HIOI,OC;Y96, 515-519 (19831
and Larval Hemoglobins during the Early Development of the Bullfrog, Rana catesbeiana PHILLIP B. MAPLES, ALLAN R. DORN, ANI) ROBERTH. BROYLES'
The main hemoglobin (Hb) found in Shumway (embryonic1 stage 25 bullfrogs is that which we have designated Td-4. The other major tadpole Hbs (Td-1, 2, and 31 predominate during Taylor and Kollros (larval) stages IIXVIII. We propose that Td-4 is an embryonic Hb, whereas T&l, 2, and 3 are larval (fetal-like) Hbs. Embryonic Hb Td-4 continues to be synthesized during the larval stages. During the larval period, the average peripheral blood Hb profile changes very little with morphological stage or general growth. However, there is great heterogeneity in the embryonic:larval Hb ratio among individual tadpoles of a given stage or weight, apparently due to differential Hb and red cell production by the two active erythropoietic sites, mesonephric kidneys (Td-41, and liver (Td-I, 2, 3).
During development, vertebrates of all classes undergo hemoglobin (Hb)” transitions. These changes are the result of switches in gene expression-the genes which code for the variety of globin chains which characterize an animal’s Hbs. The phenomenon of Hb switching is a useful model for investigating the regulation of gene expression and control of cellular phenotypes during development and differentiation. A central question is what determines the choice of the Hb type that a red blood cell (RBC) will synthesize as it differentiates from a stem cell. Studies with Rann cutesbeiawa tadpoles have indicated that factors within the erythropoietic microenvironments of the hematopoietic organs influence this choice (Broyles et ul., 1981a). During larval development of the bullfrog, RWKL catesbc~inntr, four major hemoglobins (Hbs) and one or more minor Hbs are found in the circulating blood. The four major tadpole Hbs are well characterized with regard to their structural and functional properties (see Broyles, 1981, for a review). These Hbs share no globin chains with the adult Hbs which replace them during and after metamorphosis. Previous work from this laboratory has shown that the major tadpole Hbs are heterogeneously distributed within two morphologically
distinguishable types of red blood cells (RBCs). Each of the two types of RBCs emanates from a different erythropoietic organ in the tadpole. Both erythropoietic sites, the mesonephric kidneys and the liver, are active in individual tadpoles during the larval period of development (Broyles et ul., 1981a). Ultimately, we wish to know the molecular and cellular basis of this organ-specific production of different Hb and RBC types, including how the erythropoietic potentials of the organs arise in relation to their histogenesis. In this paper, we report how the relative amounts of these tadpole Hbs change with development from embryonic to late larval stages, and discuss these findings in relation to what is known about the histogenesis of the erythropoietic organs in anurans. In light of these and previous findings we propose that Hb Td-4 be classified as an embryonic form and that Td-1, 2, and 3 be classified as larval (fetal-like) Hbs. We hypothesize that erythropoiesis of the mesonephric kidneys (producing Hb Td-4) begins before liver erythropoiesis (producing Td-1, 2, and 3) during early development. The coproduction embryonic and larval Hbs during the larval period has implications for the mechanisms of cellular differentiation and globin gene regulation, which are briefly discussed. MATERIALS
’ To whom all correspondence and reprint requests should be addressed. ” Abbreviations used: BPB, bromphenol blue; Hb, hemoglobin; PAGE, polyacrylamide gel electrophoresis; RBC, red blood cell; TK stages, Taylor and Kollros (1946) stages of larval development; Tris, tris(hgdroxgmethyl1aminomethane.
Animals Rana catesbeiana tadpoles were obtained from Howe Brothers Minnow Farm, Atlanta, Texas, and from Car olina Biological Supply Company, Burlington, North
Carolina. The conditions and procedures used to maintain the animals in the laboratory have been described previously (Broyles et al., 1981a). Tadpoles ranging from Taylor and Kollros (1946) stage I through stage XVIII were used. Developing bullfrog eggs (Shumway stage 16) and early bullfrog tadpoles (Shumway stage 22), collected from the wild, were obtained from Carolina Biological Supply Company, in late April of 1980. These embryos were reared in the laboratory using established procedures (Broyles and Strittmatter, 1973). Development of the embryos was staged according to the criteria of Shumway (1940). Chemicals
Polyacrylamide and other reagents used for polyacrylamide gel electrophoresis (PAGE) were electrophoretic grade, either obtained as such from Bio-Rad Laboratories, Richmond, California, or recrystallized in our laboratory (Broyles et al., 1979). All other chemicals (salts and buffers) were reagent grade, and all solutions were made with distilled-deionized water or twice-distilled water. Preparations
of Samples ,fzwElectrophoresis
The collection of peripheral blood RBCs, liver, and kidneys, and the preparation of clarified extracts of the RBC and organ samples have been described previously (Broyles and Frieden, 1973; Broyles et al., 1981a). Animals of very early developmental stages (Shumway stages 19 through 25) which were too small to be bled were homogenized whole, and extracts were prepared in the same manner as for organ samples. All clarified extracts were stored at -20°C until PAGE was performed. Control experiments have shown the following: (1) Storing samples at -20°C does not alter Hb patterns on PAGE. (2) Making extracts of whole animals or organs does not cause any artifactual alteration of Hb patterns on PAGE. In particular, we have made extracts of RBCs in the presence and absence of organs and have found that the Hb pattern of the RBCs is unaltered by homogenization and lysis with either liver or kidneys. We have also compared the Hb pattern of RBCs extracted in the presence and absence of whole embryos or young tadpoles and found no alterations. (3) Eluted, unstained bands from electrophoresed organ extracts have been shown to have Hb spectra identical to comparable Hbs from peripheral blood samples. Electrophoretic
Separutimz of Hemoglobins
Clarified RBC lysates and supernatant fluids from homogenized embryos or organs were electrophoresed on
VOLUME 96, 1983
7% polyacrylamide disc gels at basic pH as previously described (Broyles et al., 1979). The electrophoresis (tank) buffer was either Tris/glycine or Tris/DL-alanine (Parkinson et al., 1981). With the former buffer, the running pH is approximately 9.4; with the latter, approximately 9.6. Hemoglobins were quantitatively stained by immersing the gels in benzidine-HZ02; the staining method denis specific for Hb (Broyles et al., 1979). Quantitative sitometry of the stained Hb bands was performed as previously described (Broyles et al., 1981a). Radioactive
of Hemoglobins in Vivo
Stages X-XII tadpoles were injected with 5 &i/gm body weight of a mixture of 14C-labeled amino acids (average sp act equals 232 mCi/mmole; New England Nuclear, Inc.) The animals were given a single intramuscular injection on the dorsal side of the body. Eight hours after the injection, the animals were anesthesized and bled. Washed RBCs were obtained and lysed (Broyles and Frieden, 1973), and Hb’s were separated by electrophoresis as described above. Polyacrylamide gels were sliced serially into l-mm segments, and the individual gel slices were dissolved and counted in a scintillation system as previously described (Broyles and Frieden, 1973). RESULTS
The predominant Hb found in embryonic (Shumway) stage 25 Rana catesbeiana is Td-4 (Figs. 1 and 2A). Hemoglobins Td-1, 2, and 3 predominate during the larval (Taylor and Kollros) stages (Figs. 2A, B). We have yet to determine exactly when, between Shumway stage 25 and TK stage I, this marked change takes place and how long the animals take to accomplish it. These data lead us to propose that the tadpole Hb which we have designated Td-4 is actually an embryonic Hb, whereas Hbs Td-1,2, and 3 are truly larval or fetallike Hbs. This proposal is supported by the following additional considerations: (a) Td-4 has a unique globin chain composition different from that of the other tadpole Hbs or the adult Hbs (as reviewed by Broyles, 1981). (b) Td-4 (along with Td-3) has the highest oxygen affinity of the R. catesbeiana Hbs; the oxygen affinities of Td-1 and Td-2 are intermediate between that of Td-4 and those of adult Hbs (Watt and Riggs, 1975). Thus, as in other vertebrates, there is a sequential decrease in oxygen affinity during development from embryonic to adult Hbs. (c) In larval (Taylor and Kollros) stage animals, Td-4 is found within RBCs of primitive morphology (Broyles et al., 1981a, b), similar to the primitive RBCs of chick embryos (Lemez, 1964) and to the embryonic RBCs of the mouse (Craig and Russell, 1964).
FIG. 1. Hemoglobin patterns of Shumway stage 25 Rurcct ccctrsbcicf)~~. Samples were prepared and electrophoresed as described under Materials and Methods. (a) Stage 25 plus 9 days (pool of live animals). (b) Stage 25 plus 17 days (pool of five animals). (c) Taylor and Kollros stages X-XII (RBCs pooled from six animals). (d) Purified Hb Td-4 isolated by preparative isoelectric focusing (Darn, 1982). BPB designates the migration point of the bromphenol blue tracking dye.
Hemoglobins Td-1,2, and 3 are found within RBCs whose morphology is of the definitive type, i.e., similar to the morphology of adult frog RBCs (Broyles et ul., 1981b), as well as to the definitive RBCs of chickens (Lemez, 1964). (d) During larval development, Td-4-containing RBCs emanate from the mesonephric kidneys (Broyles Av. wt. b-s,’
et ul., 1981a), which are ontogenetically and phylogenetically a primitive erythropoietic site in vertebrates (Hollyfield, 1966; Torrey, 1971), whereas RBCs containing Td-1, 2, and 3 emanate from the liver, which in mammals is the major site of fetal erythropoiesis (Metcalf and Moore, 1971). The predominance of Hb Td-4 in Shumway stage 25 animals correlates with the period of histogenesis of the mesonephric kidneys of R. pipiem (Rugh, 1951; Horton, 1971). The presence of small amounts of Hbs Td1, 2 and 3 in some Shumway stage 25 animals (Fig. 2A) correlates with the beginning of liver hemopoiesis in R. pipiens, and the predominance of these Hbs at TK stages I-VIII and thereafter (Figs. 2A, B) correlates with the predominance of erythroblasts in R. pipiens liver at TK stages IV-X (Turpen et al., 1979). We postulate that mesonephric kidney erythropoiesis begins before liver erythropoiesis in R. ca,tesbeiunu. However, the possibility that the Hb Td-4 of stage 25 R. catesbeium is being elaborated by more than one erythropoietic site cannot be eliminated. Ventral blood island erythropoiesis precedes that of the kidneys and liver in amphibians (Hollyfield, 1966; Turpen et al., 1979). The Hb type elaborated by the ventral blood islands has not been identified. The minor Hb which we have labeled Td-5 (Figs. 2, 4) may also be an embryonic form, but any such designation is tentative until this Hb has been further characterized. Td-5 is in the same electrophoretic position relative to the major Hbs as band 4 reported by AK Wt. bms) : “:
P m 5 40z 8
FIG. 2. How cc~te.sbeic~nc~ hemoglobin patterns at different developmental stages. (A) Animals obtained from Carolina Biological Company, Burlington, North Carolina. (B) Animals obtained from Howe Brothers Minnow Farm, Atlanta, Texas. Each value shown is a mean + the standard error (SE) for the number of animals (n) in that group except for stage IV and greater in B, which are values for pooled samples. Multiple determinations on a single Hb sample agreed to within a SE of 3Y of the mean (Broyles ef crl., 1981a).
, I ' .5
30 40 Gel slice X
FIG. 3.1~ r,ivo synthesis of tadpole hemoglobins. A stage X tadpole was injected with 5 PCi of ‘“C-amino acid mixture per gram body wt as described under Materials and Methods. After 8 hr, RBCs were obtained, Hbs were electrophoretically separated, and the polyacrylamide gels were scanned, sliced, dissolved, and the radioactivity per slice was determined, all as described under Materials and Methods.
Moss and Ingram (1968), which was shown to be composed of two types of globin chains. The continued presence of embryonic Hb Td-4 during the larval period is due to new synthesis, and not merely to persistence, as shown by four observations: (1) Previous experiments have shown that organ cultures of mesonephric kidneys of TK stages X-XII tadpoles incorporate radioactive precursors in vitro into Hb Td-4 (Broyles and Frieden, 1973; see Broyles (1981), for revised nomenclature of the Hbs). (2) As shown in Fig. 3, midlarval (stage X) tadpoles incorporate 14C-labeled amino acids in viva into all tadpole Hbs, including Td-4. (3) The conclusion is implicit in the data in Fig. 2. As the animals increase in average body weight from Taylor and Kollros stage I to stage XVIII, there will be parallel increases in total blood volume, total RBCs and total hemoglobin. Although the percentage of total Hb that is Td-4 decreases slightly during this growth, it is obvious that the total amount of Td-4 must increase. (4) The life span of tadpole RBCs has been reported to be 80-100 days (Forman and Just, 1976). However, the period of larval growth between stages I and XVIII is lengthy, ranging from one (Viparina and Just, 1975) to 3 years (Collins, 1979). Since stage XVIII animals have considerable amounts of Hb Td-4, and because more than a few half-lives of RBCs have passed since they were embryos, they must be producing more RBCs containing Td-4 during larval development. As shown in Fig. 2, parts A and B, once R. catesbeiana have reached Taylor and Kollros stage I, there is very little change in the average Hb profile with further growth (increase in body weight) and morphological development (TK stages I through XVIII). In contrast to
this relative constancy of the average Hb profile among different stage and weight groups, there is a great deal of heterogeneity in the Hb profiles among individual tadpoles of a given stage: Note the size of the standard error bars in Fig. 2 and the different Hb profiles found for individual animals of TK stages I-VIII (Fig. 4, lefthand column of densitometer tracings). This polymorphism in Hb profiles is not due to a differential inheritance of globin structural genes since each of the four major Hbs (Td-1 through 4) can be found in every tadpole, either in the peripheral blood (Fig. 4, left column) or in the erythropoietic organ (kidney or liver) that elaborates it (Fig. 4, middle and right columns, respectively).
FIG. 4. Hemoglobin profiles of peripheral blood, mesonephric kidneys, and liver from individual animals of Taylor and Kollros (1946) stages I-VIII. Shown are densitometer tracings of polyacrylamide gels on which the Hbs were separated using Tris/nL-alanine as the tank buffer (Parkinson et al., 1981). Each tracing represents a single sample from one tadpole. In the left column, four typical peripheral blood Hb profiles are shown (A-D), with the number of animals (n) which exhibited each profile indicated. The middle and right columns show the Hb patterns of the erythropoietic organs from the same individuals. Different Hbs are enriched in extracts of the organs which produce them: mesonephric kidneys (Hb Td-4) or liver (Hbs Td-1 and Td-3). The arrows denote the location of the bromphenol blue tracking dye near the bottom of each disc gel.
These data, combined with those in Fig. 2, make two other possible explanations of the heterogeneity in tadpole Hb profiles less likely: (a) Since the heterogeneity in Hb profiles of animals from a single source (Fig. 4) is greater than the difference in average Hb profiles between animals from two different sources (Fig. 2A versus B), the variation among different R. catesbeiana populations would seem to be of minor importance as an explanation of this heterogeneity. (b) Nor is the heterogeneity to be explained as a previously undetected trend corresponding to morphological development or growth during the larval period (Fig. 2). The heterogeneity is evident even in the early larval stages (Fig. 4). As shown in Figs. 4A-D, four typical Hb profiles are found in peripheral RBC lysates of stage I-VIII tadpoles. Some tadpoles have a predominance of Hb Td-4 (Fig. 4A), whereas in others the peripheral blood Hb consists almost entirely of Hbs Td-1,2, and 3 (Fig. 4D). The majority of animals have intermediate patterns as exemplified by Figs. 4B, C. From the 62 animals of this range of stages, the frequency distribution of these patterns (ratio of A:(B + C):D in Fig. 4) is approximately 1:3:1, a normal distribution. Thus, during development of R. catesbeiana there is an early change in the Hb types elaborated-the larval Hbs are acquired in addition to the embryonic Hb Td4. Once this change is accomplished (by TK stage I), the circulating Hb phenotype appears to be determined by at least two distinct levels of control: (1) The choice of Hb type expressed in differentiating erythroid cells is apparently determined by factors specific to the different erythropoietic microenvironments within the mesonephric kidneys and the liver (Broyles et al., 1981a). (2) We postulate that another regulatory factor(s) modulates the relative output of the erythropoietic organs. The degree to which this modulation is influenced by genetic inheritance and the external environment is unknown; both types of influence are consistent with the normal distribution of Hb profiles described above. WC thank Stuart Berger and David Vogt for technical assistance and (‘heryl Bontempi, Shirley Logue, Darci Sargent, and Toni Johnson for typing the manuscript. We are especially indebted to Howe Brothers Minnow Farm for generously providing animals of excellent quality during the past 10 years. This work was supported in part by Grants AM 21386 and AM 21764 from the National Institutes of Health.
REFERENCES BROYLES, R. H. (1981). Changes in the blood during amphihian metamorphosis. It( “Metamorphosis, A Problem in Developmental Bi-
ology,” 2nd ed. (L. I. Gilbert and E. Frieden, eds.), Chap. 14, pp. 461-490. Plenum, New York. BROYLES, R. H., and FRIEDEN, E. (1973). Sites of haemoglobin synthesis in amphibian tadpoles. Nature New Biol. 241, 207-209. BROYLES. R. H.. and STRITTMATTER. C. F. (1973). Hexose monophosphate shunt dehydrogenases in the developing frog. Camp Biochem. Physiol. 44B, 667-676 BROYLES, R. H., JOHNSON, G. M., MAPLES, P. B., and KINDELL, G. R. (1981a). Two erythropoietic microenvironments and two larval red cell lines in bullfrog tadpoles. Dev. Biol. 81, 299-314. BROYLES, R. H., PACK, B. M., BERGER, S., and DOHN, A. R. (1979). Quantification of small amounts of hemoglobin in polyacrylamide gels with benzidine. Anal. Biochem. 94, 211-219. BROYLES, R. H., DORN, A. R., MAPLES, P. B., JOHNSON, G. M., KINDELL, G. R., and PARKINSON, A. M. (1981b). Choice of hemoglobin type in erythroid cells of Rana catesbeiana. In “Hemoglobins in Development and Differentiation” (G. Stamatoyannopoulos, and A. W. Nienhuis, eds.), pp. 1799191. Alan R. Liss, New York. COLI.INS, J. P. (1979). Intrapopulation variation in the body size at metamorphosis and timing of metamorphosis in the bullfrog, Ranu catesbeiam Ecology 60, 738-749. CRAIG, M. L. and RUSSELL, E. S. (1964). A development change in hemoglobins correlated with an embryonic red cell population in the mouse. De?). Biol. 10, 191-201. DORN, A. R. (1982). “Erythrocyte Differentiation during the Metamorphic Hemoglobin Switch of Rnnn ccctestwiuno.” Ph.D. dissertation, University of Oklahoma at Oklahoma City. FORMAN, L. J., and J~JST,J. J. (1976). The life span of red blood cells in the amphibian larvae, Rana catcsbrianu. Dw. Bid. 50, 537-540. HOLLYFIELLI, J. G. (1966). The origin of erythroblasts in &II/~ ~?i/lic%s tadpoles. Urv. Rio/. 14, 461-480. HORTON, J. D. (1971). Histogenesis of the lymphomyeloid complex in the larval leopard frog, Rnna piyiexs. J. MOT-phol. 134, l-20. LEM~Z, L. (1964). The blood of chick embryos: Quantitative emhryology at a cellular level. Adr>an. Morphqqw~ 3, 197-245. METCALF, D., and MOORE, M. A. S. (1971). “Hf,nlopoic+ic Cells.” pp. 911, 195-210. North-Holland, Amsterdam/London. Moss, B., and IN(:RAM, V. M. (1968). Hemoglobin synthesis during amphibian metamorphosis. I. Chemical studies on the hemoglohins from the larval and adult stages of Ranu ctrtc~sbcic~~~ J. Md. Bid. 32. 481-492. PARKINSON, A. M., DORN, A. R., MAPI,ES, P. B., and BROYLES, R. H. (1981). Improved polyacrylamide gel electrophoresis with different amino acids as the trailing constituent. AML Biochem. 117, 6-11. R~J~EI,R. (1951). “The Frog.” pp. 219-223,251-260. McGraw-Hill, New York. s IIIJM~AY, W. (1940). Stages in the normal development of Ram piyims. Amt. Rec. 78, 139-147. TAYI,OR, A. C., and KOLLROS, J. J. (1946). Stages in the normal devclopment of Ram pipions larva. Amt. Rw. 94, 7-23. TORREY, T. W. (1971). “Morphogenesis of the Vertebrates.” 3rd cd. pp. 368-369. Wiley, New York. TIJRPEN, J. B., TIJRPEN, C. J., and FLAJNIK, M. (1979). Experimental analysis of hematopoietic cell development in the liver of larval Rvnu pipicws. Zlev. Rid. 69, 466-479. VIPAKINA, S., and JUST, J. J. (1975) The life period, growth and differentiation of Rmo catrsbriomt larvae occurring in nature. “Copcia,” No. 1, Feh. 28, pp. 103-109. WATT, K. W. K., and RIIXS, A. (1975). Hemoglobins of the tadpole of the bullfrog, Ram cutesbeianu. Structure and function of isolated components. J. Rio!. Cherrr. 250, 5934-5944.