Changes in Milk Composition of the Egyptian Fruit Bat, Rousettus aegyptiacus (Pteropodidae), during Lactation

CHANGES IN MILK COMPOSITION OF THE EGYPTIAN FRUIT BAT, ROUSEITUS AEGYPTIACUS (PTEROPODIDAE),

DURING LACTATION

CARMI KORINE AND ZEEV ARAD

Department of Biology, Technion-Institute of Technology, Haifa 32000 Israel Present address of CK: Smithsonian Tropical Research Institute, P. O. Box 2072, Balboa, Panama

Composition of milk of the free-ranging Egyptian fruit-bat (Rousettus aegyptiacus) was studied from early to peak lactation. Carbohydrates were the main component during early lactation and fats were the main component in milk from mid- to peak lactation. Dry-matter and energy content of the milk increased during mid-lactation and stabilized at peak lac­tation. Increases were a consequence of increased fat content, but carbohydrate and protein contents remained relatively stable. We conclude that the milk composition of R. aegyptia­cus is related to diet and frequent suckling by young. We suggest that the inability of R. aegyptiacus to produce concentrated milk of high fat and energy contents and a low protein content is compensated for by prolonged lactation.

Key words: Rousettus aegyptiacus, fruit bat, lactation, milk composition

Although bats are the second largest or­der of mammals (920 species), little is known about their milk composition. The composition of milk is known for only 10 microchiropteran species, and changes in milk composition during lactation are known for three species of free-ranging in­sectivorous bats (Kunz et al., 1995), and one species of free-ranging omnivorous bat (Stem et al., 1997). The former study shows that milk is characterized by relatively high levels of protein and fat, but carbohydrate levels are relatively low. Kunz et al. (1995) suggested that phylogenetic variations in milk composition may be related to length of foraging bouts, access to water, and fre­quency of suckling. An increase in fat, dry matter and energy contents was reported for the greater spear-nosed bat, Phyllostomus hastatus (Stem et al., 1997). Stem et al. (1997) proposed that individual variations in foraging time, milk composition, and milk yield may account for large differ­ences in reproductive success of females of this species.

Information on milk composition of me­gachiropteran bats is limited to 2 of 173

Journal of Mammalogy, 80(1):53-59, 1999 53

species-the epauletted fruit bat, Epomo­phorus wahlbergi (Quicke et aI., 1984) and the grey-headed flying-fox, Pteropus poli­ocephalus (Messer and Parry-Jones, 1997). Quicke et aI. (1984) used a single sample for which the stage of lactation was not known, and six of 13 bats were lactating females in captivity. Messer and Parry­Jones (1997) sampled milk of captive fe­males at different times during the lactation period; milk composition did not change significantly throughout lactation. The di­lute low-fat milk of P. poliocephalus is pos­tulated to be associated with frequent feed­ing bouts of young.

Comparisons of milk composition be­tween insectivorous and frugivorous bats suggest that diet may contribute to relative­ly large differences. Milk of those few fru­givorous species studied is characterized by low fat, protein, and dry-matter contents and a high carbohydrate content; milk of insectivorous bats shows the opposite trend (Jenness and Studier, 1976; Kunz et al., 1983, 1995; Messer and Parry-Jones, 1997; Quicke et al., 1984; Stem et aI., 1997). However, mineral composition of the milk

54 JOURNAL OF MAMMALOGY Vol. 80, No.1

may reflect not only diet but also postnatal growth rates and need for water conserva­tion in various species (Stem et al., 1997; Studier and Kunz, 1995; Studier et al., 1995).

Megachiropteran bats are relatively large and feed mainly on ripe fruits, but plant ma­terial such as pollen, nectar, and leaves may be eaten. Adult females generally have one or two young per year, depending on the reproductive cycle, and a long gestation pe­riod; young bats are totally dependent on their mothers until weaned because they lack the ability to fly. Lactation may last 2-6 months, but young may depend on their mothers for as long as a year (Kulzer, 1979; Kurta and Kunz, 1987; Pierson and Rainey, 1992).

We assessed milk composition of a free­ranging megachiropteran bat, Rousettus ae­gyptiacus during known stages of lactation. R. aegyptiacus is the only fruit bat in Israel, common in urban areas and nature reserves. The bat feeds mainly on fleshy fruits, in particular on introduced figs (Ficus), but also may feed on leaves and pollen of other plants (Korine, 1996). R. aegyptiacus is bi­modally polyestrous in Israel, with a single young born after a 4-month pregnancy in early spring (March-April) and late sum­mer (August-September) (Makin, 1990). The offspring is totally dependent on the mother for the first 6 weeks. According to Noll (1979), eyes of captive R. aegyptiacus open after 9 days and thermoregulatory ca­pacity is achieved fully after 21 days. Dur­ing the first days of life, the female carries the young to feeding sites and probably suckles it on demand. Peak lactation occurs after 6 weeks when the young develops its ability to fly. Thereafter, the offspring typ­ically follows its mother to feeding sites. The offspring is not fully weaned until 3 months of age.

Our main objective was to determine changes in milk composition during the lac­tation period (from early to peak lactation) and establish if those changes were consis­tent with a prolonged-lactation strategy.

MATERIALS AND METHODS

To define the stage of lactation, one ideally should use predetermined postnatal growth curves. In the case of R. aegyptiacus, a postnatal growth curve was only available for captive an­imals (Noll, 1979) and therefore was not used in our present study. Length of forearm, as a mea­sure of relative age and lactation stage of young, was available from beginning to end of the lac­tation period for a large sample of R. aegiptia­cus, based on a capture-recapture study by Ma­kin (1990). Young with forearm lengths of ::546 mm were ::52 weeks old, corresponding to early lactation. Young with forearm lengths >47 mm, still caught on the mother, were from 2 to 6-7 weeks old, corresponding to mid-lactation. Young with forearm lengths >69 mm can fly on their own (Makin, 1990) but continue suckling for the next 5 weeks until they reach a forearm length of 84-85 mm. Females captured without offspring could be either at peak or toward the end of lactation; those with highly developed mammary glands that yielded milk easily and with a bald area surrounding large and soft nip­ples, were defined to be in peak lactation. In contrast, females that were toward the end of the lactation period had shrunken mammary glands from which only small amounts of milk could be expressed (which was insufficient for analy­sis), small and hard nipples, and some new growth of hair around nipples.

Milk was collected between August-October 1995, corresponding to the second reproductive cycle of R. aegyptiacus (Makin, 1990) in the Ru­pin Cave (Haifa, Israel). Lactating females with or without young were captured by mist-netting between 0300-0500 h, after bats returned from feeding. Following capture, each lactating fe­male was placed in a small cage and transferred to the laboratory (ca. 5-min drive) where mea­surements and milking were performed. While measuring and milking, bats were placed in a temperature-controlled room (22°C ± 0.2 SE, 60 ± 3% relative humidity, 12D: 12L) that mim­icked the day roost environment to avoid water loss. All females were placed together to de­crease physiological stress and maintain the so­cial environment.

Body mass and forearm length were measured separately in females and offspring. Young were removed from their mothers by placing a cotton swab dipped in sugar water at the comer of the offspring's mouth and applying gentle pressure

1999 KORINE AND ARAD-MILK COMPOSITION OF ROUSETTUS 55

until the young opened its mouth. Care was taken during this procedure to avoid injury to either young or mother. Females and offspring were weighed on an electronic balance (± 1 mg). Length of forearm was measured with calipers (±0.5 mm).

Milking procedure.-Milk was collected be­tween 0600-0700 h. First, nipples and surround­ing skin were cleaned with cotton balls soaked in 70% ethanol to avoid contamination. Milk from females with young was collected from only one nipple; milk from females without young was taken from both nipples. Oxytocin was not used in the milking procedure. Milk was expressed manually from nipples and collected with a custom-made pump. Milk samples were transferred to l-rnl Eppendorf tubes; tubes were then frozen at -20°C until analyzed. After sam­pling, females were returned to the temperature­controlled room. Females were milked again be­tween 1600-1800 h to determine if dry-matter content differed during the day. All females were returned to the day roost (Rupin Cave) and released when bats normally emerged from the cave at night.

Milk analysis.-Milk was analyzed for dry matter, total nitrogen, lactose, and fat. Because samples were small, it was necessary to pool many samples of milk, thus providing an ade­quate volume for reliable analysis. Milk of fe­males was pooled for each of the following stages: early, mid-, and peak lactation. Samples were pooled only for analysis of nitrogen, lac­tose, and fat. Dry matter was determined from unpooled samples. To save milk for chemical analyses, dry matter was not determined for each sample. Samples were warmed to 37°C and were vortexed before each analysis.

Dry-matter content was calculated from mass changes using an electronic balance (± 1 mg). Samples of 0.4-0.5 mg of milk were placed in pre-weighed aluminum flasks. Flasks were re­weighed, were transferred to an oven, and were dried for 3 h at 100°C. Samples then were cooled in a desiccator for 20 min and reweighed.

To determine total-nitrogen content, 0.5-mg samples of milk were dried for 3 h at 70°C. The nitrogen content of the dried sample was then determined (FP 228, Leco Analyzer, St. Joseph, MI) using a method based on protein combus­tion at 800°C and measurement of the released nitrogen. Crude protein was calculated as 6.38

times total nitrogen. No correction was made for non-protein nitrogen.

To determine lactose content, 0.1 ml samples of milk were diluted in 10 ml of distilled water and run through a column specific for lactose (Sugar Gap Column, HPLC 1050, Waldbronn, Germany). Although that method detected all saccharides, it quantitatively determined only lactose because of a low resolution for other sac­charides.

The rest of the milk was used to determine fat content. Samples were diluted in 10 ml of distilled water. Following disruption of fat mem­branes with aluminum hydroxide and ethyl al­cohol, each sample was extracted three times with a 4: 1 mixture of diethyl ether and petro­leum ether. The resulting organic phase was dried with sodium sulfate and filtered to a pre­weighed dry flask. That flask was then dried for 2 h at 100°C, cooled in a desiccator, and then redried for a further hour, cooled, and reweigh­ed. Fat content was calculated as the mass loss of the flask.

Energy content was calculated from the milk composition using energetic equivalents of 38.1 kJ/g, 24.68 kJ/g, and 16.74 kJ/g for fat, protein and carbohydrate, respectively (Bondi, 1987). All results are presented as means ± SD. Stu­dent's t-test and one-way analysis of variance (ANOV A) were used to test differences among lactation stages. A value of P < 0.05 was ac­cepted as significant. All analysis were per­formed using SAS procedures.

RESULTS

The 52 females of three different lacta­tion stages included 14 females in early lac­tation, 23 females in mid-lactation, and 15 females without young, at peak lactation. Dry-matter content of milk at both early morning and late afternoon (Table 1) in­creased significantly (52%) from early to mid-lactation and stabilized at that level during peak lactation. Dry-matter content increased significantly (26%) from early morning to late afternoon only at peak lac­tation. That increase may have reflected an increase in one of the ingredients of the milk, or have been a result of a decrease in water content.

An increase in dry matter content during

56 JOURNAL OF MAMMALOGY Vol. 80, No.1

TABLE I.-Mean (±SD) dry-matter content of milk of the Egyptian fruit-bat that was collected in the early morning and late afternoon from early to peak lactation. Early lactation, female with young whose forearm length was s46 mm; mid-lactation, female with young whose forearm length was >46 mm; peak lactation, female without a young, according to nipple morphology. Different super­script letters within a row indicate significant differences between lactation stages; the only signifi­cant difference between sampling times occurred during peak lactation (P < 0.05).

Early lactation Mid-lactation Peak lactation

Time of day X SD n X SD n X SD n

Early morning 12.78b 1.52 6 19.40' 2.94 12 19.60' 2.83 13 Late afternoon 11.98b 1.87 5 21.07' 3.67 6 24.75' 4.77 8

development also was seen when dry-mat­ter content was plotted against the body mass of the young from early to mid-lac­tation (Fig. 1). The dry-matter content of milk (Y) and body mass of young (X) were correlated (Y = 0.41X - 0.16, r = 0.74, n = 18, F = 19.51, P < 0.001). Thus, as lactation progresses, offspring receive more solids from milk.

Milk composition and energy content from early lactation to peak lactation are summarized in Table 2. Statistical tests

26

• 24 •

22

• • ~ 20 ~ 'E • • '0 • • 18 Q; • ::: '" E • ~ 16 Cl •

• 14

• • 12

• •

10 ---1.

30 35 40 45 50 55

Body mass of young (g)

FIG. I.-The relationship between dry-matter content of milk and body mass of young in the Egyptian fruit-bat, Rousettus aegyptiacus.

were not performed because results were based on small samples of pooled data. The HPLC analysis indicated that lactose was indeed the only saccharide present in the milk of R. aegyptiacus. Therefore, lactose values were considered as total carbohy­drate.

The major component of milk during ear­ly lactation was carbohydrate. From mid- to peak lactation, carbohydrate and protein contents remained relatively constant, while fat content increased. From mid- to peak lactation, fat was the major component in the milk (Table 2). A positive correlation was found between energy and dry matter content of milk [(kJ/g) = 0.322 (% wet mass) - 1.61, r = 0.92, n = 5, F = 4.20, P < 0.02]. That correlation may have re­flected the increase in fat content from early to peak lactation. Fat and dry-matter con­tents also were correlated [(% wet mass) = 0.7987 (% wet mass) - 7.56, r = 0.91, n = 5, F = 3.95, P < 0.02]. Protein and car­bohydrate contents were not correlated with dry-matter content.

DISCUSSION

Our results suggest that milk composition of the Egyptian fruit-bat changes from early to peak lactation. Carbohydrate and protein contents remained relatively stable, while fat content increased from early to peak lac­tation. Dry-matter and energy contents were low during early lactation, but both in­creased throughout lactation. The energy

1999 KORINE AND ARAD-MILK COMPOSITION OF ROUSE1TUS 57

TABLE 2.-Mean (±SD) composition, in percentage of wet mass, and energy content of milk from R. aegyptiacus from early lactation to peak lactation. Number of samples tested is followed by number of bats milked in parentheses.

Protein Carbohydrate Fat Energy (KJ/g)

Lactation period n X SD X SD X SD X SD

Early lactation 1(14) 2.2 5.7 2.7 2.52 Mid-lactation 2(23) 2.33 0.04 6.05 0.07 7.05 1.06 4.27 0.38 Peak lactation 2(15) 8.95 0.63 5.05 0.11 "

6 2.9 0.62 5.51 0.51

, Mean fat content was added to each of the six individual measurements of protein and lactose to yield the energy content.

content was mainly influenced by fat con­tent.

The protein and carbohydrate content of milk during peak lactation were roughly comparable with those for E. wahlbergi and P. poliocephalus (Messer and Parry-Jones, 1997; Quicke et aI., 1984). Fat and dry­matter contents were higher than in those two species. These differences in milk com­position may be attributed to the sampling procedure (definition of lactation stages and chemical analysis of the milk) and the spe­cific diet of each species. However, our findings should be interpreted with caution: first, our definition of stage of lactation was based on three relative ages; second, we an­alyzed pooled samples and therefore indi­vidual variations could not be assessed, nor statistical tests applied.

Milk of R. aegyptiacus at peak lactation was characterized by high carbohydrate (5.5), low nitrogen (2.9), relatively low fat (8.9), and low dry-matter and energy con­tents compared with the milk (range of 3.4-4.4, 7.7-10.6, 15.8-25.8, for carbohydrate, nirtogen and fat, respectively) of insectiv­orous bats (Kunz et aI., 1983, 1995). This may be caused by high carbohydrate con­tent and low fat and low protein contents of fruit (Korine et al., 1996; Mattson, 1980) eaten by R. aegyptiacus. The overall data on milk composition in megachiropteran bats suggest that the maternal diet (i.e., mainly plant material) is reflected in milk composition (Kunz and Stern, 1995; Messer and Parry-Jones, 1997). This conclusion

also holds true for the milk composition of microchiropteran plant-visiting bats such as the common fruit bat, Artibeus jamaicensis (Oftedal and Iverson, 1995), the greater spear-nosed bat, P. hastatus (Stern et al., 1997) and the long-nosed bat, Leptonycte­rus sanborni (Huibregtse, 1966)-the latter two omnivores.

Ben Shaul (1962) found a correlation be­tween milk composition and suckling be­havior. When suckling was based on a reg­ular schedule, milk was concentrated and had a high fat content. Ben Shaul (1962) suggested that the high fat content helped sustain the newborn between feedings be­cause it was energy-rich. By contrast, milk of mammals that suckle on demand is more dilute and with a low fat content (Oftedal, 1984). At early lactation, the young of R. aegyptiacus are helpless and totally depen­dent on their mother (captive bats-Kulzer, 1979; free-ranging bats-Makin, 1990). Thus, milk in this species is provided on demand, and the dilute milk with a low fat content is consistent with Ben Shaul's (1962) hypothesis. As the offspring devel­op, they begin to fly and lactation behavior changes to a frequent schedule. Thus, at this stage milk is more concentrated, and fat content increases to match the development of the young. We conclude that milk com­position and suckling behavior of R. aegyp­tiacus are consistent with Ben Shaul's (1962) hypothesis.

Our results suggest that carbohydrate was the major component in the milk at the be-

58 JOURNAL OF MAMMALOGY Vol. 80, No.1

ginning of lactation. whereas fat was the major component during mid- and peak lac­tation. The switch from carbohydrate to fat may be explained by the rapidly increasing energy requirements of offspring as they begin to fly. or by a limited capacity to use milk lipids at the beginning of lactation.

Our results are in contrast to the findings of Messer and Parry-Jones (1997) for cap­tive P. poliocephalus. They did not find any changes in milk composition. and fat re­mained the main component in the milk throughout the lactation period. Too few studies have been conducted for a pattern to be detected and differences cannot be ex­plained. Further studies on other megachi­ropteran bats are needed to determine if they switch from carbohydrate to fat. as the major energy source in milk and if changes in milk composition during lactation are general characteristics of this suborder. The relatively long lactation period of R. aegyp­tiacus may be explained by the inability to produce concentrated milk of high fat and energy contents and by the low protein con­tent of the milk. These are compensated for by a prolonged lactation period.

ACKNOWLEDGMENTS

This study was supported by the Technion­Institute of Technology. We thank M. Tishel for the milk analysis and D. Korine for her valuable help in the field. T. H. Kunz. E. H. Studier. and one anonymous reviewer provided critical com­ments on earlier versions of this manuscript.

LITERATURE CITED

BEN SHAUL, D. M. 1962. The composition of milk of wild animals. International Zoo Yearbook, 4:333-342.

BONDI, A A 1987. Animal Nutrition. John Wiley & Sons, New York.

HUIBREGTSE, W. H. 1966. Some chemical and physical properties of bat milk. Journal of Mammalogy, 47: 551-554.

JENNESS, R., AND E. H. STUDIER. 1976. Lactation and milk. pp. 201-218, in Biology of bats of the New World Family Phyllostomatidae. Part III (R. J. Ba­ker, J. K. Jones, Jr., and D. C. Carter, eds.). Special Publication, The Museum, Texas Tech University, 10:1-218.

KORINE, C. 1996. Energy, protein and water budgets of the fruit-bat Rousettus aegyptiacus. Ph.D. disser-

tation, Technion-Israel Institute of Technology, Hai­fa, Israel (in Hebrew with English summary).

KORINE, C., Z. ARAD, AND A ARIEL!. 1996. Nitrogen and energy balance of the fruit-bat, Rousettus ae­gyptiacus, on natural fruit diets. Physiological Zo­ology, 69:618-634.

KULZER, E. 1979. Ecology and geographical range in the fruit-eating cave bat Genus Rousettus Gray 1821-a review. Bonner Zoologische Beitrlige, 30: 233-275.

KUNZ, T. H., AND A A. STERN. 1995. Maternal invest­ment and post-natal growth in bats. Symposia of the Zoological Society of London, 67: 123-138.

KUNZ, T. H., M. H. STACK, AND R. JENNESS. 1983. A comparison of milk composition in Myotis lucifugus and Eptesicus fuscus (Chiroptera: Vespertilionidae). Biology of Reproduction, 28:229-234.

KUNZ, T. H., O. T. OFrEDAL, S. K. ROBSON, M. B. KRETZMANN, AND C. KIRK. 1995. Changes in milk composition during lactation in three species of in­sectivorous bats. Journal of Comparative Physiol­ogy, B. Biochemical, Systematic, and Environmental Physiology, 164:543-551.

KURTA, A, AND T. H. KUNZ. 1987. Size of bats at birth and maternal investment during pregnancy. Sym­posia of the Zoological Society of London, 57:79-106.

MAKIN, D. 1990. The biology of the fruit-bat Rousettus aegyptiacus in Israel. Ph.D. dissertation, Tel Aviv University, Israel (in Hebrew with English summa­ry).

MATTSON, W. J. 1980. Herbivory in relation to plant nitrogen content. Annual Review of Ecology and Systematics, 11: 119-161.

MESSER, M., AND K. PARRy-JONES. 1997. Milk com­position in the Grey-headed fiying-fox, Pteropus po­liocephalus (Pteropodidae: Chiroptera). Australian Journal of Zoology, 45:65-73.

NOLL, U. G. 1979. Postnatal growth and development of thermogenesis in Rousettus aegyptiacus. Com­parative Biochemistry and Physiology, A. Compar­ative Physiology, 63:89-93.

OFrEDAL, O. T. 1984. Milk composition, milk yield and energy output at peak lactation: a comparative re­view. Symposia of the Zoological Society of Lon­don, 51:33-85.

OFrEDAL, O. T., AND S. J. IVERSON. 1995. Phylogenetic variation in the gross composition of mammalian milks. pp. 749-788, in Handbook of milk compo­sition (R. G. Jensen, ed.), Academic Press, New York.

PIERSON, E. D., AND W. E. RAINEY. 1992. The biology of fiying foxes of the genus Pteropus: a review. pp. 1-17, in Proceedings of the Pacific island fiying fox conservation conference (D. E. Wilson and G. L. Graham, eds.). Biological Report, Fish and Wildlife Service, United States Department of the Interior, 90:1-176.

QurCKE, G. v., S. SOWLER, R. K. BERRY, AND M. GED­DES. 1984. Composition of mammary secretion from the epauletted fruit bat, E. wahlbergi. South African Journal of Science, 80:481-482.

STERN, A. A., T. H. KUNZ, E. H. STUDIER, AND O. T. OFrEDAL. 1997. Milk composition and lactation out­put in the greater spear-nosed bat, Phyllostomus has-

1999 KORINE AND ARAD-MILK COMPOSITION OF ROUSETTUS 59

tatus. Journal of Comparative Physiology, 167:389-398.

STUDIER, E. H., AND T. H. KUNZ. 1995. Accretion of nitrogen and minerals in suckling bats, Myotis veli­fer and Tadarida brasiliensis. Journal of Mammal­ogy, 76:32-42.

STUDIER, E. H., S. T. SEVICK, D. E. WILSON, AND A. P.

BROOKE. 1995. Concentrations of minerals and ni­trogen in milk of Carollia and other bats. Journal of Mammalogy, 76: 1186-1189.

Submitted 3 October 1997. Accepted 22 May 1998.

Associate Editor was Robert K. Rose.