Early warming stress on juvenile fish impairs testicular development and sperm quality but 1 contrastingly elicits intergenerational thermotolerance 2

1: Department of Structural and Functional Biology, Botucatu São Paulo State University (UNESP), Institute of 5 Biosciences, Botucatu, São Paulo 18618-689, Brazil; 6 2: Department of Surgery, School of Veterinary Medicine and Animal Sciences, University of São Paulo, São Paulo, 7 São Paulo 05508-270, Brazil 8 3: Laboratorio de Biología del Desarrollo, Instituto Tecnológico de Chascomús (INTECH), Consejo Nacional de 9 Investigaciones Científicas y Técnicas/Universidad Nacional de San Martín (CONICET/UNSAM). Chascomús, 10 7130, Argentina 11 4: Fishery Engineering Course and Aquaculture Centre (CAUNESP) São Paulo State University, Registro, São Paulo 12 11900-000, Brazil 13 5: Salmonid Experimental Station at Campos do Jordão, UPD-CJ, Sao Paulo Fisheries Institute (APTA/SAA), 14 Campos do Jordão, São Paulo 12460-000, Brazil 15 16 Corresponding Author: 17 * Ricardo Shohei Hattori, UPD-CJ Sao Paulo Fisheries Institute (APTA/SAA), CP 361. 12460-000. Campos do 18 Jordao, Brazil. +55 123663-1021 19


40
Water temperature comprises an important modulatory factor with critical roles on fish reproduction.

41
During early life stages, the destiny of gonadal sex differentiation in gonochoristic species can be irreversibly driven 42 towards either female or male by temperature, overcoming the predisposed sex determined by genotypic factors 1-4 .

43
The appearance of sex-reversed fish and the concomitant skews in sex ratios has great implications from ecological 44 perspectives due to their impacts on population structure 5 .

45
Another effect of temperature on reproduction occurs through the regulation of reproductive cycle, either by 46 promoting 6 or suppressing gametogenesis 7 . However, chronic exposure at those temperatures or acute thermal stress 47 at even higher temperatures can cause opposite inhibitory effects on spermatogenesis 8,9 . In ovaries, although warm 48 conditions are also able to hasten gametogenesis as in males 6 , high temperatures that do show clear inhibitory effects 49 on spermatogenesis do not necessarily induce comparable changes in oocyte development 10 . At sub-lethal, high 50 temperature conditions, the survival of testicular somatic-supporting cells as well as the germ cells can be severely 51 affected, whereby undifferentiated spermatogonia seems to be more tolerant to depletion by apoptosis than the 52 differentiated ones, such as spermatocytes, spermatids, and spermatozoa 8,9 . The mechanism of heat-induced germ 53 cell depletion is not well understood, but the Sertoli cells are likely involved, since apoptosis in these cells have been 54 detected along with germ cells 8,9 . On the other hand, undifferentiated oogonia seems to be more susceptible than the 3 differentiated oocytes upon exposure to those temperatures 10 , suggesting that high temperatures affect fish 56 reproduction in a sex-specific manner.

57
Although the effects of thermal stress on fish reproduction have been assessed in some species, the 58 implications on their reproductive capacity have not been well explored, especially in terms of gamete quality and 59 offspring performance. Furthermore, the performance of offspring produced by fish exposed to warm water 60 temperature has not been well evaluated yet. Research about how temperature acts on fish germ line, on gametes 61 production or quality, and on progenies fitness might provide important insights for the evaluation of environmental 62 changes (e.g., global warming) on wild populations and extensive aquaculture. On this regards, salmonids are an 63 excellent group of fish to evaluate the effects of increasing temperature because they include several cold-water 64 species which born in freshwater environments and then migrate downward to the river mouths until reaching the 65 sea. Also, some species present variants that spend their entire life cycle in inland waters (landlocked) such as the 66 rainbow trout (Oncorhynchus mykiss) and the Atlantic salmon (Salmo salar). But regardless of these different life 67 cycles, salmonids have a high likelihood to experience warm temperatures and hypoxia conditions during the 68 juvenile stage 11,12 . Some of those effects include impairment of steroidogenesis and vitellogenin synthesis 13 , and 69 advance or delay in oocyte maturation in females 14 . In case of males, impairment of spermatogenesis and reduced 70 milt volume are reported in fish exposed to high temperature 15 .

71
In this study, we used the rainbow trout as an experimental model to evaluate the effects of prolonged 72 treatment at warm temperatures during juvenile stage on several reproduction parameters in female and male adults.

73
We also compared survival and growth performances of the respective progenies under warm temperature in 74 juveniles and the upper thermal tolerance in adults in order to investigate the intergenerational inheritance of 75 thermotolerance.

79
Effects of warm temperature on body growth and gonads of juveniles 80 At the end of the experiment (3 months) with F0 juvenile fish, warm temperature group (WT) showed lower 81 survival rate than control group (CT) (70% and 97%, respectively), but growth parameters such as standard length 82 (mean ± SD = 21.67 ± 9.41cm and 23.57 ± 8.76cm, respectively), and body weight (81.61 ± 28.79g and 71.73 ± 83 18.26g, respectively) did not differ significantly ( Fig. 1A-B). The gonadosomatic index in females was lower in WT 84 compared to CT (0.0007 ± 0.0003 vs 0.0012; p < 0.05) whereas no difference was found for males (Fig. 1C).

85
Histological analyses of ovaries ( Fig. 1D and 1F) and testis ( Fig. 1E and 1G) did not reveal clear differences between

86
WT and CT groups.

89
Females from WT group were significantly smaller than those from CT group and presented higher 90 condition factor (K). Nevertheless, no differences were found in body weight (Suppl. Fig. 1). Fecundity rates, oocyte 4 mean weight, and the percentage of non-ovulated females did not differ between females of WT and CT group 92 (Suppl. Fig. 2; Table 1).

93
WT males were significant smaller and had lower body weight than control males; hence, condition factor 94 was significantly higher in WT group (Suppl. Fig. 1), as in females. Gonad dissection and histological analyses in 95 some of those fish revealed three patterns of testis morphology. The first one consisted in a large whitish testis, 96 similar to those of CT males ( Fig. 2A). In the second pattern, a smaller whitish testis was detected and correlated 97 with males with low relative milt volume (Fig. 2C). The third pattern was found for immature males and consisted in 98 a thinner gonad with a reddish color (Fig. 2E). Histological analyses revealed the presence of some undifferentiated 99 spermatogonia and a high quantity of spermatozoa in the first two patterns ( Fig. 2B and 2D) whereas the third pattern 100 was characterized by undifferentiated spermatogonia without any spermatozoa or spermatocytes ( Fig. 2F and 2H), as 101 revealed by immunohistochemistry analysis with an antibody for undifferentiated spermatogonia (Fig. 2G), which 102 resembled the immature testis of F0 juveniles (Fig. 1E).

103
The relative milt volume was also lower in WT than CT group (Fig. 3A), with no difference in the 104 concentration of spermatozoa (Fig. 3B). The estimated total amount of spermatozoa was reduced in about 57% in 105 WT males (Fig. 3C). A proportion of males did not show secondary sexual characters and did not release any milt.

106
These males were classified as immature males and they were not detected in any of CT males (Table 1)   127 values were 71.4% and 54.5% higher in WT group for sub-adult and adults, respectively.

130
In this study, we demonstrated that exposure to warm temperature in male juveniles affect negatively not 131 only survival, growth, and fecundity, but also the motility of spermatozoa and body formation in the respective 132 progenies. Despite of those impacts, the progenies obtained from these males presented a remarkable high survival 133 and growth rates when exposed to warm temperature compared to the control group, which supports an increased 134 thermal tolerance after a single generation.

135
Long-term exposure to temperatures above optimal ranges causes detrimental effects on growth and survival 136 in salmonids 14,16 . Another related impact comprises the depletion of the germ cells in juveniles and adults of many 137 other fish species 9,10,17,18 . In our experiment, heat treatment at juvenile stage did not affect reproductive parameters in 138 females, which was not the case of males. Those fishes showed quiescent gonads that persisted for two consecutive 139 reproductive seasons, and reduced testes, impacts that were not observed in any fish from CT group and, to the best 140 of our knowledge, in any teleost fish. Since the quiescent gonads possess germ cells, we can consider these fish as 141 infertile, but not sterile. A deep analysis on the regulatory mechanisms is required to determine how this state is 142 controlled, if these fish still possess the capacity to resume spermatogenesis, or if steroid hormones can make this 143 fish to overcome such "quiescent" condition. The significant decrease in milt volume is likely due to spermatogonial 144 apoptosis 9,10,18,19 , because thermal treatment was performed at juvenile stage, when the immature testes are composed 145 mainly by undifferentiated spermatogonia. Regarding the difference in heat sensitivity between females and males, 146 similar responses have been reported in teleost and mammals 10,20,21 , supporting the idea that spermatogonia exhibits 147 higher heat-sensitivity than the oogonia.

148
Another intriguing effect of heat treatment was observed on motility parameters of spermatozoa, whereby 149 WT group did not show any motility after 20 s (25 and 30 s). In addition, WT spermatozoa also had higher sperm 150 velocity and lower wobble compared to CT group at 20 s. Such patterns could be associated with high energy 151 consumption by WT spermatozoa at the initial 5 and 10 s, based on average motility which was about 19 to 26% 152 higher, respectively. Thus, after 20s, WT spermatozoa would have lower available energy for motility. Reduced

190
In conclusion, this study showed that warm temperature exposure in juveniles causes deleterious effects on 191 germ cells that persist even in adults by affecting gamete production in males. Apart from germ cell degeneration,

240
Total body weight (BW; g) and standard length (SL; cm) were measured for each fish. The condition factor 241 K was calculated as K=100xBW/SL 3 and the gonad weight (GW; g) was also measured for some fish to calculate the

267
In the analysis using 2 years-old-fish, a total of 12 WT males were crossed with 13 CT females and 268 conversely 6 WT females were crossed with 11 CT males. For those crosses, 27 g of oocytes were inseminated with 9 2 mL of milt. In the following year, three-years-old males from WT (n = 8) and CT (n = 12) were crossed with eggs 270 from a single CT female, in order to evaluate whether the effects observed in progenies from two-years-old-fish were 271 restricted to fish from that age. Spermatozoa were activated using a sodium bicarbonate solution (0.01%). Eggs were 272 then hydrated with rearing water during 20 min. and incubated at 11 °C in UV-treated water under constant flow, 273 following conditions described in a previous study 39 . The percentage of fertilization at eyed-egg stage, hatching rate, 274 and abnormality in eleuteroembryos were quantified for each cross. Embryos with body deformations such as twisted 275 body or with circular movements were considered as abnormal.

277
Survival and growth performance of progenies derived from WT males.

278
The survival and growth performances of F1 progenies derived from WT males were compared with those 279 from CT males. Milt from 2 WT males and 2 CT males was used to artificially inseminate the same pool of oocytes 280 for each trial. Egg incubation and larvae rearing followed the same procedure described in the previous section. For