Biosynthetic events of hydrid regeneration II. Patterns and profiles of RNA synthesis during distal morphogenesis

The freshwater cnidarian Hydra has repeatedly served as an experimental system for studies of regeneration and morphogenesis. It regenerates rapidly and consistently after distal amputation and it reforms only two types of structures, a mouth and several tentacles.

Reports from this and other laboratories have demonstrated a requirement for RNA synthesis during normal distal regeneration in hydra (Kass-Simon, 1969; Clarkson, 1969b; Lesh-Laurie & Hang, 1972; Lesh-Laurie, 1974). In each of these studies distal regeneration was examined in the presence of an inhibitor of RNA synthesis, and results ranging from the total inhibition or regeneration to an arrest after the emergence of the first tentacle pair have been obtained. Although each of these investigators concluded that RNA synthesis was necessary for regeneration, only Clarkson (1969a) attempted to measure the synthetic events of normal distal regeneration. His study, which examined only the first 2·5 h after subhypostomal excision, revealed a ten-fold increase in RNA synthesis during this time period. No effort was made to examine synthetic activity beyond 2·5 h of regeneration.

In a previous report (Lesh-Laurie, Brooks & Kaplan, 1976) we described the pattern of [³H]thymidine incorporation into DNA during the first 48 h of distal regeneration in Hydra oligactis. Peaks of incorporation occurred during the intervals 0-3, 24-27, and 33-36 h after excision. These periods corresponded temporally to wound healing, the emergence of the first tentacle pair, and the emergence of the third tentacle, respectively.

In this paper we report the results of an investigation of [³H]uridine in-corporation into RNA during the first 48 h of hydrid distal regeneration. Analyses of incorporation into total TCA-insoluble RNA revealed increases in incorporation rates corresponding to the times of wound healing and tentacle formation. In addition, fractionation of labeled RNA on polyacrylamide gels demonstrated the production of a low molecular weight RNA species preceding tentacle initiation.

H. oligactis were mass cultured at 18° ± 2 °C in 14 cm finger bowls according to the methods of Loomis & Lenhoff (1956) with distilled water substituted for tap water. Cultures were fed to repletion with Artemia salina nauplii every other day and cleaned 4-6 h later.

Non-budding animals, fed one day before subhypostomal excision, were used for all experiments. For at least 24 h prior to use hydra were cultured in sterile culture solution containing 0·05% (v/v) Kantrex (0·33 g/ml kanamycin sulfate, Bristol Laboratories, Syracuse, New York). This treatment has been shown autoradiographically to reduce bacterial precursor incorporation in the mucous coat of the animal to background levels (Lesh-Laurie et al. 1976).

Removal of the hypostome and tentacle ring was accomplished with one clean cut immediately proximal to the tentacle bases using iridectomy scissors or a scalpel. Following excision, animals were returned to sterile, kanamycin-treated culture solution (SKCS) until used.

Analyses of uridine incorporation were carried out during the first 48 h of regeneration. By this time at least two and often three tentacles were present. Groups of 15 hydra were incubated for 3 h in 0·5 ml of SKCS containing 15 μCi/ml [³H]uridine (5-[³H]uridine, 29 Ci/mM, Amersham/Searle Corporation). A group of hydra was labeled during each 3-h interval from 3 h prior to excision until 48 h after excision. Following the 3-h pulse, they were washed four times over a period of five min in non-radioactive uridine (3 mg/ml), homogenized, and the resulting material precipitated by addition of ice-cold 100% TCA to a final concentration of 5%. Precipitation was allowed to proceed for at least 1 h at 2 °C.

The RNA was extracted by the Schmidt-Thannhauser technique (1945). After washing twice with ice-cold 5% TCA, once with 95% ethanol, and once with 95% ethanol:ether (3:1), the RNA was hydrolyzed in 2ml of 0·5 N-KOH by heating at 37 °C overnight. To each sample 0·2 ml of 70% perchloric acid was added and the macromolecules were allowed to precipitate on ice for 10 min. This procedure removed the DNA, which was resistant to alkaline hydrolysis, from the hydrolyzed RNA (Munro & Fleck, 1966).

After centrifugation the supernatant was titrated with KOH to a pH of approximately 6·8, diluted to 5 ml with distilled water, and chilled on ice for 20 min. Following another centrifugation, the supernatant was divided with 1 ml counted in 10 ml of Bray’s solution (Bray, 1960) and 1·5 ml analyzed for RNA content by the Dische modification (Dische, 1954) of the orcinol reaction. Standards were prepared from yeast RNA (Type XI, Sigma Chemical Company) and absorbance at 665 nm was read in a Zeiss PMQ11 Spectrophotometer.

A specific activity (cpm/μg RNA) was then calculated for each labeling interval. In addition an overall average specific activity was calculated for each experiment. The specific activities of the individual labeling intervals for a given experiment were expressed as percentages of the average specific activity for that experiment. This procedure eliminated differences among experiments resulting from changes in the condition of the organisms and from errors in the preparation of labeled precursor solutions.

Each labeling interval is designated as a single time, that of homogenization at the end of the labeling interval. For example, animals labeled from 24 to 27 h after excision are referred to as 27-h regenerates.

To study the changes in specific RNA fractions during regeneration, groups of 50 animals were incubated for 4 h in 4 ml of SKCS containing 20 μCi/ml [³H]uridine. The longer labeling period and higher radioactive concentration were required to insure a significant level of radio-activity in the fractionated RNA. A group of animals was labeled during each 4-h interval from 4 h prior to excision until 48 h after excision. Immediately after labeling, RNA was extracted with phenol-chloroform and electrophoresed on 2·4% polyacrylamide gels as previously described (Voland, 1975).

Gels were analyzed in a Gilford Recording Spectrophotometer equipped with a linear transporter. Scanning was carried out at 260 nm at a scanning rate of 2 cm/min. Once A₂₆₀ patterns were obtained, gels were consecutively sliced into 1 mm sections, placed in glass vials and solubilized with 0·1 ml concentrated NH₄OH. The vials were capped and allowed to sit overnight at room temperature. The vials were then filled with Bray’s solution and counted in a liquid scintillation counter.

Results from the polyacrylamide gels are expressed as cpm/μ g 18S RNA.

To compute the amount of 18S RNA present, A₂₆₀ profiles were photocopied, the 18S peak cut out and weighed. This weight was normalized to a weight at a full scale deflexion of 1-5 on the Gilford Recording Spectrophotometer and the micrograms of 18S RNA determined from a standard curve (courtesy of Dr S. Macintyre) of the weight of the 18S peak at a full scale deflexion of 1-5 versus micrograms of 18S RNA.

Representative radioactive patterns are presented for selected regeneration times. Each represents an average profile for the time selected (i.e. each experiment was repeated a minimum of three times, and the pattern selected is neither the high nor the low extreme).

Preliminary experiments revealed that exposure of 15 hydra to 15 μCi/ml [³H]uridine for 3h resulted in sufficient labeling levels (> 1000cpm/ml) and RNA concentrations ( > 8 μg/ml) after TCA extraction for analysis by scintillation counting and the orcinol procedure. Using these conditions, radioactive specific activities were determined at 3-h intervals during regeneration and are shown graphically in Fig. 1.

A five-fold increase in incorporation occurred in the first 3-h interval after excision, with incorporation levels remaining high throughout the 48-h period studied. Specific activity levels began to increase again at 24 h, climbing to a peak at 30 h of regeneration. The 30-h specific activity level represented a 20-fold increase over the zero hour level.

After this peak of incorporation, the specific activity level fell to that of the 0-3-h interval, then climbed to approximately twice this value and remained nearly constant for the remainder of the period studied.

The observed specific activities from 3 to 48 h of regeneration were significantly higher (P < 0·01) than the zero-hour level when analyzed by a two-way analysis of variance with a test of contrasts (Snedecor & Cochran, 1956). By the same test, the levels at 27 and 30 h were significantly higher (P < 0·005) than those from 3 to 48 h. Although fluctuations in specific activity were observed in individual experiments after 30 h, no significant changes occurred in the average values.

Because consistent patterns of incorporation were found in TCA-insoluble material, fractionation and analysis of the newly synthesized RNA was carried out. Due to the higher radioactive concentrations and longer incubations required for these procedures, the time periods studied do not correspond exactly to those employed for the previous experiments.

Both absorbance and radioactive profiles were obtained for phenol-chloroform extracted RNA. The A₂₆₀ profiles were similar for all intervals studied regardless of the extent of regeneration. These profiles were also similar to those obtained for intact hydra (Voland, 1975), a result consistent with the hypothesis that the majority of hydrid RNA is not altered during the regeneration process.

The levels of uridine incorporation, however, showed considerable variation with time during regeneration. The qualitative changes were similar to those found for TCA-insoluble RNA, with peaks of incorporation occurring at 0-4 and 28-32 h of regeneration. An increase over the zero-hour level was also noted at 36-40 h after excision.

The basic radioactive profile was relatively constant for most times during regeneration. Representative profiles are presented in Fig. 2 for time intervals preceding and following the 28-32-h peak of incorporation. Uridine was in-corporated into ribosomal RNA (28S, 18S, 5S), transfer RNA (4S), and DNA. Incorporation into bacterial ribosomal RNA (23S and 16S) was also noted. The observed profiles showed a lower level of incorporation into the 18S and 16S species than was anticipated. The reason for this result is probably an instability of these species related to the extraction procedure.

While the profiles presented in Fig. 2 are representative of most time intervals, they do not match the profiles obtained during the intervals from 28-32 and 36-40 h of regeneration. These two profiles are shown in Fig. 3. In addition to the normal incorporation into DNA, ribosomal RNA, and transfer RNA, there is a new peak of incorporation designated 8S. This low molecular weight RNA species was seen consistently in six repeats of the 28-32-h interval and in three repeats of the 36-40-h interval. Moreover, it was never observed at any other time interval.

In this investigation, incorporation of uridine into RNA was measured, not RNA synthesis per se. In order to relate uridine incorporation to RNA synthesis, two assumptions must be made. First, the cellular precursor pools must equilibrate rapidly with each other and with the culture solution. That this is the case in hydra has been shown by Clarkson (1969a), who found that [³H]uridine incorporation increased linearly with incubation periods of from 5 min to 8 h.

The second assumption, that the specific activity of the immediate precursor pool (UTP) is constant over the 48-h period studied, is much more difficult to verify. Because hydra consists of an asynchronous population of mixed cell types, measurement of the TCA soluble pool, or even the UTP pool, would yield results which could not be interpreted with certainty (see Hauschka, 1973). Indirect evidence that the observed fluctuations in incorporation rates are not due to changes in pool specific activities can, however, be obtained by studying incorporation of different precursors into RNA. When the incorporation of [³H]adenosine or inorganic [³²P] was examined, patterns of incorporation into phenol-chloroform extracted RNA were similar to those obtained with [³H]uridine (Voland, unpublished results). Considering these results and the similarity of the uridine incorporation pattern reported here to the pattern of thymidine incorporation into DNA reported earlier (Lesh-Laurie et al. 1976), we believe that, under the conditions used in this investigation, uridine incorporation rates accurately reflect RNA synthetic rates.

The results of this investigation, then, show that amputation of distal structures in hydra results in a rapid increase in RNA synthesis. This synthesis remains high throughout the first 48 h of regeneration, with a large increase in synthesis occurring between 24 and 30 h. This peak of activity closely follows the 24-27-h peak of DNA synthesis in regenerating hydra, and im-mediately precedes the emergence of the first tentacle pair (Lesh-Laurie et al. 1976). Because actinomycin D is capable of suppressing regeneration completely, it may be presumed that RNA synthesis is necessary for regeneration and therefore involved in tentacle morphogenesis.

Examination of radioactive profiles of phenol-chloroform-extracted RNA confirmed the pattern of synthesis found using TCA-insoluble RNA. Although bacterial contamination was present, increases in hydrid RNA synthesis were seen to occur during the 0-4 and 28-32 h periods following subhypostomal excision.

Moreover, this fractionation of hydrid RNA revealed the synthesis of a low molecular weight RNA species during the 28-32 h period not present at other times. The correspondence of this novel RNA synthesis, the increase in overall RNA synthesis, and the emergence of the first tentacle pair strongly indicates an interrelationship among these events.

The 8S RNA species was present again during the 36-40-h interval, a time at which the level of RNA synthesis was relatively high, with individual experiments showing peaks of incorporation between 36 and 42 h of regeneration (see Fig. 1). This time period precedes the formation of the third tentacle, an event that shows less synchrony than the emergence of the first pair. Because of this lack of temporal consistency, the relationship of the biochemical and morphogenetic events after 30 h of regeneration is unclear.

In conclusion, this investigation has demonstrated a clear pattern of RNA synthesis following hypostomal amputation. The increases in total RNA synthesis, as well as the occurrence of the low molecular weight species revealed by RNA fractionation, preceded events of major morphogenetic significance, i. e. wound healing and tentacle initiation. The pattern of DNA synthesis in regenerating hydra, which is remarkably similar to this RNA synthetic pattern, is closely linked to tentacle morphogenesis in the animal (Lesh-Laurie et al. 1976). We believe that the RNA synthesis observed in these experiments is similarly involved in the morphogenetic process in hydra. Further investigations of this involvement employing actinomycin D to inhibit RNA synthesis have been undertaken, as have analyses of protein synthesis and mesogleal collagen secretion during regeneration. Results of these studies will be presented in subsequent papers.

The authors express their gratitude to Dr Marlene Samuelson for many helpful discussions. This work was supported by a Brown-Hazen grant from Research Corporation, an institutional grant from the American Cancer Society, and the Graduate Alumni Fund of CWRU.