The Drosophila seminal Sex Peptide can associate with rival as well as own sperm and provide function for SP in polyandrous females

In populations in which females tend to mate with more than one male, sperm competition and cryptic female choice can occur, triggering biases in sperm use and influencing males’ paternity share outcome of the mating. This competition occurs in the context of molecules and cells of male and female working interdependently towards the common goal of optimal fertilization. For example, a male’s seminal fluid molecules modify the female’s physiology to increase reproductive success. However, since some of these modifications induce long-term changes in female physiology, this can indirectly benefit rival males. Indeed, rival males can tailor their ejaculates accordingly, minimizing the energy cost of mating. Here we investigate the direct benefits that seminal fluid proteins from an ejaculate of one male can confer to sperm of a rival. We report that Sex Peptide (SP) that a female receives from one male can bind to sperm from a prior mate, that were already stored in the female. Moreover, the second male’s SP can restore fertility and facilitate efficient sperm release or utilization of sperm received from the first male that had been stored in the female. Thus, SP from one male can directly benefit another and as such is a key molecular component in the process of inter-ejaculate interaction.


Introduction 49
In many animal species, females mate with more than one male. This polyandry lays the 50 foundation for sperm competition, in which ejaculates from rival males compete for 51 fertilization opportunities [1,2]. These conflicts and associated cryptic female choice can 52 drive the evolution of male traits including optimal sperm numbers, morphology, and 53 seminal protein sequences [3][4][5]. 54 Against the backdrop of these conflicts, male and female molecules and/or cells must 55 also work together to ensure reproductive success. How efficiently sperm interact with 56 the egg and instigate successful fertilization or embryo support (where relevant) is key 57 to successful fertility. Accordingly, males have evolved molecular mechanisms to trigger 58 physiological changes in females that increase the reproductive success of the mating 59 pair. Seminal fluid proteins (SFPs) are crucial regulators of these changes. SFPs are 60 produced within glandular tissues in the male reproductive tract and are transferred to 61 females along with sperm during mating [6][7][8][9][10][11]. Within a mated female, SFPs mediate 62 an array of post-mating responses such as, in insects, changes in egg production, 63 elevated feeding rates, higher activity or reduced sleep levels, long-term memory, 64 activation of the immune system and reduced sexual receptivity [12][13][14][15][16][17][18]. 65 The ability of a male's SFPs to induce long-term changes in the mated female enhances 66 that male's reproductive success. For example, the seminal Sex Peptide (SP) of male 67 6 rematings at either 1d (Fig 1B) or 4d (Fig 1C) after the original SP-less mating. We 127 confirmed these findings with western blotting. Sperm stored in seminal receptacles of 128 females that had mated to SP-null males and subsequently remated to spermless males 129 were dissected and probed for the presence of SP. Consistent with our immunostaining 130 data, SP was detected in samples of SP-null male's sperm from females that had remated 131 to spermless males at 1d or 4d after the start of first mating (ASFM; Fig 1D, lanes 7 and 132 8). Thus, SP from a second male can bind to SP-deficient sperm stored from a prior male. 133 To see if the mating order was important, we carried out the reciprocal cross, i.e. testing 134 if SP deposited by a first male (spermless, in this scheme) could bind to sperm that were 135 subsequently introduced by a second (SP-null) male (Fig 2. Cartoon). Spermless males 136 transfer SP to the female tract after mating [32], but we did not detect any SP in females 137 mated to spermless males by 1d after the start of mating (ASM; Fig 1D. lane 4). We saw 138 no SP signal in sperm samples isolated from females that had mated to spermless males, 139 and then subsequently to SP-null males at 1d ASFM ( Fig 1D. lane 5). Our 140 immunofluorescence data were consistent with our western blots: we saw no SP-sperm 141 binding in females that mated first with a spermless male and a day later with SP-null 142 male ( Fig 2B). Therefore, if SP entered the female without sperm, it was unavailable to 143 bind to sperm from a subsequent SP-deficient male. 144 We hypothesized that we did not see SP bound to sperm in this second (reciprocal) 145 crossing scheme because by the time of the second mating SP from the spermless male 146 was no longer present in the female at 1d ASFM, since it could not be retained without 147 binding to sperm [19] and no sperm were being supplied by these first males. To 148 circumvent this, we attempted to remate females that had previously mated to 149 spermless males as soon as 3-6hrs ASFM. However, few females remated, likely due to 150 the recent experience of copulation, or to the effects of pheromones from the previous 151 mating [33,34]. In the few females that did remate, no SP-sperm binding was observed 152 ( Fig S1). Since the simplest explanation for these results was that SP transferred without 153 sperm had disappeared from females by the time of the second mating, we performed 154 western blotting to determine how long SP persists in the reproductive tract of females 155 in absence of sperm. Females mated to spermless males were flash frozen at 0'(min) 156 immediately after mating, 35'(min), 1hr, and 3hr ASM and their bursa (B) and seminal 157 receptacle (SR) were dissected and probed for the presence of SP. We detected SP in the 158 bursa protein samples at 0'(min) after mating, 35'(min), and 1hr ASM. (Fig 2D. lanes 5, 7, 159 9). However, SP was undetected in bursa or seminal receptacles of females at 3hr ASM 160 ( Fig 2D. lane 11, 12). Thus, we could not determine whether SP from mating with a 161 spermless male could bind a second male's sperm, because SP received from the first 162 mating was lost from the female reproductive tract before a second mating could occur. 163 Xue and Noll [31] reported that a similar cross (females mated first to spermless males 164 and then to Prd males) also gave no progeny (showed no copulation complementation) 165 which they proposed to be due to inactivation or early loss of SFPs in the absence of 166 sperm. Our results, showing that SP can bind to stored sperm from a prior male, provide 167 the molecular explanation for their observation. 168

SP from a second male restores fertility, inhibits receptivity and regulates 169
optimal release of the first male's sperm from storage 170 SP is needed for efficient sperm release and utilization from the female sperm storage 171 organs [6]. We tested whether SP from a second male could restore the use of a first 172 male's sperm. Females mated to spermless males have no progeny (Fig 3A). Females 173 singly-mated to SP-null males have significantly reduced numbers of progeny ( Fig 3A. SE 174 of diff = 8.043; p***=<0.001) relative to females mated to control males (Fig 3A), likely 175 because lack of SP prevents increase in egg production [17,35,36] and release of sperm 176 from storage [6]. However, females mated to SP-null males and then remated to 177 spermless males at 1d (Fig 3B; p=0.2487) and 4d (Fig 3C; p=0.8618) ASFM had progeny 178 levels similar to those of females that had mated to control (SP + ) males and were 179 subsequently remated to spermless males at the same time points. Thus, SP from a 180 second (SOT) male could rescue the fertility defects that resulted from the lack of SP 181 from an SP-null first male. 182 Reducing the likelihood of mated females to remate is another crucial postmating 183 response regulated by SP [35,37]. Females that do not receive SP generally fail to exhibit 184 this reluctance, and remate readily. We tested whether SP from a second male could 185 delay the receptivity of females that had previously mated to SP-null males. Females 186 singly-mated to SP-null males or spermless males show a significantly higher tendency 187 to remate at 1d ASM ( Fig 3D; p***=<0.001) or 4d ASM (Fig 3E; p***=<0.001) relative to 188 females mated to wt (CS) males (Fig 3D and 3E). In contrast, females mated to SP-null 189 males and then remated to spermless males at 1d ASFM (Fig 3D;p=0.43) showed 190 receptivity similar to mates of control males at 1d after the start of second mating 191 (ASSM). The effect, however, did not persist as long as after a mating to a wt male. At 4d 192 ASSM ( Fig 3E; p***=<0.001) doubly-mated females exhibited higher receptivity relative 193 to females mated to wt males but lower than those mated to spermless males. This could 194 be either because less SP from the second (spermless) mating is able to bind to stored 195 sperm from the previous mating and thus SP levels have been more depleted by 4 days 196 ASSM than after a control mating where the sperm-SP enter the female together. 197 Alternatively, the active portion of SP received from a rival male, bound to first male's 198 sperm might be released from the sperm at a higher rate. We performed western 199 blotting to determine how long SP received from the second (spermless) male persists 200 in the reproductive tract of females previously mated to SP-null males. Females singly-201 mated to CS males and those doubly-mated to SP-null males and spermless males at 1d 202 ASFM, were flash frozen at 2hr, 1d or 4d ASM/ASSM, respectively. SP signals were 203 detected in females mated to CS males at 2hr, 1d or 4d ASM ( Fig 3F. lanes 3, 4, 5). SP was 204 detected in females mated to SP-null males and then remated to spermless males at 2hr 205 and 1d ASSM ( Fig 3F. lanes 6, 7) but not (or very weakly) at 4d ASSM ( Fig 3F. lane 8). 206 Taken together, our results show that SP from a second male can rescue the receptivity 207 defects that resulted from the first male's of lack of SP but that sufficient SP for such an 208 effect is not retained for as long as in a control situation (e.g. a mating with a wt male). 209 SP is also needed for release of sperm from storage within the mated female [6]. Thus, 210 females mated to SP-null males retain significantly more sperm in their seminal 211 receptacle at 4d ASM. To test whether SP acquired from a spermless male in a second 212 mating could also rescue this defect, we counted sperm in storage after a single mating 213 seminal receptacle of females mated to SP-null; ProtB-eGFP males at 5d ASM ( Fig 3J) to 222 make sure that the evident decline in sperm counts or release of stored sperm in doubly 223 mated females (SP-null; ProtB-eGFP mates remated to spermless males that were frozen 224 at 5d ASFM or 4d ASSM) was not dependent on days after mating, rather on receipt of SP 225 from spermless males. Thus, SP from a second male can rescue the sperm release 226 defects of prior matings to males that lacked SP. 227

SP from a second male can bind to stored sperm from a previous male, while still 228
binding strongly to his own sperm 229 In the experiments described above SP was provided by a spermless second male, but in 230 nature females are much more likely to encounter a male who has his own sperm, 231 capable of binding his SP. To test whether SP from a male with sperm can still bind to 232 sperm from another male, we modified our experimental protocol such that females 233 were mated to SP-null males as described earlier, but rather than spermless males, we 234 now used ProtB-dsRed males [38] as the second male (Fig 4I. Cartoon). These second 235 males have a full suite of SFPs, sperm and their sperm-heads are labeled with ProtB-236 dsRed. This allowed us to distinguish between sperm received from SP-null males (blue 237 heads) and those received from ProtB-dsRed males (red heads). Females were frozen at 238 2hrs ASSM and sperm dissected from their seminal receptacles were probed for SP. We 239 observed anti-SP staining along the entire sperm (head and tail) from ProtB-dsRed males 240 ( Fig 4B). Sperm received from the SP-null males (blue heads) were also stained with 241 anti-SP along their length (head and tail; Fig 4B). Therefore, a control (wt) male with a 242 complete suite of SFPs and sperm of his own can also provide SP to bind to SP-deficient 243 sperm from another male. 244 The likelihood of finding an SP-null male in nature is very low. However, multiple-245 mating has been shown to deplete SFP reserves [39], so it is possible that inter-ejaculate Our immunofluorescence data showed no (or extremely weak) SP-sperm binding in 256 sperm dissected from the seminal receptacle of females mated to SFP-depleted males 257 ( Fig S2. C). Females mated to SFP-depleted CS males were then subsequently remated at 258 4d ASFM (long enough to have lost any SP signal from their first multiply-mated, mates) 259 to ProtB-dsRed males. Sperm were dissected from the seminal receptacles of these 260 females at 2hrs ASSM, and probed for SP (Fig 4II. Cartoon). We did not observe any 261 detectable SP signal on sperm stored in females singly-mated to SFP-depleted CS males 262 at 4d ASM ( Fig 4C). However, we observed anti-SP staining along the entire sperm (head 263 and tail) received by the doubly-mated female from the SFP-depleted CS male (blue 264 heads; Fig 4D) and ProtB-dsRed males (red+ blue heads; Fig 4D). 265 Thus, in a normal mating the amount of SP that a male transfers is sufficient to bind not 266 only his own sperm but also to remaining sperm from a rival. Moreover, SP from an 267 unmated control male can bind to previously stored sperm of a male that had his SFP 268 reserves depleted prior to mating with the female. 269 270 no LTR-SFPs are detectable on sperm or in female RT at 1d ASM ( Fig 5 & Fig S3). We 277 wondered whether LTR-SFPs were required from the second male in order to bind his 278 SP to the first male's sperm. 279

Sex peptide binding to sperm of a prior male does not require receipt of LTR-SFPs
We carried out experiments similar to those previously described, in which females 280 were first mated to SP-null males and then remated to spermless males at 1d ASFM. We 281 froze females at 2hr ASSM and immunostained stored sperm dissected from their 282 seminal receptacles, for the presence of LTR-SFPs that had been received from second 283  Thus sperm no longer detectably bind new LTR-SFPs after they have bound LTR-SFPs 311 from their own (SP-null) male. That LTR-SFPs are needed for SP-sperm binding, and that 312 SP from spermless male binds the first male's sperm, further suggests that the first 313 male's sperm (or the female RT) had already been primed with its own LTR-SFPs during 314 storage in the female tract. 315 Unlike the four LTR-SFPs assessed above, the two other LTR-SFPs, CG17575 and 316 seminase, do not bind to sperm, yet are crucial for SFP-sperm binding. In the absence of 317 CG17575 or seminase, SP fails to bind to sperm [10,40]. To determine if these proteins 318 were required for a second male's SP binding to a first male's sperm, we first crossed 319 females to SP-null males and then to CG17575-null or seminase-null males at 1d ASFM 320 (Fig 6. Cartoon). In this situation, CG17575 and seminase had entered the female with 321 the first male's sperm, but by the time of the second mating, were undetectable in the 322 female ( Fig S3). We examined whether in this situation SP transferred by CG17575-null 323 (or seminase-null) males would still bind to the SP-null sperm stored in the female. We 324 made use of ProtB-eGFP labelled SP-null males to differentiate between sperm received 325 from first (cyan (DAPI+ eGFP) sperm heads) and second (blue (DAPI) sperm heads) 326 males. Immunostaining and western blots for detection of SP on sperm dissected from 327 females mated to SP-null; ProtB-eGFP males and then remated to seminase-null (Fig 6. A males also showed no SP-sperm binding, as expected, due to lack of the LTR-SFP. 334 Therefore, sperm no longer require even CG17575 or seminase from the second male's 335 ejaculate, after they have received the LTR-SFPs from their own (SP-null) male. 336

Discussion 337
Ejaculate molecules, particularly the seminal fluid proteins (SFPs) that are received by 338 females during mating, play crucial roles in successful fertilization. In Drosophila they 339 induce striking changes in the physiology and behavior of females, instigating a wide 340 array of post mating responses [6,14,28,29,[40][41][42]. Some of these responses persist 341 long-term, due to binding of a male's SP to his sperm and gradual release of the SP's 342 active C-terminal region [19]. This important process is mediated by a cascade of "LTR-343 SFPs" that are needed to bind SP to sperm [10,20,21,40]. While all of the above can be 344 seen as facilitating reproductive success of the mating pair (particularly from the male's 345 perspective), SFPs also play roles in conflicts between males in species where females 346 are polyandrous. Den Boer et al [43] investigated sperm survival in monandrous and 347 polyandrous ants and bees. They observed that while seminal fluid enhanced the 348 survival of "self" sperm, it preferentially killed the sperm of rival males. In other words, 349 while SFPs worked in a cooperative interdependent way with "self" sperm, they harmed 350 rival sperm when in a situation of conflict (and cryptic female choice). Previous studies 351 have shown that males respond to threat of rivals by altering the allocation of both 352 sperm as well non-sperm components of their ejaculate (e.g for Drosophila: [11,25,26]). 353

Studies of SFP functions have tended to investigate how a male's SFPs can 354
promote the interests of his own (self) sperm. However, some data suggest that one 355 male's SFPs (ovulin, ACP36DE) can indirectly benefit a subsequent male within a 356 polyandrous female [27][28][29][30]44]. Here, we tested for direct effects of one male's SFPs on 357 another male's sperm and/or fertility. Specifically, we show that SP from a second male 358 can bind to and act with sperm received from a previous mating. Sperm stored in 359 females mated to SP-null males show no SP-sperm binding (as expected), but if these 360 mated females subsequently remate to a spermless male, his SP can bind to stored 361 sperm from the prior male. This binding of SP to the SP-null sperm restores his fertility 362 and proper sperm release dynamics. Even if a second male transfers sperm, he transfers 363 sufficient SP to bind to his own and rival sperm. Finally, our data suggest that the LTR-364 SFPs (that usually assist in binding of SP to sperm) are not required from the second 365 male for the association of his SP with sperm received from the first male (who had 366 already provided LTR-SFPs). The first male's sperm appear to be sufficiently "primed" 367 by prior receipt of their own LTR-SFPs to be able to bind SP from a second male. 368

SP from a second male can associate with a prior male's sperm that were stored 369 within the female 370
Xue and Noll [31] reported that sperm transferred to females by Prd mutant males (that 371 lack the entire suite of SFPs) were capable of fertilizing a few eggs to yield progeny, but 372 only after the females were subsequently remated to spermless males. They coined the 373 term "copulation complementation" to describe this phenomenon, and proposed that 374 SFPs from the second male might interact with the first male's sperm to yield this result. 375 Consistent with the idea of copulation complementation, several reports suggested that 376 first-males that lacked particular SFPs [6,8,[45][46][47] can have reproductive advantage in 377 terms of their higher paternity share in competitive situations; they were better 378 competitors, compared to controls, in defensive sperm competition assays [6,29,48]. A 379 simple explanation for these results, based on the findings that we report here, is that 380 the deficiency in the first male led to impaired release/use, and thus retention, of his 381 sperm, and this was complemented by receipt of the second male's SFPs, as we have 382 now shown here, for SP. 383 SP is the only SFP thus far known to persist within the Drosophila female (for 10-14 days 384 post-mating), eliciting long-term post mating responses through gradual release of its C-385 terminal portion [19]. The long-term persistence of SP on sperm made it an excellent 386 candidate to examine for interaction with rival sperm. Here, we report that SP 387 subsequently received from a spermless male binds to a first male's sperm (SP-null). 388 This association is apparent even if the second mating occurs at 1d or as long as 4d 389 ASFM, indicating that binding of SP to the first male's sperm occurs irrespective of how 390 long sperm have been in the storage organs. It remains unclear how SP received from 391 spermless (second) male enters the sperm storage organs, where sperm from the first 392 mating had been stored. However, Manier et al [38] reported that 60-90 min after the 393 start of a second mating, 26% of the resident sperm (received from the previous mating) 394 are moved from storage back into the bursa where they mix with the second male's 395 ejaculate before moving back into storage. Therefore, it is possible that SP received from 396 the spermless male binds to the first male's sperm that relocated to the bursa, and the 397 newly SP-bound sperm are then transferred back into storage in the seminal receptacle. 398

The binding of SP received from one male to sperm of another can restore defects 399 that resulted from lack of SP from the first male 400
In the absence of sperm, or if SP is not bound to sperm, females do not maintain post-401 mating responses and fail to efficiently release sperm from storage resulting in fewer 402 sperm available for fertilization and fewer progeny [6,10,17,35]. We observed that these 403 defects were restored when SP was received by females in a remating with spermless 404 males. Thus, the second male's SP bound to the first male's sperm is functional. The 405 rescue of the phenotype, however, was not as long lasting as in a normal single mating 406 with SP transfer, wearing off by 4d postmating rather than the normal ~10d. This could 407 be because only fewer sperm relocated from storage to the bursa [38], so they may not 408 carry sufficient SP back into storage to associate with SP-null sperm. Consistent with 409 this, the levels of SP that we see stained in these situations are lower than those in a wild 410 type mating. 411

An unmated male transfers sufficient SP to bind to his own as well rival sperm 412
We did not know whether the amount of SP that is transferred during mating is more 413 than the available binding sites on sperm. Here, we observed that an unmated control 414 male does transfer enough SP to bind his own as well as pre-stored sperm (SP-null) in a 415 previously mated female. Consistent with our findings, several reports suggest that in 416 response to potential threats of sperm competition and conflicts, males allocate the 417 levels of SFPs and transfer more SP, yet less ovulin, to previously mated females [11,26]. 418

Rubinstein et al [28] demonstrated that ovulin induces ovulation, acting through 419
octopamine (OA) neuronal signalling and increases the number of synapses that the 420 female's Tdc2 neurons make on the musculature of the oviduct. This latter effect 421 persisting could benefit rivals too, so males may thus be able to mitigate the levels of 422 ovulin in their ejaculate. But the question remains that if SP from one male's ejaculate 423 can bind to and assist another's sperm, why do males not lower the amount of SP 424 transferred while mating? A potential explanation is that a male would still benefit by 425 transferring enough SP to ensure that his own sperm remains saturated with SP, even at 426 a cost of part of his SP binding to another male's sperm. SFPs. We observed that sperm stored by subsequent (7 th ) females mated to these 441 multiply-mated males had undetectable SP signals. However, when these females were 442 remated to unmated control males, strong SP signals were detected on both the SFP-443 depleted sperm received from the previous mating and the newly received rival sperm. 444 Therefore, our results support the idea that in nature males who have multiply-mated 445 might get some help from the SFPs of subsequent, less depleted, males. Interestingly, 446 this inter-ejaculate interaction might confer an added advantage to the second male. 447 More of the second male's SP will be retained in the female reproductive tract, for even 448 longer, if it binds to previously-stored sperm in addition to his own sperm. This could 449 allow the post-mating responses in polyandrous females to be maintained for longer 450 than in singly-mated females. 451

Association of a second male's SP to sperm received from a prior male does not 452 require the receipt of LTR-SFPs from the second male 453
Binding of SP to sperm is facilitated by a network of LTR-SFPs [10]. Two LTR-SFPs, 454 CG17575 and seminase, do not themselves bind to sperm, whereas other LTR-SFPs bind 455 sperm transiently (CG1652, CG1656, CG9997, antares). CG17575 and seminase localize 456 the other LTR-SFPs, and SP, to sperm storage organs [10,20,40,42,49]. We found that SP 457 from a second male (spermless or control) can associate with sperm from the first male 458 (SP-null) even if it enters the female in absence of its own LTR-SFPs. This suggested that 459 SP-null sperm (or the mated female RT) had already received modifications ("priming") 460 from its own LTR-SFPs that were required for SP binding. This further suggests that 461 once primed, a sperm can bind SP from a rival's ejaculate without the need for 462 additional LTR-SFPs, and can restore its own post-mating dynamics. 463 Thus, we find that a critical SFP from one male can associate and offer direct benefits to 464 sperm from another male, restoring the SP function to the previously stored sperm. Our 465 work shows that SP is a crucial candidate for copulation complementation in Drosophila,466 and that sperm in storage (or the female RT) are primed for SP binding by the first 467 male's LTR-SFPs. Thus, despite potential competition between males, there could be 468 subtle cooperation between males as well. In addition, the allocation of resources by, 469 and effects on, rival males that mate to polyandrous females, should be viewed in light of 470 not only sexual conflicts, but also both direct and indirect effects of SFPs. 471

Fly strains 473
Spermless males, [sons of tudor, (sot) that lack sperm but produce and transfer a 474 served as controls. Sperm-heads of these control males were tagged with ProtB-eGFP, 484 but the males had normal levels of SP ( Fig S4). All flies were reared under a 12:12h light-485 dark cycle at 22±1°C on standard yeast-glucose medium. Mating experiments were 486 carried out by single-pair mating 3-5 day old virgin CS females to 3-5 day old unmated 487 males of genotypes indicated in the text and remating the same female 1 day or 4 days 488 after the start of first mating (ASFM) to age matched unmated males of the genotypes 489 indicated in the text. 490

Crossing scheme to study first male's sperm and rival's SP binding 491
Xue and Noll [31] reported copulation complementation in females mated to Prd males 492 (which produce sperm but lack SFPs) remated to spermless males (sot, which produce 493 SFPs). We followed a similar scheme but to focus on SP specifically, we used SP-null 494 males as the first male. As described in Results, we then remated these females to 495 spermless males, which make SFPs but not sperm. We attempted to do the reciprocal 496 experiment, where females were mated to spermless males and then remated to SP-null 497 males, but consistent with what was reported by Xue and Noll [31], we could not detect 498 copulation complementation in this direction for technical reasons: SP from the 499 spermless male did not persist long enough in the mated female to interact with the 500 second male's sperm (see Results). We carried out rematings at three time points, 3-6 501 hrs, 1d, and 4d AFSM. We assessed results at 2hr after the start of the second mating 502 (ASSM). 503

Fertility 504
The reproductive performances of singly-mated or doubly-mated females were assayed 505 by analyzing fertility (numbers of progeny eclosed over ten days) [32]. Briefly, the 506 fertility assays were carried out with (A). "Single matings": Females were singly mated 507 to (i) spermless males, (ii) SP-null males, or their (iii) TM3 siblings (genetically-matched 508 control males) in three individual sub-batches, and (B). "Rematings": Females were 509 mated to SP-null males or their TM3 siblings (SP + ) and were then subsequently remated 510 to spermless males at 1d and 4d ASFM. Matings that lasted 15 mins or more were 511 considered successful. At the end of a mating, males were removed from the vials and 512 females were allowed to lay eggs for 10 days after the start of mating (ASM) in the first 513 batch and after the start of second mating (ASSM) in the second batch. Females were 514 transferred to fresh food vials every three days. Flies emerging from each vial were 515 counted. Fertility is represented as total progeny number produced by each female over 516 a period of 10 days. The differences in fertility were analyzed through One way Analysis 517 of Variance (ANOVA) followed by Tukey's multiple comparison tests for single-matings 518 and Mann Whitney U-tests for rematings. All assays were repeated more than two times 519 and comprised of two technical replicates, with each group consisting of a minimum 520 sample size of 15-20. 521

Receptivity 522
To determine the propensity of females to remate, receptivity assays [17] were set for 523 females singly mated to SP-null, spermless or CS males and females mated to SP-null 524 males and then subsequently remated to spermless males at 1d ASFM. For the assay, 525 females from singly-mated and doubly-mated groups were then provided with (CS) 526 males at 1d and 4d ASM or ASSM, respectively. We determined the number of females 527 that mated within 1hr from when the CS male was introduced within the vial. Assays 528 were repeated more than two times, with each group consisting of a minimum sample 529 size of 15-20. The data were analyzed by Fisher exact tests and Chi-squared group 530 analyses. 531

Sperm utilization/ release from sperm storage organs in females 532
To study the effect of first male's sperm and rival male's SP binding on sperm utilization 533 and release, we generated SP-null males whose sperm-heads are labelled with ProtB-534 eGFP [38]. Females were mated to SP-null; ProtB-eGFP or SP + ; ProtB-eGFP (control) 535 males. Some of the mated females were frozen at 4d ASM (or 5d ASM) for sperm counts. 536 The remaining mates of SP-null; ProtB-eGFP males were remated to spermless males at 537 1d ASFM. These flies were flash-frozen at 4d ASSM. Subsequently, seminal receptacles of 538 females singly-mated to SP-null; ProtB-eGFP and SP + ; ProtB-eGFP, or doubly-mated to SP-539 null; ProtB-eGFP and spermless males, were dissected and eGFP sperm were counted (at 540 a total magnification of 200X, with FITC filter on an Echo-Revolve microscope). Mature 541 sperm in the seminal receptacles of mated females were counted twice and groups were 542 blinded to ensure reproducibility and avoid bias. The percent accuracy was 90-94%. 543 Assays were repeated more than two times, with two technical replicates. Every group 544 contained a minimum sample size of 15-25. Differences in the sperm counts between 545 groups were analyzed statistically through One way ANOVA followed by Tukey's 546 multiple comparison tests. 547

Brood matings 548
Control (CS) males were subjected to brood matings [51,52] to deplete SFPs, as their 549 levels are known to become exhausted at a higher rate than sperm numbers [39]. 550 Briefly, three day old control males were mated to CS females in two broods (each 551 consisting of three virgin females) over two days. The first mating of both broods was 552 observed. On the third day, previously mated females were removed and the male was 553 provided with an additional virgin female (7 th mate), matings were observed and 554 depleted CS males were removed. Half of the 7 th mated females were frozen at 4d ASM, 555 while the others were subsequently remated to control (ProtB-dsRed) males at 4d ASFM, 556 and then frozen at 2hr ASSM. Sperm stored in the seminal receptacle of the frozen flies 557 were dissected and immunostained for SP. 558

Sample preparation and Western blotting 575
To further examine transfer, persistence or binding of SP to sperm stored in singly-576 mated or doubly-mated females, the lower reproductive tract (RT) or sperm stored (SS) 577 in seminal receptacles of mated female were dissected. The dissected tissues (lower RT, 578 n=5-10 or sperm, n =20-30) were suspended in 5µl of homogenization buffer (5% 1M 579 Tris; pH 6.8, 2% 0.5M EDTA) and processed further according to the protocol of Ravi

Conflict of interest statement 596
The authors declare no conflict of interests.  sperm dissected from SR of 20 females mated to wild type (CS) males at 2hr ASM, 4: SP-718 null @ 2hr, sperm dissected from SR of 20 females mated to SP-null males at 2hr ASM, 5: 719 SP-null @1d, sperm dissected from SR of 20 females mated to SP-null males at 1d ASM, 720 6: SOT@35', reproductive tract of 3 females mated to spermless males at 35'ASM 721 (positive control), 7: SP-null, SOT @ 2hr, sperm dissected from SR of 20 females mated 722 to SP-null males and then remated to spermless males at 1d ASFM, and frozen at 2hr 723 ASSM. Lanes were probed for LTR-SFPs CG9997, CG1656, antares and CG1652 and SP as 724 described in the text. Actin served as loading control. Sperm dissected from females mated to unmated males, frozen at 2hrs ASM. (C) Sperm 768 dissected from females mated to multiply mated males, frozen at 2hrs ASM. Sperm 769 heads were stained with DAPI (blue) and presence of SP (green) detected with Alexa 770 fluor 488 (n=5; Bar=20µm). 771 Bibliography 792