Placental mitochondrial function, nutrient transporters, metabolic signalling and steroid metabolism relate to fetal size and sex in mice

Fetal growth depends on placental function, which requires energy supplied by mitochondria. Here we investigated whether mitochondrial function in the placenta relates to growth of the lightest and heaviest fetuses of each sex within the litter of mice. Placentas from the lightest and heaviest fetuses were taken to evaluate placenta morphology (stereology), mitochondrial energetics (high-resolution respirometry), and mitochondrial regulators, nutrient transporters, hormone handling and signalling pathways (qPCR and western blotting). We found that mitochondrial complex I and II oxygen consumption rate was greater for placentas supporting the lightest female fetuses, although placental complex I abundance of the lightest females and complexes III and V of the lightest males were decreased compared to their heaviest counterparts. Expression of mitochondrial biogenesis (Nrf1) and fission (Drp1 and Fis1) genes was lower in the placenta from the lightest females, whilst biogenesis-related gene Tfam was greater in the placenta of the lightest male fetuses. Additionally, placental morphology and steroidogenic gene (Cyp17a1 and Cyp11a1) expression were aberrant for the lightest females, but glucose transporter (Glut1) expression lower in only the lightest males versus their heaviest counterparts. Differences in intra-litter placental phenotype were related to sex-dependent changes in the expression of hormone responsive (androgen receptor) and metabolic signalling pathways (AMPK, AKT, PPARγ). Thus, in normal mouse pregnancy, placental structure, function and mitochondrial phenotype are differentially responsive to growth of the female and the male fetus. This study may inform the design of sex-specific therapies for placental insufficiency and fetal growth abnormalities with life-long benefits for the offspring.


Introduction
Morphology of the Lz was assessed by double-labelling placental sections with cytokeratin and lectin antibodies 183 to identify trophoblast and fetal capillaries, respectively. Details about the staining protocol have been described 184 in detail elsewhere 27 . Stained sections were then scanned using a NanoZoomer 2.0-RS Digital Pathology 185 System (NDP Scan Hamamatsu, Japan) and stereological analysis of the Lz was performed as described 186 previously 27 .

198
Conceptus biometry for all female versus all males within the litter 199 Considering all fetuses together, conceptus biometric data were different between females and males within the 200 litter at gestational day 18 (term on day 20; Figure S1). In particular, fetus, placenta and Lz weights were lower 201 in females when compared with male fetuses (p=0.03, p<0.001 and p=0.02, respectively). However, there were 202 no differences in fetal brain and liver weights (relative to body weight), or placental efficiency (p=0.06), 203 calculated as the ratio of fetal weight to placental weight between females and males ( Figure S1).   215 To understand why there may be a sex-related difference in the relationship between placental weight and fetal 216 weight, the lightest and heaviest fetuses from each litter were selected and conceptus biometry were compared 217 for each sex separately ( Figure 2). As expected, fetal weight was lower for the lightest compared to the heaviest 218 for each fetal sex in the litter (Figure 2A, females; p=0.002; males; p<0.0001), and the mean weight difference 219 between them were similar for females and males (14.1% and 13.6% less than heaviest, respectively). Fetal 220 brain and liver weights as a proportion of body weight did not vary, which suggests that the lightest fetuses are Stereological analysis of the placental Lz zone revealed that there were no differences in trophoblast and fetal 226 capillary volumes ( Figure 2I-K). However, there was less maternal blood spaces in the placental Lz of the 227 lightest females, compared to the heaviest females, and this difference was not found for the males ( Figure 2M). 228 Similarly, maternal blood space surface area was lower in the lightest, compared to heaviest female fetuses, an 229 effect not observed for the male fetuses ( Figure 2N). The surface area of the fetal capillaries ( Figure 2L) and 230 barrier thickness ( Figure 2O) of the Lz did not vary between the lightest and the heaviest fetuses, for either 231 females or males. To investigate whether sex-dependent structural changes in the Lz zone between the lightest and heaviest 236 fetuses may be related to mitochondrial functional alterations, high resolution respirometry was performed 237 ( Figure 3A). Oxygen flux rate analysis revealed that in LEAK state, mitochondrial CI related oxygen 238 consumption was ~60% greater for the placental Lz of lightest compared to the heaviest females, but no effect 239 was seen for the males ( Figure 3B, p=0.003). Whilst CI oxygen flux under OXPHOS state was not different between the lightest and heaviest fetuses of either sex ( Figure 3C), after adding succinate, Lz CI+CII oxygen 241 consumption rate was ~44% greater in the lightest compared to the heaviest females within the litter (p=0.01); a 242 difference that was not observed for the males ( Figure 3D). Fatty acid oxidation (FAO), total ETS capacity and 243 CIV associated oxygen consumption by the placental Lz were not different between lightest and heaviest fetuses 244 for either fetal sex ( Figure 3E-G). When oxygen consumption rates for CI in LEAK state and CI+II in OXPHOS 245 state were corrected to total ETS oxygen flux to provide a qualitative indication of changes in mitochondrial 246 function per mitochondrial unit, these values were also increased in only the lightest compared to heaviest 247 females (not different for the lightest compared to heaviest males) ( Figure 3H, p=0.02; 3I and 3J, p=0.03). In 248 addition, calculation of 1-P/E indicated that ETS excess capacity was lower in the lightest females compared to 249 the heaviest (p=0.03), again a difference not identified for the males ( Figure 3K). To gain further information on the sex-dependent differences in placental mitochondrial respiratory capacity, 254 western blotting and qPCR was performed to determine the expression ETS complex proteins (CI-V), 255 biogenesis, fusion and fission genes, and additional mitochondrial regulatory proteins in the placental Lz of the 256 lightest and heaviest fetuses for both sexes (Figure 4). These analyses revealed that CI abundance was lower in 257 the lightest compared to heaviest females ( Figure 4A), meanwhile CIII and CV were lower only in the lightest 258 compared to the heaviest males ( Figure 4B). In addition, the expression of mitochondria biogenesis gene, Nrf1 259 and mitochondrial fission genes, Drp1 and Fis1, was lower in the Lz of the lightest females, when comparing 260 with the heaviest females ( Figure 4C). Whereas the expression of Tfam, a mitochondria biogenesis transcription 261 factor gene, was greater in the Lz of the lightest males versus the heaviest males ( Figure 4D). Mitochondrial 262 content, informed by citrate synthase abundance, did not vary in the Lz between the lightest and heaviest fetuses 263 within the litter, regardless of fetal sex ( Figure 4E). In addition, abundance of proteins involved in 264 mitochondrial biogenesis (PGC-1α), fusion (MNF2 and OPA1), heat shock (HSP60, HSP70) and chaperone 265 (TID1) proteins did not differ in the Lz between the lightest and heaviest fetuses within the litter, in either sex 266 ( Figures 4E). However, CLPP a key protease involved in mitochondrial protein clearance and a marker of the 267 mitochondrial unfolded protein response (UPRmt), was lower in the lightest females compared with the heaviest 268 females; an effect not seen for males ( Figure 4E).

lightest versus the heaviest fetuses of each sex
Since the energy provided by mitochondria helps to fuel placenta transport and endocrine function, we 273 evaluated whether sex-related variations found on mitochondria functional capacity (respiratory function, gene 274 and protein regulators) are associated with the expression of nutrient transporter and steroidogenic genes 275 between the lightest and heaviest of each sex within the litter. In particular, the mRNA expression of key 276 transporters for glucose (Slc2a1 and Slc2a3), amino acid (Slc38a1, Slc38a2, Slc38a4, Slc7a5 and Slc3a2) and 277 lipids (Fatp1, Fatp3, Fatp4, Fatp6 and Cd36) were quantified in the placental Lz zone by RT-qPCR ( Figure 5). 278 We also evaluated the expression of genes involved in steroid hormone production (Star, Cyp11a1 and 279 Cy17a1), glucocorticoid metabolism (11bhsd1 and 11bhsd2) and steroid hormone signalling (Esr2 and Ar) in 280 the placental Lz using qPCR. These analyses showed that the expression of Slc2a1 mRNA was ~20% lower for 281 the lightest compared to the heaviest males ( Figure 5B, p = 0.021), however this difference was not observed for 282 the lightest versus the heaviest females. In addition, no differences were found between the lightest and the   p=0.005 and p=0.01, respectively), however this was not related to a significant change in activation of AMPKα (abundance of phosphorylated normalized to total AMPKα, Figures 6B and D). While the total abundance of 302 AKT did not vary between the lightest and the heaviest fetuses, activated AKT (phosphorylated to total) was 303 ~32% lower in the Lz zone supporting the lightest compared to the heaviest males (p=0.032), but no difference 304 was found for the females (Figures 6B and D). The abundance and activation of p44/42 MAPK and p38 MAPK 305 was not different between the lightest and the heaviest fetuses, irrespective of fetal sex ( Figure 6A and D). 306 Interestingly, PPARγ an important transcription factor involved mitochondrial metabolism and lipid synthesis, 307 was greater in the lightest female compared to the heaviest female, whereas PPARγ was lower in the lightest 308 males when compared to the heaviest males from the litter ( Figure 6E, p=0.04 and p < 0.05, respectively).  Table 1). These data showed that the heaviest males were ~5% heavier 314 than the heaviest female fetuses within the litter (p <0.05). The placenta of the heaviest males in the litter was 315 also greater by ~13% when compared to the heaviest females (p=0.04), with a tendency for this to also vary 316 with sex for the lightest female littermates (p=0.054). The placental expression of glucose (Slc2a1: -37%, 317 p=0.003) and lipid (Fatp1: +17%, tendency p=0.07) transporter genes were also differentially expressed 318 between the lightest (but not heaviest) male and female fetuses of the litter. Placental respirometry rates 319 associated with CI Oxphos (+39%, tendency p=0.05) and with CI+CII Oxphos /Total ETS (+30%, tendency p=0.08) 320 together with biogenesis (Nrf1: +37%, p=0.03; Tfam: +25%, tendency p=0.06) and dynamic (Opa1: +13%, 321 tendency p=0.06, Mfn1: +11%, p=0.05) genes were all greater in the lightest males compared to the lightest 322 females. Meanwhile, biogenesis genes Pparγ (-38%, p=0.03) and Tfam (-19%, tendency p=0.07) were 323 decreased in the heaviest males compared to females. Finally, the expression of the steroidogenic gene Cyp11a1 324 was lower (-38%, p=0.02) in the lightest males compared to females and in heaviest fetuses, gene expression of 325 Cyp17a1 (-41%, p=0.01) were decreased on males compared to females. There was also no effect of fetal sex in 326 the placental Lz morphology of the lightest and heaviest fetuses. The present study in mice shows that placental mitochondrial functional capacity varies in relation to natural 330 differences in the weight of females and male fetuses within the litter. The placenta Lz of both the lightest female and male fetuses showed altered abundances of ETS complexes and mitochondrial biogenesis genes 332 between the lightest and heaviest fetuses, although the specific nature of these changes depended on fetal sex. 333 Moreover, the morphology, respiratory capacity, mitochondrial fission, and misfolded protein regulators of the 334 placental Lz differed between the lightest and heaviest females, but not males. The level of nutrient (glucose) 335 transporters varied between the lightest and heaviest males, but not females, whereas the ability to produce 336 steroid hormones differed only between the lightest and heaviest females within the litter. There were also sex-337 dependent changes in the expression of hormone responsive, growth and metabolic signalling pathways in the 338 placental Lz between the lightest and heaviest fetuses. However, despite these sex-related variations, the 339 average weight differences between the lightest and heaviest fetuses were similar for both sexes. Together, these 340 data suggest that in normal mouse pregnancy, placental structure, function and mitochondrial phenotype appears  properties and efficiency of their placentas in these normal, healthy pregnancies. Interestingly, previous work 351 exploring the implications of natural intra-litter variability of placental weight, rather than fetal weight in mice, 352 has found morphological differences between the lightest and the heaviest placentas, which included a greater 353 Lz volume and an increased surface area for exchange 3 . Functional adaptations were also found, with a greater 354 rate of amino acid transfer and enhanced expression of sodium-dependent neutral amino acid transporter-2 355 (Slc38a2) by the lightest versus the heaviest placentas 3 . Similarly, calcium transfer across lightest placenta was 356 higher than the heaviest placentas within the litter, resulting in similar calcium accretion levels in the fetus 6 .

357
Variations in placental structure and transport within the litter were related to an increase in placental efficiency 358 in both of these previous studies 3,6 . However, in our study the placentas sustaining the lightest or the heaviest 359 fetus were not necessarily the lightest or the heaviest placenta within the litter. In addition, a key strength of our 360 study is that the lightest and heaviest fetuses of each sex were analysed. Segregating the data by sex identified 361 that there was a positive correlation between placental and fetal weight for only in the lightest females in the litter. These data indicate that there may be differences in the way in which the placenta may be supporting 363 growth of the female and male fetuses within the litter.

365
While there was no difference in placental fetal capillaries, trophoblast volume or barrier thickness between the 366 lightest and heaviest of either fetal sex, maternal blood space volume and surface area of the placenta was lower 367 in the lightest compared to the heaviest female fetuses. These differences in placental structure suggest mal-368 perfusion of the placenta of the lightest females and would be expected to decrease the delivery of nutrient and 369 oxygen to fetus and could explain the weight discrepancy when comparing to the heaviest females in the litter.

370
Previous studies in pigs have shown that compared to placentas supporting fetuses weighing closest to the litter 371 mean, placentas supplying the lightest fetuses within the litter have impaired angiogenesis 28 . Moreover, there 372 are differences in the extent of endothelial cell branching morphogenesis with conditioned media from the 373 placenta of females compared to males 28 . Work in rats has suggested that an angiogenic imbalance may 374 underlie regional differences in uteroplacental vascularization and fetoplacental development within the litter 29 .

375
Our study was not designed to assess the contribution of uterine position to the sex-related differences in 376 placental morphology of the lightest versus heaviest fetuses in the litter, which could be a focus of future work.  In the lightest female, but not lightest males, there was lower CLPP protein abundance when compared to the 406 heaviest female fetuses of the litter. CLPP is also decreased along with mitochondrial complex abundance in the 407 placenta of preeclamptic women delivering FGR babies 33 . However, no other members associated with the 408 UPRmt pathway that was analysed varied in the placenta between the lightest and heaviest fetuses (HSP60, 409 HSP70 and TID1), which may be expected given that our study was performed on mouse litters from normal,  However, it would be beneficial to assess glycolysis and glycolytic enzyme expression in the placenta to assess 448 whether there are intra-litter differences for females or males. In addition to its role in energy sensing, AMPK  489 Previous work has shown there are changes in placental Lz morphology, function, mitochondrial respiratory 490 capacity and mitochondrial-related regulators that help to support the growth demands of the fetus during 491 normal late mouse pregnancy 23 . In the current study the heaviest males and their placentas were heavier than 492 the heaviest females within the litter, although no differences were found in placental morphology, 493 mitochondrial functional capacity, or transport/hormone genes between them. In contrast, the lightest males and 494 their placentas did not differ in weight when compared to the lightest females (tendency for placental weight to      533 The authors declare that no conflicts of interest/competing interests exist.