effects of fasting and exogenous melatonin on annual rhythms

www.elsevier.com/locate/cbpa

Comparative Biochemistry and Physiol

Effects of fasting and exogenous melatonin on annual rhythms in the blue

fox (Alopex lagopus)

Petteri Nieminena,*, Teija Pyykfnenb, Juha Asikainena, Jaakko Mononenb, Anne-Mari Mustonena

aDepartment of Biology, University of Joensuu, P.O. Box 111, FIN-80101, Joensuu, FinlandbInstitute of Applied Biotechnology, University of Kuopio, P.O. Box 1627, FIN-70211, Kuopio, Finland

Received 3 June 2004; received in revised form 26 August 2004; accepted 1 September 2004

Abstract

The arctic fox (Alopex lagopus) is a winter-active inhabitant of the high arctic with extreme fluctuations in photoperiod and food

availability. The blue fox is a semi-domesticated variant of the wild arctic fox reared for the fur industry. In this study, 48 blue foxes were

followed for a year in order to determine the effects of exogenous melatonin and wintertime food deprivation on their reproductive and

thyroid axes. Half of the animals were treated with continuous-release melatonin capsules in July 2002, and in November–January, the

animals were divided into three groups and either fed continuously or fasted for one or two 22-day periods. Food deprivation decreased the

plasma triiodothyronine and thyroxine concentrations probably in order to preserve energy due to a decreased metabolic rate. The same was

observed in the plasma testosterone levels of the males but not in the plasma estradiol concentrations of the females. Exogenous melatonin

advanced the autumn moult and seasonal changes in the voluntary food intake. It also advanced the onset of the testosterone peak in the

males. The plasma estradiol levels of the females were unaffected, but the progesterone levels peaked more steeply in the sham-operated

females. Melatonin exerted a strong influence not only on the reproductive axis of the males but also on the seasonal food intake. The species

seemed quite resistant to periodic involuntary food deprivation.

D 2004 Elsevier Inc. All rights reserved.

Keywords: Alopex lagopus; Blue fox; Fasting; Melatonin; Photoperiod; Sex steroids; Thyroid hormones

1. Introduction

The arctic fox (Alopex lagopus L. 1758; Canidae) is a

middle-sized carnivore (3–3.5 kg; Prestrud and Nilssen,

1995) with a circumpolar distribution. The blue fox is a

semi-domesticated, blue-gray color type of the arctic fox

reared for the fur industry. Similar color types with bluish

winter fur can be encountered also in nature in Alaska and

Greenland (Nes et al., 1988). The blue fox is larger than the

arctic fox (6–9 kg, N10 kg in late autumn), but it is an easily

accessible animal that can be used as a model to study the

seasonal physiology of the arctic fox. Due to its distribution

in the high arctic, the arctic fox is an extremely photo-

1095-6433/$ - see front matter D 2004 Elsevier Inc. All rights reserved.

doi:10.1016/j.cbpb.2004.09.002

* Corresponding author. Tel.: +358 13 251 3576; fax: +358 13 251

3590.

E-mail address: [email protected] (P. Nieminen).

periodic species experiencing wide variations in daylength

from total darkness to 24-h daylight (Fuglei, 2000). Photo-

period and its effector hormone, melatonin, are the main

determinants in the timing of the reproductive season of the

species. Similar to hibernating animals, the blue fox and the

arctic fox have a great capability to gather extensive adipose

tissue stores in the autumn, but they do not utilize passive

wintering strategy, remaining active throughout the winter

(Pulliainen, 1993).

Male blue foxes experience rapid increases in testicular

mass, volume and spermatogenesis between January and

March and a decline between April and October (Smith et

al., 1984). Early spring is also characterized by an increase

in the circulating concentrations of luteinizing hormone

(LH), follicle-stimulating hormone and androgens (Smith et

al., 1985). In females, oestrus with increased LH, estradiol

and progesterone concentrations usually takes place

between March and April (Møller et al., 1984), and if

ogy, Part A 139 (2004) 183–197

P. Nieminen et al. / Comparative Biochemistry and Physiology, Part A 139 (2004) 183–197184

pregnancy occurs, the plasma prolactin concentrations

increase until parturition takes place usually in June and

remain high during lactation (Mondain-Monval et al., 1985).

In the male blue fox, darkness treatment between August

and March causes an advancement of the mating season

(Szeleszczuk, 1992). Repeated treatments with exogenous

melatonin in August, February and March allow normal

testicular development but prevent testicular regression in

the spring (Smith et al., 1987). Unfortunately, data on the

effects of photoperiod and melatonin on the reproductive

cycle of the female blue fox are scarce or nonexistent.

Another crucial determinant of the reproductive pro-

cesses of the arctic fox is food availability. In nature, a large

population of small rodents increases the litter size of the

arctic fox, but it has been hypothesized that on fur farms,

intraspecific social dynamics, health and stress could be

more important factors determining the number of offspring

in the blue fox (Frafjord, 1993). It has been observed that

wild arctic foxes have larger litters if given supplementary

feeding in the summer and winter (Angerbjorn et al., 1995).

In the wild, but also on farms, the animals gather a large

amount of white adipose tissue (WAT) in the autumn to be

mobilized during the cold season and in the spring

(Korhonen, 1988; Prestrud and Nilssen, 1992, 1995).

The aim of the experiment was to study the interactions

between photoperiod and nutrition in the seasonal repro-

duction of the blue fox. Excessive obesity seems to be

detrimental to the reproduction of farmed blue foxes

(Hernesniemi and Knutar, 2000) and restricted feeding in

the autumn and winter could enhance reproductive success

and reduce feeding costs on fur farms. In the case of wild

arctic foxes, the results of this study could have practical

applications in wildlife management and in the protection of

the threatened Fennoscandian arctic fox population.

Fig. 1. Diurnal ambient temperature at the study area (638N, 288E) between Jul

2. Materials and methods

Forty-eight farm-bred blue foxes (A. lagopus, 24

males and 24 females) born in May 2002 were randomly

divided into six study groups consisting of four males

and four females. At the beginning of the study on July

23rd 2002, three of the study groups received a

continuous-release melatonin implant (12.0 mg PRIME-

XR melatonin implant, Wildlife Pharmaceuticals, Fort

Collins, CO, USA) inserted into the interscapular

subcutaneous WAT under sterile conditions (MEL ani-

mals). The other groups were sham-operated with placebo

capsules (SHAM animals). In the blue fox, these

implants are known to advance the physiological changes

connected to wintering, such as the autumn molt

(Parkanyi et al., 1993).

The experimental animals were fed with commercial fur

animal diets according to the common farming practices and

water or ice was available ad lib. In autumn 2002, the blue

foxes were fed ad libitum with a diet containing 1500–1900

kcal metabolizable energy kg�1. From November 1st 2002,

to the end of the study, June 24th 2003, they were fed with

1000 kcal animal�1 day�1. Two of the study groups were

fed throughout the study, two study groups were fasted for

22 days between November 26th and December 17th 2002,

and two study groups were fasted twice for 22 days

(November 26th to December 17th 2002, and January

14th to February 4th 2003). The blue foxes were food-

deprived in order to simulate the periodic wintertime

nutritional scarcity experienced by arctic foxes in the wild

(Fuglei, 2000). The 3-week fasting period has been

previously tolerated well by the farmed blue fox (Nieminen

et al., 2001). After the fasting periods, the fasted animals

were fed with a small amount of feed for a couple of days,

y 2002 and June 2003 and the timetable of the experimental procedures.

P. Nieminen et al. / Comparative Biochemistry and Physiology, Part A 139 (2004) 183–197 185

and thereafter their feeding level was returned to the level of

1000 kcal animal�1 day�1.

The treatments of the study groups were as follows;

group 1: sham-operated and fed throughout the winter

(n=8); group 2: sham-operated and fasted for 22 days in the

winter (n=8); group 3: sham-operated and fasted for two 22-

day periods in the winter (n=8); group 4: melatonin-treated

and fed throughout the winter (n=8); group 5: melatonin-

treated and fasted for 22 days in the winter (n=8); group 6:

melatonin-treated and fasted for two 22-day periods in the

winter (n=8).

The blue foxes were housed in cages (113�108�72 cm)

at natural temperature (Fig. 1) and photoperiod at the

Juankoski Research Station fur farm, Juankoski, Finland

Fig. 2. (a,b) Body mass (BM) of blue foxes (A. lagopus) in (a) different feeding

(feeding groups combined), meanFS.E.M. *Significant difference between the ex

(638N; 288E) according to standard fur farm practices.

During the study, one female from group 1 and one female

from group 2 died, but the postmortem examinations of the

National Veterinary and Food Research Institute of Finland

revealed no obvious cause of death. The rest of the animals

remained healthy. The experiment was approved by the

Animal Care and Use Committee of the University of

Joensuu and complied with the current laws of Finland.

During the reproductive season, the animals were mated

within an experimental group with monogamy by placing a

male and a female fox into the same cage. The onset of

oestrus was determined by measuring the electrical resist-

ance of the vaginal mucosa daily around oestrus with an

ohmmeter (Møller et al., 1984).

treatments (SHAM/MEL combined) and (b) SHAM vs. MEL treatments

perimental groups (two-way ANOVA, Pb0.05).


Blood samples (1–5 mL) were drawn from a superficial

vein of the left hind leg under sterile conditions without

anesthesia but with minimal disturbance between 0900 and

1100 h before the animals were fed. The samples were taken

in the shadowhouses at the ambient temperature (Fig. 1) to

avoid any effects of rapid temperature fluctuations on the

biochemical variables.

The blue foxes were sampled approximately every third

week in autumn 2002 and in spring–summer 2003 (Fig. 1).

During food deprivation, the samples were taken before the

fasting periods and after 48 h, 8 days, 15 days and 22 days

of fasting. After the 4-week recovery period with two blood

samplings, the fasting and sampling procedures were

repeated for groups 3 and 6. Due to the relatively small

Fig. 3. (a,b) Body mass index (BMI) of blue foxes (A. lagopus) in (a) different feed

(feeding groups combined), meanFS.E.M. *Significant difference between the ex

amount of plasma obtained during the fasting periods, only

the thyroid hormone concentrations were determined from

all sampling dates and plasma testosterone and estradiol

concentrations were measured only after 22 days of fasting.

Body masses (BM: kg) and lengths (m) from the tip of

the nose to the anus along the ventral midline were

measured at blood samplings. Body mass indices (BMIs)

reflecting body adiposity were calculated by the formula:

BM/body length3 used previously for the species (Nieminen

et al., 2001). The formula correlated perfectly with the

obesity index derived experimentally for a fur-bearing canid

of similar size (Korhonen et al., 1982). Maturing of the

winter pelage was estimated on October 29th and November

20th 2002, by the same specialist. The maturing of the guard

ing treatments (SHAM/MEL combined) and (b) SHAM vs. MEL treatments

perimental groups (two-way ANOVA, Pb0.05).

Fig. 4. Voluntary energy intake of the SHAM- and MEL-treated blue foxes, meanFS.E.M. *Significant difference between the experimental groups (two-way

ANOVA, Pb0.05).


hairs and the underfur was determined with a scale from one

to five (Asikainen et al., 2003). The completion of the

spring moult was estimated on June 24th 2003 by dividing

the loss of old winter hair into three categories as follows:

one, full winter pelage; two, winter pelage partially lost;

three, full summer pelage.

The plasma testosterone (intraassay variation 3.8–7.5%

CV, interassay variation 4.8–7.0% CV; total testosterone,

i.e., protein-bound and non protein-bound), estradiol (2.9–

9.7 and 2.3–10.2% CV; total estradiol 17h), progesterone(2.9–5.8 and 4.7–5.1% CV), thyroxine (T4; 3.3–6.8 and

Fig. 5. Plasma melatonin concentrations of the SHAM- and MEL-treated blue foxes

the experimental groups (two-way ANOVA, Pb0.05).

3.3–8.0% CV), and triiodothyronine (T3; 3.3–6.1 and 4.5–

7.5% CV) concentrations were measured with the Spectria

[125I] Coated Tube Radioimmunoassay kits (Orion Diag-

nostica, Espoo, Finland). The diurnal plasma melatonin

concentrations were determined to verify the presence of

excess melatonin in the circulation of the MEL animals with

a Melatonin RIA kit (4.3–7.4 and 11.7–21.1% CV; DLD

Diagnostika, Hamburg, Germany).

In the thyroid hormones and the BM data, there were no

differences between the two sexes, and these data were

analyzed together.

(feeding groups combined), meanFS.E.M. *Significant difference between


2.1. Statistics

Multiple comparisons were performed with the two-way

analysis of variance (ANOVA) with the SPSS program with

food deprivation and melatonin treatment as determinants,

and with the post hoc Duncan’s test. For the time-series

analyses, the repeated-measures ANOVA with combined

groups was used. If the criteria for ANOVA were not met,

the nonparametric Kruskal–Wallis one-way ANOVA with a

post hoc Dunn’s test was performed. The P value less than

0.05 was considered to be statistically significant. As the

two-way ANOVA revealed no significant interactions

between the experimental procedures of melatonin treatment

and food deprivation, some of the results were pooled across

Fig. 6. (a–b) Plasma testosterone concentrations of the male blue foxes (A. lagopus

vs. MEL treatments (feeding groups combined), meanFS.E.M. *Significant diffe

the experimental regimes. Correlations were calculated with

the Spearman correlation coefficient. The results are

presented as the meanFS.E.M.

3. Results

Mean BM of the blue foxes increased from the initial

3.1F0.1 to 13.2F0.3 kg between July and November 2002

(repeated-measures ANOVA, Pb0.05; Fig. 2). At the

beginning of the first fast (November 26th 2002), mean

BM of the fasted blue foxes was 13.2 kg, and it decreased to

10.5 kg (�20.3%) after 22 days of fasting (repeated-

measures ANOVA, Pb0.05; Fig. 2a). At the same time,

) in (a) different feeding treatments (SHAM/MEL combined) and (b) SHAM

rence between experimental groups (two-way ANOVA, Pb0.05).


their BMIs decreased by 20.3% (repeated-measures

ANOVA, Pb0.05; Fig. 3a). After 6 days of re-feeding,

BM increased to 11.0 kg (+5.1%; repeated-measures

ANOVA, Pb0.05). At the beginning of the second fast

(January 14th 2003), the fasted blue foxes weighed 11.4 kg

(Fig. 2a). A mass loss of 2.4 kg (�21.4%) led to the final

BM of 9.0 kg after 22 days of fasting (repeated-measures

ANOVA, Pb0.05; Fig. 2a). After a recovery period of 7

days, BM of the blue foxes had increased to 9.5 kg (+5.5%;

repeated-measures ANOVA, Pb0.05). In the spring, mean

BM of all animals decreased to the value of 7.2F1.1 kg on

June 24th 2003 (repeated-measures ANOVA, Pb0.05). The

fasted blue foxes had lower BM than the fed foxes from

December 5th to December 23rd 2002, on January 17th and

from January 23rd to February 11th 2003 (two-way

Fig. 7. (a–b) Plasma estradiol concentrations of the female blue foxes (A. lagopus)

vs. MEL treatments (feeding groups combined), meanFS.E.M. *Significant diffe

ANOVA, Pb0.05; Fig. 2a). BMIs were lower in the fasted

animals between December 5th and December 23rd 2002,

and again between January 17th and June 24th 2003 (two-

way ANOVA, Pb0.05), except that the difference between

experimental groups did not reach significance on March

4th (Fig. 3a).

Melatonin treatment had no acute effects on the BM of

the experimental animals (Fig. 2b). However, in spring

2003, the BM of the MEL groups was lower than in the

SHAM groups between March 4th and April 15th (two-way

ANOVA, Pb0.05). The BMI of the MEL groups was higher

than in the SHAM groups on June 24th 2003 (two-way

ANOVA, Pb0.05; Fig. 3b). The voluntary food intake of the

MEL groups reached its peak values in late Nov and in the

SHAM groups about one month later (repeated-measures

in (a) different feeding treatments (SHAM/MEL combined) and (b) SHAM

rence between the experimental groups (two-way ANOVA, Pb0.05).


ANOVA, Pb0.05; Fig. 4). In a similar manner, the spring-

time increase in the voluntary food intake was advanced by

approximately a month due to melatonin treatment.

The winter pelage matured earlier in the MEL groups

compared to the SHAM groups. On October 29th 2002, the

guard hairs and the underfur of the MEL animals were more

mature compared to the SHAM groups (scapular guard hairs

4.13F0.14 vs. 3.17F0.17; hip guard hairs 4.38F0.15 vs.

3.29F0.16; scapular underfur 3.29F0.13 vs. 2.29F0.14;

hip underfur 3.75F0.24 vs. 2.67F0.13; Kruskal–Wallis

one-way ANOVA, Pb0.001). On November 20th 2002, the

winter pelage of almost all individuals had matured and

there were no differences between the experimental groups.

On June 24th 2003, the loss of winter pelage was

Fig. 8. (a–b) Plasma progesterone concentrations of the female blue foxes (A. lag

SHAM vs. MEL treatments (feeding groups combined), meanFS.E.M. *Significan

incomplete in the MEL animals compared to the SHAM

animals with the scale from one to three (1.54F0.13 vs.

2.38F0.15; Kruskal–Wallis one-way ANOVA, Pb0.001).

Plasma melatonin concentrations were higher in the MEL

groups during all the sampling dates following the insertion

of the capsules (Fig. 5; two-way ANOVA, Pb0.0001).

Plasma testosterone concentrations of the males were low

between July and October 2002 (Fig. 6a–b). Food depriva-

tion caused a transitory decrease in the plasma testosterone

concentrations after 22 days of fasting during both of the

food deprivation periods (repeated-measures ANOVA,

Pb0.05; Fig. 6a). One week after the fasts the testosterone

levels of the fasted males had returned to the level of the fed

males. In early November, testosterone levels of the MEL

opus) in (a) different feeding treatments (SHAM/MEL combined) and (b)

t difference between the experimental groups (two-way ANOVA, Pb0.05).


males started to increase and were higher than in the SHAM

males on November 5th and November 11th 2002 (two-way

ANOVA, Pb0.05; Fig. 6b). The testosterone peak of the

MEL males occurred in early March and in the SHAM

males about one month later (repeated-measures ANOVA,

Pb0.05). The springtime decrease in plasma testosterone

concentrations was also advanced by a month in the MEL

males, so the SHAM males had higher testosterone

concentrations on March 24th 2003 (two-way ANOVA,

Pb0.05). By June, testosterone levels in all male blue foxes

had decreased back to low concentrations.

Plasma estradiol concentrations of all females were very

low in early autumn 2002 and unaffected by the fasting

regimes (Fig. 7a–b). In the MEL females, the estradiol

concentrations of individual animals peaked betweenDecem-

Fig. 9. (a–b) Plasma T3 concentrations of the blue foxes (A. lagopus) in (a) diffe

treatments (feeding groups combined), meanFS.E.M. *Significant difference betw

ber 17th 2002 and April 15th 2003 (Fig. 7b). The plasma

estradiol concentrations of the MEL females were lower than

in the SHAM females on September 3rd (4.11F0.35 vs.

5.24F0.29 pmol L�1; Kruskal–Wallis one-way ANOVA,

Pb0.03) and on November 5th 2002 (5.38F0.41 vs.

7.20F0.57 pmol L�1; Kruskal–Wallis one-way ANOVA,

Pb0.03). However, on December 17th 2002, their estradiol

levels were significantly higher than in the SHAM females

(33.59F18.19 vs. 7.61F0.90 pmol L�1; Kruskal–Wallis one-

way ANOVA, Pb0.01; Fig. 7b). The interindividual variation

in the timing of the estradiol peak was so great that the

seasonal pattern of plasma estradiol concentrations was less

obvious than in plasma testosterone levels. However, several

females experienced increased plasma estradiol levels

between December 2002 and April 2003.

rent feeding treatments (SHAM/MEL combined) and (b) SHAM vs. MEL

een the experimental groups (two-way ANOVA, Pb0.05).


The fasting experiments did not affect the plasma

progesterone concentrations measured after the cessation of

the second fast (Fig. 8a). The plasma progesterone concen-

trations of theMEL females started to increase in earlyMarch

and those of the SHAM females in lateMarch 2003 (repeated-

measures ANOVA, Pb0.05; Fig. 8b). The MEL females had

lower plasma progesterone concentrations than the SHAM

females on May 13th 2003 (two-way ANOVA, Pb0.05). In

the MEL females, plasma progesterone concentrations

increased significantly between March 4th and March 24th

2003 (repeated-measures ANOVA, Pb0.05). Progesterone

levels increased in the SHAM females between March 24th

and April 15th 2003 and decreased again between May 13th

and June 24th 2003 (repeated-measures ANOVA, Pb0.05).

Fig. 10. (a–b) Plasma T4 concentrations of the blue foxes (A. lagopus) in (a) diff


A slight trend of decreasing plasma T3 levels could be

observed during the whole experiment (Fig. 9a). Food

deprivation caused a transitory decrease in plasma T3

concentrations, which were lower in the fasted animals

between December 3rd and 17th 2002 (8–22 days of

fasting during the first fast) and between January 21st and

February 4th (8–22 days of fasting during the second fast),

and on February 11th 2003 (two-way ANOVA, Pb0.05).

Plasma T3 concentrations were unaffected by the melatonin

treatment (Fig. 9b). The plasma T4 concentrations of all

animals decreased between October and December 2002

and increased again in February–March 2003 (repeated-

measures ANOVA, Pb0.05; Fig. 10a–b). T4 concentrations

of the fasted animals were transiently lower than in the fed

erent feeding treatments (SHAM/MEL combined) and (b) SHAM vs. MEL



groups during the second food deprivation period between

January 28th and February 4th 2003 (15–22 days of

fasting; two-way ANOVA, Pb0.05; Fig. 10a). In the MEL

animals, plasma T4 levels were lower than in the SHAM

animals between October 15th and November 27th 2002

and again on April 15th 2003 (two-way ANOVA, Pb0.05;

Fig. 10b). The plasma ratio of T3 to T4 was unaffected by

food deprivation (Fig. 11a) but higher in the MEL groups

than in the SHAM groups on October 15th and December

3rd 2002 and on April 15th 2003 (two-way ANOVA,

Pb0.05; Fig. 11b).

Reproduction of the blue foxes was unsuccessful. Of the

22 females only six came into heat and only one female

from group 2 eventually gave birth to pups. The onset of the

Fig. 11. (a–b) Plasma T3 T4�1 ratios of the blue foxes (A. lagopus) in (a) differ


heat or the proportion of the females that came into heat did

not differ between the experimental groups. Neither was the

descent of the testes in the male blue foxes influenced by

food deprivation or melatonin treatment.

4. Discussion

4.1. General remarks

As a winter-active middle-sized predator, the arctic fox

experiences extreme seasonal fluctuations in prey avail-

ability, temperature and photoperiod in its natural habitat

(Fuglei, 2000). In autumn, it has the potential for seasonal

ent feeding treatments (SHAM/MEL combined) and (b) SHAM vs. MEL



obesity (Prestrud and Nilssen, 1992, 1995) shared by other

boreal and arctic carnivores. In the wild arctic fox, WAT

represents only 6% of BM in the summer but as much as

20% in the early winter (Prestrud and Nilssen, 1992). The

farmed blue fox, on the other hand, has a reliable food

source and can thus exhibit extreme seasonal obesity

(Korhonen, 1988) as also evidenced by the autumnal

fattening of the experimental animals of this study. In

nature, the involuntary food deprivation caused by periodic

lack of nutrition must be one of the main determinants of the

wintering strategy of the arctic fox. Thus, the species has to

maintain a delicate balance between energy preservation to

minimize the consumption of fat reserves during unfavor-

able periods, and alertness in order to forage effectively

when opportunity presents itself.

4.2. Effects of seasonality and food deprivation

BM of wild arctic foxes is higher in the late autumn than

in the spring (Fuglei and Øritsland, 1999). This seasonal

cycle of autumnal fat storage and wintertime and vernal use

of stored energy to attain a leaner BM in the spring and

summer seems to be preserved in the farmed blue foxes as

evidenced by the results of this study. Food deprivation

caused an expected decrease in BM and BMI of the fasted

animals. However, they were eventually able to return to the

seasonal BM cycle of the fed animals and there were no

differences in BM or BMI of the experimental groups at the

end of the study. These data suggest that the innate

mechanisms of seasonal BM cycles have evolved in the

arctic fox to cope with quite long periods of food

deprivation without any effect on BM or body fat content

during the spring months. The same has been observed in

raccoon dogs (Nyctereutes procyonoides) fasted for 2

months in mid-winter (Mustonen et al., 2004). The annual

rhythms of BM in these canids may be regulated by weight-

regulatory hormones such as leptin, ghrelin and growth

hormone interacting with melatonin and photoperiod (Nie-

minen et al., 2001, 2002; Fuglei et al., 2004).

In the thyroid axis, the results of this study mostly

confirm previous observations on the effects of food

deprivation on the wild arctic fox (Fuglei et al., 2000).

The decrease in the circulating T3 levels during fasting has

been previously detected in wild arctic foxes on Svalbard in

November and in May (Fuglei et al., 2000). This has been

explained as an attribute of the reduced metabolic rate.

Furthermore, the increase in the T3 levels during re-feeding

was also previously observed in wild arctic foxes. The

vernal reduction in T3 levels of all blue foxes of this study

has also been indirectly observed by Fuglei et al. (2000)

with lower plasma T3 levels in May than in November.

In a similar manner, the results of the plasma T4 levels

also conform to earlier studies. Fuglei et al. (2000) did not

notice any seasonal differences in plasma T4 concentrations

of wild arctic foxes between November and May. Also in

this study, plasma T4 concentrations were approximately on

the same level in November and May, but a decrease could

be observed during the mid-winter. In a previous study

(Fuglei et al., 2000), the wintertime T4 values were

measured only in November and thus it is probable that a

wintertime decrease in the T4 levels remained undetected. In

farmed blue foxes, however, the autumnal decrease in T4

concentrations has been observed previously in November

(Nieminen et al., 2001).

The decrease of plasma T4 concentrations during the

second food deprivation period in Jan 2003 has not been

described previously for the species. Fuglei et al. (2000)

found no effect of a 13-day fast on plasma T4 levels of wild

arctic foxes. In a similar manner, no decrease in plasma T4

levels has been detected in farmed blue foxes after a 3-week

food deprivation period (Nieminen et al., 2001). However,

the results of this study also show that it was only during the

second fasting period when the T4 levels of the fasted

animals decreased significantly. It seems that in the thyroid

axis, T3 responds more rapidly to food deprivation and

levels of T4 decline only after more prolonged fasting. Of

course, decreased thyroid hormone levels most probably act

to reduce metabolic rate and thus preserve energy during

starvation (Bruck, 1983) to enhance the survival of the arctic

fox through unfavorable periods. The same has been

observed previously in many carnivores such as the raccoon

dog (Mustonen et al., 2004), the domestic dog (Canis

familiaris, de Bruijne et al., 1981), the European brown bear

(Ursus arctos arctos, Hissa et al., 1994), and the American

badger (Taxidea taxus, Harlow and Seal, 1981).

The thyroid hormone results of this study are difficult to

interpret as, e.g., free thyroid hormone or binding protein

concentrations, monodeiodinase activities, cellular receptor

levels or thyroid hormone utilization rates indicating

hormone function (Tomasi, 1991) could not be monitored

in the blue foxes of this study. However, it is possible that

the first 3-week fast did not inhibit the hypothalamic and

pituitary regulatory hormones controlling the secretion of T3

and T4. Instead, the observed decrease in the plasma T3

concentrations may have derived from the suppression of

the peripheral conversion of T4 into T3. On the contrary, the

second fast decreased the plasma concentrations of both T3

and T4, which could have resulted from downregulation of

hormone synthesis and release by the thyroid gland. Long-

term food deprivation probably suppressed the thyroid

activity via the hypothalamus–pituitary axis (Hugues et

al., 1983; Tveit and Larsen, 1983).

The sex steroid concentrations of the blue foxes were less

affected by food deprivation than the thyroid axis. The most

obvious effect was a decrease in the plasma testosterone

concentrations of the fasted males measured after three

weeks of food deprivation. The effects of fasting on the sex

steroids of the species have not been studied previously.

However, there exist some data on related canid species. In

the farmed raccoon dog, the plasma testosterone levels of

males are mostly unaffected by wintertime fasting periods

(J. Asikainen et al., unpublished), but on some occasions,


the plasma testosterone levels of fasted raccoon dog males

have been higher than in fed males during the mating season

(Asikainen et al., 2003). The decrease in plasma testosterone

concentrations due to food deprivation is a phenomenon

shared by other mammals (Ahima et al., 1996). It is

reasonable to assume that reproduction during nutritional

scarcity would not be successful and the sex steroid levels

decrease due to fasting. In species with a passive wintering

strategy such as the raccoon dog with elaborate adaptations

to wintertime fasting, this mechanism would not be

necessary as prolonged periods of voluntary food depriva-

tion are a natural part of their life history. However, the

testosterone results of the blue foxes should be interpreted

with caution. Alternative explanations for the observed

differences between the experimental groups could be, e.g.,

changes in the testosterone carrier protein content. As these

levels could not be determined in this study, they will have

to be investigated further in the future.

In the female blue foxes, the great individual variation

in the timing of oestrus makes the interpretation of the

estradiol and progesterone data confusing. However, at no

point of the study were there any effects of food

deprivation on the concentrations of these hormones.

Previous studies have shown that food availability exerts

a strong influence on the reproduction of wild arctic

foxes. According to the results of this study, it is more

probable that in the wild nutritional scarcity affects, e.g.,

the number of ova at ovulation, embryos that implant or

the number of cubs born, instead of influencing the onset

of oestrus (see also Angerbjorn et al., 1995). Possibly, the

arctic fox and the blue fox have an optimal BM or

adipose tissue mass for reproduction balanced between the

scarcity of wild animals and the excessive obesity of

farmed foxes as observed previously in the raccoon dog

(Asikainen et al., 2002).

4.3. Effects of continuous melatonin treatment

The observed higher melatonin levels of the MEL blue

foxes show that excess melatonin was present in their

circulation for the whole duration of the study. The results

suggest that the effects of exogenous melatonin could have

been mostly due to an advancement of innate seasonal

physiological cycles caused by the initial melatonin surge

from the capsules causing a long-lasting phase-shift of

seasonal physiological phenomena (see also Nieminen et al.,

2002 for the raccoon dog). The advancement of seasonal

rhythms could be very clearly observed in the food intake of

the blue foxes, as both the wintertime decrease and the

increase in the food intake in the following spring were

advanced by approximately a month in the MEL groups.

This suggests that the appetite of the melatonin-treated

animals had switched to a new timetable and followed it

thereafter in spite of their continuously higher plasma

melatonin concentrations. The same has been observed

previously in the raccoon dog after the insertion of

continuous-release melatonin capsules in the late summer

(Nieminen et al., 2002).

In nature, the fur growth of arctic foxes follows the

photoperiod (Underwood and Reynolds, 1980). Moreover,

priming of the winter pelage is a well-known effect of

exogenous melatonin treatment in diverse mammalian

species (Allain and Rougeot, 1980; Forsberg and Madej,

1990; Xiao, 1996). This phenomenon could be reproduced

in this study as the winter pelage of the MEL blue foxes

matured earlier than in the SHAM groups. In addition,

continuous melatonin treatment delayed the loss of the thick

winter pelage in the following spring as observed also

previously by Smith et al. (1987) in the blue fox. This effect

does not fit the pattern mentioned above of an initial

triggering effect of exogenous melatonin being responsible

for the observed phase-shifts. However, the delayed spring

molt can be due to the lack of decreasing melatonin

concentrations encountered normally in spring simultane-

ously with increasing daylength. Plasma melatonin concen-

trations of the MEL group were relatively high and stable

after the initial melatonin peak and, thus, no significant

decrease in the melatonin levels could be observed during

the next spring and summer. It is known that in the spring

decreasing melatonin levels cause an increase in circulating

prolactin concentrations leading to the spring moult in some

fur-bearing species (Martinet et al., 1983).

A phase-shift could be observed also in the plasma

testosterone concentrations of the male blue foxes due to

exogenous melatonin. The testosterone levels of the MEL

males started to increase in early November, about a month

earlier than in the SHAM males. In a similar manner, the

peak testosterone values of the MEL males were observed in

early March, almost a month earlier than in the SHAM

males. This melatonin-induced advancement of the mating

season has also been described previously in the male silver

fox (Vulpes vulpes, Forsberg et al., 1989; Forsberg and

Madej, 1990) and in the male raccoon dog (Xiao, 1996;

Asikainen et al., 2003).

The case of the female blue fox is, again, more complex.

Previously, advancement of the mating season due to

continuous melatonin treatment has been observed in the

female raccoon dog (Asikainen et al., 2003) and in the

female ferret (Mustela putorius furo; Nixon et al., 1995) but

no previous findings in the blue fox are available. The

plasma progesterone concentrations of the females of this

study offer some indications that melatonin would be a

principal regulator in the timing of their seasonal reproduc-

tion, too. The plasma progesterone levels of the SHAM

females peaked in May, but no clear progesterone peak

could be detected in the melatonin-treated females. Yet, their

plasma progesterone concentrations started to increase

significantly in March. The same occurred in the SHAM

females a month later, in March–April. This suggests that

melatonin also times the onset of the reproductive period in

the female of the species. A more consistent pattern of

advancement of the reproductive season and the corre-


sponding progesterone peak has been observed in the

raccoon dog (Asikainen et al., 2003). In the case of the

blue fox, the individual variation in the timing of oestrus

was so great that no consistent or simultaneous estradiol

peaks could be detected in the experimental females. Due to

practical reasons sampling could not be performed as often

as would have been required to obtain precise data on the

individual progesterone and estradiol peaks of the females.

In the thyroid axis, the melatonin-treated animals had

frequently lower plasma T4 concentrations than the SHAM

animals. The same has been observed previously in the

silver fox, with decreased T4 concentrations due to

exogenous melatonin treatment (Forsberg and Madej,

1990) and in rodents with inactivation of the thyroid gland

due to short daylength or melatonin (Vriend, 1983). Future

studies will determine whether the observed decrease in the

T4 levels is a direct effect of melatonin or caused by an

advancement of an innate seasonal rhythm of T4 concen-

trations. T4 levels of the melatonin-treated animals started

to decrease between August and September 2002, and those

of the SHAM animals in October 2002. Thus, it is possible

that the lower T4 values in the MEL animals were caused

by an earlier transition into wintertime energy preservation

characterized by low T4 levels.

4.4. Reproduction

The female blue foxes of this study were young

nulliparous animals entering their first reproductive period

in the spring 2003. Only six of them came into heat and only

one female eventually delivered cubs. This can be partly due

to the young age of the females, as older and more

experienced blue fox females usually have a higher number

of cubs and reproduce more successfully (Hernesniemi,

2000). In a similar manner, some of the juvenile male blue

foxes of this study were quite unresponsive to their female

companions and did not even attempt copulation (T.

Pyykfnen, personal communication, 2004).

Due to these difficulties, the effects of wintertime fasting

or melatonin on the reproductive performance of captive

blue foxes could not be determined in this study. There are

some indications that excessive autumnal fattening and a

very high adipose tissue mass can have detrimental effects

on blue fox reproduction (Hernesniemi and Knutar, 2000). It

is possible that wintertime food restriction or deprivation

and the leaner BM attained by these methods could be

beneficial for their reproductive performance, but it will

have to be determined in the future with a larger number of

experimental animals and artificial insemination.

Acknowledgements

This study was supported financially by the Academy of

Finland, the Faculty of Science of the University of Joensuu,

the Helve Foundation and the Maj and Tor Nessling

Foundation. We sincerely thank Mrs. Maija Miskala and

the other staff at the Juankoski Research Station for taking

care of the foxes. The assistance, strength and speed of Mr.

Heino Pirinen and Mr. Harri Kirjavainen are also highly

appreciated.

References

Ahima, R.S., Prabakaran, D., Mantzoros, C., Qu, D., Lowell, B., Maratos-

Flier, E., Flier, J.S., 1996. Role of leptin in the neuroendocrine response

to fasting. Nature 382, 250–252.

Allain, D., Rougeot, J., 1980. Induction of autumn moult in mink (Mustela

vison Peale and Beauvois) with melatonin. Reprod. Nutr. Dev. 20,

197–201.

Angerbjfrn, A., Tannerfeldt, M., Bj7rvall, A., Ericson, M., From, J., Noren,

E., 1995. Dynamics of the arctic fox population in Sweden. Ann. Zool.

Fenn. 32, 55–68.

Asikainen, J., Mustonen, A.-M., Nieminen, P., Pasanen, S., Araja-

Matilainen, H., Hyv7rinen, H., 2002. Reproduction of the raccoon

dog (Nyctereutes procyonoides) after feeding or food deprivation in

winter. J. Anim. Physiol. Anim. Nutr. 86, 367–375.

Asikainen, J., Mustonen, A.-M., Hyv7rinen, H., Nieminen, P., 2003.

Seasonal reproductive endocrine profile of the raccoon dog (Nyctereutes

procyonoides)—effects of melatonin and food deprivation. J. Exp.

Zool. A299, 180–187.

Brqck, K., 1983. Functions of the endocrine system. In: Schmidt, R.F.,

Thews, G. (Eds.), Human Physiology. Springer-Verlag, Heidelberg,

Germany, pp. 658–687.

de Bruijne, J.J., Altszuler, N., Hampshire, J., Visser, T.J., Hackeng, W.H.L.,

1981. Fat mobilization and plasma hormone levels in fasted dogs.

Metabolism 30, 190–194.

Forsberg, M., Madej, A., 1990. Effects of melatonin implants on plasma

concentrations of testosterone, thyroxine and prolactin in the male silver

fox (Vulpes vulpes). J. Reprod. Fertil. 89, 351–358.

Forsberg, M., Fougner, J.A., Hofmo, P.O., Madej, M., Einarsson, E.J.,

1989. Photoperiodic regulation of reproduction in the male silver fox

(Vulpes vulpes). J. Reprod. Fertil. 87, 115–123.

Frafjord, K., 1993. Reproductive effort in the arctic fox Alopex lagopus, a

review. Nor. J. Agric. Sci. 7, 301–309.

Fuglei, E., 2000. Physiological adaptations of the arctic fox to high Arctic

conditions. PhD thesis, University of Oslo, Oslo, Norway.

Fuglei, E., aritsland, N.A., 1999. Seasonal trends in body mass, food intake

and restingmetabolic rate, and induction ofmetabolic depression in arctic

foxes (Alopex lagopus) at Svalbard. J. Comp. Physiol. B169, 361–369.

Fuglei, E., Aanestad, M., Berg, J.P., 2000. Hormones and metabolites of

arctic foxes (Alopex lagopus) in response to season, starvation and re-

feeding. Comp. Biochem. Physiol. A126, 287–294.

Fuglei, E., Mustonen, A.-M., Nieminen, P., 2004. Effects of season, food

deprivation and re-feeding on leptin, ghrelin and growth hormone in

arctic foxes (Alopex lagopus) on Svalbard, Norway. J. Comp. Physiol.

B174, 157–162.

Harlow, H.J., Seal, U.S., 1981. Changes in hematology and metabolites in

the serum and urine of the badger, Taxidea taxus, during food

deprivation. Can. J. Zool. 59, 2123–2128.

Hernesniemi, T., 2000. Siitosel7inten valinta ja jalostusel7inten k7yttf(Selection of animals for breeding and their use in stock improvement).

In: Hernesniemi, T. (Ed.), Ketunkasvatuksen taito (The Art of Fox

Farming). Finnish National Board of Education, pp. 122–138. in

Finnish.

Hernesniemi, T., Knutar, T.-A., 2000. Kettujen hoito eri tuotantokausina

(Care of foxes during various productive seasons). In: Hernesniemi, T.

(Ed.), Ketunkasvatuksen taito (The Art of Fox Farming). Finnish

National Board of Education, pp. 68–105. In Finnish.

Hissa, R., Siekkinen, J., Hohtola, E., Saarela, S., Hakala, A., Pudas, J.,

1994. Seasonal patterns in the physiology of the European brown bear


(Ursus arctos arctos) in Finland. Comp. Biochem. Physiol. A109,

781–791.

Hugues, J.-L., Burger, A.G., Grouselle, D., Voirol, M.-J., Chabert, P.,

Modigliani, E., Sebaoun, J., 1983. Evidence of a thyrotropin-releasing

hormone-dependent increase in plasma thyrotropin during refeeding of

starved rats. Endocrinology 112, 715–719.

Korhonen, H., 1988. Regulation of energy economy in raccoon dogs and

blue foxes: a process of dynamic interactions. Comp. Biochem. Physiol.

A91, 263–268.

Korhonen, H., Harri, M., Asikainen, J., 1982. Effects of various diets and

energy levels on the growth of farmed raccoon dogs. Savonia 5, 1–9.

Martinet, L., Allain, D., Meunier, M., 1983. Regulation in pregnant mink

(Mustela vison) of plasma progesterone and prolactin concentrations

and regulation of onset of the spring moult by daylight ratio and

melatonin injections. Can. J. Zool. 61, 1959–1963.

Mbller, O.M., Mondain-Monval, M., Smith, A., Metzger, E., Scholler,

R., 1984. Temporal relationships between hormonal concentrations

and the electrical resistance of the vaginal tract of blue foxes

(Alopex lagopus) at pro-oestrus and oestrus. J. Reprod. Fertil. 70,

15–24.

Mondain-Monval, M., Mbller, O.M., Smith, A.J., McNeilly, A.S., Scholler,

R., 1985. Seasonal variations of plasma prolactin and LH concen-

trations in the female blue fox (Alopex lagopus). J. Reprod. Fertil. 74,

439–448.

Mustonen, A.-M., Nieminen, P., Puukka, M., Asikainen, J., Saarela, S.,

Karonen, S.-L., Kukkonen, J.V.K., Hyv7rinen, H., 2004. Physiologicaladaptations of the raccoon dog (Nyctereutes procyonoides) to seasonal

fasting—fat and nitrogen metabolism and influence of continuous

melatonin treatment. J. Comp. Physiol. B174, 1–12.

Nes, N., Einarsson, E.J., Lohi, O., 1988. Vackra p7lsdjur-och dessas

f7rggenetik (Beautiful Fur Animals and Their Colour Genetics).

Scientifur, Hillerbd, Denmark. 271 pp. (in Swedish).

Nieminen, P., Asikainen, J., Hyv7rinen, H., 2001. Effects of seasonality andfasting on the plasma leptin and thyroxin levels of the raccoon dog

(Nyctereutes procyonoides) and the blue fox (Alopex lagopus). J. Exp.

Zool. 289, 109–118.

Nieminen, P., Mustonen, A.-M., Asikainen, J., Hyv7rinen, H., 2002.

Seasonal weight regulation of the raccoon dog (Nyctereutes procyo-

noides): interactions between melatonin, leptin, ghrelin, and growth

hormone. J. Biol. Rhythms 17, 155–163.

Nixon, A.J., Ashby, M.G., Saywell, D.P., Pearson, A.J., 1995. Seasonal

fiber growth cycles of ferrets (Mustela putorius furo) and long-term

effects of melatonin treatment. J. Exp. Zool. 272, 435–445.

Parkanyi, V., Zeman, M., Tocka, I., Mertin, D., Dravcoca, E., 1993. Effect

of melatonin on change of winter fur and spermatogenesis in male polar

foxes. Scientifur 17, 17–23.

Prestrud, P., Nilssen, K., 1992. Fat deposition and seasonal variation in

body composition of arctic foxes in Svalbard. J. Wildl. Manage. 56,

221–233.

Prestrud, P., Nilssen, K., 1995. Growth, size, and sexual dimorphism in

arctic foxes. J. Mammal. 76, 522–530.

Pulliainen, E., 1993. Alopex lagopus (Linnaeus, 1758)-Eisfuchs. In: Stubbe,

M., Krapp, F. (Eds.), Handbuch der S7ugetiere Europas, Raubs7uger-Carnivora (Fissipedia), Teil I: Canidae, Ursidae, Procyonidae, Muste-

lidae 1. Aula-Verlag, pp. 195–214.

Smith, A.J., Clausen, O.P.F., Kirkhus, B., Jahnsen, T., Mbller, O.M.,

Hansson, V., 1984. Seasonal changes in spermatogenesis in the blue fox

(Alopex lagopus), quantified by DNA flow cytometry and measurement

of soluble Mn2+-dependent adenylate cyclase activity. J. Reprod. Fertil.

72, 453–461.

Smith, A.J., Mondain-Monval, M., Mbller, O.M., Scholler, R., Hansson, V.,

1985. Seasonal variations of LH, prolactin, androstenedione, testoste-

rone and testicular FSH binding in the male blue fox (Alopex lagopus).

J. Reprod. Fertil. 74, 449–458.

Smith, A.J., Mondain-Monval, M., Andersen Berg, K., Simon, P., Forsberg,

M., Clausen, O.P.F., Hansen, T., Mbller, O.M., Scholler, R., 1987.

Effects of melatonin implantation on spermatogenesis, the moulting

cycle and plasma concentrations of melatonin, LH, prolactin and

testosterone in the male blue fox (Alopex lagopus). J. Reprod. Fertil. 79,

379–390.

Szeleszczuk, O., 1992. The effect of management system on the

developmental changes in the testes and testosterone level in the blood

sera of blue foxes (Alopex lagopus L.). Nor. J. Agric. Sci. (Suppl 9),

128–136.

Tomasi, T., 1991. Utilization rates of thyroid hormones in mammals. Comp.

Biochem. Physiol. A100, 503–516.

Tveit, B., Larsen, F., 1983. Suppression and stimulation of TSH and thyroid

hormones in bulls during starvation and refeeding. Acta Endocrinol.

103, 223–226.

Underwood, L.S., Reynolds, P., 1980. Photoperiod and fur lengths in the

arctic fox (Alopex lagopus L.). Int. J. Biometeorol. 24, 39–48.

Vriend, J., 1983. Evidence for pineal gland modulation of the neuro-

endocrine–thyroid axis. Neuroendocrinology 36, 68–78.

Xiao, Y., 1996. Seasonal testicular and moulting cycles in the adult male

raccoon dog (Nyctereutes procyonoides) and the effects of melatonin

implants. PhD thesis, University of Kuopio, Kuopio, Finland.

effects of fasting and exogenous melatonin on annual rhythms

Documents