It is possible that phenylhydrazine hydrochloride, the drug used to induce hemolytic anemia, interferes with the erythroid cells of these anurans, inducing the formation of Heinz bodies associated with RNA or RNP, as recently observed by Moenner et al. (1998) in human erythrocytes.On the other hand, the cytoplasmic inclusions should also represent excess ribosomal RNA (rRNA) synthesized in both diploid/tetraploid erythroid cells under physiological stress.This response may represent a regulatory system for the production of hemoglobin.Such regulation has been described in polyploid fishes, where hidden breaks are responsible for the accumulation of nontranslated rRNA transcribed in excess (Leipoldt and Engel, 1983).Further studies are necessary to elucidate these events.
We described erythropoiesis in vivo in diploid and tetraploid cryptic species of O. americanus frogs, and showed differences at the physiological and molecular levels.These data support the idea that speciation is in course in this population.The complexity of the mechanisms which coordinate the metabolic functions responsible for the survival of these tetraploids should be examined further, particularly in view of the similarity between amphibian and mammalian erythropoiesis (Cianciarullo et al., 1989).

ACKNOWLEDGMENTS
The authors thank Miss Marta A. Santiago for operational assistance with the flow cytometer and Conselho Nacional de Desenvolvimento Científico e Tecnológico (CNPq), Fundação Oswaldo Cruz (FIOCRUZ) and Fundação Butantan for the financial support.Publication supported by FAPESP.

Figure 1 -
Figure 1 -Histogram of the relative cell size in spherical-made Odontophrynus americanus erythroid cells, during the different maturation stages of erythropoiesis.Arbitrary units; SD -standard deviation ( I ).

Figure 2 -Figure 3
Figure 2 -Variations detected in the DNA (A), RNA (B), and mitochondrial (C) content during erythropoiesis in Odontophrynus americanus, measured as fluorescence intensity and days after anemia induction.Arbitrary units; SD -standard deviation (I).