The ontogenesis of skin and organ characteristics in the Syrian golden hamster

Exp Toxic Pathol 1992; 44: 113-124 Gustav Fischer Verlag lena

Central Animal Laboratory, University of Essen, Essen, Germany

The ontogenesis of skin and organ characteristics in the Syrian golden hamster ffi. Ontogenetic and intraspecific allometry for strain and sex as well as body weight- and age-dependent correlations K. MILITZER, D. BUTTNER and E. MOOG With 8 figures and 2 tables Received: December 23, 1990; Accepted: January 14, 1991

Address for correspondence: Priv-Doz. Dr. med. vet. KLAUS MILITZER, Zentrales Tierlaboratorium am Universitatsklinikum, HufelandstraBe 55, D- W -4300 Essen 1, Deutschland Key words: skin, ontogenesis; ontogenesis, skin; Syrian golden hamster; allometric relations; body weight relationships; age-dependent correlations; sex correlations; strain correlations

Summary

The allometric relations, i.e. bodily characteristics, body weight-relationships as well as age-dependent organ and biochemical data were studied in a total of 464 golden hamsters of both sexes of an acromelanic white inbred and agouti coloured outbred strain. In the 3 age sections studied (I = day 1-20, II = day 25-100, III = day 110-365) the body weight- and age-dependent relations were found to be altered between and within the various characteristics. The body weight correlations predominated in the case of organ weights and skin muscle thickness. By contrast, age correlations were seen above all in the skin compartments with cyclic growth, hair follicle density and reticular thickness in the age sections I and II. Papillary thickness showed a positive relation to body weight and age after weaning, but no long-term relations were observed with both plasma insulin and blood glucose levels. The allometric behaviour of skin compartments could be explained particularly by thermoregulatory and "geometric" similarities and that of the organs by metabolic and other similarities. Most sex and strain differences in the absolute data, except for the kidney and adrenal weights, disappeared on allometric analysis and were thus mainly due to differences in body weight. For long-term toxicological investigations, the documentation of age and body weight as well as the determination of ontogenetic allometry is a "must".

Introduction Ontogenetic studies within the framework of pathological or toxicological investigations usually consider age of the species tested to be the most important parameter. This has already been

described for the golden hamster in part I and II (MILITZER et al. 1990, 1991). However, it appears doubfful whether all characteristics are indeed exclusively subjected to age-dependent developments during maturation processes. Then, finally, body weight itself changes considerably during ontogenesis and not always linearly over longer periods of life. Moreover, the body weight is a characteristic that, in contrast to age, can be readily determined. A suitable method of evaluation is the use of allometry, i. e. the mathematical determination of body weightbodily characteristic relations (HUXLEY 1932). With the aid of this, it is possible to show quantitative relations between organ characteristics and other parameters, in the main, weight, length and time (GUNTHER and MORGADO 1982). The allometric results obtained during the course of development provide a good basis for the comparison of data from different groups of strain, sex and age as well as other experimental studies. In this part (III) the correlations of all previously described characteristics of Syrian hamsters with age and body weight are now given as well as those between these characteristics. At the same time, it has been determined whether comparable growth strategies for the characteristics and age periods can be observed. The comparison between intra-, interspecific and ontogenetic allometric data should reveal what common factors exist with respect to growth behaviour of the laboratory animals.

Material and Methods A total of 464 Syrian hamsters, males and females, was studied. Approximately half the animals were from an agouticoloured outbred stock (HAN: AURA) and the rest from an acromelanic white inbred strain (Bio l.5). Details concerning Exp. Toxic. Pathol. 44 (1992) 3

113

animals, methodology as well as data analysis are given in parts I and II (MILITZER et al. 1990, 1991). The following characteristics were determined for 25 age groups of maximally 5 hamsters between days 1- 365 of life: Epidermal, papillary, reticular, and skin muscle (panniculus carnosus) thickness, hair follicle density, inguinal adipocyte diameter, body, liver, adrenal, kidney, testes and ovarian weight, blood glucose and plasma insulin levels. The statistical analysis is based on the individual data from each animal. The power of the relationship and the type of association between the various characteristics were studied by the means of correlation and regression analysis. The allometric regression line is calculated according to HUXLEY (1932): log Y = log a + b . log X (Y = characteristics, a = weight coefficient, b = weight exponent, X = body weight). Intraspecies allometry was calculated separately for each strain and sex over the overall experimental period (days 1-365). Ontogenetic allometry was determined each of the age sections (I = day 1-20, II = day 25-100, III = day 110-365). The significance of the slope b of the allometric coefficient against the null hypothesis was determined by the t-test (SACHS 1984).

more age - rather than body weight dependent; in adult hamsters, however, the negative epidermal thickness - body weight relationship prevails. Papillary thickness shows a positive correlation both to body weight as well as to age during the overall experiment (fig. 1). On subdivision into the 3 age sections (table 1), however, there was a negative correlation both with age and body weight during the infancy phase (I). By contrast, distinctly positive relations to body weight were observed in the juvenile (lI) as well as adult (III) period. x-Variable: Age

o -.5

+.5

-1.:.-..L..L----~-:;....=====EL~+1

x-Variable: Body weight

Results Correlations From the age profiles already described (MILITZER et al. 1990, 1991) it was evident that most characteristics were altered with age in the golden hamster. Fig. 1 shows the dependences existing for the various characteristics during the overall experiment with respect to either age or body weight. Positive correlations are predominant in both, but higher coefficients are found in the case of body weight instead of age. Epidermal - and reticular thickness as well as hair follicle density show negative correlations. No definite correlations could be observed for the biochemical parameter blood glucose and plasma insulin levels (fig. 1). The overall results on all animals investigated during the course of the experiment, however, also mask important differences in the age groups, since a number of characteristics and ontogenetic phases change in relation to age and body weight as is also the case for the orientation of correlation gradients. The significance of simple and partial correlation analysis in table 1 is explained by the following example. For the characteristic "follicle density in age section I", a simple positive coefficient (r = 0.72, 0.65) is seen with respect to age (AGE) and body weight (BW). Both the basic variables AGE and BW, on their part, are closely correlated with one another (r = 0.94) and together influence the relation to hair follicle density. By excluding body weight influences, the partial correlation between follicle density and age is largely retained (CHAR. to AGE: r = 0.41). By contrast, the correlation with body weight is lost when age effects are not taken into consideration (CHAR. to BW: r = -0.10). Thus, follicle density shows a stronger positive age than body weight relation in the postnatal period. No distinct relations to both variables are seen in age section II, and in the case of age section III, again age relations predominate for follicle density. The behaviour of epidermal thickness in juvenile animals is 114

Exp. Toxic. Pathol. 44 (1992) 3

o

Fig. 1. Relation between the individual parameters of the overall experiment and the 2 variables, age and body weight, for all golden hamsters. The correlation coefficient r obtained by logarithmic calculation is shown as an angle. The negative and the positive r are given on the left and right side respectively. - - - - - - r = not significant; ALT = age, E = epidermal thickness, FO = follicle density, FT = adipocyte diameter, G = glucose level, H = testes weight, IN = insulin level, KM = body weight, L = liver weight, M = skin muscle thickness, N = kidney weight, NB = adrenal weight, 0 = ovarian weight, P = papillary thickness, R = reticular thickness.

The reticular layer decreases in thickness with increasing age; concomitantly, the more heavy hamsters in adult age, however, show an increased reticular thickness than lighter animals. This is indicated in section III by the negative correlation with age and a slight, positive one to body weight (table 1). Also skin muscle thickness is increasingly more affected by body weight than age correlations during ontogenesis. The adipocyte diameter of adult animals is mainly, on one hand, positively related to body weight, but, on the other negatively correlated with age. .

Table 1. Simple (simp.) and partial (part.) linear correlation coefficients for histometrical and biochemical characteristics of all data in the experiment. Characteristics (CHAR.)

Coeff. r

CHAR. to AGE I

CHAR. to BODY WEIGHT (BW)

II

III -.27* -.28* -.09 -.04

II

III

I

II

III

.65* -.10

-.30* -.ll

-.02 .04

.94* .89*

.82* .80*

.21 * .22*

-.63* .12

.08 -.23*

-.25* -.23*

.94* .90*

.82* .83*

.21 * .20

.28* .22*

-.31 * -.05

.63* .31 *

.42* .38*

.94* .93*

.82* .70*

.21 * .ll

-.42* -.47*

.23* .00

.13 .25*

.94* .93*

.82* .82*

.21 * .30*

.60* .27*

.31 * .34*

.94* .94*

.82* .72*

.22* .26*

.61 * .29*

.31 * .39*

.82* .72*

.21 * .33*

Follicle density

simp. part.

.72* .41 *

-.29* -.08

Epidermal thickness

simp. part.

-.71 * -.42*

.25* .33*

Papillary thickness

simp. part.

-.31 * -.07

.60* .18

Reticular thickness

simp. part.

.25* .09

Skin muscle thickness

simp. part.

-.48* -.44*

Adipocyte diameter

simp. part.

Glucose level

simp. part.

.19 .39

-.08 -.04

-.25* -.23

Insulin level

simp. part.

.45* -.38*

-.02 .03

-.10 -.08

Age sections I

= day

1-20, II

= day 25-100,

III

-.03 .01 .58* .19

-.08 -.16

.58* .18

-.27* -.37*

= day

-.36* .30*

.06 -.35 .59* .55*

-.05 -.04

-.07 .00

-.11 -.06

.94* .95*

.82* .81 *

.21 .19

-.04 -.05

-.12 -.10

.94* .93*

.82* .82*

.21 * .20

110-365; *significantly different from 0, p"5 0.01; - no data available.

In the case of glucose level, no conclusive relations either to age or body weight are to be seen in all three age periods. During the postnatal period (I), the insulin level shows a distinct relation to both basic variables but varying orientation of the regression lines for body weight and age (table 1). Distinctly positive relations to body weight are predominantly seen for all organ characteristics such as liver, kidney, adrenals, testes and ovarians as well as skin muscle thickness and adipocyte diameter.

Ontogenetic allometries Age-dependent ontogenetic allometries for all characteristics are given in table 2. By considering the allometry exponent b, complex but orientatingly varying body weight relationships are seen in the histometric characteristics. The follicle density and thickness of the stratum reticulare always develop more slowly than body weight (hypo allometry according to CALDER 1984) after the infancy phase either with no or negative relation to body weight (fig. 2; table 2). The epidermal-, papillary- and skin muscle thickness in baby hamsters with a body weight from 7 to 20 g grows more slowly than the body weight (figs. 3, 4 above; table 2). A negative correlation for these parameters is observed for body weights from 26 to 36 g (subadult phase II). Even later, the thickness of the stratum papillare and skin muscle develops hypoallometrically, i.e. smaller than the body weight. In the case of epidermal thickness, the allometric regression line falls off to the right in both the age sections I and III (fig. 3 above). 1*

AGE to BW

Hypoallometric relations apply to the adipocyte diameter in the subadult phase (II) as well as in the fully-grown hamsters (III) (fig. 4 below; table 2). Liver weight develops relatively more slowly than the body weight in sub adult hamsters (II). At the adult stage (III), an identical growth rate of organ and body weight (b = 1.0), isometry, can be seen (fig. 5 above). The increase in kidney weight occurs more slC!wly in both age periods than that of the body weight (hypoallometry: fig. 5 below). Hyperallometry becomes apparent in the sub adult phase from days 25 to 100 (II) during the development of the adrenals (fig. 6 above, table 2) and testes (fig. 7 above). Thereafter the allometric lines for testes, however, flatten off markedly to b = 0.59. The lines for the ovaries are approximately isometric (fig. 7 below). Other biochemical characteristics from table 2, a definite body weight dependent relation can only be observed for the insulin level in the juvenile phase (I) of the hamsters (fig. 6, below).

Allometric sex - and strain differences After adjustment of allometric data to uniform body weights, the allometric gradients of the stratum papillare for female hamsters were shifted towards the higher layer thicknesses in comparison to the males (fig. 8, above). Thus, the sex-dependent thickness of the stratum papillare in the lighter female hamsters is, on average, relatively larger than in the heavier males. The adrenals from male hamsters of both strains are sexspecifically heavier than those from female animals in age sections II and III after allometric evaluation (fig. 8, below). Exp. Toxic. Pathol. 44 (1992) 3

115

Table 2. Allometry coefficient a and exponential b ±

Sb for histometrical, morphometrical (weight) and biochemical characteristics of 2 hamster strains, correlated with x-variable body weight (kg).

Characteristics

Dimensions

Age section l )

Age in days

Coeff. a

Exponent b ± Sb

Follicle density

number/ 736 [lm

I II III

1-20 25-100 110-365

1640.3 7.8 14.0

.71 ± .06]] -.46± .11 -.05 ± .17

Epidermal thickness

[lm

I II III

1-20 25-100 110-365

5.3 12.6 8.9

-.25± .02] .02± .03] -.17 ± .05

Papillary thickness

[lm

I II III

1-20 25-100 110-365

58.9 253.1 264.0

-.06± .03]] .45 ± .04 .40± .07

Reticular thickness

[lm

I II III

1-20 25-100 110-365

573.0 134.2 194.2

.18 ± .03] -.05 ± .09 .27± .14

Skin muscle thickness

[lm

I II III

1-20 25-100 110-365

12.6 162.8 121.6

-.18±.03]] .51 ± .05 .39± .09

Adipocyte diameter

[lm

II III

25-100 110-365

117.7 122.7

.33 ± .03] .22± .05

Liver weight

kg

II III

25-100 110-365

.02 .05

.74± .03] 1.08 ± .05

Kidney weight

kg

II III

25-100 110-365

.003 .004

.46 ± .02] .64± .04

Adrenal weight

kg

II III

25-100 110-365

.0003 .0002

1.23 ± .08 1.23 ± .16

Testes weight

kg

II III

25-100 110-365

.25 .01

2.11±.21] .59± .14

Ovarian weight

kg

II III

25-100 110-365

.0005 .0003

1.08±.13 .84± .14

Glucose level

mg/dl

I II III

1-20 25-100 110-365

157.4 151.0 110.0

.05 ± .06 -.03 ± .06 -.10± .10

Insulin level

mg/dl

I II III

1-20 25-100 110-365

416.8 106.4 20.5

.54± .lj .09± .17] -.50± .28

I = day 1-20; II = day 25-100; III = day 110-365;

1 significant differences (t-test, p :5 0.05) between age sections.

For the remaining characteristics testes and ovarian weight as well as epidermal and skin muscle thickness, all previously described absolute sex differences (MILITZER et al. 1990, 1991) have been found to solely express body weight differences. Even the strain differences in the absolute data (between the agouti-coloured strain HAN: AURA and acromelanic white hamsters Bio 1, 5), did not stand the test of allometry in any single case.

116

Exp. Toxic. Pathol. 44 (1992) 3

Discussion Up to now, ontogenetic allometries have seldom been carried out systematically on several characteristics in small rodents. The varying organ-body weight relation during the course of ageing has repeatedly been described for individual characteristics but never explained on a causative basis (SCHAFER et al. 1970, TRIEB et al. 1976). However, our expansive material obtained concurrently presents itself for such attempts at explanation. Thus, the allometric exponents deter-

NE

.....E

~r-----------------------------------------'

o

~

~

iO

c ~

GI

300

;g :E 100

30

1.5

50

10

400 Body weight (g)

E .:. In In GI

.,.c

(J

5 :u :;

450

'

..

I

•

••

•

.... • <'0

(J

ia:

......

'0

I) ':."+

eO.",: '!

I ••

..-:::~: ... :

.. : '" .O:'!'':'.

II --- ~·.. -.-.-~U~ ! ..... ~.~::~:,. . . '. '.' .~-f.f."':-."". (

120

•

•

•

•

0°

Fig.2. Hair follicle density (above) and thickness

1I.1.... -::~ ...::· ;

... :=;~~:~. ;. .-.' ..: .

30~----------~--------~----------~ 1.5

10

400

50

of the reticular layer (below) in the hamster's skin as a function of body weight. Allometric linear regression for the age periods I (day 1-20), II (day 25-100) and III (day 110-365) with respect to the data from all hamsters. - - significant (P:2i 0.05), - - - - - - not significant.

Body weight (g)

mined for golden hamsters (table 2) are comparable to empirical b-values from the literature. Agreements with the following allometric models can then be interpreted as an expression of certain growth strategies: "Geometric" similarities (JURGENS 1989, p. 32) that take distances and surface relations into account, metabolic similarities that relate metabolically induced alterations to organs with those of by weight (REFINETTI 1989), thermoregulatory similarities that describe physiological or structural changes as a means for maintaining homeothermia (WEBB and MCCLURE 1989). Apart from these, there exists a multitude of other possible comparisons which can be alluded to only marginally. All attempts at explanation only cover partial aspects since developmental processes are always characterized by a high degree of complexity (JURGENS 1989, p. 29).

Allometry and growth strategies 1. Skin compartmenK' For the skin compartments the change from surface to body weight correlated relationships proves to be the decisive

developmental strategy during the maturation from the infant to the subadult stage in the golden hamsters. The skin thickness, e.g. the main compartments such as epidermis, stratum reticulare and skin muscle, was correlated to the surface after birth and which can be explained in terms of an insulating effect. After all, small animals possess a relatively large body surface area which must be considered to be the critical parameter during this age period. In the heavier hamsters with a relatively small surface (SCHMIDT-NIELSEN 1984, p. 200), however, the weight-dependent relations for skin thickness predominate later.

In the postnatal period (1) the course of allometric gradients for epidermal, skin muscle and reticular thickness according to SCHMIDT-NIELSEN (1984, p. 197) corresponded to thermoregulatory constants such as heat production and surfacerelated heat conduction and insulation (Epidermis: b = -.25, skin muscle: b = -.18, stratum reticulare: b = .18). By contrast, the papillary layer, insignificant with respect to thickness, showed no allometric or thermoregulatory relationship. The low postnatal hair - and follicle density did not yet protect sufficiently from heat losses, so that only an allometric relation to heat-producing metabolism could be realized (b = .75: GUNTHER 1971). Exp, Toxic. Pathol. 44 (1992) 3

117

E

.=. (/) (/)

CD .><

c:

35

:s CJ

OJ

E

~ .Q. "0

w

12

5~----------~----------~----------~ 1.5

10

50

400

Body weight (g)

E

.=..

.

CD c: .><

200

CJ

S

~

~ Ci.

as

Q.

75

Fig.3. Epidermal thickness (above) and thickness

~~--------------------~----------~ 1.5 10 50 400

of the papillary layer (below) in hamster's skin as a function of body weight (for details see fig. 2).

Body weight (g)

During the juvenile development (II) of hair follicles, papillary and skin muscle thickness, thermoregulatory allometries continued to dominate. However, relationships to body weight dependent constants of heat conduction and capillary blood flow but not to body surface area could be shown (b = .45 to .51: CALDER 1984, p. 118, SCHMIDT-NIELSEN 1984, p. 138). In contrast to that of the stratum reticulare the slight epidermal thickness was of no account for heat insulation. Even in the case of follicle density, which in this phase of development is better represented by the reticular thickness, no allometric correlation was able to be observed. The diameter of the adipocytes satisfied typical "geometric" or "mechanical" criteria of similarity with a length-body weight correlation of b = .33 (JURGENS 1989, GUNTHER 1971). Such simple length-weight correlations also prevailed in the adult stage (III) for such major skin compartments as the papillary -, reticular - and skin muscle thickness. Only the outermost layer, the epidermal thickness and the size of the adipocytes, which is a measure of the thickness of the insulating layer, were still keeping in with the allometric constant for heat conduction per unit area (b = .17 to .20). By contrast, all body weight relations were lacking for hair follicle numbers in hamsters older than 110 days. After the 118

Exp. Toxic. Pathol. 44 (1992) 3

completion of fur formation, the body temperature could apparently be regulated more flexibly through alterations in intermediary metabolism and blood circulation than by the coupling of hair and body surface development. The main strategy of the growth of skin compartments in the juvenile phase was directed towards optimal heat insulation and formation of border structures. Even in the case of golden hamster skin, an evolutionary optimum was achieved for a system that was manifest in empirical allometric relationships (JURGENS 1989, p. 30).

2. Liver and testes For methodological reasons, we concentrated on skin and biochemical characteristics. Therefore, no organ weight determinations were carried out in golden hamsters during infancy, since data for agouti coloured hamsters had been published by SCHUMACHER et al. (1965 a -c). These indicate that both the liver and testes show an overproportional growth in relation to body weight during infancy (I). These allometric functions allowed the constants for circulatory functions and blood supply to be compared (b = 1.15 for the cardiac performance: GUNTHER 1971), but not those of simple metabolic constants.

E

3



c:

-'" (.)

85

~

G>

0

'" E :J

c: :i

(f)

30

..

10 1.5

10

50

400 Body weight (g)

E 3

~

~

E '6

"

120

G>

~

8. (.)

'6

< 60

30 ~----------~~--------~------------~ 10 50 1.5 400

Fig. 4. Skin muscle thickness (above) and adipocyte diameter (below) as a function of body weight (for details see fig. 2).

Body weight (g)

In the subadult hamsters (II) the liver allometric exponent from our findings corresponded to the constants of the basal turnover of equally warm animals (KLEIBER 1967, p. 172). Later, the initial extremely high organ metabolism was adapted to an intermediate bodily requirement, so that linear liver-body weight relations were found in the adult section III (GUNTHER 1971), and agreed with the findings of CALDER (1984, p. 48) for mammals. By contrast, the liver exponent of .74 reported by FRAHM (1971) for 3 and by VIEREGGE et al. (1986) for 4 adult species of hamster was comparable to our juvenile animals. This leads one to suspect that the hamster studied by these authors were not fully grown.

3. Ovaries In comparison to testicular growth, which even in j u v e nil e hamsters was directed towards a rapid attainment of sexual function, ovarian weight developed in relation to the surface area of the body. By contrast, the growth of infant kidneys was found to be dependent "on body weight (SCHUMACHER et al. 1965b, d). In the j u v e nile s tag e (II), a rapid growth of the testes and

ovaries was observed with exponents of more than 2.0. These extremely large intraspecific allometric exponents have been confirmed for subadult rats (SCHAFER et al. 1970). Our own results corresponded to the interspecific allometric exponent for ovaries and testes in hamsters reported by FRAHM (1971); extreme testicular growth with b-values greater than 2.8 have also been reported for subadult primates (LARSON 1985). During the phase up to the onset of sexual maturity there is apparently such an overproportional growth of the gonads that the growth strategies discussed up to now did not apply! The development of the kidneys in our 25 -100 day old hamsters, however, could not be brought into a functional relation with circulatory constants (b = .41 to ,61: GUNTHER 1971). The interspecific allometric exponent b of ,68 ± .04 for male hamster kidneys from 4 species (HACKBARTH and GARTNER 1989) agreed with our intraspecific findings (b = .64 ± .04). In the adult stage (III), both testes and kidney development in our hamsters could be explained by surface-body weight relationships (b = .59). VIEREGGE et al. (1986) reported a comparable kidney exponent of b = .71 for 4 species of hamsters. In addition, the size of the ovaries at the time of sexual maturation could be explained by simple metabolic relations. The development of the gonads and kidneys in the adult was Exp. Toxic. Pathol. 44 (1992) 3

119

~

.§. 1:aI 'ij

1500

III

~

>CD C

"tI

i:

750

300

1.5

10

50

400 Body weight (9)

~

.§. E

aI 'ij ~

9000

III

Qj

> ::i

3000

II

1000~------------~----------~------------~ 1.5

10

50

Fig. 5. Liver weight (above) and kidney weight (below) as a function of body weight (for details see fig. 2).

400

Body weight (g)

thus only still orientated towards the characteristics of intermediate metabolic performance.

4. Adrenals During in fan c y, body weight - surface area relationships have been found for adrenal ontogenesis (SCHUMACHER et al. 1965b, d) . In the first days of life , the body surface area thus proved to be the limiting parameter to which hormone production was directed in light-weight hamsters. In the juvenile and ad u It stage, the adrenal allometries even of our golden hamsters, however, were found to be comparable to circulatory parameters (GUNTHER 1971). As a result, the organ-body weight relationship in hamsters considerably exceeded the reported interspecific allometric exponent of b = .80 for other mammals (CALDER 1984, p. 48) . This may possibly be reflected in a heavier burden to the adrenals of the originally solitary-living golden hamster caused by laboratory socialization. Our ontogenetic findings in general confinn those of LARSON (1985) for adrenals ami ovaries from primates: each organ development is influenced more effectively by the prevailing physiological condition of the organism than by simple body 120

Exp. Toxic. Pathol. 44 (1992) 3

weight relations. The growth behaviour of the skin and organ characteristics of the golden hamster can be attributed to achieving an optimized adaptation to the functional demands of the respective age phases and the limiting metabolism. In this way, each parameter was adapted to the respective ontogenetic conditions through a series of explicable metabolic, surface area, weight related systems as well as thermoregulatory ones.

Body weight versus age relation

The description of ' the normal development of skin and organ characteristics in laboratory animals has preferentially been carried out on age profiles. In animal experiments, however, other emphases have been placed. Thus, in general, higher priority is given to the body weight than age specification in experimental pathological and toxicological studies (LANG and VESELL 1976). At the same time, the warning by JURGENS (19.89, p. 125) must be heeded that varying similarities can also occur in different classes of body mass, i.e. temporally different phases of ontogenesis require special interpretations. Of all 13 organ characteristics investigated in the 2 strains of golden hamster, the body weight relations were clearly predominant. This has also been reported by other authors for other

50

8

II

1.5~----------~----------~----------~

1.5

10

50

400 Body weight (g)

E

...... :;)

700r-------------------------------------,

3

~

.S :; .,

.: OJ

'

.

120

E

=

it

25

Fig. 6. Adrenal weight (above) and plasma insu5

~----------~----------~----------~

1.5

10

50

400

lin level (below) as a function of body weight (for details see fig. 2).

Body weight (g)

hamster species (SCHUMACHER et al. 1965a-d; FRAHM 1971; VIEREGGE et al. 1986). The only difference seen was the course of organ development (table 2): For liver, kidneys and testes of the golden hamster, the more rapid organ instead of body weight growth during the juvenile phase was found to be inversely related to age. Even in the case of mice and rats, the liver, kidneys and testes could be grouped together because of the similar ontogenetic mass development (SCHAFER et al. 1970; TRIEB et al. 1976). HACKBARTH and GARTNER (1989) found no changes in slope and intercept for kidneys on comparing the findings from 30-50 day old and adult female rats. Despite overall narrow body weight relations, a divergent growth behaviour characterized the development of adrenal - and ovarian weight. A slow organ development in the first weeks of infancy was followed by a constant to more rapid organ than body weight growth in age section II. Based on the figures by TRIEB et al. (1976), this development could also be confirmed for the adrenal and ovarian weights in female rats during the course of ageing. The diameter of sul?eutaneous adipocytes is also one of the parameters that show a greater body weight than age-dependence in the golden hamster. This also applied to the size of adipocytes or the lipid content per cell in other small rodents

(Mouse: EISEN et al. 1978; rat: GREENWOOD and HIRSCH 1974; HA USMAN et al. 1981) and in primates (J EN et al. 1985). In baby hamsters, a positive body weight relation was seen only in the case of the insulin but not plasma glucose level. Similar observations have been made in primates (JEN et al. 1985), rat foetuses (GIRARD et al. 1976) and adult rats (CODINA et aI. 1980). The glucose level in adult golden hamsters (III), however, showed a negative correlation with age. The negative course of the allometric gradient for the hamster (a = 110, b = .10; table 2) was found to be identical to the interspecific "mouse to elephant"-allometry for glucose (a = 115, b = .101: UMMINGER 1975). Also CALDER (1984, p. 118) emphasized the fact that the glucose and insulin blood levels are characterized less by the basic variables age and body weight than by factors of transport capacity such as blood volume, vascular system and circulatory parameters. . For characteristics that are continuously subjected to growth processes such as organ weights, it is recommended to use the body weight relation over long life phases. Even the course of skin muscle thickness development over longer age periods strongly paralleled that of the body weight. By contrast, in golden hamsters characteristics were always Exp. Toxic. Pathol. 44 (1992) 3

121

Ci 8000

.s E

C>

Qi ~

en CD

iii CD

f-

1200

"300

II

50 1.5

10

50

400 Body weight (g)

Ci

.s E

C>

'Gj ~

c: .!!!

60

lii >

0

15 II

3

~----------~--------~

1.5

10

__________~

50

400

Fig. 7. Testes weight (above) and ovarian weight (below) as a function of body weight (for details see fig. 2).

Body weight (g)

found to be age-dependent only during short periods of life. Skin characteristics with cyclic growth behaviour, in particular hair follicle density and reticular thickness, showed an age instead of body weight relation during the juvenile phase. A positive relation to both weight as well as age was observed in the case of papillary thickness after infancy. The repeated discussions as to whether age or body weight is suitable as an independent variable in ontogenetic studies thus still remain unresolved (GERMAN and MEYERS 1989a, b). For long-term toxicological studies in hamsters this means that age-dependent changes in references points may become necessary (WIESER 1984). The regular documentation of age and body weight thus proves to be a "must" in all doubtful cases.

Sex- and strain differences Distinct strain and sex differences between golden hamsters were observed from an assessment of the data of absolute organ weights (MILITZER et al. }990, 1991). The present allometric analysis, however, shows that these variations are due to differences in body weight. From our results, it must be assumed that the strain differences described by SHAW and TURTON 122

Exp. Toxic. Pathol. 44 (1992) 3

(1979, cit. by HOGER et al. 1983) in the development of liver weight in agouti and acromelanic white inbred ALAe/Lac hamsters can also only be explained by differences in body weight. Also sex differences in the absolute data should only be definitely evaluated following allometric calculaJion. HOGER et al. (1983) thus considered absolute differences in kidney weights in female hamsters to be the result of an age-dependent kidney amyloidosis. Indeed, we also observed the sex-specific heavier kidneys in females after allometric assessment, but were unable to find any increase in amyloidosis (BUSCH 1988). Furthermore, HACKBARTH and GARTNER (1989) found heavier kidneys in female hamsters in an interspecific investigation but without being statistically significant. Sex differences were found to remain only for adrenal weights after allometric calculation; here, the values for male hamsters were always greater than those of the females as previously described for the absolute data (MILITZER et al. 1991). Pharmacological-toxicological statements as to sex and strain differences in laboratory animals should thus always take the allometric concept of comparison into account. This applies particularly when varyingly physically developed out- and inbred strains are to be compared.

E

..

200

3

II

...

II

c: \)

S

~

100

01

= a. 01

IL

9 50

(j

10 15

c;;

.§. E

250

80

30

Body weight (g) 50

III~"

"

CII

'0; ~

.,

OJ

c: .t;

0' __ ..::..:~__ -----,

<

10

Fig. 8. Thickness of the papillary layer (above)

2~--

____

~~

15

Acknowledgement

30

__________

~

______________

80

~

250

and adrenal weight (below) as allometric linear regressions, given seperately for the 2 hamster strains, sexes and age periods II (day 25-100) and III (day 110-365). 0 Sl = acromelanic white hamsters; .. agouti coloured hamsters ; - - significant (p ~ 0.05), - - - - - - not significant.

,=

Body weight (g)

The authors thank Dr. PETER TAMULEVICIUS; University of Essen, for criticism and translation of part I - III.

References BUSCH R: Histometrische und histopathologische Charakterisierung altersabhiingiger Nierenveriinderungen an den Malpighischen Korperchen bei zwei Stiimmen des Syrischen Goldhamsters (Mesocricetus a uratus W.). Veteriniinnedizinische Dissertation, Giel3en 1988. CALDER WA: Size, Function, and Life History. University Press, Harvard 1984. CODINA J, VALL M, HERRERA E: Changes in plasma glucose, insulin and glucagon levels , glucose tolerance tests and insulin sensitivity with age in the rat. Diabete Metab (Paris) 1980; 6: 135-139. EISEN EJ, HAYES JF, ALLEN CE, et al.: Cellular characteristics of gonadal fat pads , livers and kidneys in two strains of mice selected for rapid growth. Growth 1978; 42: 7-25. FRAHM H: Metrische Untersuchungen an den Organen von Hamstern der Gattung Phodopus , Me s ocricetu s und Cricetus. Mathematisch-naturwissenschaftliche Dissertation , Kiel 1971. GERMAN RZ, MEYERS LL: The role of time and size in ontogenetic allometry: I. Review. Growth, Devel Aging 1989a; 53: 101-106. - - The role of time and.size in ontogenetic allometry: II. An empirical study of humane growth. Growth , Devel Aging 1989b; 53: 107-115 . GIRARD JR, RIEUTORT M, KERVRAN A, JOST A: Hpnnonal control of fetal growth with particular references to insulin on growth hormone , p.

197 - 202. In: Perinatal Medicine (eds.: Rooth G, Bratteby L-E). European Congress of Perinatal Medicine, Uppsala 1976. GREENWOOD, MRC, HIRSCH J: Postnatal development of adipocyte cellularity in the nonnal rat. J Lipid Res 1974; 15: 474-483. GUNTHER B: Stoffwechsel und Korpergrol3e: Dimensionsanalyse und Similaritiitstheorien, p. 117-151. In: Physiologie des Menschen , Band 2: Energiehaushalt und Temperaturregulation (eds.: Aschoff J, Gunther B , Kramer K [siehe Girard]). Urban & Schwarzenberg, Berlin 1971. MORGADO E: Dimensional analysis and theory of biological si milarity. Exp Bioi Med 1982; 7: 12-20. HACKBARTH H, GARTNER K: Intraspecific allometry: The kidney. Z Siiugetierkd 1989; 54: 397-405. HAUSMAN GJ, CAMPION DR, RICHARDSON RL, MARTIN RJ: Adipocyte development in the rat hypodennis. Am J Anat 1981; 161: 85 - 100. HOGER H, GIALAMAS J, ADAMIKER D: Massenentwicklung und Organmasse beim Syrischen Goldhamster mit Berucksichtigung altersbedingter Organveriinderungen. Z Versuchstierkd 1983; 25: 317-331. HUXLEY J: Problems of Relative Growth. Methuen, London 1932. Cit. by Hackbarth & Giirtner (1989). JEN KLC , HANSEN BC, METZGER BL: Adiposity , anthropometric measures and plasma insulin levels of rhesus monkeys. Int JObes 1985 ; 9: 213-224. JURGENS KD: Allometrie als Konzept des Interspeziesvergleiches von physiologischen Grol3en. Schriftenreihe Versuchstierkd. Heft 15 : Parey, Berlin 1989. KLEIBER M: Die Grundumsatzrate als Potenzfunktion des Korpergewichtes, In: Der Energiehaushalt von Mensch und Haustier. pp. 172-1 93. Parey, Hamburg 1967. Exp. Toxic . Pathol. 44 (1992) 3

123

LANG M, VESEL ES: Environmental and genetic factors affecting laboratory animals. Impact on biomedical research. Fed Proc 1976; 35: 1123-1124. LARSON SG: Ontogenetic and interspecific organ weight allometTY in old world monkeys. Am J Phys Anthropol 1985; 64: 59-67. MILITZER K, HIRCHE H, MOOG E: The ontogenesis of skin and organ chamcteristics in the Syrian golden hamster. 1. Skin compartments and subcutaneous adipocytes. Exp Pathol 1990; 40: 77-93. HERBERG L, BUTTNER D: The ontogenesis of skin and organ characteristics in the Syrian golden hamster. ll. Body and organ weights as well as blood glucose and plasma insulin levels. Exp Pathol 1990; 40: 139-153. REFINETTI R: Body size and metabolic rate in the laboratory rat. Exp. Bioi 1989; 48: 291-294. SACHS L: Applied Statistics - A Handbook of Techniques. Springer, New York 1984. SCHAFER A, MARTIN H, ROTZSCH W: Allometrie, Organ- und K6rperwachsturn zweier Stamme von Laboratoriumsratten. Bioi Zent.bl 1970; 89: 751-764. SCHMIDT-NIELSEN K: Scaling. Why is Animal Size so Important? University Press, Cambridge 1984. SCHUMACHER GH, WOLFF E, JUTZI E: Quantitative Untersuchungen uber das postnatale Organwachstum des Goldhamsters (M e soc ric e t us auratus Wtrh.). 1. K6rpergewicht-Herz. Gegenbaurs morphol Jahrb (Leipzig) 1965a; 107: 550-567. - -: Quantitative Untersuchungen uber das postnatale Organwachstum

des Goldhamsters (Mesocricetus auratus Wtrh.). ll. LungeLeber-Milz. Gegenbauers morphol Jahrb (Leipzig) 1965 b; 108: 18-40. - -: Quantitative Untersuchungen uber das postnatale Organwachstum des Goldhamsters (Mesocricetus auratus Wtrh.). III. ThymusSchilddruse-Nebenniere-Mundspeicheldriise-Oesophagus-MagenDarm. Gegenbauers morphol Jahrb (Leipzig) 1965c; 108: 41-66. - -: Quantitative Untersuchungen uber das postnatale Organwachstum des Goldhamsters (Mesocricetus auratus Wtrh.). IV. Besprechung der Ergebnisse (SchluB). Gegenbaurs morphol Jahrb (Leipzig) 1965d; 108: 123-128. TRIEB G, PAPPRITZ G, LUTZEN L.: Allometric analysis of organ weights. 1. Rats. Toxicol Appl Pharmacol1976; 35: 531-542. UMMINGER BL: Body size and whole blood sugar concentration in mammals. Comp Biochem Physiol1975; 52A: 455-456. Cit. by Calder, p. 122 (1984). VIEREGGE TH, HACKBARTH H, SCHINKEL K, FRANKE P, MESSOW C: Bodyweight-power-functions of organ weights and histometrical parameters in 4 different species of hamster. Z Versuchstierk 1986; 28: 236. WEBB DR, MCCLURE PA: Development of heat production in altricial and precocial rodents: Implications for the energy allocation hypothesis. Physiol Zool 1989; 6: 1293-1315. WIESNER W: A distinction must be made between the ontogeny and the phylogeny of metabolism in order to understand the mass exponent of energy metabolism. Respir Physiol 1984 ; 55: 1-9.

Exp Toxic Patho1 1992; 44: 124 Gustav Fischer Verlag Jena

Information

Gesellschaft ffir Toxikologische Pathologie 1992 Award for Scientific Publication In 1992 the Gesellschaft fUr Toxikologische Pathologie is offering its first award combined with the sum of DM 1000.for a scientific publication by a member, with the aim of promoting scientific activities and the publication of their results by young scientists. Original papers on morphological aspects of toxicological pathology will be considered which were published or submitted for pu~lication no earlier than 1990 or are now ready for submission.

124

Exp. Toxic. Pathol. 44 (1992) 3

Manuscripts and applications should be sent by authors or co-authors to the chairman of the award committee, Prof. Dr. R. Hess, Fluhweg 11, CH-4l43 Domach by July 31, 1992. Members and associate members of the Gesellschaft fUr Toxikologische Pathologie under 40 years of age are eligible to apply. The selection will be made by the award committee elected by the Gesellschaft. The presentation will take place at the members' meeting in Berlin on October 31, 1992. Prof. Dr. R. Hess Dr. E. Karbe Award Committee Chairman President