Temperature modulates the osmosensitivity of tilapia prolactin cells

In euryhaline fish, prolactin (Prl) plays an essential role in freshwater (FW) acclimation. In the euryhaline and eurythermal Mozambique tilapia, Oreochromis mossambicus, Prl cells are model osmoreceptors, recently described to be thermosensitive. To investigate the effects of temperature on osmoreception, we incubated Prl cells of tilapia acclimated to either FW or seawater (SW) in different temperature (20, 26 and 32°C) and osmolality (280, 330 and 420 mOsm/kg) combinations for 6 h. Release of both Prl isoforms, Prl188 and Prl177, increased in hyposmotic media and were further augmented with a rise in temperature. Hyposmotically-induced release of Prl188 was inhibited at 20°C. In SW fish, mRNA expression of prl188 and prl177 showed direct and inverse relationships with temperature, respectively. In SW-acclimated tilapia Prl cells incubated in hyperosmotic media, Prl receptors, prlr1 and prlr2, and the stretch-activated Ca2+ channel, trpv4, were inhibited at 32°C, suggesting the presence of a cellular mechanism to compensate for elevated Prl release. Transcription factors, pou1f1, pou2f1b, creb3l1, cebpb, stat3, stat1a and nfat1c, known to regulate prl188 and prl177, were also downregulated at 32°C. Our findings provide evidence that osmoreception is modulated by temperature, and that both thermal and osmotic responses vary with acclimation salinity.


Introduction
In vertebrates, hydromineral balance is maintained through osmoregulation. Osmoregulatory processes, in turn, are largely mediated through osmosensitive cells and the neuroendocrine system 1,2 . Euryhaline shes, which are characterized by their capacity to thrive in a wide range of environmental salinities, have been employed to elucidate the mechanisms underlying the transduction of osmotic stimuli [3][4][5][6][7] . More recently, in light of impending climate-change driven changes in environmental temperature and salinity, a need for cellular and organismal models where the integration of distinct thermal and osmotic stimuli can be studied has emerged 8, 9 .
Prolactin (Prl) is a pleiotropic hormone that exerts hundreds of physiological functions in vertebrates including lactation, osmoregulation, growth, reproduction and immune function [10][11][12] . In euryhaline sh, the main function of Prl is to stimulate ion absorption and retention in osmoregulatory tissues to maintain osmotic balance in fresh water (FW) 13,14 . Mozambique tilapia (Oreochromis mossambicus) has been widely used to study the effects of Prl on osmoregulation due to its euryhalinity and the morphology of Prl secreting cells, which comprise a nearly homogeneous portion of the rostral pars distalis (RPD) of the pituitary 15,16 . Consistent with its role in FW adaptation, plasma Prl levels are high in FW and its release increases in pituitaries and dispersed Prl cells incubated in hyposmotic media [17][18][19] . Tilapia Prl cells secrete two isoforms of Prl, Prl 188 and Prl 177 , which are encoded by separate genes 20,21 . Both Prl isoforms act through Prl receptors, Prlr1 and Prlr2, which have been shown to exert distinct downstream effects through JAK/STAT activation and differentially respond to changes in extracellular osmolality 17,22 . Prl 188 responds more robustly to hyposmotic stimuli than Prl 177 17,23 . Due to their importance in FW adaptation, both prl 188 and prl 177 are found to be 10-30 times higher in Prl cells in FWacclimated tilapia compared with their seawater (SW) counterparts 24,25 . Consequently, hyposmotically-induced prl expression is more responsive in tilapia acclimated to SW than those in FW 17,26 . Recently, we reported that tilapia Prl cells are also thermosensitive 27 . Insamuch as Mozambique tilapia is both euryhaline and eurythermal, surviving in salinities ranging from FW to over double-strength SW and temperatures between 10-38°C 28 , it is likely that both thermal and osmotic stimuli interact during adaptive hormonal responses.
The cellular mechanisms underlying osmoreception in tilapia Prl cells have been recently reviewed 29 . Brie y, when extracellular osmolality drops, water enters the Prl cell through aquaporin 3 channels (Aqp3) leading to an increase in cell volume and activation of the stretch-activated ion channel, transient receptor potential vanilloid 4 (Trpv4), which enables extracellular Ca 2+ into the cell [30][31][32][33][34] . An increase in intracellular [Ca 2+ ] alone, or through the activation of the cyclic AMP (cAMP) secondary messenger system, increases Prl release 30,35 . Prl also exerts autocrine responses on Prl cells which are in turn modulated by extracellular osmolality 36 . Thermally-induced Prl release also appears to operate through a cell-volume dependent mechanism 27 . The mechanistic commonalities between hyposmotically-and thermally-induced release of Prls, raise the question of whether similar mechanisms are present in the regulation of prl genes.
Recently, several putative transcription factors (TF) predicted to bind promoter regions of prl 177 and prl 188 were identi ed 29 and their activities measured in the tilapia Prl cell model 37 . Among them, several POU family TFs, such as pituitary transcription factor 1 (Pit1, also known as Pou1f1), a key regulator of pituitary cell differentiation 38 , were directly responsive to changes in extracellular osmolality 37 . Pit1 shares a common binding site on the tilapia prl 188 promoter region with Octamer 1 (Oct1, also known as Pou2f1), another POU family TF 29 activated by stressors [39][40][41] . The roles of cAMP and Ca 2+ second messenger systems in cellular signaling have been studied, including downstream activation of CAAT/enhancer binding protein (CEBP) and cAMP response element binding protein (CREB) 42,43 , two TFs also predicted to bind prl 188 and prl 177 promoter regions 29 and recently shown to be hyposmotically induced in tilapia Prl cells 37 . On the other hand, the nuclear factor of activated T cells (NFAT) is hyperosmosensitive, leading to the production of secondary metabolites 44,45 ; binding sites for NFAT were found in the promoter region of tilapia prl 177  The effects of temperature and osmolality on Prl 188 and Prl 177 released from Prl cell incubations of tilapia acclimated to FW and SW by 1 h are shown in Fig. 1. In Prl cells of SW-acclimated tilapia, effects of both osmolality and temperature were seen in Prl 188 release by 1 h (Fig. 1A). By 1 h, hyposmotically-induced Prl 188 release was only observed at 32°C. In FW-acclimated tilapia Prl cells, only an osmotic effect was seen by 1 h (Fig. 1B), with hyposmotically-induced Prl 188 release observed at all incubation temperatures. In SW sh, only osmolality had an effect on Prl 177 release by 1 h (Fig. 1C); Prl 177 release was inversely related to extracellular osmolaity at 20 and 26°C. By contrast, Prl 177 release from FW-tilapia Prl cells showed both thermal and osmotic effects (Fig. 1D). Hyposmotically-induced Prl release was seen at both 20°C and 26°C, while a rise in temperature inhibited Prl 177 release in isosmotic and hyperosmotic conditions.
The patterns of Prl release from Prl cell incubations by 6 h were more evident and consistent than those observed by 1h (Fig. 2). Prl 188 released from both SW-and FW-tilapia Prl cells showed thermal, osmotic and interaction effects. In SW sh, a hyposmotic effect was only observed at 32°C, while Prl release increased at 32°C compared with other incubation temperatures in hyposmotic and isosmotic media ( Fig. 2A). In FW sh, the thermally-induced Prl 188 release was observed in cells incubated in all osmotic conditions (Fig. 2B); a ve-fold rise in Prl 188 release was seen in cells incubated in hyposmotic media at 32°C compared with those at 20°C.
Hyposmotically-induced Prl 188 realese, however, was only observed at 26 and 32°C (Fig. 2B). By contrast, an inverse relationship between Prl 177 release and medium osmolality was seen in both SW and FW sh at all temperatures ( Fig. 2C and 2D). In both SW-and FW-acclimated sh, Prl release was increased in hyposmotic media and inhibited by a drop in temperature.

Effects of temperature and osmolality on prl mRNA expression
The mRNA expression of prl 188 and prl 177 in tilapia Prl cells incubated for 6 h are shown in Fig. 3. In SWacclimated tilapia, prl 188 expression was inversely related to media osmolality at all temperatures, and directly related to temperature in isosmotic and hypoosmotic conditions (Fig. 3A). By contrast, in FW-acclimated sh, prl 188 did not vary among treatments (Fig. 3B). In both SW-and FW-acclimated sh, temperature was the only factor affecting prl 177 mRNA expression ( Fig. 3C and D). In SW-sh, prl 177 in isosmotic and hyperosmotic media was higher at 20°C than at 32°C, while in FW-sh, prl 177 in hyposmotic and hyperosmotic conditions were higher at 20°C compared with 26°C.

Effects of temperature and osmolality on prlr mRNA expression
Main effects of temperature and osmolality were observed in the transcription of prlr1 and prlr2 from Prl cell incubations (Fig. 4). In SW-acclimated tilapia, prlr1 was downregulated at 32°C compared with other temperatures, while the osmotic effect changed according to temperature (Fig. 4A). In FW-acclimated sh, prlr1 was upregulated as media osmolality increased at all temperatures, while inversely related with temperature in hypo-and hyperosmotic incubations (Fig. 4B). A notable increase of prlr2 (up to ~ 3-fold) was observed in Prl cells of both SW-and FW-acclimated sh incubated in hyperosmotic media at all temperatures ( Fig. 4C and D).
At 32°C, expression of prlr2 in Prl cells of SW sh was downregulated at all media osmolalities compared with the other incubation temperatures (Fig. 4C).

Effects of temperature and osmolality on trpv4 mRNA expression
There were main effects of osmolality and temperature in trpv4 expression in Prl cells of tilapia acclimated to both FW and SW (Fig. 5). In Prl cells of SW-acclimated tilapia, trpv4 expression was higher in hyperosmotic media, with the exception of incubations carried out at 32°C, where expression was highest in isosmotic conditions (Fig. 5A). In FW-sh, trpv4 expression was increased by rises in extracellular osmolality (Fig. 5B). The expression of trpv4 was inhibited in Prl cells incubated at 20°C compared with that at 26°C in sh acclimated to FW, but not those in SW.  6F) were highly expressed at 26°C; expression in both high and low temperatures was lower than isothermal controls. The expression of stat3 was inversely related with osmolality at all temperatures; stat1a expression was elevated by hyposmotic media only at 32°C. Similarly, nfatc1 expression was suppressed in hyperosmotic media ( Fig. 6G). High temperature inhibited nfatc1 in isosmotic and hyperosmotic media, while both high and low temperatures suppressed hyposmotically-induced nfatc1 expression.

Effects
In Prl cells of FW-acclimated tilapia, pou1f1 expression was inversely related with extracellular osmolality at 20°C and 26°C (Fig. 7A). A thermal effect on pou1f1 was only seen in isosmotic conditions, where it was downregulated at 32°C. In isothermal conditions, pou2f1b expression was not affected by extracellular osmolality (Fig. 7B). At 20°C, pou2f1b was inversely related to osmolality, however, at 32°C, it increased with osmolality. The expression of pou2f1b in hyposmotic media was inhibited by a rise temperature, while its expression in hyperosmotic media was elevated at 32°C. As temperature rose, creb3l1 was downregulated in hyposmotic media (Fig. 7C). There was no temperature effect on cebpb expression; hyperosmotically-induced transcription was observed at all temperatures (Fig. 7D). Hyposmotically-induced stat3 expression was observed at 20°C and 26°C, while transcripts in hyposmotic and isosmotic conditions were inhibited at 32°C (Fig. 7E). Medium osmolality did not affect stat1a expression (Fig. 7F); transcription was inhibited by low temperature at all media osmolalities. Similarly, nfatc1 was not affected by osmolality, but was inhibited by lower temperatures (Fig. 7G).

Discussion
Stemming from the recent nding that tilapia Prl cells are thermosensitive 27 in addition to their well established role in osmoreception, the present study examined the interactions between osmotic and thermal stimuli in Prl cells of Mozambique tilapia acclimated to either FW or SW. Our ndings indicate that: 1) A rise in temperature increases Prl release in-vitro as early as 1 h; 2) The osmotic-sensitivity of Prl release is lost at 20°C; 3) Tilapia acclimated to SW are more responsive to changes in temperature than those acclimated to FW; 4) prlr2 expression is inversely related with circulating levels of Prl and is inhibited by a rise in temperature; 5) trpv4 responded differentially to temperature depending on the acclimation salinity of sh; 6) Most of the TF transcripts in Prl cells of SW-acclimated tilapia decrease their mRNA levels in response to an elevation in termperature.
Mozambique tilapia Prl cells are osmoreceptors 7 which have been recently described to also respond to physiologically relevant increases in temperature by increasing Prl release 27 . The control of Prl release by environmental salinity, in vivo, and extracellular osmolality, in vitro, is well studied 17,19,49,50 . Prl cells from FWacclimated tilapia release more Prl than their SW-counterparts, and respond more robustly to changes in extracellular osmolality 17,26 . As expected, robust hyposmotically-induced Prl 188 release was observed in FWacclimated tilapia Prl cells, especially by 6 h of incubation; a rise in temperature ampli ed this effect. In our previous study, both dispersed Prl cells and RPD organoids responded to higher temperatures by elevating Prl 188 release by 6 h of incubation (24). The present study, however, is the rst to test the response of dispersed tilapia Prl cells subjected to 20°C, which interestingly blocked hyposmotically-induced release of Prl 188 , but not Prl 177 . A previous in-vivo study showed no changes in plasma Prl 188 when tilapia were exposed to temperatures ranging between 20°C and 35°C, though plasma cortisol decreased at higher temperatures 51 . Inasmuch as cortisol has been reported to inhibit Prl release 52-54 ,the thermally-induced rise in Prl release observed in the present study is consistent with lower cortisol in circulation. Moreover, because a rise in temperature also increases Prl cell volume, which mediates Prl release 27 , the observed inhibition of Prl release at 20°C by 6 h may be directly linked to cell volume change.
Inasmuch as Prl is pleiotropic, a rise in temperature might affect several other key functions of Prl, including growth and reproduction. In fact, a previous study has shown that warmer water (32°C) increases growth, while temperatures as low as 22°C resulted in stunted growth 55 . Moreover, Prl has been linked with testosterone production and gonadal activity of tilapia 56 underscoring the linkage between elevated Prl at high temperatures and increased sexual maturity. While the osmotic sensitivity of Prl 188 release was lost at 20°C, it did not affect the osmotic responsiveness of Prl 177 release.. Prl 177 also exerts somatotropic actions in tilapia 57 and while a reduction in Prl 177 at 20°C is consistent with lower growth, the retention of its hyposmotic response at that temperature may be vital for FW acclimation in cooler temperatures. The differential responsiveness of Prl 188 and Prl 177 to extracellular osmolality has also been suggested to underlie the observed differences in salinity tolerance between Mozambique tilapia and its congener Nile tilapia, Oreochromis niloticus 58 . Based on the thermal modulation of osmotic responses observed in the present study, it would also be tenable that variations in temperature act in concert with changes in salinity in determining the species-speci c environmental regulation of Prl in teleosts.
In FW-acclimated tilapia, prl mRNA did not show any osmosensitivity, consistent with previous studies 17, 59 and the notion that prl mRNA levels in FW-sh may be at or near the maximum transcriptional activity and therby unresponsive to further osmotic stimulation 26 . On the other hand, Prl cells from SW-acclimated tilapia contain low levels of Prls, and therefore, activate prl 177 and prl 188 transcription in hyposmotic conditions 26,37,59 .
Accordingly, we observed prl 188 to be responsive to osmotic stimuli in Prl cells of SW-tilapia. Furthermore, in SWtilapia, prl 177 was not as osmotically sensitive as prl 188 , consistent with previous observations 17 . The two prl transcripts showed opposite expression patterns in response to thermal stimuli. The expression of prl 188 peaked with a rise in temperature in hyposmotic media, indicating that a combination of heat and low osmolality synergizes to maximally induce prl 188 transcription. Thermally-induced Prl release was recently shown to be mediated, at least partially, by a cell-volume dependent mechanism, similar to that involved in osmotically induced Prl release 27 . Consistently, the transcription of prl 188 may be activated in similar fashions by thermal and osmotic stimuli, and further augmented in environments that are both hyposmotic and warm.
In Mozambique tilapia, the biological effects of Prls are mediated by Prlr1 and Prlr2, whose transcription in target tissues is also characterized by high osmotic sensitivity 17,60,61 . Expression of both prlrs was affected by temperature and osmolality. Consistent with previous studies 17,22 , the relationship of prlr2 was inversely related to extracellular osmolality in Prl cells of both SW-and FW-acclimated tilapia. Both prlrs were decreased by incubation at 32°C compared with cooler temperatures (Fig. 4C and 4D). Regardless of the circulating levels of Prls, the environmental control of their receptors are implicated in modulating the hormonal actions 17,61 . The observed decreases in prlrs with a rise in temperature, especially in Prl cells of SW-acclimated sh incubated in hyperosmotic media, suggests that Prl's effects in high temperature may be attenuated.
The transduction of hyposmotic stimuli in tilapia Prl cells is dependent on the entry of extracellular Ca 2+ 50,62,63 through trpv4 channels 31,34 . Trpv4 is sensitive to many stimuli including osmotic pressure and heat 64,65 . We observed an increase in trpv4 proportional to that of extracellular osmolality, though this relation was attenuated at the highest incubation temperature. The responses of trpv4 to thermal and osmotic sensitivity differed between Prl cells from FW-and SW-acclimated tilapia, though genereally, the transcript was most highly expressed at 26°C. In our previous study, Prl cells of FW-acclimated tilapia increased trpv4 in response to an elevation in temperature 27 . In the present study, this trend was con rmed, but only when comparing cells inclubated at 20°C and 26°C. In SW-acclimated tilapia, however, there were no clear effects of thermal regulation of trpv4 expression. Acclimation history plays a vital role in trpv4 expression and it has been reported that Prl cells of SW-acclimated tilapia express four-fold higher trpv4 than their FW counterparts 59 . Hence, a decreased in trpv4 expression observed in Prl cells of SW-acclimated sh incubated at 32°C may indicate the attenuation of cellular sensitivity to extracellular Ca 2+ entry in response to environmental stimuli, similar to the response of prlrs. It is well accepted that strict regulation of Ca 2+ concentrations in the cytosol is important in Ca 2+ -mediated cell signaling 66 ; a rise in cellular Ca 2+ concentration beyond optimum levels may lead to cytotoxicity and cellular apoptosis 67 . Hence the thermally-induced downregulation of trpv4 in SW-acclimated tilapia could also serve as a protection mechanism to prevent Ca 2+ toxicity.
The transduction of osmotic stimuli into the activation of prl transcription is largely regulated by the activity of TFs and TF modules (TFMs) that operate in the promoter regions of prl 188 and prl 177 genes 29,37 . In Prl cells of tilapia acclimated to SW, expression of TF transcripts was more sensitive to both thermal and osmotic stimuli compared with those in FW. Pit1 and Oct1 have been reported to regulate prl transcription in sh and mammalian models 38,68,69 . In the tilapia RPD, pou1f1 and pou2f1b were the most highly expressed transcripts of Pit and Oct1, respectively 29 . Moreover, pou1f1 expression was inversely related with osmolality in SWacclimated tilapia 37 . Both pou1f1 and pou2f1b were inhibited by a rise in temperature. Inhibition of these TFs by high temperature reinforces the notion of a compensatory mechanism that attenuates thermally-induced Prl release. The osmotic sensitivity observed at 32°C indicates that Prl cells are capable of retaining osmoreceptive functions at higher temperatures. In FW-sh, both pou1f1 and pou2f1b were inversely related to osmolality at 20°C. At this low temperature, Prl 188 release was minimal and unresponsive to changes in media osmolality; prl 188 expression was unresponsive to both osmotic and thermal stimuli by 6 h. Collectively, these results suggest that, in FW-acclimated tilapia, Prl cells maintained their osmosensitivity through pou1f1 and pou2f1b at 20°C even though prl 188 mRNA was unchanged across treatments, possibly as result of pre-existing elevated levels of transcripts and stored Prl 188 .
Hyposmotically-induced Prl release has also been shown to involve the cAMP second messenger system 35,70 .
To address downstream changes in this second messenger system, we characterized the response of two transcripts of CREB and CEBP, creb3l1 and cebpb, respectively, which are prevalent in tilapia Prl cells 29 . Similar to the pattern of expression observed for POU genes, a rise in temperature inhibited creb3l1 expression in Prl cells of both SW-and FW-acclimated tilapia. In SW sh, creb3l1 increased in hyperosmotic media at colder temperatures and was attenuated at 32°C. This expression pattern was quite similar to the expression of trpv4, suggesting a linkage between Ca 2+ and cAMP second messenger systems in the integration of thermal and osmotic responses. By contrast, creb3l1 was not affected by media osmolality in Prl cells of FW-acclimated tilapia; the only effect observed was the downregulation of the transcript with rising temperature in hyposmotic media. Previously, we reported that trpv4 increased from 26°C to 32°C in isosmotic conditions, but did not affect and stat3 expression at the warmer temperature, as a long-term negative feedback response. The thermal response of stat1a was similar to that of stat3, although the similarity in osmotic sensitivity was only observed at 32°C. These results indicate that the responses of stat1a to environmental changes may not be as sensitive as those of stat3, and suggest that during downstream signaling it may be largely sensitive to autocrine regulation by Prls. In Prl cells of FW-acclimated tilapia, stat3 showed similar osmotic sensitivity as their SW counterparts at lower temperatures. At 32°C, however, osmotic responses were abolished or attenuated in a similar manner as observed with prlr2, suggesting that this receptor isoform and stat3 may be linked during the downstream activation of autocrine signaling. Stat1 is activated by heat in mammalian cell models 73,74 , though downstream signaling effects may differ if Stat1 dimerizes or binds with Stat3 75 . Earlier we found stat3 levels to be similar in RPDs of SW-and FW-acclimated tilapia but stat1a levels to be higher in SW sh 29 . Therefore, the nuances we observe in stat transcripton may be tied with acclimation salinity.
Finally, NFATs have been reported to be activated following rapid Ca 2+ in ux 76 and in response to hyperosmotic stress in mammalian cell models and in gills of Atlantic salmon, Salmo salar 44,77,78 . Also, NFAT is reported to form TFMs with AP1, a TF that is sensitive to both hypo-and hyperosmotic stress 76, [79][80][81] . Recently, we reported that the TFM, NFAT_AP1F is activated by both hypo-and hyperosmotic stimuli in tilapia Prl cells 37 . In the present study, nfatc1 expression was reduced in Prl cells of SW-acclimated tilapia by hyperosmotic conditions. The induction of nfatc1 at lower media osmolalities may occur, therefore, in response to hyposmotically-induced Ca 2+ entry. Furthermore, nfatc1 transcription was attenuated by heat. At 32°C, trpv4 was also inhibited, suggesting that the attenuation of nfatc1 could be linked to a reduction in Ca 2+ in ux. At 32°C, similar patterns of transcription were observed in trpv4, creb3l1, cebpb and nfatc1, underscoring the importance of free Ca 2+ entry to activate prl transcription. In FW sh, nfatc1 expression was reduced at 20°C and unresponsive to osmotic stimuli. Similarly, trpv4 expression was lower at 20°C compared with other incubation temperatures.
Together, these results are consistent with the notion that extracellular Ca 2+ entry into the intracellular space is important to upregulate nfatc1.
This study unveils the transcriptional responses to temperature and salinity of molecular regulators involved in prl transcription and Prl release in a euryhaline and eurythermal sh model that is highly adaptable to environmental uctuations. Even though teleosts are considered ectotherms, these results provide evidence of cellular mechanisms of a pleiotropic endocrine system that sense and respond to both thermal and osmotic stimuli. Rises in temperature further augmented hyposmotically-induced Prl release while at the same time attenuating the transcription of TFs and prlrs involved in the osmoreceptive and autocrine responses of Prl cells, indicating thereby that both temperature and extracelluar osmolality modulate Prl cell responses in concert. As a result, multiple physiological processes such as growth, development, reproduction and osmoregulation are likely modi ed following the integrated adaptive responses to changes in environmental temperature and salinity. In light of the predicted increase in frequencies of extreme whether events leading to rising sea surface temperatures and uctuating salinities 82,83 , these ndings provide insight on how sh may be capable of integrating and responding to these various environmental cues simultaneously.  Table 1. The geometric mean of three reference genes (ef1-α, 18S, and β-actin) was used to normalize target genes. Data are expressed as mean fold-change ± SEM (n = 8) from the isosmotic-isothermal treatment (330 mOsm/kg at 26°C).

Statistics
Data from static incubations of Prl cells were analyzed by two-way ANOVA with osmolality and temperature as main effects. Signi cant effects of medium osmolality and temperature were followed up by protected Fisher's LSD test. When necessary, data were log-transformed to satisfy normality and homogeneity of variance requirements prior to statistical analysis. All statistics were performed using Prism 9 (GraphPad, La Jolla, CA) and data are reported as means ± SEM.

Declarations
Declaration of competing of interest The authors declare that they have no con icts of interest.

Ethics Declaration
All experimental procedures and methods were conducted in accordance with the ARRIVE guidelines and all experiments and methods used were approved by the Institutional Animal Care and Use Committee, University of Hawai'i.
(Interact), group comparisons were conducted using protected Fisher's LSD test. Groups not sharing uppercase letters indicate signi cant (P<0.05) mean differences in response to incubation temperatures and groups not sharing lowercase letters re ect signi cant (P<0.05) mean differences in response to media osmolality.   Effects of incubation osmolality and temperature on the mRNA expression of prlr1and prlr2 in SW-acclimated tilapia (A and C) and FW-acclimated tilapia (B and D) Prl cells after 6 h of incubation. Data are expressed as mean fold change from the isosmotic and isothermal (330 mOsm/kg:26 °C) group ± SEM (n=6-8). The effects of osmolality and temperature were analyzed by two-way ANOVA (*P<0.05, **P<0.01, ***P<0.001). When there was a signi cant effect of temperature (Temp), media osmolality (Osm) or interaction (Interact), group comparisons were conducted using protected Fisher's LSD test. Groups not sharing uppercase letters indicate signi cant (P<0.05) mean differences in response to incubation temperatures and groups not sharing lowercase letters re ect signi cant (P<0.05) mean differences in response to media osmolality.   sharing lowercase letters re ect signi cant (P<0.05) mean differences in response to media osmolality.