Growth and transgenerational acclimatization of juvenile Pocillopora damicornis

Global carbon emissions and associated increase in ocean temperatures are understood to be the main driving force in the degradation of coral reefs. Elevated temperatures impact various life stages of scleractinian corals, from the free-floating planulae of brooding corals to older, sexually viable individuals. With global warming, questions have arisen over whether organismal adaptation will be enough to keep up with the pace of environmental change. Researchers have pursued investigations of whether or not rapid acclimatization, through transgenerational plasticity, can help protect populations until genetic adaptation occurs. Acclimatization in corals has been widely studied in all life stages of corals, with the important exception of recently settled juveniles. In this study, I built upon past research by exposing adult Pocillopora damicornis colonies to elevated (28.5°C) or ambient (25.5°C) temperatures and examining the settlement ability and growth of their planulae ex situ. Juveniles from preconditioned parents fared better in higher temperatures compared to their naïve counterparts. Lunar timing of planula release between treatments peaked at different times in the lunar cycle. Peak planula release occurred on lunar day 23 for prestressed corals and on lunar day 7 for corals from ambient temperature seawaters. While future projects should follow up on these preliminary trials with in situ experiments to assess this phenomenon in the field, this study represents an important step in understanding how corals may be able to acclimatize and eventually adapt to climate change.

9 nutrient puck comprised of Spirulina, mysids, and Artemia, as well as ~175 ml of Continuum Aquatics Coral 0 Exponential Amino Acids (Continuum Aquatics, Fort Payne, AL, USA) every Monday, Wednesday, and Friday.

2 Planula Collection
3 Corals were transferred to an outdoor planulation table on June 17 th , 2019. Each coral was placed into an individual 4 plastic mixing bowl, so that flowing water allowed the buoyant planulae to flow over the smooth handle and into 5 collection beakers with plankton mesh sides. Beakers were checked every morning for planulae, and planula 6 release was tracked for 3.5 months. Planulae were transferred into and stored in 1000ml Erlenmeyer flasks and 7 kept from settling with magnetic stir plate and stir bar. Seawater in the flasks was changed every week. Water 8 temperature on the planula table was measured daily and fluctuated between ~26 and 27ºC. Mean lunar day of 9 release of planulae was determined using circular statistics. Statistical analysis was performed in R(18). 0 1 Settlement 2 Four 5-liter tanks were used for settlement of planulae. Seawater from a supply well was filtered using a bag and 3 sand filter in series. Aragonite tiles, which had been cured in flowing sea water for the three months, were arranged 4 in a 3 by 4 grid on the bottom of the tanks. The bottoms of the tanks between the tiles were filled with sand to 5 encourage settlement on tiles. Holes were drilled into the side of the tanks and covered with plankton mesh to 6 allow water to flow through the tanks. Water flow was kept at ~1.5 l min -1 . Planulae were divided into one of four 7 groups: planulae from preconditioned parents to be settled in heated seawater (n=7), planulae from preconditioned 8 parents to be settled in ambient temperature seawater (n=8), planulae from naïve parents in heated seawater (n=46), 9 and planulae from naïve parents in ambient temperature seawater (n=46). Two header systems were built to allow 0 even heating of seawater using two 1000-watt titanium heater. On settlement day, seawater in the tanks was 1 lowered to one-third capacity and flow was turned off to ensure planulae did not settle on the upper glass of the 2 tanks and were not washed through the tanks. Temperatures were taken every fifteen minutes with an electronic 3 thermometer. Water level was kept at one-third capacity from 11am to 4pm, at which point flow was restored.

5 Juvenile Measurements
6 After one month of growth, juvenile corals were photographed weekly using a Moticam X 3 Wi-Fi camera (Motic, 7 Kowloon Bay, Kowloon, Hong Kong) and compound microscope. Growth was subsequently measured for 6 weeks 8 using built in Moticam software. After six weeks, photophysiology of Symbiodinacea was measured as the 9 maximum relative rate of electron transport (ETR) for a standardized 40 mol m -2 s -1 . Photosynthetic 0 efficiency was measured during rapid light curves (RLC) using a pulse-amplitude-modulated (PAM) fluorometer 1 (Diving-PAM, Walz GmbH, Germany; methods drawn from (19) and (20)) and WinControl-3.30 software with 2 saturating pulses at 0:10 intervals and the following increasing actinic light values: 10, 20, 30, 40, 50, 60, 70, 80, 3 90, 100 mol photons m -2 s -1 . The intensity of the red light-emitting diode (655 nm peak, <0.15 m -2 s -1 , at 1.6 kHz) 4 was set to 8 and was too low to induce fluorescence when used as a probe light. Maximum fluorescence (F m ) was 5 measured during a saturating light pulse from a halogen lamp (0.8 seconds, at >5,000 mol m -2 s -1 ; KL1500, Schott, 6 Mainz, Germany) exposed to the sample via an optical fiber. Aragonite tiles were scrubbed free of algal growth 7 around the new recruits to prevent measurement errors. The diving-PAM probe was fitted with a black plastic tube 8 to minimize light leakage. Corals were re-measured using the diving-PAM 48 and 96 hours after initial 9 measurements. Missing ETR values, likely a result of low baseline values (F 0 or F t ), were manually calculated 0 using the equation: . Relative ETR (rETR), calculated as: 1 transformation were carried out using Python (23).

7
Peak planula production for prestressed adults occurred on lunar day 23 with a circular standard deviation 8 of 8.4, while peak production for naïve adults occurred on lunar day 7 with a circular standard deviation of 3.6 ( Fig   9  1). A two-way ANOVA test determined that pre-conditioning of parental corals was found to be independently 0 significant examining its effect on juvenile growth (p-value = 0.04; Table 1), while temperature of current 1 environment (ambient or elevated) was not a significant factor (p-value = 0.20; Table 1). However, it was found 2 that there was no significant interaction between environmental temperature and preconditioning on juvenile 3 growth (p-value = 0.80; Table 1). Change in area was, on average, higher in juveniles from preconditioned adults 4 for both the elevated and ambient temperature groups (Fig 2). Meanwhile, growth at ambient temperature was 5 higher than growth in elevated temperatures for both the preconditioned and naïve juveniles (Fig 2).  The effects of the interaction of adult preconditioning history and experimental heat stress 3 on change in area, electron transport rate (ETR), and relative electron transport rate (rETR). 4 Juvenile response analyzed with two-way ANOVA. Bold indicates significance. 5 6 7 Figure 2. Response of juvenile P. damicornis to heat stress following pre-conditioning. Larvae were pooled by 8 preconditioning treatments and held at either elevated or ambient temperature prior to settlement. Juveniles were 9 tracked for 8 weeks before measuring: a) growth rate as a change in area; b) electron transport rate using PAM 0 fluorometry; c) relative electron transport rate calculated using PAM fluorometry. Significant effects of 1 preconditioning were found for ETR, growth rate and rETR. Growth rate, ETR, and rETR were lower after 2 elevated temperature exposure for both groups, but higher on average in the preconditioned group than the naïve 3 group. For ETR values, both preconditioning of parental corals (p-value = 0.007; Table 1) and temperature of 7 current environment (p-value = 0.0006; Table 1) were found to be independently significant of each other. There 8 was no significant interaction between environmental temperature and preconditioning on ETR (p-value = 0.189; 9 Table 1). As with growth rate, ETR was higher in juveniles from preconditioned adults for both the elevated and 0 ambient temperature groups, and also higher at ambient temperature for both naïve and preconditioned juveniles 1 (Fig 2).
2 For rETR, both preconditioning of parental corals (p-value = 0.03; Table 1) and temperature of current 3 environment (p-value = 0.017; Table 1) were also found to be independently significant of each other, similar to 4 ETR. There was no significant interaction between environmental temperature and preconditioning on rETR (p-5 value = 0.843; Table 1). rETR was higher in juveniles from preconditioned adults for both high and ambient 6 temperature groups, and also higher at ambient temperatures for both naïve and preconditioned juveniles (Fig 2).

7
This followed similar patterns to both ETR and growth rates. 6 was higher in recruits from preconditioned parents, as well as in both preconditioned and naïve juveniles in 7 ambient temperature seawaters (Fig 2). The same trends were seen for rETR, with higher levels in preconditioned 8 individuals as well as in both preconditioned and naïve juveniles grown in ambient seawaters. rETR also had less 9 variability than growth numbers, indicating that this assay may have been a better indicator of relative health of 0 juveniles than a simple measurement of juvenile area. These results provide evidence that bolsters the idea that 1 cross-generational acclimatization is a process that has the potential to mitigate the effects of a changing climate on 2 multiple life stages of corals.
3 rETR of corals settled in elevated temperature treatments was significantly higher for juveniles from 4 preconditioned parents compared to juveniles from naïve parents. Two potential mechanisms could explain a trans-5 generational shift that would cause differences between the naïve and preconditioned juveniles. The first is the 6 development rate of brooding planulae and the timing of planula release. Temperature has been shown to reduce 7 the number and timing of reproductive events in Pocillopora damicornis (24). The peak days of planulation for 8 prestressed adults occurred on lunar day 23. This average was likely driven by the single large planulation event on 9 the 23 rd lunar day in August. Naïve adults, meanwhile, had a peak release on lunar day 7 (Fig 1). They also had 0 lower, more consistent peak release numbers, rarely releasing more than 25 planulae at a time. Reproductive 1 processes in corals have been shown to be plastic (25), and it is possible that this plasticity contributed to 2 acclimatization in juveniles. By the time planulae were released, peaking later in the month for prestressed corals, 3 any traits that would have failed to survive higher temperature conditions would have been weeded out, as planulae 4 possessing these traits failed to survive the brooding process.

5
A second possibility is that, rather than acclimatization weeding out unfit planulae during brooding, adult 6 corals passed resistance directly to offspring. Temperature stress has been shown to reduce important biochemical 7 components in adult corals such as lipids, proteins, and mycosporine-like amino acids, the reduction of which was 8 amplified in offspring from bleached parents (26). A reduction in lipid content in P. damicornis also coincided 9 with a decrease in planulae and gametes during planulation events (27). It is possible that the vertical transmission 0 of Symbiodinacea from parent colonies conditioned against temperature stress was picked up by PAM fluorometry 1 and manifested as the change in rETR values observed here. If thermal stress limited which corals were able to 2 planulate due to a lack of sufficient resources to put toward reproduction, this might provide another explanation 3 for why the planulae that came from pre-stressed corals had higher photochemical responses.

4
There are other pathways of acclimatization that might explain the difference between the responses of 5 corals from naïve and preconditioned treatments but were outside the scope of this experiment to test. One such 6 direction for future research is the role of mycosporine-like amino acids (MAA). These molecules are small, water 7 soluble, UV radiation absorbers found in a wide variety of marine organisms (28). Historically, research on MAAs 8 has focused on their usefulness as UV protectors (29-31). However, evidence has emerged that they could 9 additionally act as secondary metabolites, serving a multipurpose role (28). A 2009 study by Yakovleva et al.
0 found that some MAAs provide rapid protection during thermal stress before antioxidant enzymes are activated 1 (32). Typically, the costs associated with symbiosis are not examined, as it is thought to be a mutualistic process 2 that benefits both parties. However, this study found that Acropora intermedia larvae containing zooxanthellae 3 showed decreases in specific MAAs thought to reduce oxidative stress, decreases in survival rates, and increases in 4 oxidative cellular damage when subjected to elevated temperatures. Because this study was conducted on such an 5 early life stage, there is a compelling case that MAAs might explain at least some of the rapid acclimatization to 6 thermal stress in Pocillopora damicornis and other coral species.

7
It is possible that tank effects could have influenced the outcomes of this study and that the experimental 8 design involved pseudoreplication, since all adult corals were held in just two separate tanks and juveniles were 9 held in just four separate tanks. However, due to limitations in available resources necessary to maintain corals in 0 individual tanks, the relatively small biomass of corals used coupled with large tank volume and high water flow 1 (4-6 l min -1 for adult tanks, 1.5 l min -1 for juvenile tanks) should have minimized any effects of pseudoreplication.
2 Growth of algae in the juvenile tanks could have also led to an increase in dissolved organic carbon, facilitating 3 microbial growth and reducing oxygen availability for juveniles. However, daily cleaning coupled with high flow 4 and regular turnover of water in the tanks should have been enough to keep microbial buildup below harmful 5 levels.

6
There were limitations to this investigation that could be improved upon in future studies. Among 7 treatments, numbers of planulae settled were uneven, with two groups of n=46 planulae settled from naïve parents, 2 Photosymbiotic algae are integral to reef building corals growing in oligotrophic waters, as they provide 3 photosynthate to help meet daily nutritional demands (21). The development of PAM fluorometry techniques has 4 allowed for a non-invasive measurement of the relative health of corals by quantifying metrics such as the potential 5 quantum yield of photosystem II and the estimated rates of photosynthetic electron transport, but the majority of 6 testing to date has been performed on adult corals. PAM fluorometers measure the relative quantum yield of Chl a 7 by targeting samples with excitation pulses, and probing the probability of light energy being re-emitted (19). The 8 ETR max measured here, which peaked somewhere between 40 to 60 mol m -2 s -1 for all samples, is the maximum 9 amplitude at which a re-emission response was recorded. For measurements taken on December 16 th , sufficient 0 data were obtained to produce ETR and Y(II) values, which were calculated automatically by the diving-PAM.
1 However, for subsequent measurements taken 48 and 96 hours later, the diving-PAM began to regularly fail to 2 automatically calculate ETR and Y(II). Using the equations Φ PSII = (F m' -F t )/F m' and rETR = Φ PSII * PAR made it 3 possible to fill in these gaps manually. Values calculated for December 18 th and 20 th closely matched rETR and 4 ETR outputs from December 16 th calculated automatically by the PAM fluorometer, but this most likely 5 oversimplifies relative photobiological health. Fluorescence was significantly lower in these subsequent tests, and 6 because Φ PSII is calculated as a ratio of these two numbers ((F m' -F t )/F m' ), this would not have shown up in the Y(II) 7 values. Instead, calculated Y(II) values would have masked any sharp reductions in F m' and F t , which would have 8 indicated changes in photobiological processes. Although there is no known evidence to support the potential for 9 damage to juveniles from saturating light pulses (>10,000 mol photons m -2 s -1 (22)), investigation of this topic 0 could produce interesting and novel results. More work is clearly needed to determine if the efficacy of using PAM 1 fluorometry as a technique extends to coral individuals at the juvenile stage.