Some Questions Concerning the Syngnathidae Brood Pouch

BULLETIN OF MARINE SCIENCE, 49(3): 741-747,1991

SOME QUESTIONS CONCERNING THE SYNGNATHIDAE BROOD POUCH

Marie Y. Azzarello

ABSTRACT For more than a century the physiological role of the Syngnathidae brood pouch has been the subject of scientific interest and debate. Some of the earliest investigators purported that the highly vascular brood pouch was physiologically adapted for the reception of fertilized eggs and for the sustenance of the embryos (i.e., a "pseudo-placenta"). Others posited that the brood pouch served as an osmoregulatory organ for the developing embryos. To determine whether the primary physiological role of the brood pouch is one of nutrition or osmoregulation, Syngnathus scove/li embryos were removed from the brood pouch at different developmental stages (4.0-13.0 mm TL), placed in sterilized, aerated, artificial seawater hyper- or iso-osmotic to the blood and pouch fluid, to which no nutritive substances were added. In hyperosmotic media 25.7% of the in vitro embryos completed their normal gestation versus 18.7% in iso-osmotic media. These results appear to indicate that the male Syng- nathidae brood pouch serves neither as the primary nutritional source nor as an osmotic buffer for the developing embryos after a length of 4.0 mm TL is achieved.

Members of the Syngnathidae manifest atypical reproductive behavior and parental care. Reproduction is ovoviviparous with a complete reversal of the usual maternal and paternal brooding roles. Large telolecithal eggs produced by the female are fertilized by the male the moment they are deposited in his brood pouch. Embryos are then incubated throughout their entire gestation period, which varies depending upon genus and species, in the paternal brood pouch (Gill, 1905; Hubbs, 1943; Breder and Rosen, 1966). The physiological function of the brood pouch has been an area of much spec- ulation and investigation having involved the work of physiologists, embryolo- gists, anatomists, and histologists, for more than a century; its intriguing history is briefly described. Inasmuch as eggs and embryos of seahorses would not develop in seawater after their removal from the brood pouch, some of the earliest investigators concluded that the male provided a source of sustenance for the charges in his marsupial pouch. Lockwood (1867) observed that upon the male's reception of the eggs, the wall of the pouch became enriched internally with fat, and that upon birth the pouch hung flaccid becoming a thin membrane. Utilizing a sectioning technique, Huot (1902) demonstrated that each eggwas encompassed by epithelium perfused with a rich network of blood vessels supplying nourishment presumably by os- mOSIS. Gill's observations (1905) of the thickened, vascular, ovigerous pouch, led him to concur with earlier investigators that the male was not only physiologically adapted for the reception of eggs, but for their sustenance as well. Based solely on histological evidence, Gudger (1905) believed he had established evidence confirming the transfer of nutrients from the male pipefish's marsupium to its developing embryos. Thevenin (1936) noted considerable changes in the con- junctive tissue lining the pouch at the moment of conception. Capillaries multi- plied and enlarged causing the tissue to swell and become sponge-like. Within this "pseudo-placental" tissue the embryos remained quiescent until all their yolk was absorbed. Hudson and Hardy (1975) purported that the narrow end of the seahorse egg embeds in the wall of the brood pouch and that the transfer of nutrients takes place through this part of the egg.

741 742 BULLETIN OF MARINE SCIENCE, YOLo 49, NO.3, 1991

More recently, Haresign and Shumway (1980) introduced the use of a radio- active tracer to demonstrate the transfer of low molecular weight compounds across the epithelium of the male's brood pouch. They injected a non-metabolized radioactively labeled amino acid (alpha-amino isobutyric acid) into the marsupium of Syngnathus scovelli. Although their data indicated embryonic incorpo- ration of the labeled amino acid, these investigators were entremely cautious in drawing conclusions as to the permeability of the marsupium to nutrient exchange, having noted the likelihood of contamination with the methods they employed. The nutritive role was first questioned by Leiner (1934) who examined the osmotic pressure of the pouch fluid of Hippocampus brevirostris and H. guttulatus. He found that during early incubation, this pressure was similar to that of the male's blood while it approached that of seawater late in gestation. Leiner was also successful in getting a few embryos to develop partially in dilute seawater. An implication of these findings was that the male provided an optimum osmotic chamber for the developing young. The completely fused brood pouch of the seahorse, with its internal walls densely perfused with blood vessels, according to Nikolsky (1963), undoubtedly provided a means to supply oxygen to the developing embryos during gestation. Thirty years after Leiner's study, Linton and Soloff (1964) demonstrated that while the serum Na+ remained relatively constant, the Na+ ion concentration in the brood pouch fluid increased as the brood pouch epithelium actively trans- ported Na+. After parturition, the brood pouch Na+ concentration dropped back to blood Na+ levels. This low Na+ level represented the normal resting state signifying that the male was preparing for the reception of eggs. Calcium, on the other hand, being incorporated into the embryonic skeleton as well as the dermal exoskeleton, showed a decrease in concentration in the pouch fluid during gestation. This implied that the pouch epithelium may thus serve as a selective osmoregulatory organ protecting the embryos from a hyperosmotic environment by introducing them to higher salinities as their osmoregulatory systems (kidneys and gills) develop. Moreover, the fact that a small percentage of embryos survived to the stage of complete yolk absorption, after being removed from the brood pouch and introduced into 0.4 seawater (Linton and Soloff), seemed to preclude a dependence of the young on the male for nutrition. Spannhof and Bremer (1969), utilizing histophysiological techniques, noted functional changes in epithelial cells lining the brood pouch. They observed an increase in vascularization and that the epithelium consisted of thick cuboidal cells resembling cells having a secretory function. They concluded that the main function of the brood pouch was to ensure sufficient embryonal gas exchange. Kronester- Frei (1975) described the morphological changes in the brood pouch epithelium of Nerophis, a related genus, and concluded that the morphological prerequisites for transport of inorganic ions from the male's circulatory system to the embryos seemed to exist. In 1980, Quast and Howe investigated the osmotic role of the brood pouch in Syngnathus scovelli and determined that throughout incubation the pouch fluid osmolality was regulated near blood osmolality (370 (± 5) mOsm· kg-I). It was their surmise that the role of the brood pouch was to acclimate the embryos by regulating its osmolality during incubation to simulate that of the external environment. In contrast to the proposed nutritional and osmoregulatory theories as described, Gross and Sargent (1985) and others examined the costs, benefits, and net advantage to both sexes during procreation. According to these two investigators, the one benefit incurred by parental care is increased survivorship of the AZZARELLO: SYNGNATHIDAE BROOD POUCH 743 young. However, three potential costs to care are incurred: a mating cost, a survivorship cost, and a future fertility cost. Since only the male may lose spawnings by giving care, males alone have a potential mating cost. In general, the costs of care are probably less in males because male fertility is not a function of body growth. By contrast, fecundity accelerates with female size, which in turn increases the female's future fertility cost. Thus, male parental care is explained by these investigators, in terms of the male paying a smaller future cost relative to the female. Clearly, the scientific literature is divided as to the question of whether the role of the brood pouch is that of nutrition or ionic-osmotic regulation. The objectives of this study were two-fold: (I) to determine whether sustenance is the primary physiological function of the brood pouch, and (2) to determine whether hyper- and iso-osmotic incubating media have different effects upon the survival of embryos excised from the brood pouch.

METHODOLOGY

To test these objectives the following simplified procedure was devised: (I) gravid male Syngnathus scovelli were collected and kept in 38- and 76-liter aquaria filled with artificial seawater (30.60/00or 12.90/00at 24.5°C). Their diet consisted of newly hatched Anemia salina nauplii supplemented with zooplankton on alternate days; (2) pipe fishes were acclimated in their respective salinities for 24-96 h, depending upon the developmental stage of the embryos at the time of collection; (3) a sterilized scalpel was used to separate the flaps and expose the fertilized ova (the membranous lateral flaps of the brood pouch meet medially but are unfused); (4) using a sterilized I" x 1/100 ml pipette, the embryos were suctioned out and transferred to a 400-ml jar containing a solution of sterilized, aerated, artificial seawater at the respective salinity of the incubating male; (5) the excised embryos were incubated at 24.5°C for the duration of the natural gestation as indicated by the embryos left intact in the male's marsupium; (6) the male was returned to the holding tank and observed until parturition took place. This procedure was repeated 22 times. When embryos were excised, those which did not have an intact yolk sac and whose survivability was highly improbable, were preserved in neutrally buffered formalin for the purpose of measuring the embryos at the time of removal. These embryos were not counted as part of the total number removed when determining percent survival.

RESULTS Ten of a total of 22 experiments yielded embryos which survived in vitro after removal from the male pipefish's brood pouch. Of those 10 experiments, 7 were under hyperosmotic conditions. The percent survival of all excised embryos which were incubated in media hyperosmotic (30.60/00) to the blood and pouch fluid was 25.7% compared to 18.7% of those incubated in an iso-osmotic media (12.90/00) (Table 1). [Survivors are defined as those embryos which completed their gestation and subsisted solely on the yolk sac to meet their energy needs. Before parturition of the in vivo embryos, the in vitro embryos were swimming and osmoregulating.] The time embryos developed in vitro ranged from 1 to 9 days, corresponding to a size of 11.0 mm and 5.0 rom TL, respectively, at the time of their removal. There seems to be no correlation between percent survival and the length of the embryos at the time of removal, or with time (month) of the breeding season. Mortality was caused by damage to the yolk sac or the excised embryos, or for unknown reasons (Table 2). Upon parturition, during one hyperosmotic run, the yolk sac was completely absorbed in the embryos which emerged from the brood pouch, whereas the yolk sacs of the in vitro embryos were still apparent. 744 BULLETIN OF MARINE SCIENCE, VOL. 49, NO.3, 1991

I""- I""- 1""-0\0\0\ .,;.,; ";NNN I""- I""- 1""-101010 00 00 00 ~Mf'4"'l

10 10 10 10 10 00.....•.....•.....•00 00 •..•.•00 00•..•.• I""-'

o~ 10 00

___oN'o t::..=-=-=- -O-M

E E E E E E E E E E E <"\ E~ I "i - 0\ N -0 -

10 10 10 00.....•.....•.....•00 00 '

Table 2. Summary of results of embryos excised from the brood pouch of Syngnathus scovelli which died in 30.60/00(874.7 mOsm/kg) and 12.90/00(362.9 mOsm/kg) incubating media

Number of embryos mOsm/kg Date excised Size TL excised osmolality Date died Reason for mortality 06/27/86 05-06 mm 4 874.7 06/29/86 yolk sac damaged 07/30/86 04-05 mm 4 874.7 07/31/86 yolk sac damaged 07/31/86 03mm 4 874.7 08/02/86 unknown 08/06/86 06-08 mm 6 874.7 08/08/86 unknown 08/10/86 05 mm 4 874.7 08/12/86 unknown 08/17/86 na 4 874.7 08/20/86 unknown 08/21/86 04-05 mm 5 874.7 08/23/86 unknown 09/23/86 01 mm 6 362.9 09/23/86 embryos disintegrated 09/29/86 02mm 6 362.9 10/02/86 embryos damaged when removed 10/22/86 03mm 13 362.9 10/25/86 unknown 10/22/86 04mm 10 362.9 10/24/86 unknown 11/10/86 na 8 362.9 11/11/86 unknown

"'"Data not available.

DISCUSSION No nutritive substances were added to the incubating media of the in vitro embryos, yet these excised embryos completed their development outside the brood pouch. In terms of survival these results indicate that the male is not the primary source of nutrition and that the yolk sac is the principal, if not the sole, source of energy for the young. In one experiment, seven embryos which measured 4.0-5.0 mm TL upon removal were incubated for the duration of their gestation (6 days) in a hyperosmotic solution. At the time of parturition, five of the seven embryos which survived measured circa 10.0 mm TL, while embryos incubated in vivo measured circa 13.0 mm TL. From a replication of these conditions, that is, of seven embryos (measuring 5.0 mm TL) incubated in hyperosmotic media, one survived and grew to a length of 12.0+ mm TL. At parturition the in vivo juveniles emerged at 13.0 mm TL. There are several possible explanations for these length differences: (1) Though the brood pouch of the male may not play an essential role in providing nutrition for the survival ofthe embryos, it is possible there is a transfer of growth factors (i.e., hormones) from male to embryos across the brood pouch epithelium. (2) The fact that the in vitro embryos were making attempts to swim as early as 3 days before parturition, with intermittent swimming to the surface of the jar observed during the next 48 h, raises the possibility that the energy they used to locomote may also have deprived them of their full growth. (3) The removal of embryos from the brood pouch and their introduction into a hyperosmotic incubating media may have created a stressful situation thereby stunting their growth. Moreover, under these atypical osmotic conditions it would be necessary to shunt some of the total energy to osmoregulation depriving them of their full growth potential. It is worth noting that those embryos incubated in vitro in hyperosmotic media were also smaller than those embryos incubated in an iso-osmotic media. This observation supports assumptions (2) and (3) for the following reasons. In vitro embryos incubated in both the hyper- and iso-osmotic solutions were expending energy on locomotion. However, embryos incubated in hyperosmotic media were diverting energy into osmoregulation as well. These embryos were smaller by 0.5- 3.0 mm (mean difference 0.4 mm). A t-test confirmed that the difference in size 746 BULLETIN OF MARINE SCIENCE, VOL. 49, NO.3, 1991

at parturition between the hyper- and iso-osmotic groups was significant (t(38) = - 15.03, P < 0.5). In regard to the presence of a yolk sac in the in vitro embryos but not in the in vivo embryos at parturition, the interpretion is not clear-cut. However, it was not unusual to observe some in vivo embryos of the same brood, emerge with and others without a yolk sac. It has been cited in the literature that male pipe fishes will accept eggs from several females, resulting in embryos at different stages of development during anyone gestation period. Gudger (1905) observed as many as three groups of embryos at different developmental stages in the brood pouch of Syngnathusfloridae. Herald (1941) considered the frequency of occurrence of embryos at different stages to be common among pipefishes. This may explain why I observed, on three separate occasions, juveniles of the same brood, emerge 24 h apart. Furthermore, contortions of the male during parturition may induce juveniles with yolk sacs to emerge prematurely before their full gestation period. Hasse (1974) reported the absence of a yolk sac on Syngnathoides biaculeatus juveniles emerging from the brood pouch. He noted that Sudarsan (1968) incor- rectly described newly emerged S. biacu/eatus as having a yolk sac, when, in fact, Sudarsan's specimen corresponded to the developmental stage Hasse observed 5 days before parturition at which time an embryo was accidentally removed from the brood pouch. Assuming the role of the brood pouch is to maintain a constant osmotic pressure of the pouch fluid, despite fluctuating conditions in the environment, one would expect that the in vitro embryos, which were introduced into an incubating me- dium, whose osmolality was nearly two and one-half times greater than that of the pouch fluid, to have a smaller percentage of survivors. However, the Mann- Whitney Wilcoxon test indicated (P = 0.628) that there was no significant difference between the percent survival of embryos incubated at the higher compared to the lower osmolality. [These results reflect the fact that at both osmolalities the number of times there were zero survivors equalled the times there were survivors (numbering 1 to 10).] Though a larger sample size may yield results indicative of a significant difference in the ability to survive under different salinity gradients, these preliminary findings indicated that survivability is not a function of osmolality. Consequently, the role of the brood pouch as an osmoregulatory organ, which provides an osmotic buffer to the extremes of the environment after a length of 4.0-5.0 mm TL is achieved, is questioned.

CONCLUSIONS Excised embryos, ranging in size from 4.0-13.0 mm TL, completed their normal development in vitro in both an iso-osmotic and hyperosmotic, sterilized, solution of artificial seawater, to which no nutritive substances were added. Between one to nine days later, the intact embryos within the brood pouch emerged. In two hyperosmotic experiments, it was noted that at parturition, although the in vitro embryos were 0.5-3.0 mm smaller than those incubated in vivo, the former appeared to have developed normally, that is, they were feeding, swimming, and osmoregulating. Implications for this difference in length were ascribed to a possible nutritional advantage in terms of a transfer of growth factors from the male to the in vivo embryos, and/or energy diverted into locomotion and osmoregulation which deprived the in vitro embryos from reaching their full growth potential. The results of this study indicated that the male is not essential for nutritional survival, and that the yolk sac is the primary energy source for the developing embryos. Though it is apparent that one cannot rule out the role of the brood pouch as being an osmoregulatory organ for embryos smaller than 4.0 mm TL, AZZARELLO:SYNGNATHIDAEBROODPOUCH 747 that is, the period preceding the development of the embryos' gills, chloride cells, and kidneys, the question arises as to the function of the brood pouch after this critical size is attained. Clearly, one needs to investigate the survivability of eggs fertilized in vitro to rule out early osmoregulation as the function of the brood pouch before this critical size is reached. Addressing the question of the function of the brood pouch after the embryos are capable of independent osmoregulation, one wonders whether the brood pouch simply provides a safe haven for the developing embryos. Is it a means to achieve maximum growth potential to reduce predatory pressure once they emerge as juveniles? Or, perhaps, one needs to reconsider the ramifications of Gross and Sargent's (1985) work, that is, rather than searching for the physiological reason, one should investigate this rather atypical mode of reproduction in terms of energetics.

ACKNOWLEDGMENT

I wish to thank J. C. Briggs for his support throughout this work.

LITERATURE CITED

Breder, C. M., Jr. and D. E. Rosen. 1966. Modes of reproduction in fishes. Natural History Press, Garden City, New York. 941 pp. Gill, T. 1905. The life history of the sea-horses (Hippocampids). Proc. U.S. Natl. Mus. 28: 805- 814. Gross, M. R. and R. C. Sargent. 1985. The evolution of male and female parental care in fishes. Amer. Zool. 25: 807-822. Gudger, E. W. 1905. The breeding habits and segmentation of the eggs of the pipefish, Siphostoma floridae. Proc. U.S. Natl. Mus. 29: 447-499. Haresign, T. W. and S. E. Shumway. 1980. Permeability of the marsupium of the pipefish Syngnathus Juscus to I"C-a1pha amino isobutyric acid. Compo Biochem. Physiol. 69a: 603-604. Hasse, J. J. 1974. Observations on egg hatching of Syngnathoides biaculeatus (Bloch) (Pisces: Syng- nathidae). Micronesica 10(2): 279-283. Herald, E. S. 1941. A systematic analysis of variation in the western American pipefish, Syngnathus californiensis. Stanford 1chthyol. Bull. 2: 49-73. Hubbs, C. L. 1943. Terminology of the early stages of fishes. Copeia 1943(4): 260. Hudson, L. L. and J. D. Hardy, Jr. 1975. Eggs and larvae of the Atlantic seahorse, Hippocampus hudsonius. Univ. Maryland, CEES, Ref. 175-12 CBL. Huot, A. 1902. Recherches sur les poissons 10phobranches. Ann. Sci. Nat. (Zool.) 8(14): 197-288. Kronester-Frei, A. 1975. Light and electron microscopia1 studies of the brood epithelium of the male Nerophis lumbriciformis (Pennant 1716), Syngnathidae, with special reference to the struc- tural changes in the egg membrane (zona radiata). Forma functio 8: 419-462. Leiner, V. M. 1934. Der osmotisches Druck in den Bruttaschen der Syngnathiden. Zool. Anz. 108: 278-289. Linton, J. R. and B. L. Soloff. 1964. The physiology of the brood pouch of the male seahorse, Hippocampus erect us. Bull. Mar. Sci. Gulf Caribb. 14: 45-61. Lockwood, S. 1867. The seahorse (Hippocampus hudsonius) and its young. Amer. Nat. 1(5): 225- 234. Nikolsky, G. V. 1963. Ecology of fishes. Academic Press, Inc. (London) Ltd., London. 352 pp. Quast, W. D. and N. R. Howe. 1980. The osmotic role of the brood pouch in the pipefish Syngnathus scovelli. Compo Biochem. Physiol. 67a: 675-678. Spannhof, L. and H. Bremer. 1969. Histophysio10gische Untersuchungen zur Brutpflege bei Syng- nathiden. Limnol. (Berlin) 7(1): 163-166. Sudarsan, D. 1968. On the early development of the pipefish Syngnathoides biaculeatus (Bloch). J. Mar. BioI. Assoc. India 8(1): 222-224. Thevenin, R. 1936. The curious life habits of the sea horse. Nat. Hist. March 1936: 211-222.

DATEACCEPTED: January 14, 1991.

ADDRESS: University oj South Florida, Department oj Marine Science, 140 Seventh Avenue South. St. Petersburg, Florida 33701; PRESENTADDRESS:Allegheny College, Department oj Biology, Meadville, Pennsylvania 16335.