0. MORRHUA) GOD THE RELATION BETWEEN THE CONCENTRATION OF SOME hIAlN ELEMENTS AND THE STAGES OF MATURATION OF OVARIES

THE RELATION BETWEEN THE CONCENTRATION OF SOME hIAlN ELEMENTS AND THE STAGES OF

MATURATION OF OVARIES I N GOD (GAD US MORRHUA)

BY

K. JULSHAMN and 0. R. BRBKRAN Institute of Vitamin Research Directorate of Fisheries, Bergen

The contents of the major elements sodium, potassiuizi, calcium and magnesium, and the minor elements iron, manganese, zinc, copper, cobalt, lead, cadmium and mercury have been determined in relation to the reproductive cycle of ovaries in cod (Gadus ~norr/~zln). I n addition were determined thc cobalt-containing vitaillin B ,,by nlicrobiological assay of some of the samples, All ovaries were collected from cod caught off the Western coast of Norway, and aI1 analyses were bascd on freeze-dried samples.

Tlie results are reported on the basis of wet weight and duy matter, and the relations to the reproductive cycle are given graphically. The contents of calciun~, potassium, magnesium and zinc, showed different but significant decreases during the reproductivc cycle, whereas sodiuliz, manganese and copper showed significant increases. Calculation of cobalt derived from vitamin B,, compared with total cobalt showed a high degree of correlation, indicating that cobalt in the ovaries is present as part of vitamin B,,. The levels of lead, cadmium and izlercury were well below the limits given as provisional tolerable weekly intakes proposed by FAO/WHO.

INTRODUCTION

With improved methods for the determination of minerals and in particular trace elements, t l ~ e r e has beell a n increased interest in their occurrence in marine organisms and their physiological significance.

The present study reports on the relatioil betw~eeiz major and minor elements in ovaries of cod (Gadus mon;lzua) and the stage of the repro- ductive cycle. I n a previous commuilication (JUL~HAMN & BRAEKKAN,

1975) we reported or1 the relation between cobalt in ovaries from salmon (Salmo solar) and the stages of maturation. No illlormation is available on the content of metals during the development of ovaries in fish. However,

O G I N ~ & UASUDA (1962) have investigated several inorganic constituei~ts in eggs from railzbow trout (Salmo gairdneri) from before fertilization to the larval stage.

Vitamin B

,,

determinations were included in the present study, to find if the unique relation between cobalt from vitamin B

,,

and total cobalt found in ovaries from salmon (Snlmo snlar) could be observed in cod (JULSEIAMN & BKAEKKAN, 1975).

METHODS

COLLECTION AND T R E A T M E N T OF T H E SAMPILES

Cod roes were collected during the first three months of 1974, from fish caught off the Western coast of Norway and brought alive to the fish market in Bergen. The fish were weighed, gutted and the ovaries taken out, brought to the laboratory and weighed. The state of maturation was classified according to the metliod described by SIVERTSEN (1 935, 1939).

The size of the ovaries as cxpressed by percentage of the body-weight was used as the criterion for tlie determination of maturation. The stages noted with Roman numerals I -VIII refers to groupings <0.5, 0.5 -2, 2 -4, 4 -6, 6 -8, 8 -10, 10 -13 and

>

13

%.

When possible, the ovaries from each individual fish were analyzed, with the exception of juvenile ovaries, where five to ten ovaries were pooled. Each ovary or sample was homo- genized, an at least 50 g material was taken out for freeze-drying. The freeze-dried samples were homogellized in a bleiidor and stored in tightly closed jars until analysis,

All glasswares were wasl~ed carefully i11 Thernards mixture (60

%

destilled water, 30

O/'

conc, hydrochloric acid, 10

%

hydrogen peroxide (30

%)),

followed by rinsing with diluted l~ydroclzloric acid, very careful rinsing with deionized water and drying a t 90 OC.Al1 reagents were ana- lytical grade.

ANALYTICAL PROCEDURE

Destruction of organic matter in the freeze-dried samples was based on wet-digestion, utilizing a combination of heat and pressure. A mixture of 4 ml HNO,/EICLO, (1 : 1) was added to tlie bottles, and the samples predigested over ilight. Four replicates from each samples and blanks were then placed in a pressure boiler and heated a t llO°C ior 2 hours. After cooling and addition of water, tlie sol~ttions were warmed to expel excess volatile aci. The solutions were transferred to 25 ml volu- metric flasks and made up to volume with redistilled water.

.

ⁱ

..

6

? - . .

2

. ..

C

- _.

2

z 2.

a

::

E

2

ii in

... U

g

... c:

E

3

.$ $

8

⁰

2

.$

E

Y

$ 2

5 ;

. ..

2 6 ci ⁷³

$ gi

. 4 0 ) "

, ? z

&5

6

"O

% 1

8

iV) i

B g

-

U L3

.- E

U 42 *

ii(

* V)

C

ii

s

U

f;l

i 0

i i ii

n H

- >

13

5

>

+. =:

= : -

u

-4 i

LQ

F 3 .

=

1

z

M

0 )

-

0

8

%

2

d w

8

1

6

35 - D R Y MATTER

30 -

Z S T A G E A V E R A G E S . n 25-

w -

k- -

k- -

$ 20:

> ^-

Ili -

"

^15:

s

^-

10 - Y.955 l o g X r 2 0 0 6 ( r = O 9 7 )

5 ^/ / / I ,1111 I 1 i I I t d I 1 I I

01 0 5 1 5 10 50 100

S T A G E O F R E P R O D U C T I V E CYCLE ( # ~ ~ ~ t ~ . l ~ ~ ~

Fig. 1. Relation betwecil percciltage total dry matter and stage of reproductive cycle in cod (Gadus nzoi rhzlzla)

.

each single observatioli aiid the other on the stage averages given in Tables 1 and 2. T l ~ e curves were fitted by a method of least squares to the logarithmic functioli, by the followiiig equation :

Y ⁼ ax

.

log,, X

+-

^a,

Figure 3 shows tlie relation between pmol sodiuin and ,urn01 potassium during the maturation of the ovaries. The calculation of the least-square curve tlirough the set of poiiits were based on the simple linear regression equation, Otfierwise, statistical treatment as I-test was performed on each element.

Total dp matter (Table 1 a i d Fig. 1) increased from 16.7

+

0.80 at the first stage of the reproductive cycle (juvenile ovarics) aiid 19.5 & 5.51 a t stage 11 (early maturation) to values in the order of 28 to 30 a t stages I V to V I I I . A t-test sllowed no significant difference between stage I and 11, but highly sigliificaiit differciices between stages I and 111, and stages I1 and V (0,001 <p<0.01). During the stages V to V I I I the total dry matter remained ratlier constant. Tlie general trend coilfirmed the findings reported by BRAEKICAN

PL

BOGE (1959, 1962) in their studies oil vitamins and the reproductive cycle of ovaries iii cod. They pointecl out that the increase in total dry matter during tlie first three stages of maturatioii mainly constitutes ail increase in the protein content. Just before spawnii~g a decrease in dry-matter coulci be observed. This decrease was not very marlted, as the material did not complise samples a t i n ~ m i ~ ~ e n t spawning stages.

111 Fig. 1 may be i~otecl that the curve based on single observatio~is and the curve based on stage averages both gave the same correlatioii coeffi- cient (r = 0.97).

Fig. 2 . Relations bet-

2.0 SODIUM ween mgNa/g and mg

S T A G E A V E R A G E S :

0.5 Y = - 1 30 log X+359 (r:-0.99)

1 l 1 l , , , d # , , , , I

0 1 0.5 1 5 10

5 2 8J

50 I 0 0 S T A G E OF R E P R O D U C T I V E CYCLE

(Ez%ge

* l o o )

K/g wet weight and stage o f reproductive cycle i n cod (Gadus morrl~na)

.

7.5-

?-

>

o

.

m 0 1.0-

Z .

F

0.5

Sodium values are given in Table 1 and Fig. 2. Based on wet weight sodium showed a slow increase from juvenile ovaries (stage I) to stage V from 0.87 & 0,099 to 1.41 & 0.192 mg/g. From stage V the content re- mained fairly constant. t-Test showed no significant difference between stage I and 11 (0.05 <p<O.l). The difference between stage I1 and V was llighly significant ( p

<

0.00 l ) , while the differences between later stages were not significant. The curve based on single observatiolls (Fig. 2) showed a correlatiolz coefficient r = 0.72, wliereas calculations based on stage averages showed a n improved correlatiolz ( r = 0.97). Calculatiolzs to a linear as well as a n exponential fttllction were tried, but did not give improved correlations.

Potassium. The levels of potassium (Table 1 and Fig. 2) showed a n opposite effect of maturation, compared to sodium. This element decreas- ed markedly through the maturation stages. T h e table showed a rapid and significant decrease from 4.2 0.078 mg/g wet weight in juveiiile to 3.31 & 0.323 mg/g potassium in stage 11, further to 1.90 -& 0.19 at the end of the maturation ill stage V I I I . t-Test on the analyzed valries gave

L

Y = 0.4410 - -

-

:

- Y = 0.42 l o g X + 0 9 8 (r-0.971

highly significant differences between stages I arrd I1 (0.001 <p<0.01), stages I1 and V, and V and V I I I . A logarithmic function gave the best relationship with a correlation of r = -0.67, and a least square treatment on the stage averages gave a very good correlation of r = -0.996. Fitting of observations to a linear aiid exponeiltial function gave for the single observatiovls correlations of -0.62 and -0.19, and stage averages

-0.95 and --0.67, respectively.

I n Fig. 3 are plotted the relation between pmol potassium and sodium per g wet weight in tlze ovaries. LEIVESTAD (1965) reported on the K / N a ratio ill tlie muscle of different fish caught a t different localities, and found the sum of the cations to be reasovlably constant, distributed around 145 ,~cmol/g. Similar values were reported in cod muscle by ELIASSEN et al.

(1960) and in rat and frog sarcoplasm by HODGIN (195 1). A least square treatment of the same relatioils in the ovaries from cod gave a regression coefficient of -0.50 with r = -0.56. Altl~ough there was a n apparent inverse rclation between the elements during tlie maturation, a coilstancy of the

M

f Na concelitrations could not be observed. I t sllould, however, be pointed out that the present values refer to the whole ovaries avid do not represent the egg cells alone.

Calcium levels are given in Table I and Fig. 4. They showed a slight but sigiiificant decrease (on 1

%

level) from 150 f 22 / ~ g / g in juvenile to 127 f 61 pglg fresh weight a t stage 11. The correspolldiilg values as pg/g

1 6 0 -

140

120

,.too-

L

d >

SODIUM / POTASSIUM

'\,

-

' '

T h e o r e t i c a l c u r v e

\ * y I

Y = - 1.0 X + 145.0

- o j u v e n i l e

m a t u r e

Z

Y = -0.50 X c91.76 ( r - - 0 . 5 6 )

Fig. 3. Relation bet- ^\

ween pmol Na/g and ^{I I I I I I I I I I I I I \ . I I}

pmol K/g ovary dur- 20 40 60 8 0 100 120 140 160 ing the maturation. ^{,urn01 K} / g ovary

C A L C I U M I

2501 S l N G L E OBSERVATIONS:

-.

^"

.

Y =-52.9 log X+151.8 (rz-0.49)

7

. .

S T A G E A V E R A G E S . A e

..

L

S I N G L E OBSERVATIONS:

250- Y=-97.9 IogX+214.2 ( r = - 0 9 7 )

X - 200:

m .

.

_{m 150-} ^.

STAGE AVERAGES;

1001 Y=-105.2 logX+218.1 ( r:-098)

50 - 300,

Fig. 4. Relations bet- ween , L L ~ Ca/g and pg Mg/g wet weight: and stage of reproductive cycle in cod (Gadus nzorrlzzcn)

.

1 1 1 , , 1 1 1 / , , ^{I I} , , , I I , , , M A G N E S I U M

0.1 0.5 I 5 10 50 100

S T A G E OF R E P R O D U C T I V E C Y C L E ( O V C I ' ~ M g h t . 1 0 0 ) Fish weight

dry matter wcre, however, of tlie same order. No significant differences were fo~113d between stages I 1 and V nor between stages V and V I I I . The observed values sliowed very high deviations within the stages as well as betweeii different stages. Tlie hest correlation of the observations as fitted.

to a logarithmic functio~i based oil single values as well as the stage averages gave rather low r-values, 0.45 and 0.89, respectively. Other fuiictio~is were tried but with even lower correlation.

Magnesium levels were of tlie same order as tliose found for calcium.

Magnesium decreased from 267 & 39 pg/g in juvellile to 197 & 67 pg/g at stage 11, based on. wet weight. Tlie decrease co~iti~iued to 116 & 27 / ~ g / g a t stage V and to 103 f 36 /rg/g at tlie elid of the maturation. A similar trend was observed on the basis of dry matter, from 1600 ,ug/g in juvenile to 367 pg/g at stage V I I I (Table 1).

Juveiiile ovaries (I) and ovaries in the first stage of maturation (11) showed no significant difference on 0.05 level in a t-test. The difference between stages 11 and V was highly significant (0.005 <p<0.01) whereas no significant difference was found betweeii stages V and V I I I . TIie

0

-

2 2 ,

I + C ? m V ) w

=? .'l 0

-

Ci) w C9 b!

- a m

I

+ $

w

-

2 2

- ^*

e

M.

- O m

I -H?

-ad-

2 2

0 0

'4

-

_I h! ⁸ _{S i r '} ^w m m d -

C?

- -

":

r. m

z z m

I ii?

0 m m

? '4

- -

%? 2

-;do?

I Sic?

m z *

2

L r-. m N "!

- o m

I SIT

N Q d -

2 2

3 m

% 8

^*

I

-He:

o * m

2 2

d

a

_I

*

Si

3 &s!:

3 2

M

SINGLE OBSERVATIONS:

1

^{, \.} Y=-31821ogXi85.51 ( r = - 0 . 7 7 )

C O P P E R

...

0,

2.5 S I N G L E O B S E R V A T I O N S :

Y = 0.55 l o g X + 1.00 ( r ~ 0 . 6 2 ^{) %} 1

Y =O.L5 l o g X 11.02 ( r = 0 . 8 4 )

0.1 0.5 1 5 10 50 100

ST A G E OF RE PRODUCTIVE CYCLE ( ~ ~ ~ ~ t ~ . i ~ ~ )

Fig. 6. Relations between pg Zn/g and pg Cu/g wet weight and stage o f reproductive cycle in cod (Gadus morrhzla).

Manganese lcvels showed ail increasc tlirough the maturation of ovaries from 0.45 &- 0.15 at stage I to 0.88

+

0.18 pg/g wet weight in stage V I I I . t-Tests showed iio significant differences between stages I and 11, I1 and V, and V and V I I I , but a significant difference between stages I1 and V I I I on the 0.01 level. Calcrrlated 011 tlie basis on dry matter, tlie values for manganese showed a decrease from the juvenile stage to stage V followed by a slow increase. The best relation between manganese and the stages of maturation was given by a linear regression curve based on stage averages with a correlation coefficient of 0.81.

Both manganese and iron lcvels were higher ill roe compared to filet and soft roe of cod, and lower than that in liver (JULSHAMN & BRAEKKEN, 1975).

Zinc. The content of zinc in the ovaries of cod are given in Table 2, and related to the stages of the maturation in Fig. 6. This element showed a considerable decrease from stage I to 11, and less markedly from I1 to V

and remained fairly unchanged from V to VLII. The same trend was found from tlie values based on dry matter, t-Tests applied on the diffe- rent stages showed highly significant differences between stages I and 11,

"

as well as 11 ancl

V

(0.001 <p<0.01), but no significa~it difference bet- weeii stages V alicl V I I I , (p>0.1). Froni Fig. 6 may be noted the good correlation based on stage averages r = 0.96, whereas the vegression curve lsased on siligle observations gave r = 0.77.

The levels of zinc were exeptioiially high, 5 to 10 times higher than that of iron ancl furtl~er the zinc level in the ovary was higher than in any other organ of cod (JULSHAMN & BRAEICICAW, 1975). BERTRAWD and V L A D I ~ C ~ , as early as 1922, reported that zinc plays a physiological role in fertilization; during spawning, zinc was transferred from tlie niuscles to the sexual products of tlie males.

Cobper levels are given in Table 2 ancl Fig. 6. The contents of coppcr seemed to increase cluri~ig tlie maturation, and a t-test showed higlily significant differences lset\veen stage I1 and V (0.001 < p <0.01). T h e correlatioii was low based on single (r = 0.62) as wcll as stage averages ( r = 0.82), nevertheless tlie logaritlii~iic fuiictioli gave tlie hest fitting to a mathematical functio~i. VIWOGRADOV (1953) reported that the ratio F ~ / C L I in the tissues of vertebrates remains constant, but this did not liold for ovaries of cod, tlie ratio decreased durilig maturation.

Cobalt is sunimarized in Table 2 and Fig. 7. The content of cobalt showed a decrease Irom 0.0252 & 0.0034 at stage I to 0.0044

+

^0.0010

pg/g wet weight in stage V I I I . Between stages I and 11, as well as between stages I1 aiicl V, and between V and VIII, very highly sigliificaiit diffc- reiices (p<0.001) were found. This decrease was more marked than that observed in ovaries of salmoii (Salmo sala~ )(JULSHAMN & BRAEICICAN,

1975). Tlie least square calculatioli gave a correlatioii coefficient of r = 0.93 values and of r = 0.99 based on stage averages (Fig. 7).

Vitamin B,, was determined in 2 to 3 samples at eacli stage of matura- tion by microbiological assay. T h e values foulid correspo~ided wcll with those reported by BRAEKKAN & BOGE (1962). Cobalt calculatecl on a molecular basis deriving from vitamiii B,,, showed closely tlie same order of magnitude as well as the same relation during tlie clifferelit stages of maturation. Wlien tlic 88 values reported by BRAEICKAN & BOGE were used as basis for calculatioli, thc same close relation was found, with de\7iatioiis of less than 10

%.

Thus the same relation between total cobalt aiid cobalt derived from vitamiii B,, could be observed in cod ovaries as reportecl for maturiiig ovaries iiz salnioii (Salmo salar) by JULSHAMN & BRAEICKAN (1 975).

I n conclusion, tlie followiiig pattern is seen: Sodium, copper aiid manganese concentrations increased during the maturation of the roe.

C O B A L T STAGE A V E R A G E S :

1

Y = -0011 log X +0.0186 ( r = -0.93)

-*x-

I / / / / I , ^{t I b 8 ! , I}

0 5 1 5 10 50 I00

S T A G E OF REPRODUCTIVE CYCLE ( $ / ~ & ~ ~ ~ . I o o )

Fig. 7. Relation between pg Co/g and stage of reproductive cycle in cod (Gadus morrhz~a).

Tlie levels ol iroii varied somewhat, but generally increased in the two first stages and tlien decreased. The intracellular elements calcium, potassium, magnesium as well as zinc decreased in conceiztration during the whole period of maturation, thereby giving decreasing levels relative to the protein contents of the roe. Tlie best fitting of tlie curves to the observations were to logarithmic functions, when the stage averages were used. This indicated statistically t l ~ a t iio systematic failure were intro- duced through collection and preparation ol the samples.

The last three elenzents, lead, cadmium and mercury, are widely distributed in Nature, lsut considered to some extent contaminants arising from enviroilrnental poll~ttion (Table 3).

l e a d slzotved low, but sornewllat fluctuatilzg values based o n wet weight. Calculated on dry lnatter a decrease may be observed.

Cadmiurn showed similar low values, but with a clearer decrease, and tlie values a t stages V I to V I I I were below the detection limit of tlie metliod.

Mercugi waq oiily analyscd in samples a t stage I and V I I I , the values being 0.041 f 0.005 in juvenile and 0.019 & 0.009 pg/g wet weight in mature ovaries.

Tlie values for lead, cadrilium and mercury in cod roe in tlie present stvdy Tvere well be10147 the limits given as provisional tolerable weekly intake proposed by tlie Joint FAO/WIIO Expert Committee on Food Additives (1 972).

Table 3. The distribution of the trace elements, lead, cadmium and mercury during the maturation of the ovary in cod (Gadu rnorrhuas).

I I1 I I I I V V VI VII VIII

No. of samples 3 7 11 14 11 9 5 I!

Dry matter

g/100g+s.d. 16.710.8 19.5A5.1 24.5k3.12 27.9h3.11 30.1&1.14 29.6h1.79 29.6A1.58 27.9k3.80 Pb Min.-max. 0.08-0.1 0.10-0.64 tr. -0.98 tr.-0.48 tr.-0.30 tr. -0.73 tr.-0.35 tr.-0.30

~ g / g K 5 s . d . 0.10+0.0! 0.2510.165 0.27&0.24 0.15+0.20 0.10&0.!2 0.1 7h0.24 0.101 0.10 0.15&0.10

-

bfd 0.60 1.28 1.08 0.54 0.34 0.57 0.34 0.54 ^.p

Cd Min.-max. 0.01 -0.05 tr. - 0.03 ir.-0.04 tr.-0.03 tr.-0.01

pg/g A4,ks.d. 0.03+0.01 0.01 &0.010 0.02&0.02 0.01&0.02 > 0.01 N.D. N.D. N.D.

Md 0.1 1 0.05 0.096 0.05

Hg Min.-max.

pg/g M,+s.d. 0.041&0.005 0.019+0.009

M~ 0.25 0.068

REFERENCES

BRAEKKAN, 0. R. and BOGE, G. 1962. Report on TechnoIogicaI Research concerning Norwegian Fish Industry. 4: No 2.

GOLDSGI~MIDT, U. M. 1954. Geochemistry Clarendon Press, Oxford. 730 p.

JULSHAMN, K. and BRAEKKAN, 0. R. 1974. At. Absorption Ncwslett. 4: 139.

- 1975. Gomp. Biochem. Physiol. 503: in press.

- 1975 b. At. Absorption Newslett. 3: 49.

LEIVES~AD, 1-I. 1965. Intet~lational Commission for the Nortllrvcst Atlantic Fisheries Special P~lblication No. 6.

MUNNS, R. I<. 1972. Laboratory i~lforrnation Bulletin: No 1500, 8 pp.

OGINO, C. and YASUDA, S. 1962. Bull. Jap. Soc. Scient. Fish. 28: 788.

SIVERTSEN, E. 1935. Report on Norwegian Fishery and Marine Investigation. 4: No 10.

- 1939. Ibid. 5 : N o 3.

THOMPSON, H. T., DIETRICH, L. S. and ELVEEIJEM, C. A. 1950. J. Biol. Clienl. 184: 175.

FAO/WHO 16th Report of the Joint Expcrt Coliirnittee on Food Additives, Rome, 1972.