Physiological Responses of Larval and Postmetamorphic Rana pipiens to Growth Hormone and Prolactin

BENJAMIN W. SNYDER',' AND B. E. FRYE Department of Zoology, University of Michigan, Ann Arbor, Michigan

ABSTRACT Bovine prolactin stimulates growth of tadpoles, whereas bovine growth hormone stimulates growth of postmetamorphic frogs. The biochemical composition of liver, muscle, and fat body were examined to determine whether there were any changes in carbohydrate, protein, fat, or nucleic acids which might correlate significantly with the growth effects of the hormones in the two stages of development. In the frog, the major effect of growth hormone was to depress carbohydrate and lipid stores, with little or no effect on protein and nucleic acids in most experiments. Prolactin had similar but smaller effects on carbohydrate, and no effect on fat body lipid in the frog. In tadpoles prolactin suppressed muscle glycogen, but otherwise did not affect tissue composition. Growth hormone had no effects in the tadpole. Ways in which the interrelated developmental and metabolic effects of growth hormone and prolactin could increase the adaptive significance of distinct larval and postmetamorphic growth-regulating hormones in the amphibians are discussed.

Amphibians exhibit a unique mecha- the gland (Gona, '67, '68), and by nism for regulation of growth in that inhibiting the action of thyroxin in the prolactin stimulates growth in larval tissues (Bern et al., '67; Derby and Et- stages of development, whereas growth kin, '68). Since several other antithyroid hormone stimulates growth of the post- compounds stimulate growth in tadpoles metamorphic animals (Brown and Frye, (Steinmetz, '54; Brown and Frye, '69a), '69b). Since the first experiment suggest- it is possible that the growth promoting ing that prolactin causes growth in tad- effect of prolactin is a consequence of poles (Etkm and Lehrer, '60), the growth its antithyroid action. However, Brown effect of prolactin in larval amphibians and Frye ('69a) showed that prolactin has been substantiated in several spe- treatment of thyroidectomized or thioura- cies, including Rana catesbeiana (Berman cil treated animals caused a further stim- et al., '64; Etkin and Gona, '67a,b), Alytes ulation of growth beyond that of control obstretricians (Remy and Bounhiol, '65, values. This indicates that the hormone '66), and Rana pipiens (Etkin and Gona, has an additional growth effect which is '67b; Brown and Frye, '69a). In these independent of its antithyroid activity. studies injections of purified prolactin Fewer growth studies have been done preparations caused an increase in length, with postmetamorphic amphibians than wet weight, and dry weight, while growth with the larval stages, but the reports hormone under the same conditions had which have been published to date indi- little or no effect." cate that, in contrast to the larval situa- Although the effect of prolactin on lar- tion, growth hormone is the primary val growth has been well documented,

the mechanisms by which this growth is 1 Supported by National Institutes of Health Train- achieved is not entirely clear. Recent stud- ing grant NIH 5T01 GM 989. 2 Present address: Laboratory of Human Reproduc- ies indicate that prolactin stimulates tion and Reproductive Biology, Harvard Medical School, growth in part by acting as an antithy- 45 Shattuck Street, Boston, Massachusetts, 02115. roid agent. Prolactin has been shown to 3 In one exceptional case (Enmar et al., '68) both prolactin and growth hormone were effective in in- act as an antithyroid agent in two ways: creasing length and weight of Rana temporaria larvae, by blocking the output of thyroxin from the latter being the more effective of the two hormones.

J. EXP. ZOOL.,179: 299-314. 299 300 BENJAMIN W. SNYDER AND B. E. FRYE growth hormone of the postmetamorphic MATERIALS AND METHODS frog. Growth hormone was shown to in- Animals crease weight and length in young Rana pipiens (Brown and Frye, '69b) and in the Frogs of the species Rana pipiens were toads Bufo boreas and Bufo marinus (Zip- used in all experiments. Postmetamorphic ser et al., '69). Prolactin appears to have frogs were obtained from either Lember- very little or no effect on either Rana or ger Company, Oshkosh, Wisconsin or col- Bufo postmetamorphic animals. lected locally. The frogs used were of In mammals, the growth promoting ef- medium length (6-7 cm) and at the start fects of growth hormone are accompanied of most experiments weighed between 15 by extensive and interrelated effects upon and 30 grams. Prior to use they were fed the of carbohydrate, fat, and crickets ad lib. two to three times/week. protein. Specifically, growth hormone pro- They were fed up to the beginning of the motes nitrogen retention and protein syn- experiment and then fasted four to five thesis (Lee and Ayres, '36); is glycostatic days prior to the sacrifice of the animals. as reflected by maintenance of glycogen During experiments frogs were placed in levels of muscle and liver in fasting hy- individual plastic refrigerator dishes lined pophysectomized and normal rats (Russell with wet paper toweling and kept in an and Bloom, '56; Altszuler et al., '68); and incubator at 21-23°C on a 12-hour photo- is lipolytic in that it increases free fatty period. acid concentration in plasma, fatty acid Tadpoles were obtained by ovulation turnover, and oxidation to carbon dioxide and artificial insemination by the method (Winkler et al., '64). of Wright and Flathers ('61). The tad- Although mammalian growth hormone poles were raised in aerated tap water and prolactin are known to be growth in enamel pans and fed boiled spinach. promoting in amphibian postmetamorphic During experiments tadpoles were main- and larval stages, as described above, tained individually in four-inch finger- there has been very little study of the bowls and kept at 21-23°C on a 12 hour related metabolic effects. The purpose of photoperiod. Tadpoles were selected for this investigation was to examine the experimental use at Taylor-Kollros stages biochemical basis of the growth promot- X-XI11 (Taylor and Kollros, '46). ing effects of GH and prolactin, and to General procedures determine what effects they might have upon overall composition of carbohydrate, Hypophysectomy of frogs was performed fat, and protein compared to the effects through a hole drilled in the sphenoid in mammals. Specifically, this paper . They were allowed to recuperate reports the effects of prolactin and growth from the surgery for four to seven days hormone upon the biochemical composi- before use in an experiment. Hypophysec- tion of various tissues after long term tomized frogs were kept in 0.3-0.5% am- treatment. The principal tissues analyzed phibian Ringers. were liver and , and the Hormones were obtained from the Na- substances measured incIude water, car- tional Institutes of Health as crystalline bohydrate, fat, protein, and nucleic acids. bovine growth hormone (NIH-GH-B12: The purpose of the biochemical analyses 0.97 USP units/mg, and NIH-GH-B14: was to compare the effects of prolactin 1.04 USP units/mg) and bovine prolactin and growth hormone in amphibians, and (NIH-P-32: 19.9 IU/mg). They were dis- in addition to compare the effects of these solved in 0.7% NaC1, pH 9.3-9.6. Injec- hormones in amphibians with their ac- tions were made intraperitoneally through tion in mammals. the ventral abdominal body wall in frogs A subsequent paper will describe the and through the tail muscle into the effects of prolactin and growth hormone abdomen in tadpoles. upon the in vivo incorporation of radio- Growth was determined by length and active amino acid into the protein frac- weight increment. Frog length was taken tion of various tadpole and frog tissues as total snout-urostyle length, or as length after prolactin and growth hormone in- of the tibia. Tadpole length was taken as jection. total body plus tail length. Weights were HORMONE REGULATION OF FROG GROWTH 30 1 taken after blotting animals uniformly was considered significant. Values in fig- on a dry cotton towel. ures and tables are given as mean & standard error of the mean. The numbers Chemical procedures in parentheses indicate the size of the groups. Tissue fractionation. The method of Shibko et al. ('67) was used for fractiona- Experimental design tion of tissues to be analyzed for chemical composition. and modified to fit the The results reported in this paper are amount of tissue and design of each based on two experiments with frogs and experiment. In general, the method four experiments with tadpoles. The plan proved satisfactory except that significant of these experiments is summarized as amounts of protein were extracted into follows: the lipid fraction. The lost protein was Frog experiment 1 : Hypophysectomized corrected for by chemically measuring animals. the amount of protein in the lipid frac- Hpx - hypophysectomized frogs (8 ani- tion and adding it to the protein fraction. mals), injected with 0.7% NaCl per day Chemical assays. Protein was deter- for six weeks. mined by the biuret method of Wanne- Hpx + GH - hypophysectomized frogs macher et al. ('65) for concentrated solu- (7 animals), injected with 3.6 pg growth tions, or the method of Lowry ('51) for hormone/gm body weight per day for six dilute samples. weeks. The carbohydrate fraction was taken Hpx + P - hypophysectomized frogs as the acid-soluble fraction after PCA (6 animals), injected with 3.6 pg prolac- or TCA treatment of the tissue homoge- tin/gm body weight, per day for six weeks. nate. The phenol-sulfuric acid reaction of Sham - sham operated injected with Ashwell ('66) was used to estimate the 0.7% NaCl per day for six weeks (7 ani- total carbohydrate of the sample. mals). RNA concentrations were determined Frog Experiment 2: Intact animals by reading the samples in a Beckman (9 animalslgroup). spectrophotometer at 260 mp referenced C - 100 pl 0.7% NaCl per day for to a standard solution of hydrolyzed five weeks. RNA. Diphenylamine reagent was used GH - 100 pg growth hormone per day to determine DNA (Burton, '56). These for five weeks. results were usually consistent with those P - 100 pg prolactin per day for five obtained by reading the DNA solution in weeks. a Beckman spectrophotometer at 268 mp The animals used in Experiment 1 were referenced to a standard of hydrolyzed collected in mid-winter (December) ap- salmon sperm DNA. proximately two weeks prior to the begin- Frog liver lipid was determined gravi- ning of the experiment. Frogs used in metrically after isolation according to the the second experiment were collected in method of Shibko et al. ('67). Lipid con- spring at the time of emergence from tent of tadpole liver was determined gravi- hibernation. metrically after homogenizing the tissue Tadpole Experiments 1, 2, and 3. in ch1oroform:methanol (2: 1) followed by C - 0.7% NaCl/day (15 animals). two more extractions with the same sol- GH - 5 pg growth hormonelday (12 vent. animals). The per cent water in liver and muscle P - 5 pg prolactin/day (16 animals). tissues was measured by placing the The animals were weighed and mea- weighed tissue in a tared vessel and dry- sured at the beginning and end of the ing to constant weight at 60°C. experiment. They were injected for either Statistics nine or ten days and killed approximately 24 hours after the last injection. Food Student's t test was used to analyze was withdrawn at the time of the last the data (Snedecor and Cochran, '69). injection in Experiments 1 and 3 but not A probability value (P) of 5% or less Experiment 2. 302 BENJAMIN W. SNYDER AND B. E. FRYE Tadpole Experiment 4. Same design as Experiment 1 except the growth hormone and prolactin groups were injected with 5 pg hormone for three days, then 10 pg for six days,

RESULTS

Length and weight changes in tadpoles and frogs. The results of this study con- firm the previous reports (cited in the introduction) that prolactin promotes - CONT. growth in tadpoles whereas growth hor- ...... 100yg PRO1 --- 100pg GH I mone stimulates growth in postmetamor- c /--t phic frogs (figs. 1, 2). Figures 1 and 2 ccc-ce represent growth data from Frog Experi- ment 2 and Tadpole Experiment 1 respec- .,? ...... 1 tively. Frog Liver. In the first experiment (hy- pophysectomized animals) growth hor- mone injections decreased both absolute and relative liver weight (fig. 3). Mean liver weight of growth hormone-treated animals was little more than one-half that of the hypophysectomized controls (P < 0.001) and only two-thirds as large as the prolactin-treated and sham control groups (P < 0.001). Neither hypophysec- tomy nor prolactin treatment significantly WEEKS increased liver weight above sham con- Fig. 1 Changes in tibia length and body weight trol values. in intact frogs during five weeks of treatment with Reduction of liver size by growth hor- growth hormone and prolactin (frog experiment 2). mone could be due to either a selective depletion of some liver components or a periment 1 are shown in table 1. In terms general depletion of all liver components. of absolute amounts, the only two compo- An examination of the amount of each nents affected by growth hormone were biochemical component in the livers was total carbohydrate and water. The hy- undertaken to distinguish between the pophysectomized control had nearly seven above two alternatives. The results of Ex- times as much carbohydrate as the growth

1.3 1.5 c I L - 1.1 I I 0 1.1 ...... u4 ..... 0 ..... z 1.0 I .. < ..... I ..... U 0.9 I .....

...... ff...... 9.. 0.6 ...... C OH P C OH P Fig. 2 Changes in total length and weight of tadpoles after nine days of treatment with growth hormone and prolactin (tadpole experiment 1). C, control; GH, 5 pg growth hormone/ day; P, 5 pg prolactinlday. HORMONE REGULATION OF FROG GROWTH 303 hormone-treated animals. Even though different in any of the four groups. Thus, livers of prolactin-treated and sham con- the reduced levels of carbohydrate and trol animals had over four times as much water in the livers of growth hormone- carbohydrate as the growth hormone treated animals account for the differ- group, they had significantly less carbo- ence in absolute weight between this and hydrate than the hypophysectomized con- the other groups. The same is true of any trol group (P < 0.05). As would be two groups: the difference in absolute expected, the total amount of water in liver weights can be accounted for en- the liver corresponded closely to the ab- tirely by the different amounts of carbo- solute liver size. Therefore, the absolute hydrate and water. water content of the growth hormone When the data from this experiment group was also significantly below the are expressed as relative values (mg of hypophysectomized control group (P < substancelgm liver tissue; this calcula- 0.001). The absolute amounts of protein, tion can be made from data in table 1) RNA, DNA, and lipid are not significantly the relative concentrations of protein,

TABLE 1

Absolute composition of frog liver after hypophysectomy and hormone replacement Recovery mg recovered Water Carbohydrate Protein RNA DNA Lipid mg/liver mglliver mglliver mglliver mglliver mglliver aye. liver wt. HYPX + 0.717

GH 547 f 42 2 35 f 8.7 2 95.4 f8 4.43 f0.35 3.81 f 0.26 33 & 11 ~ (7) 0.749

HYPX + 1.09 Prolactin 823 .+. 74 141 f28 1 98.8 f 9 4.97 f 0.28 4.21 f0.33 26 .+.2 - (6) 1.12 HYPX + 1.40 Saline 1021 f 85 239 f 30 104.5 f 8 4.81 f 0.32 4.20 f0.38 26 .+. 3 - (8) 1.42 Sham 1.07 operated 794 k 65 157 2 18 1 96.4 f 6 4.30 f 0.28 3.99 f0.32 20 +- 7 - (7) 1.13

1 p < 0.05. 2 p < 0.001

SHAM HPX HPX HPX+ SHAM ~rx HPX+ HPX OH P GH P

Fig. 3 Absolute and relative liver weights of hypophysectomized and sham-operated frogs after six weeks of treatment with growth hormone and prolactin (frog experiment 1). 304 BENJAMIN W. SNYDER AND B. E. FRYE RNA, DNA, and lipid in the livers of mized animals in Experiment 1 and there- growth hormone-treated animals were fore requires further verification. highly elevated above the values seen in Frog gastrocnemius muscle. In addi- the other three experimental groups, tion to liver, frog gastrocnemius muscle whereas carbohydrate was lowered. How- was analyzed after hormone treatment. ever, since changes in carbohydrate and The data from Frog Experiments 1 and 2 water account almost exactly for hormone are summarized in tables 3 and 4. The effects on total liver size, we conclude results obtained with muscle are pre- that the hormonal effects are more accu- sented as relative values (mg chemical/ rately represented in terms of absolute

changes and therefore that there was no c no I n20 real effect of growth hormone on protein, RNA, DMA, and lipid. In Experiment 2 (intact animals) the mean liver size of the growth hormone- 150 treated animals was again significantly -v00 below prolactin-treated (P < 0.05) and zoo , control (P < 0.001) values (table 2). Pro- lactin treatment also reduced absolute '700 and relative liver size below control I 130 . values (P < 0.001). As was the case in Experiment 1, the absolute values for '500 carbohydrate and water in Experiment 2 : were the most drastically altered cate- 100 . gories and accounted for most of the difference in liver weights between dif- ferent groups (table 2, fig. 4). In addi- 50 . tion, growth hormone treatment of intact animals decreased total protein content below the controls (P < 0.025). Prolac- tin treatment produced results intermedi- ate between the growth hormone and Fig. 4 Composition of frog liver after hypophy- control groups. The total amount of DNA/ sectomy and hormone replacement (frog experi- liver was elevated by growth hormone ment 1). CHO, carbohydrate; GH, growth hor- mone treated; P, prolactin treated; SHAM, sham injections (P < 0.05). This differed from hypophysectomized controls; HPX, hypophysec- the results obtained with hypophysecto- tomized without hormone treatment.

TABLE 2 Absolute composition of intact frog liver after hormone treatment

~ ~~ Recovery mg recovered Water Carbohydrate Protein RNA DNA mgiliver mglliver mg/liver mg/liver mgiliver ave. liver wt. GH 1.01 769k433 982123 134t62 5.47t0.35 5.14k0.29 1 __ (9) 1.12

P 1.12

839 f49 3 128 t 15 146 f 9 5.94 2 0.44 4.95 f 0.44 ~ (9) 1.28

C 1.67 1207 261 298 2 24 158 -1-8 5.78 ~0.26 4.16 20.31 - (9) 1.82

I p < 0.05. 2 p < 0.025. 3p < 0.001. HORMONE REGULATION OF FROG GROWTH 305 gm muscle). Data for absolute quantities Tadpole liver. The effect of hormone of substances in muscle show the same injections on tadpole liver in Tadpole Ex- result. In Experiments 1 and 2 there was periment 1 are shown in table 5. The no difference in total or relative (mglgm) relative liver weight was significantly re- muscle weight among any of the groups duced by prolactin treatment (P < 0.001). (tables 3, 4). However, this probably re- However, prolactin did not reduce the flects the wide range of terminal body absolute liver size. Growth hormone had sizes rather than the failure of growth no effect on either relative or absolute hormone to stimulate muscle growth. If liver size. Comparison of the hormone- initial and terminal muscle weights were treated with the control groups reveals compared, a growth hormone stimulation no significant differences in the amounts probably would be observed. of any major biochemical components. Hypophysectomy reduced the amount Carbohydrate, lipid, protein, RNA, and of muscle carbohydrate (P < 0.025) and water concentrations were not different in DNA (P < 0.025), but not RNA (P < any of the three groups. The smaller 0.08) compared to the sham operated absolute size of the livers of prolactin- group. Neither growth hormone nor pro- treated animals appears to reflect a fail- lactin had any effect on muscle protein ure of the liver to grow in proportion to concentration. Growth hormone injected body growth, and not to a specific deple- into hypophysectomized frogs elevated the tion of any component of the liver. Essen- levels of RNA and DNA above the sham tial similar results were obtained in the control group while prolactin maintained repeat experiments. One exception was these compounds at about the level of the observed in Experiment 4, in which liver intact animal. Growth hormone reduced RNA concentration of the growth hormone muscle carbohydrate below the already and the prolactin-treated groups was re- reduced hypophysectomized control values duced below control values. (P < 0.05). Prolactin treatment did not Tadpole tail muscle. Because carbo- lower carbohydrate significantly below the hydrate and RNA levels were most respon- hypophysectomized control. sive to hormone treatment in frog muscle, The experiment with intact frogs (ta- these parameters were measured in tad- ble 4) yielded similar results except that pole muscle. The results from the two growth hormone did not increase DNA tadpole experiments are summarized in concentrations significantly above con- table 6. Contrary to the results obtained trols. In Experiment 2 muscle water was with frogs, prolactin rather than growth measured and was found to be increased hormone lowered carbohydrate. Prolactin by treatment with growth hormone (P < reduced carbohydrate to nearly half the 0.001) while prolactin had no effect on control level in Experiment 2 (P < 0.001). this parameter. In Tadpole Experiment 4 prolactin-induced

TABLE 3

Chemical composition of frog gastrocnemius muscle after hypophysectomy and hormone replacement

Carbohydrate Protein RNA DNA mg Muscle/ Total mg/m mglgm ygkm pLp/gm gm body muscle wt. HYPX + GH 6.04 k0.58 1 116 i3 864k522 749k601 19.8k1.0 472k43 (7) HYPX + prolactin 6.63 k0.83 120 k2 797 k60 704 k43 18.1 k0.7 424 k42 (6) HYPX + saline 8.12 k0.65 122 k3 716 k 31 630 k20 18.0 k 1.9 453 +23 (8) Sham operated 12.28k1.62 116k3 806i-27 710k21 2 17.9k0.6 414k27 (7)

1 p < 0.05. 2 p < 0.025. 306 BENJAMIN W. SNYDER AND B. E. FRYE

TABLE 4 Chemical composition of intact frog gastrocnemius muscle after hormone treatment

Water Carbohydrate Protein RNA DNA mg muscle/ Total wlgm mgkm mdgm pglgm pglgm gm body muscle wt. GH 784.510.82 11 kO.61 156 21.2 675k341 605231 49.6k1.62 1674k82 (9) Prolactin 779 22.1 10.4k1 1 157.722.3 621k20 541k17 47.7k0.86 1544k56 (9) Saline 780 21.3 14 k0.9 152.521.4 564k27 561219 45.921.04 1461k62 (9)

1 p < 0.025. 2 p < 0.001.

TABLE 5

Relative composition of intact tadpole liver after hormone treatment Liver Water Carbohydrate Protein RNA Lipid weight mdgm mglm mglm mgkm rngigm mgigm body

GH 13.722.1 91.5k2.2 8.60k0.26 72.2k6.4 24.0k1.5 (10) P 812213 15.421.2 88.4k1.4 8.44k0.30 67.1k5 20.5k0.771 (10) C 795?8 13.921.5 86.521.5 8.55k0.23 76.915 27.5k1.5 (10)

1 p < 0.001.

TABLE 6 Comparison of carbohydrate and RNA levels in tadpole tail muscle after treatment with prolactin and growth hormone

Carbohydrate (mglgm muscle RNA (mgigm muscle)

Exp. 2 Exp. 3 Exp. 2 Exp. 3

Control 8.31 20.61 3.09 20.27 1.21 k0.06 1.13k0.04 Prolactin 4.72 k0.47 2 2.45 20.10 1 1.20k0.09 1.11 &0.03 GH 6.54 k 0.75 3.41 k 0.28 1.31 k0.06 1.16k0.04

I p < 0.05. 2 p < 0,001 reduction of carbohydrate was not great metamorphic animals. In Frog Experi- but was significant (P < 0.05). The dif- ment 2, but not Frog Experiment 1, ference between the results of the two growth hormone significantly (P < 0.05) experiments was probably due to the fact reduced the ratio of fat body weight that in Experiment 4 the carbohydrate (figs. 5a,b). Prolactin had no effect on had been depleted by the previous 24-hour fat body size in either experiment. The fast. The effect of growth hormone upon absolute fat body sizes in these two ex- carbohydrate was not significant in either periments generally paralleled the rela- experiment. Neither hormone had any tive fat body values. effect on the levels of RNA in tail muscle. Fat bodies from Tadpole Experiment 4 Fat body size. Growth hormone is were also measured (fig. 6). Neither pro- known to decrease lipid stores in some lactin nor growth hormone treatment sig- animals. Fat bodies of growth hormone nificantly altered the relative (mglgm and prolactin-treated animals were body) or absolute fat body weight. These weighed in order to compare the effects experiments indicate that growth hor- of these two hormones upon amphibians mone tends to decrease lipid stores in and higher animals and in pre- and post- frogs but not in tadpoles. Prolactin has HORMONE REGULATION OF FROG GROWTH 307 no significant effect on fat stores in either of glucose in tadpoles is only about frog or tadpole. half that of frogs. Blood glucose. Because growth hor- mone has been shown to increase blood DISCUSSION glucose levels in animals, pre- and post- metamorphic Rana pipiens were tested I. General body growth after growth hormone and prolactin in- jections. Animals tested were from Frog Until recently, little information was Experiment 2 and Tadpole Experiments available concerning growth regulation 2 and 4. Growth hormone injected into by pituitary hormones in lower verte- intact frogs increased blood sugar levels brates. However, recent work (cited in by nearly 10 mg% over controls (P < the Introduction) has established a role 0.005). The prolactin group mean was of the pituitary in growth regulation in somewhat higher than controls but not amphibians, and has brought to light the significantly so (fig. 7a). Neither growth interesting observation that larval growth hormone nor prolactin have any effect on is regulated by prolactin while postmeta- tadpole blood glucose in these two experi- morphic frogs grow primarily in response ments (fig. 7b). The normal (control) level to growth hormone. The results of this

EXP 1 fXP 1

- L n

0 .. n ..... -rw.... : 8- ...... E ...... ,- D ...... : 5 71 ...... > ...... a I ...... 6 ...... ::...... 2 P ...... c 2 *g ...... Y ...... Y ...... nrx WPX+ nrx+ C GH P GH P Fig. 5 Fat body weights of intact and hypophysectomized frogs treated with growth hormone and prolactin (frog experiments 1, 2).

6- -m 0 - I: 5- SI- E !? 4- ...... Y : ...... z 0 ...... 39 ...... 0, ...... :2 ...... L .. .I .. I ...... 2 9 1 ...... C OH P C OH P Fig. 6 Fat body weights of tadpoles after treatment with growth hormone and prolactin (tadpole experiment 4). 308 BENJAMIN W. SNYDER AND B. E. FRYE

FROG TADPOLE

EX? 4

12 - 20 d m 18 -E w 16 n 0 LJ I4

d 0 12

0 2 10 P 8

C OH P C OH P C OH P Fig. 7 Blood glucose levels of intact frogs (experiment 2) and tadpoles (experiments 2, 4) after treatment with growth hormone and prolactin. study confirm the previous reports dem- or starvation are these effects of growth onstrating growth effects of growth hor- hormone counteracted, presumably by mone and prolactin in frogs and tadpoles suppression of growth hormone secretion respectively. Interpretation of these re- and accelerated secretion of glucocorti- sults must be qualified since all work to coids which promote protein degradation. date has been done with mammalian hor- A major objective of this study has been mones. We presume that homologous to determine whether growth stimulation molecules, similar to prolactin and growth by growth hormone and prolactin is ac- hormone are the physiological growth reg- companied by a similar pattern of meta- ulating hormones of the pre- and post- bolic effects conducive to protein anabo- metamorphic stages of amphibian develop- lism and carbohydrate conservation. On ment (Brown and Frye, '69a,b). the basis of such information we hoped to see a relationship between the growth 11. Chemicals composition effects and the metabolic effects of these hormones which might be related to dif- In mammals, the physiological role of ferences in the biology of the larval and growth hormone is viewed to be one of postmetamorphic stages of development. regulating a balance in the metabolism Liver. In mammals, growth hormone of protein, carbohydrate, and fat. Regu- causes increases in absolute liver weight, lation of the exact proportionality of this ratio of liver weight to body weight, total balance is not well understood, but ap- protein content (Li and Evans, '48), and pears to depend upon the nutritional liver glycogen (Russell and Bloom, '56; balance of the animal, the hormonal back- Altszuler et al., '68). The results of the ground against which growth hormone present study on frogs are in marked acts, and the developmental state of the contrast to these observations. Growth animal. For example, on a limited diet, hormone brought about a decrease in too small to support growth, net growth liver size, carbohydrate, and in one ex- can be induced by growth hormone, which periment, total protein. Prolactin also re- promotes the utilization of lipid stores and duced liver size and carbohydrate, though thus spares ingested protein for growth to a smaller extent than growth hormone, (Lee and Ayres, '36). Similarly, during but had no effect on protein content. starvation, growth hormone spares glu- Zipser et al. ('69) found that both growth cose and amino acids at the expense of hormone and prolactin treatment of Bufo body fat. Only during advanced fasting boreas and Bufo marinus reduced the to- HORMONE REGULATION OF FROG GROWTH 309 tal liver size, although prolactin was only tomy reduced muscle glycogen, as it does about one-fifth as potent as growth hor- in mammalian muscle, but unlike the mone. These authors suggested that the mammalian response, growth hormone effect of prolactin was probably due to lowered the carbohydrate levels even be- growth hormone contamination of their low the level of hypophysectomized ani- prolactin preparations (which could also mals. Muscle carbohydrate of intact frogs be true in our experiments). In B. boreas was also reduced by growth hormone. glycogen content was reduced to less than A striking difference was observed be- one-half the control value, which would tween the response of tadpole and frog probably account for the reduced liver muscle in that tadpole tail muscle did size. In B. marinus they found that the not respond to growth hormone in any of reduction in liver size was due to loss of the parameters measured. Prolactin low- lipid. ered carbohydrate levels in both frogs and In tadpoles prolactin had no effect upon tadpoles but stimulated growth only in either liver carbohydrate or absolute liver tadpoles. Thus, there is clearly a differ- size. It is noteworthy that prolactin failed ence in the hormone sensitivity of tadpole to elicit an increase in liver size in tad- tail muscle and frog gastrocnemius mus- poles under conditions when an overall cle. Unfortunately, it was not possible to increase in body size was stimulated. As examine homologous muscles in the two noted earlier, prolactin tended to have the developmental stages. Consequently it is same effects as growth hormone on frog not possible to determine whether the dif- liver, although the magnitude was usu- ference in response is inherent in the two ally smaller. Reports in the literature types of muscle studied, or whether there concerning the effects of prolactin on is a change in the hormone sensitivity of mammalian liver size and glycogen con- skeletal muscle in general at metamor- tent are conflicting (Elghamry et al., phosis. '66; Tutwiller, '70) and comparisons with Lipid stores. Injection of growth hor- amphibians will not be made at this time. mones into mammals has long been Muscle. Mammals are known to re- known to reduce carcass lipid stores (Lee spond to growth hormone stimulation with and Ayres, '36; Li et al., '49; Li and increases in total muscle size paralleled Evans, '48). Liver lipid of hypophysecto- by an increase in content of protein, mized rats is also reduced by growth hor- water, and nucleic acid (Li et al., '49; mone injections (Li et al., '49). In this Scow and Hagan, '65; Cheek et al., '65; study, growth hormone treatment reduced Reid, '56; Di Stefan0 et al., '53). In fat body lipid but had no effect on liver addition growth hormone has been shown lipid of frogs. Prolactin did not signifi- to have a glycostatic effect on muscle of cantly change frog fat body or liver lipid fasting intact and hypophysectomized levels. This result is similar to the obser- mammals (Russell and Bloom, '56). In the vation of Zipser et al. ('69) that growth present study, growth hormone did not hormone reduces fat body weights in Bufo increase total muscle size or total protein boreas and Bufo marinus. In B. marinus, content significantly, but this was prob- but not in B. boreas, liver lipid was also ably due to the fact that initial and reduced by growth hormone. Prolactin final muscle size could not be compared had no effect on toad lipid when given at rather than to the lack of growth hor- the same dose as growth hormone. Thus, mone effect. Growth hormone did increase in frogs, toads, and mammals, there is the levels of nucleic acids and water in a correlation between growth stimulation frog muscle, and in that a growth by growth hormone and reduction in total effect was observed. In addition, in anoth- energy stores, but no such correlation er phase of this study to be reported exists between prolactin-induced growth separately, stimulation of incorporation and energy stores in the tadpole. of '%-labeled amino acid into muscle In mammals, the lipolytic effect of protein was observed (Snyder, '70). growth hormone has secondary effects on Frog and mammalian muscle carbohy- carbohydrate metabolism (Randle, '63). drate respond differently to growth hor- It is thought that growth hormone pro- mone stimulation. In frogs, hypophysec- motes the mobilization of free fatty acids 310 BENJAMIN W. SNYDER AND B. E. FRYE (FFA) from lipid stores which in muscle through glycolysis and the Cori cycle as are oxidized preferentially over glucose. the major source of energy for contraction. Tissue glycogen and blood glucose levels c. Frogs undergo prolonged fasting are partially spared during fasting, exer- during winter hibernation, during which cise, etc. This metabolic interrelationship time they maintain high levels of liver between lipid and carbohydrate metabo- and muscle glycogen (Farrer, thesis in lism would appear not to occur in am- preparation; Mizell, * ’65). Calculations phibians since growth hormone simul- based on the metabolic rate and the total taneously reduces fat body lipid and fat and glucogen stores indicate that high carbohydrate stores. Although the mecha- levels of liver and muscle glycogen must nism by which growth hormone induces be maintained by gluconeogenesis (Far- carbohydrate and lipid breakdown in frogs rer, thesis in Preparation). is not known, the results of this study d. Blood glucose values are low and suggest that it must be different from variable. Typical values for postmetamor- mammals. phic amphibians range from 1540 mg% with values sometimes reported to be un- measurably low (Wright, ’59; Bartell, ’69; 111. Biological significance of the growth Farrer, thesis in preparation). and metabolic effects of growth On the basis of these points we can hormone and prolactin in theorize that the major mechanism for amphibians regulating carbohydrate content of the body in frogs is gluconeogenesis and that In view of the adaptive significance a growth hormone glycostatic mechanism which has been attached to the inter- as observed in mammals would be rela- related growth and metabolic effects of tively unimportant. The relative unimpor- growth hormone in mammals, departures tance of the glycostatic mechanism is from this pattern, especially with respect suggested by the observation of high tis- to the glycogenolytic rather than glyco- sue levels of glycogen in both intact and static effect, are surprising and require hypophysectomized frogs, and by the fact explanation. At present there are very that even at unmeasurably low levels of few known facts about amphibian carbo- blood glucose adverse effects upon the hydrate metabolism from which to weave central are not observed. a consistent interpretation of the glyco- In short, glucose per se may not be an genolytic, lipolytic, and protein anabolic important or essential substrate for ner- effects of growth hormone. The pertinent vous tissues as it is in mammals and, points which are presently known, at least therefore, the importance of conserving tentatively, which bear upon this point glycogen and blood glucose during fasting are: may be slight. In fact, if one presumes a. The primary diet of frogs is insects that carbohydrate is always produced in of which the two largest components are abundance through gluconeogenesis, then protein and fat. As is characteristic of the glycogenolytic effect of growth hor- carnivores in general, carbohydrate is a mone seen in frogs may be viewed as a relatively smaller proportion of the caloric mechanism for diverting carbohydrate into intake than in herbivores. energetic pathways, including growth, b. Glycogen is an important metabolic while conserving tissue protein. This idea substrate, perhaps the major one in cer- is consistent with the observation that tain tissues. This may be surmised from liver and muscle glycogen are lowest in two observations: that glycogen levels of the summer when growth and presuma- liver and skeletal muscle are unusually bly growth hormone secretion is highest high compared to mammals (Farrer, the- (Farrer, thesis in preparation). Nothing sis in preparation); and that frog skeletal is known, of course, about seasonal pat- muscle is “white” muscle which is very terns of growth hormone secretion in am- poorly vascularized, fatigues rapidly, and phibians, but in terms of correlation generates large anounts of lactic acid between growth hormone and the inter- during exercise. It presumably depends, related metabolic effects described in this therefore, upon glycogen metabolism study, it would be expected to be highest HORMONE REGULATION OF FROG GROWTH 31 1 during summer when food is most abun- and prolactin on fat and carbohydrate dant. stores. There are several ways in which A mechanism must exist by which the tadpole energy metabolism differs from protein anabolic effects of growth hor- that of frog. First, tadpoles do not nor- mone and the protein catabolic effects of mally experience periods of starvation - gluconeogenic hormones are kept from food is always present in the digestive being antagonistic. In the frog, the bal- system. Secondly, anuran tadpoles are her- ance between protein anabolism (under bivores and consume relatively more car- the influence of growth hormone) and bohydrate and less protein than frogs. protein catabolism (presumably under the Thus, hormonal mechanisms for regulat- control of ) could be regu- ing carbohydrate homeostasis and gluco- lated by seasonal variations in the rela- neogenesis may not be as important in tive activity of these two hormones. Al- tadpoles as in frogs. Third, lipid is the though existing evidence does not allow primary form of energy storage in tad- for positive identification of the hormones poles (cf. table 7 and figs. 23, 24). This responsible for gluconeogenesis, the glu- is probably advantageous during the ener- cocorticoids are implicated by the fact getically demanding period of metamor- that they are gluconeogenic in mammals, phosis since fat is the most efficient way and by the fact that adrenalectomy causes to store large amounts of energy in a a depletion of frog muscle and liver gly- small space. In view of this, it is adap- cogen (Gorbman, '64). Once growth hor- tively advantageous that tadpole growth mone and the gluconeogenic hormones is regulated by prolactin which is con- set the pattern of metabolism other hor- ducive to lipid deposition rather than by mones may operate to augment these growth hormone which causes fat break- patterns. Thus, in Rana pipiens, sub- down. strates have been shown to be metabo- In addition, at least part of the growth lized primarily to glycogen in winter promoting effect of prolactin in tadpoles months and toward oxidation in summer. is caused by suppression of thyroid hor- Insulin increased both summer and win- mone secretion or antagonism of thyroid ter depending upon the season it was hormone action in target tissues (Gona, tested (Gourley et al., '69). '67; Bern et al., '67). Selection of prolac- The findings of this study additionally tin as the larval growth regulator may suggest that growth hormone stimulates have been favored by the fact that this net protein catabolism in liver while stim- molecule combines the properties of ulating overall growth and protein syn- blocking metamorphosis and stimulating thesis in the rest of the body. Growth growth - processes that are antagonistic hormone appears to combine the proper- during larval life. ties of promoting protein breakdown in One of the most important problems the major gluconeogenic tissue, liver, in biology involves the changing patterns while promoting growth in other tissues. of gene expression during development. Thus, carcus protein can be conserved The fact that amphibians appear to have under the influence of growth hormone different growth hormones at different while dietary protein catabolism is aug- stages of development implies that a de- mented, a condition which would appear velopmental change in hormone sensitiv- to be necessary in an organism which ity occurs at metamorphosis. This study uses large amounts of dietary protein as documents this change by demonstrating an energy substrate and as a source of an alteration in hormone sensitivity of amino acids for growth. muscle, liver, and fat body. The mecha- One of the most interesting aspects of nism by which this changeover occurs is amphibian growth concerns the reason of considerable interest. One possibility why there should be different growth is that different tissues in tadpoles and factors for the pre- and postmetamorphic frogs are sensitive to prolactin and growth stages of development. While there is no hormone respectively, and it is the rela- satisfactory explanation at this time, the tive proportion of these tissues that difference in the two stages may be re- change at metamorphosis. Thus, the large lated to the effects of growth hormone tail of the tadpole might be sensitive to 312 BENJAMIN W. SNYDER AND B. E. FRYE prolactin while the skeletal muscle and ferences in the responses of tadpoles and liver of frog may respond to growth hor- frogs to prolactin and growth hormone. mone. In this case, changing patterns of prolactin and growth hormone sensitivity LITERATURE CITED at metamorphosis would not involve cellu- Altszuler, N., I. Rathgeb, B. Winkler and R. C. lar differentiation per se, but would re- de Bod0 1968 The effects of growth hormone flect a decline in numbers of prolactin- on carbohydrate and lipid metabolism in the sensitive cells, and a rise in numbers of dog. Ann. N. Y. Acad. Sci., 148: 441-458. Ashwell, G. 1966 New colorimetric methods of growth hormone sensitive cells compris- sugar analysis. VII. The phenol-sulfuric acid ing the hormone sensitive tissues. reaction for carbohydrates. In: Methods in Alternatively, at metamorphosis a Enzymology. Vol. 8. E. F. Neufeld and V. change in hormone sensitivity may occur Ginsburg, eds. Academic Press, New York, pp. 93-94. within the individual cells of prolactin Bartell, M. 1969 The role of the pituitary in and growth hormone responsive tissues. blood glucose regulation in larval and adult This would amount to a process of cellu- salamanders, Ambystomn tigiinum. Ph.D. Dis- lar differentiation in which a growth-hor- sertation, Department of Zoology, University of Michigan, Ann Arbor, Michigan, 145 pp. mone response mechanism would be syn- Berman, R., R. C. Strohman, C. S. Nicoll and thesized and/or activated and the prolactin H. Bern 1964 Growth-promoting effects of response mechanism inactivated at the mammalian prolactin and growth hormone in time of metamorphosis. There are of tadpoles of Ram c<[tesbeinnu.J. Exp. Zool., 156: 353-360. course a large number of levels and ways Bern, H. A., C. S. Nicoll and R. C. Strohman in which such a change might be ef- 1967 Prolactin and tadpole growth. Proc. SOC. fected, ranging from changes in DNA ac- Exptl. Biol. Med., 126: 518-520. tivity to changes in membrane structure Blatt, L. M., K. A. Slickers and K. H. Kim 1969 Effect of prolactin on thyroxin-induced meta- or protein conformation. If we assume morphosis. Endocrinol., 85: 1213-1215. that the same populations of cells which Brown, P., and B. E. Frye 1969a Effects of comprise the liver and the fat body of the prolactin and growth hormone on growth and tadpole, constitute these same structures metamorphosis of tadpoles of the frog, Rnnn pipiens. Gen. Comp. Endocrinol., 13: 126-138. in the frog, then the results of this study 1969b Effects of hypophysectomy, pro- suggest that the mechanism of change in lactin, and growth hormone on growth of post- prolactin and growth hormone sensitivity metamorphic frogs. Gen. Comp. Endocrinol., is indeed one of cellular differentiation. 13: 139-145. Burton, K. 1956 A study of the conditions and The assumption that the bulk of the cells mechanism of the diphenylamine reaction for of the larval tissues do persist and form the colorimetric estimation of deoxyribonucleic the liver and fat bodies of the frog has acid. Biochem. J., 62: 315-323. not, to our knowledge, been critically Cheek, D. B., G. K. Powell and R. E. Scott tested. 1965 Growth of muscle cells (size and num- ber) and liver DNA in rats and Snell Smith Finally, the change in hormone sensi- mice with insufficient pituitary, thyroid, or tes- tivity at metamorphosis could reflect a ticular function. Bull. Johns Hopkins Hosp., change in sensitivity to, or secretion of, 117: 30G-321. another hormone which modifies the ac- Derby, A,, and W. Etkin 1968 Thyroxine in- duced tail resorption in vitl’o as affected by tion of growth hormone or prolactin. For anterior pituitary hormones. J. Exp. Zool., 169: instance, prolactin may produce part of 1-8. its growth effect in tadpoles by inhibiting Di Stefano, H. S., H. F. Diermeier and J. Tepper- man 1953 Effects of growth hormone on the thyroxin-induced breakdown of tad- nucleic acid and protein content of rat liver pole tissues [note the lack of stimulation cells. Endocrinol., 57: 158-167. of amino acid incorporation obtained in Elghamry, M. I., A. Said and S. A. Elmougy this study and the prolactin inhibition of 1966 The effect of lactogenic hormone on liver glucogen and blood glucose in ovariecto- thyroxin-induced tail breakdown observed mized mice. Naturwissenschaften, 53: 530. by Blatt et al. (‘69) and Derby and Et- Enemar, A,, B. Essvik and R. Klang 1968 kin (‘68)j. Although prolactin appears to Growth-promoting effects of ovine somatotro- have a growth effect independent of its phin and prolactin in tadpoles of Rnm trm- pol’nrin. Gen. Comp. Endocrinol., 11; 328-331. antithyroid effect (Brown and Frye, ’69), Etkin, W., and R. Lehrer 1960 Excess growth hormone interactions of this type cannot in tadpoles after transplantation of the adeno- yet be discounted in explaining the dif- hypophysis. Endocrinol., 67: 457466. HORMONE REGULATION OF FROG GROWTH 313

Etkin, W., and A. G. Gona 1967a Antithyroid Scow, R. O., and S. N. Hagan 1965 Effect of action of prolactin in the frog. Life Sci., 6: testosterone proprionate and growth hormone 703-707. on growth and chemical composition of mus- 1967b Antagonism between prolactin cle and other tissues in hypophysectomized and thyroid hormone in amphibian develop- male rats. Endocrinol., 77: 852-858. ment. J. Exp. Zool., 165: 249-258. Shibko, S., P. Kiovistoinen, C. A. Tratnyek, A. R. Gona, A. G. 1968 Radioiodine studies on pro- Newhall and L. Friedman 1967 A method lactin action in tadpoles. Gen. Comp. Endo- for sequential quantitative separation and deter- crinol., 11: 278-283. mination of protein, RNA, DNA, lipid, and gly- 1967 Prolactin as a goitrogenic agent cogen from a single rat liver homogenate or in amphibia. Endocrinol., 81 : 748-754. from a subcellular fraction. Anal. Biochem., Gorbman, A. 1964 Endocrinology of the am- 19: 514-528. phibia. In: Physiology of the Amphibia. J. A. Snedecor, G. W., and W. G. Cochran 1967 Moore, ed. Academic Press, New York, pp. Statistical Methods. Iowa State University Press, 371425. Ames, Iowa. Gourley, D. R. H., T. K. Suh and L. L. Brunton 1969 Seasonal differences and the effect of Snyder, B. W. 1970 Physiological responses of insulin on pyruvate uptake, oxidation and syn- larval and postmetamorphic Rana pipiens to thesis to glycogen by frog skeletal muscle. Comp. growth hormone and prolactin. Ph.D. Disserta- Biochem. Physiol., 29: 509-524. tion, Department of Zoology, University of Lee, M. O., and G. B. Ayres 1936 The compo- Michigan, Ann Arbor, Michigan, 95 pp. sition of weight lost and the nitrogen partition Steinmetz, C. 1954 Some effects of thyroxin of tissues in rats after hypophysectomy. Endo- and antithyroid compounds on tadpoles and crinol., 20: 489-495. their relation to hormonal control of growth. Li, C. H., and H. M. Evans 1948 The bio- Physiol. Zool., 27: 2840. chemistry of pituitary growth hormone. Recent Taylor, A. C., and J. J. Kollros 1946 Stages Progr. Horm. Res., 3: 344. in the normal development of Rana pipiens Li, C. H., M. E. Simpson and H. M. Evans larvae. Anat. Rec., 94: 7-23. 1949 The growth of hypophysectomized fe- Tutwiler, G. F. 1970 Physiochemical and bio- male rats following chronic treatment with pure logical characterization of a diabetogenic pep- pituitary growth hormone. Growth, 13: 171- tide from bovine adenohypophysis. Ph.D. Dis- 174. sertation, Department of Biological Chemistry, Lowry, 0. H., N. J. Rosebrough, A. L. Farr and University of Michigan, Ann Arbor, Michigan, R. J. Randall 1951 Protein measurement 159 pp. with the Folin phenol reagent. J. Biol. Chem., Wannemacher, R. W., W. L. Banks and W. H. 193: 265275. Wunner 1965 Use of a single tissue extract Mizell, S. 1965 Seasonal changes in energy re- to determine cellular protein and nucleic acid serves in the common frog, Rana pipiens. J. concentrations and rate of amino acid incor- Cell. and Comp. Physiol., 66: 251-258. poration. Anal. Biochem., 11 : 320-326. Randle, P. J. 1963 Endocrine control of metab- olism. Ann. Rev. of Physiol., 25: 291-324. Winkler, B., R. Steele, N. Altszuler and R. C. Reid, E. 1956 Hormonal effects on liver and De Bod0 1964 Effect of growth hormone on kidney cytoplasm. J. Endocrinol., 13: 319-329. fatty acid metabolism. Am. J. Physiol., 206: Remy, C., and J. J. Bounhiol 1965 Croissance 174-1 78. exageree des tetards de crapaud accoucher en- Wright, P. A. 1959 Blood sugar studies in the tires ou prives de leur hypophyses et subissant bullfrog Rana catesbeiana. Endocrinol., 64. 551- un traitement par la prolactine. Comp. Rend. 558. SOC.Biol., 159: 1532-1535. Wright, P. A,, and A. R. Flathers 1961 Fa- Remy, C., and J. J. Bounhiol 1966 Gigantisme cilitation of pituitary induced ovulation by pro- experimental obtenu par intervention, chiruge- gesterone in early fall. Proc. SOC.Exptl. Biol. cale et traitements hormonaux chez le tetard Med., 106: 346-347. der crapaud accoucher. Ann. Endocrinol., Zipser, R. D., P. Licht and H. Bern 1969 Com- Paris, 27: Suppl., 377-382. parative effects of mammalian prolactin and Russell, J. A., and W. Bloom 1956 Hormonal growth hormone on growth in the toads Bufo control of glycogen in the heart and other tis- boreas and Bufo marinus. Gen. Comp. Endo- sues in rats. Endocrinol., 58: 83-94. crinol., 13: 383-391.