Temperate zone birds are highly seasonal in many aspects of their physiology. In mammals, but not in birds, thepineal gland is an important component regulating seasonal patterns of primary gonadal functions. Pineal mela-tonin in birds instead affects seasonal changes in brain song control structures, suggesting the pineal gland reg-ulates seasonal song behavior. The present study tests the hypothesis that the pineal gland transducesphotoperiodic information to the control of seasonal song behavior to synchronize this important behavior tothe appropriate phenology. House sparrows, Passer domesticus, expressed a rich array of vocalizations rangingfrom calls to multisyllabic songs and motifs of songs that varied under a regimen of different photoperiodic con-ditions that were simulated at different times of year. Control (SHAM) birds exhibited increases in song behaviorwhen they were experimentally transferred from short days, simulating winter, to equinoctial and long days,simulating summer, and decreased vocalization when they were transferred back to short days. When main-tained in long days for longer periods, the birds became reproductively photorefractory as measured by theyellowing of the birds' bills; however, song behavior persisted in the SHAMbirds, suggesting a dissociation of re-production from the song functions. Pinealectomized (PINX) birds expressed larger, more rapid increases in dailyvocal rate and song repertoire size than did the SHAM birds during the long summer days. These increases grad-ually declined upon the extension of the long days and did not respond to the transfer to short days as was ob-served in the SHAM birds, suggesting that the pineal gland conveys photoperiodic information to the vocalcontrol system, which in turn regulates song behavior.
In temperate zone animals, annual cycles of many processes, includ-ing reproduction, metabolism, immune function, migration and song,are temporally controlled by an endogenous biological timekeeping sys-tem or clock (Dawson et al., 2001; Goldman, 2001; Pittendrigh, 1993).This biological clock interprets environmental time—such as thatmarked by annual changes in photoperiod, temperature and other fac-tors—and activates or terminates these processes at the appropriatetime of the year. In seasonally breeding mammals, the pineal glandplays an important role in photoperiodic regulation of annual gonadalcycles (Hoffman and Reiter, 1965, 1966; Reiter, 1973a,b). For example,pinealectomy (PINX) prevents gonadal regression in response to shortwinter-like photoperiods in both male and female hamsters (Hoffmanand Reiter, 1966; Reiter, 1973a), and PINX hamsters remain reproduc-tively competent in short-day environments (Reiter, 1973b). The dailyduration of nocturnal melatonin released by the pineal gland reflectsthe photoperiod, such that melatonin duration is long during the longnights and short days of winter, whereas melatonin duration is short
ne).
during the summer. This change in melatonin duration mediates thephotoperiodic response of the hypothalamo-pituitary-gonadal (HPG)axis, which in turn regulates seasonal reproduction (Barrett et al.,2003; Dupre et al., 2008; Goldman, 2001; Lincoln, 2006) as well asother processes, such as pelage coloration and metabolism (Bartnesset al., 1993).
The pineal gland is important for the generation and regulation ofcircadian rhythms in oscine passerine birds (Cassone, 1990; Gwinnerand Hau, 2000). Although PINX house sparrows, Passer domesticus, en-train to light:dark (L:D) cycles, the surgery eliminates the expressionof free-running circadian patterns of locomotor behavior, song, bodytemperature and brain metabolism in house sparrows and severalother species of passerine birds when they are maintained in constantenvironmental lighting such as constant darkness (D:D) or constantdim light (Binkley et al., 1971; Gaston and Menaker, 1968; Lu andCassone, 1993a; Wang et al., 2012). Transplantation of a pineal glandinto arrhythmic, PINX house sparrows confers both rhythmicity andtime of day information (Zimmerman and Menaker, 1979). These ef-fects are likely due to the circadian secretion of melatonin by the pinealgland, which, in contrast to galliform and columbiform birds, producesall of the circulating hormone in house sparrows (Janik et al., 1992).The pineal glands of all birds studied to date express circadian patterns
373G. Wang et al. / Hormones and Behavior 65 (2014) 372–379
of biosynthesis and secretion of melatonin in D:D that entrain to light:dark cycles (L:D) such that melatonin is released during the night(Brandstätter et al., 2000; Takahashi et al., 1980), and that rhythmic ad-ministration of the melatonin entrains locomotor rhythms of arrhyth-mic PINX birds (Cassone et al., 2008; Heigl and Gwinner, 1995; Lu andCassone, 1993b; Wang et al., 2012).
In contrast to mammals, photoperiodic regulation of seasonal go-nadal activity in birds is largely independent of the pineal melatoninsystem insofar as PINX has little effect on gonad size or activity in re-sponse to changes in photoperiod (Bentley, 2001; Dawson et al., 2001;Wilson, 1991). Nonetheless, the house sparrow pineal gland, as in sea-sonally reproducing mammals such as hamsters, expresses seasonalchanges in melatonin duration in vivo and in vitro based upon the pho-toperiod in which birds were raised (Brandstätter et al., 2000). Thus,whereas the avian pineal gland is important for circadian rhythms andwhereas circadian rhythms are important for seasonal reproduction,the pineal gland is not directly involved in the regulation of the HPGaxis (Bentley, 2001; Pant and Chandola-Saklani, 1992; Wilson, 1991).
As with the gonads (Dawson et al., 2001), the size and complexity ofthe song control system within the brains of male oscine passeriformbirds change depending on the photoperiod (Ball et al., 2004;Brenowitz, 1997; Whitfield-Rucker and Cassone, 2000). The song con-trol nucleus “HVC” (formerly “high vocal center”), the robust nucleusof the archipallium (RA), Area X and the lateral magnocellular nucleusof the nidopallium (lMAN) are small during the short days of winter(when the bird is said to be “photosensitive”) and grow 10–30% in vol-ume as the photoperiod increases in the spring and summer (i.e., thebird becomes “photostimulated”) (Brenowitz, 1997; Whitfield-Ruckerand Cassone, 2000). If birds are experimentally maintained in long pho-toperiods for longer periods of time (10–20 weeks, depending on thespecies and photoperiod; Dawson et al., 2001), their gonads and songcontrol systemsbecome insensitive to the previously stimulatory effectsof long photoperiod and spontaneously regress, a life history stageknown as “photorefractory”. Secretion of gonadal and neural steroidsis important for changes in song control structures (Brenowitz, 1997;Tramontin et al., 2003), although they are not the only processes con-trolling song; castration has little effect on the photostimulatory effectsof long photoperiods on song control structures (Ball et al., 2004;Bernard et al., 1997; Whitfield-Rucker and Cassone, 2000).
One candidate molecule contributing to non-gonadal song controlregulation is pineal melatonin. The distribution of melatonin bindingsites and melatonin receptors in the avian brain suggests that the pinealhormonemay regulate the song control systemdirectly (Gahr andKosar,1996; Whitfield-Rucker and Cassone, 1996). In vitro binding of the mel-atonin agonist 2[125I]-iodomelatonin (IMEL) and autoradiography in themale house sparrow, the zebra finch, Taeniopygia guttata, and theEuropean starling, Sturnus vulgaris, reveal high-affinity IMEL binding inbrain structures associated with the song control system, includingArea X, HVC, lMAN and RA (Bentley and Ball, 2000; Gahr and Kosar,1996; Whitfield-Rucker and Cassone, 1996). In house sparrows, atleast, IMEL binding in the song control nuclei is not affected by castration(Whitfield-Rucker and Cassone, 1996). Although all three melatonin-receptor sub-types, Mel1A, Mel1B, and Mel1C, are expressed in the songcontrol system, Mel1B is the predominant receptor sub-type in the songcontrol system (Bentley et al., 2013; Jansen et al., 2005).
Melatonin affects the size and complexity of brain song control nu-clei (Bentley et al., 1999; Cassone et al., 2008). Continuous melatoninadministration to European starlings, which abolishes seasonal changesinmelatonin duration, also decreases the amplitude of seasonal changesin the volume of the song control nuclei (Bentley et al., 1999). Rhythmicmelatonin administrationwith long durations, as exemplified bywinterconditions, entrains locomotor rhythms of house sparrows maintainedin constant light (L:L) and decreases the sizes of HVC and RA towinter-like volumes (Cassone et al., 2008).
To examine the pineal gland's role in seasonal changes of song be-havior, we investigated the effect of PINX on song behavior under
changing photoperiodic conditions in male house sparrows. Season-al changes in song rates in wild populations have been observed inthis species, with rate peaks occurring between late March andearly August and nadirs around November (Hegner and Wingfield,1986; Lowther and Cink, 2006). The present study investigateswhether PINX affects photoperiodic changes in song behavior of sin-gly housed male house sparrows. We hypothesize that the pinealgland, through the secretion of melatonin—whichmirrors the photo-period in duration—transduces such photoperiodic information toaccurately synchronize the expression of avian vocal behavior to anappropriate time of year.
Materials and methods
Animals and data acquisition
Adult male house sparrows, P. domesticus (N = 22), were capturedin August, 2012, around Lexington, KY, and were group housed withfood (2:1 millet to chick starter) and water ad libitum in an aviary ex-posed to a natural photoperiod at the Ecological Research Facility, Uni-versity of Kentucky in Lexington. In Winter 2012, birds at theirphotosensitive stage were transferred into individual cages in isolationcabinets with food and water ad libitum under L:D 6:18 (light at40 μW/cm2, lights on at 6 AM and lights off at 12 PM) conditions withconstant background white noise (average power = 21 dB, frequencyrange from 0 to 10 kHz, measured within the cabinets) for acoustic iso-lation before the onset of the experiments. In each cabinet, both loco-motor activity and vocalizations were continuously monitored andrecorded. Locomotor activitywas recordedwith an infraredmotion sen-sor located in the inside of the cabinet. Activity data were recorded byVitalView data acquisition equipment and software (Mini-Mitter Co.,Sunriver, OR), and activity records were subsequently single plotted inactogram format or exported as ASCII files for further analysis usingActiView software (Fig. 1; Mini-Mitter Co.). Vocalization data from 22individual audio channels were simultaneously acquired over 2 daysevery twoweekswith 11 RodeNT3 and 11 AKGPerception 170 cardioidcondenser analog microphones in each chamber and then integratedusing three MOTU 8 pre audio interface recording units (MOTU, Inc.,Cambridge, MA).
Surgeries and experimental protocol
After acclimation to the isolation cabinets for 3 weeks, birds weresubjected to either pinealectomy (PINX; n = 11) or a sham surgery(SHAM; n = 11), as previously described (Lu and Cassone, 1993a).Each bird was anesthetized with 90 mg/kg ketamine and 10 mg/kgxylazine, and the headwas secured in a Stoelting stereotaxic apparatus.A scalpel was used to make a longitudinal incision in the scalp, and theexposed skullcap was removed using a dental drill and forceps. For thePINX birds, the dura mater was then incised and retracted, and the ex-posed pineal glandwas grasped and removed. The skullcapwas then re-placed, and the scalpwas sutured shut. For the SHAMbirds, the skullcapwas immediately replaced after its removal and the dura materwas leftintact. Birds were allowed to recover in the cabinets.
Two weeks after the surgery (Fig. 1), birds were transferred intoequinoctial L:D 12:12 conditions for 3 weeks and then transferred intolong-day (L; L:D 18:6) conditions for 4 weeks (Fig. 1) before half ofthe birds (S–L–S group; n = 6 for PINX; n = 5 for SHAM) were trans-ferred back to short day (S; L:D 6:18) photoperiod conditions. The re-maining birds (S–L group; n = 6 for PINX; n = 5 for SHAM) werekept under L:D 18:6 conditions for the next 14 weeks (Fig. 1). Birdswere sacrificed at the end of the experiment. Brains were removedand sectioned to validate the success of surgery histologically, and testisvolume was calculated to confirm the reproductive stage.
Fig. 1. The single-plotted actograms of the SHAM and PINX birds under the photoperiodic manipulations of either the S–L or S–L–S condition. The different photoperiodic conditions areindicated to the left. Light or dark periods are indicated on the top inwhite (light), black (dark) and gray (light or dark, depending on the photoperiodic conditions). The time of lights-onand lights-off are indicated, respectively, by upward and downward arrows on the first cycle (marked by the points of the arrows) of each photoperiodic condition. The “X” on eachactogram indicates the time of surgery for the corresponding bird. Abbreviations: SHAM, sham-operated; PINX, pinealectomized; S–L, short-day to long-day photoperiodic manipulation;S–L–S, short-day to long-day and back to short-day photoperiodic manipulation.
374 G. Wang et al. / Hormones and Behavior 65 (2014) 372–379
Sound analysis
Vocalizationswere analyzed using Raven PRO 1.4 (Cornell Lab of Or-nithology, Ithaca, NY). The sampling frequency was 20 kHz, and thesound files were saved in 16-bit wave format with 30-second bins.Both songs and calls were extracted from sound files with the “bandlimited energy detector” function of Raven Pro 1.4, as previously de-scribed (Wang et al., 2012). According to the acoustic characteristicsof both songs and calls from house sparrows, we established target sig-nal parameters (including the frequency range, duration range, mini-mum separation time, signal-to-noise ratio, etc.) and automaticallyextracted songs and calls with time-stamps and quantified their acous-tic properties (duration, peak frequency, entropy, etc.). Sonograms ofthe extracted songs and callswere generated and screened for accuracy;the automated extractions were successful at greater than 95% com-pared with manual methods.
Due to similarities of sound characteristics (duration, peak frequen-cy, frequency modulation, entropy, etc.) between songs and calls inhouse sparrows, we did not differentiate between songs and callswhen investigating the daily patterns of vocalizations and vocal ratesunder different photoperiodic conditions. Because the absolute num-bers of vocalizations varied among the birds, the average daily vocaliza-tion rate (vocalizations/minute) was calculated for each bird for 2consecutive days during the post-surgical period in L:D 6:18. Then,every 2 weeks thereafter, recordingsweremade continuously for 2 con-secutive days, and the average vocalizationwas divided by the initial re-cordings to achieve a ratio for each bird, which was designated the“normalized vocalization rate”.
When examining the repertoire size under these photoperiodic con-ditions, we produced sonograms for all extracted vocal signals for each
individual bird, and these sonograms were visually examined by twoseparate investigators blind to the specific photoperiodic condition orexperiment group. Then, repertoire sizes were calculated separatelyfor calls, double-syllable (DS) songs, multiple-syllable (MS) songs andmotifs according to the following criteria: (1) calls were 1-syllablesounds with simple frequency modulation, whereas (2) songs weremulti-syllable sounds commonly having 2 to 4 syllables and complexfrequency modulations, and (3) complex song motifs were series ofsongs and calls temporally arranged in specific sequences and producedby individual birds at least 3 times under certain photoperiodic condi-tions. Finally, the repertoire sizes for calls, DS songs, MS songs, and mo-tifs were calculated by two investigators, averaged, and then subjectedto further statistical analyses.
Rhythm analysis
Locomotor activity data were recorded continuously throughout theexperiment using a Mini-Mitter VitalView data acquisition and controlsystem (Bend, OR). The data were graphically represented in actogramformat as described in Wang et al. (2012) (Fig. 1). The daily distribu-tions of locomotor activity and vocalization data for each bird were ar-ranged in hourly ratios of total daily occurrences for all observationperiods (Fig. 4). Under different photoperiodic conditions, the hourly ra-tios of occurrences for behaviors from representative 2-day (48 h) cy-cles were also subjected to cosinor analysis by CircWave 1.4 timeseries analysis software (R.A. Hut, Department of Chronobiology, Uni-versity of Groningen, Netherlands) (Wang et al., 2012). CircWave is anextension of cosinor analysis designed for estimation of circadian com-ponents of time-series. It produces a Fourier curve that describes thedata better than a simple linear square cosinor curve by adding as
375G. Wang et al. / Hormones and Behavior 65 (2014) 372–379
many harmonics to the wave fit as the data allow. When a time-seriespassed the cosinor analysis (P≤ 0.05), the data were considered rhyth-mic, and a cosinor–wave fit curve was provided. Otherwise, if a time-series did not pass the analysis (P N 0.05), the data were considered ar-rhythmic, and a line with a slope of zero was provided.
Measurement of bill darkness
As a proximate indicator for circulating testosterone levels in malehouse sparrows in which bill darkness is positively correlated with cir-culating testosterone levels (Laucht et al., 2010;Murton andWestwood,1977), changes in bill darkness were traced under the different photo-periodic conditions. During each observation period, birds wereremoved from their isolation chamber during the light phase, andtheir billswere photographed against awhite backgroundunder consis-tent lighting conditions. Then, the color photographs of bills were con-verted to grayscale images. By using Adobe Photoshop CS 6 software,the bill region and white background region of each photograph wereselected and the values of average darkness were measured. The billdarkness for each bird was represented by the ratio of the bill darknessrelative to the background darkness.
Analytical statistics
Bill darkness, locomotor activity rate, vocalization rate, and reper-toire size changes of call, DS song, MS song, and motifs were plottedusing representative 2-day average values during each observation pe-riod. Daily locomotor activity and vocalization of each bird were nor-malized to their respective initial values at L:D 6:18 after surgery toaccount for individual differences in the endogenousmotivations for lo-comotion and vocalization. From the initial L:D 6:18 conditions to4 weeks under the L:D 18:6 conditions, data were analyzed for the
Fig. 2. Representative sonograms of calls, double-syllable songs (DS), multiple-syllable songs (Mtoire sizes of the corresponding vocal types when considering all house sparrows used in this e
PINX and SHAM groups. From 4 weeks under L:D 18:6 to the end ofthe experiment, the datawere separated for the PINX and SHAMgroupsinto each specific photoperiodic subgroup (S–L versus S–L–S). However,the statistical analyses were performed separately, with correspondingsample sizes for these two photoperiodic manipulation periods.
Data were first tested for normality with a Kolmogorov–Smirnovtest before being subjected to the appropriate statistical analyses. Billdarkness, daily locomotor activity rate, daily vocal rate and repertoiresize change in call, DS song,MS song, andmotif datawere first subjectedto matched 2-way ANOVA with Bonferroni post-hoc tests to examinethe effects of surgery and time period. Differences between the SHAMand PINX groups were also tested at each time period separately—from the initial L:D 6:18 conditions to 4 weeks after L:D 18:6 conditionsand from 4 weeks after L:D 18:6 conditions to the end of the experi-ment. In the latter time interval, separate 2-way ANOVA tests were per-formed for the S–L and the S–L–S groups. When there was a significanteffect on time period, matched 1-way ANOVA followed by a Bonferronipost-hoc test was used within the SHAM and PINX groups. All data arepresented as the means ± SEMs, except for the daily pattern data.Data were considered significant when P≤ 0.05. All statistics were per-formed using GraphPad Prism (GraphPad Software, San Diego, CA).
Results
The song repertoire of the house sparrow
As a “close-ended learner”, the house sparrow is commonly consid-ered to have rather simple call and song repertoire, both of which arecomposed of single or a series of “cheep” or “chirrup” notes (Lowtherand Cink, 2006). However, systematic analyses of vocalizations inmale house sparrows reveal a larger and more complex repertoire inboth call and song (Nivison, 1978), which this study corroborates. In
S) andmotifs in house sparrows. The number at the top right indicates the overall reper-xperiment.
376 G. Wang et al. / Hormones and Behavior 65 (2014) 372–379
this study, individual house sparrowswere acoustically and visually iso-lated. We identified complex call and song behavior in our captive malehouse sparrows. The house sparrow's call, expressed by both males andfemales, is commonly a single-syllable sound with simple frequencymodulation (Fig. 2). The male house sparrow's song comprises a seriesof multi-syllabic, high-pitched sounds that are repeated multipletimes in a bout of singing. The type of song can be divided intodouble-syllable (DS) songs and multiple-syllable (MS) (commonly 3to 4 syllables) songs (Fig. 2). Overall, we detected 22 types of call sylla-bles, 72 types of DS songs and 23 types of MS songs in the birds used inthis study. In addition,male house sparrows sometimes expressed com-plex songmotifs, and thesemotif songswere composed ofmultiple callsand song syllables (Fig. 2). Further work in the presence of potentialmates and in more natural conditions is underway.
Effects of PINX on photoperiodic changes in vocalizations
Significant seasonal variationwas observed in the normalized vocal-ization rate (including both songs and calls) of the SHAM birds (Fig. 3).The SHAM birds were more vocal once the photoperiod increased fromshort days (S) to equinoctial conditions, and the high daily vocalizationrate wasmaintained after the photoperiod increased to long days (L) inS–L group (Fig. 3A left). Vocalization rates in the SHAM birds decreasedimmediately after the photoperiodwas switched back to short days, andthe low rate in S–L–S group was maintained (Fig. 3A right). The PINXbirds exhibited increases in their daily vocalization rates more rapidly
A
B
Fig. 3. The changes in normalized vocalization rate and DS song repertoire size under differentpost-surgery LD 6:18 conditions, and changes of DS song repertoire (B) from initial post-surgercircles) birds under either S–L or S–L–S photoperiodicmanipulation. For all figures, significant chby different letters (for the SHAMbirds) or letters with apostrophes (for the PINX birds). Differedate of sampling below, on the x-axis. Data are represented by means + SEMs. Abbreviations:back to short-day photoperiodic manipulation. S = short days, L:D 6:18; E = equinoctial phot
than did the SHAM birds after the photoperiod increased from shortto equinoctial conditions; in fact, the PINX birds actually increased invocalization following transfer from long to short days (Fig. 3A right).The high vocalization rate gradually decreased to the initial low levels,irrespective of the birds' placement in the S–L or S–L–S group (Fig. 3Aright).
Effects of PINX on photoperiodic changes in song repertoire
To test the effect of PINX on the seasonal changes of vocal repertoire,we more closely scrutinized the DS song category (Fig. 2), which wasthemost variable of the vocalizations. In the SHAM birds in the S–L pro-tocol, the number of different DS vocalizations was relatively steadywhen birds were transferred to L:D 12:12, and these numbers persistedat high levels when these birds transferred to long days (Fig. 3B left). In-terestingly, a peak in DS songs occurred 14 weeks following transfer tolong days. In the SHAM birds in the S–L–S protocol, the number of DSvocalizations persisted in response to increased photoperiod (Fig. 3Bright) but decreased when birds were transferred to short days.
The number of DS songs increased rapidly in the PINX birds in the S–L protocol for 3 weeks under long-day conditions, gradually decliningafter 12 weeks under the L:D 18:6 conditions (Fig. 3B left). In the S–L–S protocol, the PINX birds significantly increased their numbers of DSsongs relative to the SHAM birds (Fig. 3B right). However, in contrastto the SHAMbirds, the PINX birds failed to respond to transfer to shorterdays (L:D 8:16) and continued singing at a high rate. In comparison to
photoperiodic and surgical manipulations. The vocalization rate (A), normalized to initialy LD 6:18 conditions were plotted for both the SHAM (filled circles) and the PINX (unfilledanges of values (compared to the preceding observation) along the time axis are indicatednt photoperiodic conditions are indicated at the top of the figures, with the actual calendarS–L, short-day to long-day photoperiodic manipulation; S–L–S, short-day to long-day andoperiods, L:D 12:12; L = long days, L:D 18:6.
Fig. 4. Daily distributions of locomotor activity and vocalization under different photope-riodic conditions. The daily patterns of hourly occurrences in locomotor activity (upperpanels, white background) and vocalization (lower panels, gray background) are plottedfor both the SHAM (filled circles) and the PINX (unfilled circles) birds for the initial pre-surgery period (A and B), post-surgery period (C and D), and equinoctial period (E andF). The photoperiodic conditions are indicated below each pair. The asterisk indicates asignificant effect of surgery at the corresponding time as found by matched 2-wayANOVA followed by Bonferroni post-hoc test on the original hourly percentages of dailybehavioral occurrences.
377G. Wang et al. / Hormones and Behavior 65 (2014) 372–379
DS song repertoire size, the repertoire sizes of calls andmotifs were alsoseasonally regulated but to a lesser extent; here, PINX did not have anobvious effect.
Daily distributions of vocalization and locomotor activity
Photoperiodic differences in the daily distributions of vocalizationand locomotor activity were observed (Fig. 4; Supplemental Fig. 1). Inshort days before surgery, all birds exhibited a trimodal distributionwith one bout of vocalization occurring before lights-on (dawn), onebout at dawn and a third bout at mid-day (Fig. 4A). Following SHAMsurgery, this trimodal pattern of vocalization persisted, but PINX result-ed in a loss of the anticipatory predawn bout of vocalization (Fig. 4C).Transfer to L:D 12:12 resulted in a transition to a unimodal pattern ofvocalization in all birds (Fig. 4E). This pattern persisted in the PINXbirds throughout both the S–L and S–L–S protocols, whereas theSHAM birds returned to multimodal patterns of vocalization (Supple-mental Fig. 1).
In short days, the distribution of locomotor activity exhibited aunimodal pattern with peak activity occurring at dawn in both theSHAM and PINX birds (Figs. 1, 4B, D). When the birds were transferredto L:D 12:12 and then to L:D 16:8, both the SHAM and PINX birds exhib-ited bimodal patterns of activity with a major peak in activity at dawnand a smaller peak at dusk (Fig. 1, 4F). This pattern persisted for allbirds for the duration of the S–L protocol. When the SHAM and PINXbirds were transferred to short days in the S–L–S protocol, the distribu-tion of locomotor activity coalesced to a single peak at dawn but retaineda compressed bimodal pattern over the subsequent 14 weeks. Some sta-tistically significant differences in the amplitudes of these peakswere de-tected, but these were small (Supplemental Fig. 1).
Effects of pinealectomy on photoperiodic changes in bill darkness and testisvolume
When the photoperiod increased from L:D 6:18 to L:D 12:12 and, fi-nally, to L:D 18:6, a significant increase in bill darkness was observedfrom L:D 12:12 to L:D 18:6 in all birds, and there was no difference be-tween the SHAM and PINX groups with regard to changes of bill dark-ness (Fig. 5B), a known proxy for circulating testosterone, with a directrelationship between dark bills and high testosterone (Laucht et al.,2010; Murton and Westwood, 1977). Irrespective of whether the birdswere maintained in L:D 18:6 (S–L) or were placed in the transferredback to short days after 5 weeks under L:D 18:6 conditions (S–L–S),bill darkness gradually decreased in both the SHAM and PINX groups(Fig. 5B), but the decreases were more rapid and significant in the laterstage in the PINX birds than in the SHAM birds (Fig. 5B), which is consis-tent with their testis volumes at the end of the experiment (Fig. 5C).Overall, PINX did not affect bill darkness during the period of photoperi-odic increase but exacerbated changes in bill darkness in the last stage ofboth the S–L and S–L–S groups.
Discussion
In oscine passerine birds, the photoperiodic time-measurement sys-tem controlling seasonal aspects of primary reproductive function, suchas gonadal regression and recrudescence, has in part been functionallyseparated from the photoperiodic control of complex secondary sexualcharacteristics, such as song (Bentley, 2001; Cassone et al., 2009). Thereis certainly a significant amount of overlap, however. For example, songcontrol nuclei express androgen and estrogen receptors (Ball et al.,2004), and gonadal and neural steroids influence the development ofsong control circuits (Brenowitz, 1997). Testosterone and other andro-gens are necessary for the crystallization of learned song, and circulatingsteroids modulate the seasonal acquisition of song behavior (Ball et al.,2004; Hall and MacDougall-Shackleton, 2012; Strand and Deviche,2007).
However, there is also evidence for non-steroidal or, at least, non-gonadal regulation of the song control system on a seasonal basis, andthese are associated with pineal melatonin (Cassone et al., 2009). First,castration of male American tree sparrows, Spizella arborea, and ofmale house sparrows does not completely block the seasonal recrudes-cence of song control nuclei (Bernard et al., 1997;Whitfield-Rucker andCassone, 2000). Second, continuous melatonin administration bluntsthe annual changes in song control nucleus volume in European
8
7
6
5
4
3
2
1
Bill
Dar
knes
s
Date6/1 7/1 8/1 9/1 10/1 11/1 12/1
Date6/1 7/1 8/1 9/1 10/1 11/1 12/1
100
80
60
40
20
0SHAM PINX SHAM PINX
Est
. Tes
tis
Vol
ume
(mm
3)
C
B
A
Fig. 5. Changes in bill darkness and testis size under different photoperiodic conditions. Representative images and corresponding quantifiedmeasures of darkness in bill color are shown(A). Changes in bill darkness are plotted for both the SHAM(filled circles) and the PINX (unfilled circles) birds housed in S–L (B, left panel) and S–L–S (B, right panel) conditions. Significantchanges in values (compared to the preceding observation) along the time axis are indicated by different letters (for the SHAM birds) or letters with apostrophes (for the PINX birds).
378 G. Wang et al. / Hormones and Behavior 65 (2014) 372–379
starlings without affecting gonad size (Bentley et al., 1999). Third,rhythmic administration of long durations of melatonin to house spar-rows held in L:L decreases HVC and RA volume with virtually no effecton testicular volume (Cassone et al., 2008).
Conversely, PINX has little effect on photoperiodic stimulation of theHPG-axis and gonadal recrudescence (Dawson et al., 2001). PINXhas noeffect on changes in gonadal size in tree sparrows (Wilson, 1991), andexperimentalmanipulation in the duration of melatonin administrationhas no effect of testis size (Cassone et al., 2008). One caveat to this no-tion is that pineal melatoninmay regulate sex steroid secretion throughthe action of gonadotropin inhibitory hormone (GnIH) (McGuire et al.,2011). Melatonin increases GnIH expression in and secretion from thetuberal hypothalamus, which, in turn, causes gonadal regression(Ubuka et al., 2005). This inhibition is believed to have a role in fine-tuning reproductive processes late in the breeding season by acting onboth the hypothalamus and gonads (McGuire et al., 2011; Ubuka et al.,2005).
To differentiate the gonadal-dependent and gonadal-independentroles of the pineal gland, we measured bill darkness, which has beenshown to be an indicator for circulating testosterone in male housesparrows (Laucht et al., 2010; Murton and Westwood, 1977). The dataindicate that bills of PINX birds became lighter than those of theSHAM birds more rapidly as they regressed toward shorter days or be-came photorefractory in extended long days (Fig. 5B). This observationwas corroborated by testis volume measurements following the study(Fig. 5), in which the testis volumes of the PINX birds was lower thanthose of the SHAM birds, which had already regressed (Fig. 4C). Furtherinvestigation into the short- and long-term effects of pinealectomy on
theHPG axis or overt reproductive statemay help to elucidate this com-plicated interaction.
In this study, we showed that PINX interferes withmale house spar-rows' ability to respond to changes in photoperiod in terms of theamount of vocalization and its quality. Our study showed significant ef-fects of PINX on photoperiodic variation of the vocal rate and vocal rep-ertoire without affecting the circulating testosterone level as assessedfrom bill darkness in house sparrows, which suggests an independentrole of the pineal gland in the seasonal regulation of avian vocalization.The data are consistent with the notion that photoperiodic changes inthe duration of pineal melatonin induce changes in song behavior andsong control nuclei that synchronize seasonally appropriate behaviors,such as bird song (Cassone et al., 2008). This would be analogous tothe situation in seasonally breedingmammalswhereinmelatonin dura-tion signals changes in deiodinase activity in the pituitary pars tuberalis,which, in turn, affects gonadotropin release patterns (Lincoln, 2006).Similarly, pineal melatonin does not necessarily induce or suppresssong in birds or reproduction in mammals; it merely increases the like-lihood that these important events occur at appropriate times relative tothe seasons (Cassone, 2014). Current studies in our laboratory are ad-dressing the role of pineal melatonin in the timing of such importantevents.
In contrast to previous studies in other close-ended learners, such asthe rufous-sided towhee, Pipilo erythrophthalmus, and the song sparrow,Melospiza melodia (Brenowitz et al., 1991; Smith et al., 1995), we ob-served the photoperiodic changes of DS song repertoire size in housesparrows, similar to the open-ended learner European starlings, a spe-cies in which male adult birds seasonally modulate their repertoire
379G. Wang et al. / Hormones and Behavior 65 (2014) 372–379
size (Van Hout et al., 2009). It is not clear whether this is due to theirunique vocal behaviors or the maintenance of vocal learning ability inadult house sparrows. Multiple-year studies of individual birds wouldbe required to answer this question. Surprisingly, we also observedthe up-regulation of call and DS song repertoire sizes in the SHAMbirds—but not in the PINX birds—under long-term exposure to long-day conditions, which suggests the involvement of the pineal gland inthe modification of vocal plasticity under these extended conditions.
In summary, our study clearly showed extensive effects of PINX onthe photoperiodic responses of vocalization and indicates a gonadal-independent role for the pineal gland in mediating seasonal changesof vocal behavior in house sparrows. Overall, our study suggests thatthe pineal gland acts as an indirect photoperiodic regulator for seasonalreproduction through its effects on seasonal changes of vocalization andcourtship behavior.
Supplementary data to this article can be found online at http://dx.doi.org/10.1016/j.yhbeh.2014.02.008.
Acknowledgments
The authors thank Michael Mina for assistance with the bill coloranalyses, Gregory Artiushin for assistance in capturing sparrows, Ye Lifor comments on themanuscript andMelissaWhitfield-Rucker for tech-nical assistance.
References
Ball, G.F., Auger, C.J., Bernard, D.J., Charlier, T.D., Sartor, J.J., Riters, L.V., Bathazart, J., 2004.Seasonal plasticity in the song control system: multiple brain sites of steroid hor-mone action and the importance of variation in song behavior. Ann. N. Y. Acad. Sci.1016, 586–610.
Barrett, P., Schuster, C., Mercer, J., Morgan, J.P., 2003. Sensitization: a mechanism for mel-atonin action in the pars tuberalis. J. Neuroendocrinol. 15, 415–421.
Bartness, T.J., Powers, J.B., Hastings, M.H., Bittman, E.L., Goldman, B.D., 1993. The timed in-fusion paradigm for melatonin delivery: what has it taught us about the melatoninsignal, its reception, and the photoperiodic control of seasonal responses? J. PinealRes. 15 (4), 161–190.
Bentley, G.E., 2001. Unraveling the enigma: the role of melatonin in seasonal processes inbirds. Microsc. Res. Tech. 53 (1), 63–71.
Bentley, G.E., Ball, G.F., 2000. Photoperiod-dependent and -independent regulation ofmel-atonin receptors in the forebrain of songbirds. J. Neuroendocrinol. 12 (8), 745–752.
Bentley, G.E., Van't Hof, T.J., Ball, G.F., 1999. Seasonal neuroplasticity in the songbird telen-cephalon: a role for melatonin. Proc. Natl. Acad. Sci. U. S. A. 96 (8), 4674–4679.
Bentley, G.E., Perfito, N., Calisi, R.M., 2013. Season- and context-dependent sex differences inmelatonin receptor activity in a forebrain songcontrol nucleus.Horm.Behav. 63, 829–835.
Bernard, D.J., Wilson, F.E., Ball, G.F., 1997. Testis-dependent and -independent effects ofphotoperiod on volumes of song control nuclei in American tree sparrows (Spizellaarborea). Brain Res. 760 (1–2), 163–169.
Binkley, S., Kluth, E., Menaker, M., 1971. Pineal function in sparrows: circadian rhythmsand body temperature. Science 174 (4006), 311–314.
Brandstätter, R., Kumar, V., Abraham, U., Gwinner, E., 2000. Photoperiodic information ac-quired and stored in vivo is retained in vitro by a circadian oscillator, the avian pinealgland. Proc. Natl. Acad. Sci. U. S. A. 97 (22), 12324–12328.
Brenowitz, E.A., 1997. Comparative approaches to the avian song system. J. Neurobiol. 33(5), 517–531.
Brenowitz, E.A., Nalls, B.,Wingfield, J.C., Kroodsma, D.E., 1991. Seasonal changes in avian songnuclei without seasonal changes in song repertoire. J. Neurosci. 11 (5), 1367–1374.
Cassone, V.M., 1990. Effects of melatonin on vertebrate circadian systems. TrendsNeurosci. 13 (11), 457–464.
Cassone, V.M., 2014. Avian circadian organization: a chorus of clocks. Front.Neuroendocrinol. 35, 76–88.
Cassone, V.M., Bartell, P.A., Earnest, B.J., Kumar, V., 2008. Duration of melatonin regulatesseasonal changes in song control nuclei of the house sparrow, Passer domesticus: in-dependence from gonads and circadian entrainment. J. Biol. Rhythms 23 (1), 49–58.
Cassone, V.M., Paulose, J.K., Whitfield-Rucker, M.G., Peters, J.L., 2009. Time's arrow flieslike a bird: two paradoxes for avian circadian biology. Gen. Comp. Endocrinol. 163(1–2), 109–116.
Dawson, A., King, V.M., Bentley, G.E., Ball, G.F., 2001. Photoperiodic control of seasonalityin birds. J. Biol. Rhythms 16 (4), 365–380.
Dupre, S.M., Burt, D.W., Talbot, R., Downing, A., Mouzaki, D., Waddington, D., Malpaux, B.,Davis, J.R., Lincoln, G.A., Loudon, A.S., 2008. Identification of melatonin regulatedgenes in the ovine pars tuberalis, a target site for seasonal hormonal control. Endocri-nology 149, 5527–5539.
Gahr, M., Kosar, E., 1996. Identification, distribution and developmental changes of a mel-atonin binding site in the song control system of the zebra finch. J. Comp. Neurol. 367(2), 308–318.
Gaston, S., Menaker, M., 1968. Pineal function: the biological clock in the sparrow? Sci-ence 160 (3832), 1125–1127.
Goldman, B.D., 2001. Mammalian photoperiodic system: formal properties and neuro-endocrine mechanisms of photoperiodic time measurement. J. Biol. Rhythms 16 (4),283–301.
Gwinner, E., Hau, M., 2000. Pineal gland, circadian rhythms and photoperiodism. In:Whittow, C.G. (Ed.), Sturkie's Avian, Physiology, pp. 557–658.
Hall, Z.J., Macdougall-Shackleton, S.A., 2012. Influence of testosterone metabolites onsong-control system neuroplasticity during photostimulation in adult European star-lings (Sturnus vulgaris). PLoS One 7 (7), e40060. http://dx.doi.org/10.1371/journal.pone.0040060.
Hegner, R.E., Wingfield, J.C., 1986. Behavioral and endocrine correlates of multiplebrooding in the semi-colonial house sparrow, Passer domesticus. I. Males. Horm.Behav. 20, 294–312.
Heigl, S., Gwinner, E., 1995. Synchronization of circadian rhythms of house sparrows byoral melatonin: effects of changing period. J. Biol. Rhythms. 10 (3), 225–233.
Hoffman, R.A., Reiter, R.J., 1965. Pineal gland: influence on gonads of male hamsters. Sci-ence 148 (3677), 1609–1611.
Hoffman, R.A., Reiter, R.J., 1966. Responses of some endocrine organs of female hamstersto pinealectomy and light. Life Sci. 5 (12), 1147–1151.
Janik, D., Dittami, J., Gwinner, E., 1992. The effect of pinealectomy on circadian plasmamelatonin levels in house sparrows and European starlings. J. Biol. Rhythms 7 (4),277–286.
Jansen, R., Metzdorf, R., van der Roest, M., Fusani, L., ter Maat, A., Gahr, M., 2005. Melatoninaffects the temporal organization of the song of the zebrafinch. FASEB J. 19 (7), 848–850.
Laucht, S., Kempenaers, B., Dale, J., 2010. Bill color, not badge size, indicates testosterone-related information in house sparrows. Behav. Ecol. Sociobiol. 64 (9), 1461–1471.
Lincoln, G.A., 2006. Decoding the nightly melatonin signal through circadian clock work.Mol. Cell. Endocrinol. 252, 69–73.
Lowther, P.E., Cink, C.L., 2006. House sparrow (Passer domesticus). In: Poole, A. (Ed.), TheBirds of North America Online. Cornell Lab of Ornithology, Ithaca, NY.
Lu, J., Cassone, V.M., 1993a. Pineal regulation of circadian rhythms of 2-deoxy[14C]glucoseuptake and 2[125I]iodomelatonin binding in the visual system of the house sparrowPasser domesticus. J. Comp. Physiol. A 173, 765–774.
Lu, J., Cassone, 1993b. Daily melatonin administration synchronizes circadian patterns ofbrain metabolism and behavior in pinealectomized house sparrows, Passerdomesticus. J. Comp. Physiol. A. 173, 775–782.
McGuire, N.L., Kangas, K., Bentley, G.E., 2011. Effects of melatonin on peripheral reproduc-tive function: regulation of testicular GnIH and testosterone. Endocrinology 152 (9),3461–3470.
Murton, R.K., Westwood, N.J., 1977. Avian Breeding Cycles. Oxford University Press.Nivison, J.J., 1978. The vocal behavior and displays of the house sparrow, Passer domesticus
L., in the United States. Ph.D. dissertation Wayne State University.Pant, K., Chandola-Saklani, A., 1992. Pinealectomy and LL abolished circadian perching
rhythms but did not alter circannual reproductive or fattening rhythms in finches.Chronobiol. Int. 9 (6), 413–420.
Pittendrigh, C.S., 1993. Temporal organization: reflections of a Darwinian clock-watcher.Annu. Rev. Physiol. 55, 16–54.
Reiter, R.J., 1973a. Pineal control of seasonal reproductive rhythm inmale goldenhamstersexposed to natural daylight and temperature. Endocrinology 92, 423–430.
Reiter, R.J., 1973b. Influence of pinealectomy on the breeding capability of hamstersmain-tained under natural photoperiodic and temperature conditions. Neuroendocrinolo-gy 13, 366–370.
Smith, G.T., Brenowitz, E.A., Wingfield, J.C., Baptista, L.F., 1995. Seasonal changes in songnuclei and song behavior in Gambel's white-crowned sparrows. J. Neurobiol. 28(1), 114–125.
Strand, C.R., Deviche, P., 2007. Hormonal and environmental control of song control re-gion growth and new neuron addition in adult male house finches, Carpodacusmexicanus. Dev. Neurobiol. 67 (6), 827–837.
Takahashi, J.S., Hamm, H., Menaker, M., 1980. Circadian rhythms of melatonin releasefrom individual superfused chicken pineal glands in vitro. Proc. Natl. Acad. Sci. U. S.A. 77 (4), 2319–2322.
Tramontin, A.D., Wingfield, J.C., Brenowitz, E.A., 2003. Androgens and estrogens induceseasonal-like growth of song nuclei in the adult songbird brain. J. Neurobiol. 57 (2),130–140.
Ubuka, T., Bentley, G.E., Ukena, K., Wingfield, J.C., Tsutsui, K., 2005. Melatonin induces theexpression of gonadotropin-inhibitory hormone in the avian brain. Proc. Natl. Acad.Sci. U. S. A. 102 (8), 3052–3057.
Van Hout, A.J., Eens, M., Balthazart, J., Pinxten, R., 2009. Complex modulation of singingbehavior by testosterone in an open-ended learner, the European starling. Horm.Behav. 56 (5), 564–573.
Wang, G., Harpole, C.E., Trivedi, A.K., Cassone, V.M., 2012. Circadian regulation of birdsong, call, and locomotor behavior by pineal melatonin in the zebra finch. J. Biol.Rhythms 27 (2), 145–155.
Whitfield-Rucker, M.G., Cassone, V.M., 1996. Melatonin binding in the house sparrowsong control system: sexual dimorphism and the effect of photoperiod. Horm.Behav. 30 (4), 528–537.
Whitfield-Rucker, M.G., Cassone, V.M., 2000. Photoperiodic regulation of the male housesparrow song control system: gonadal dependent and independent mechanisms.Gen. Comp. Endocrinol. 118 (1), 173–183.
Wilson, F.E., 1991. Neither retinal nor pineal photoreceptors mediate photoperiodic con-trol of seasonal reproduction in American tree sparrows (Spizella arborea). J. Exp.Zool. 259 (1), 117–127.
Zimmerman, N.H., Menaker, M., 1979. The pineal gland: a pacemaker within the circadiansystem of the house sparrow. Proc. Natl. Acad. Sci. U. S. A. 76 (2), 999–1003.
