Metabolism Clinical and Experimental AUGUST
VOL 40, NO 8
Evidence That Potassium Reduced Circulating Levels
Deficiency Induces of Growth Hormone
Allan Flyvbjerg,
lnge DWp,
1991
Growth Retardation Through and Insulin-like Growth Factor
I
Maria E. Everts, and Hans Orskov
Growth retardation and impaired protein synthesis are major characteristics of potassium (K)-deficiency in animals and man. We have evaluated the effect of K-deficiency on growth, serum growth hormone (s-GH), insulin-like growth factor I (s-IGF-I), and insulin (s-insulin) in young rats. After 10 days on K-deficient fodder, 4’/z-week-old rats showed a 54% reduction in serum potassium (s-K) and a weight gain that was reduced by 97%. compared with pair-fed controls. In addition, tail length, tibia length, and muscle weight of soleus in K-depleted animals were all significantly reduced compared with pair-fed controls. The growth retardation was accompanied by a 46% reduction in s-IGF-I, while s-insulin showed no decrease. K-repletion in animals depleted for 7 days showed complete normalization of s-K within 24 hours, in addition to a significant increase in both s-IGF-I and weight. In 4-week-old rats maintained on K-deficient fodder with variable K-content (1 to 260 mmol/kg) for 1 week, a strong correlation between the K-content of fodder and s-IGF-I could be established (r = 68.P < .OOl), as well as between s-IGF-I and weight gain (I = .90, P < .OOl). Furthermore, a stepwise reduction in basal s-GH was seen with the graded reduction of dietary K-content. The peak in s-GH after stimulation with growth hormone-releasing factor (GRF 40) was 255 ? 25 pg/mL in 4’/2-week-old rats that had been K-depleted for 10 days, while age-matched pair-fed controls showed a peak of 478 ? 50 pg/L (P < .OOl). In conclusion, K-depletion leads to growth retardation that is accompanied by reduced GH response to GRF and reduced circulating IGF-I, but unchanged insulin levels. The reduction in s-GH and IGF-I could not be attributed to reduced caloric intake, but seems to be a more specific effect of nutritional deficiency of K. It is suggested that the growth retardation seen in even mildly K-depleted animals may be mediated through suppressed levels of circulating GH and IGF-I. Copyright 0 1997 by W.B. Saunders Company
OTASSIUM is essential for growth in animals’~Z and in man, as it appears in children with Bartters syndrome.’ Furthermore, part of the growth retardation seen in children with protein-caloric malnutrition may be caused by accompanying potassium (K)-deficiency.4,5 In rats, nutritional deficiency of K leads to a progressive and marked reduction in weight gain and protein synthesis, which is reversible following K-repletion.‘,’ Since protein synthesis even in cell-free systems seems to depend on the presence of K,6 it would appear plausible that the inhibition seen in the intact K-deficient animal is secondary to a decreased intracellular concentration of K. However, earlier studies performed in this laboratory have shown that the incorporation of [3H]leucine into muscle protein, as well as weight gain, was significantly reduced in rats maintained on moderately K-deficient fodder, despite almost undetectable reductions in muscle K-content.’ Furthermore, growth retardation was evident already within the first days on K-deficient fodder, before any significant cellular K-deficiency had developed.’ Therefore, the impairment of growth cannot readily be accounted for as a result of reduced availability of K in the major pools of protein synthesizing cells. This suggests that insufficient supplies of K may influence
P
Metabolism,
Vol40, No
8
(August), 1991: pp 769-775
growth via endocrine regulatory systems. Growth hormone (GH), insulin-like growth factor I (IGF-I), and insulin are hormones that are interlinked at many levels and of great importance for normal growth. The somatogenic actions of GH on the peripheral tissues is thought to be mediated in part by circulating levels of IGF-I,7 60% of which is estimated to be of hepatic origin.8 The present study was performed to measure the serum levels of GH, IGF-I, and insulin during dietary K-deficiency From the Medical Department M (Diabetes and Endocrinologyl, Aarhus Kommunehospital, Institute of Physiology, Universily of Aarhus; and the Institute of Experimental Clinical Research, Universityof Aarhus, Aarhus, Denmark. Supported by grants from DANIDA, the Danish Diabetes Association, the Danish Medical Research Council (Grant No. 128461) the Ruth Kanig Petersen Foundation, the Aa. Louis-Hansen Memorial Foundation and the Nordic Insulin Foundation. Inge D&up held a research fellowshipfrom Alfred Benzon’s Foundation. Address reprint requests to Allan Flyvbjerg, MD, Medical Depatiment M (Diabetes and Endocrinology), Aarhus Kommunehospital, DK-8000Aarhus C, Denmark. Copyright 0 1991 by U?B. Saunders Company 0026-0495/9114008-0001$03.00/0 769
FLYVBJERG
770
in rats and compare K and growth.
these parameters
MATERIALS
to changes
in serum
AND METHODS
Animals Since the growth-retarding effect of K-depletion is particularly pronounced during the rapid growth phase, all experiments were performed using young male or female Wistar rats, 3 to 6 weeks old at the start of the experiment. However, it is known that serum IGF-I levels increase considerably with age, in the range 3 to 10 weeks,” and therefore the experimental and control animals in each experiment were carefully matched with respect to age and body weight. The animals were kept at constant temperature (23°C). humidity (53%). and a 12:12 hour artificial light cycle (6:00 AM to 6:00 PM). Stainless steel frames were positioned in the bottom of the cages to prevent the animals from having access to urine or feces. Until used for experiments, all rats were maintained on a standard fodder containing 260 mmol Wkg (Altromin. Lage, Germany).
K-deficiency was induced by maintaining the animals on distilled water and a semisynthetic K-free fodder with the same caloric content as control fodder (13.8 kJ/g) (Altromin). The K-content of all batches of the fodder was analyzed and found to be on average 1.15 + 0.17 mmolikg (mean + SEM). With the exception of the two pair-feeding experiments (see below), control animals were maintained on distilled water and K-free fodder supplemented with KCI to a final concentration of 260 mmol K/kg fodder. Three days before the start of the experiments, the rats were accustomed to the semisynthetic fodder by feeding the K-supplemented fodder. After this adjustment period, they were randomly divided into the experimental groups. K-repletion after a depletion period was performed by giving the K-depleted animals free or restricted access (see below) to K-enriched fodder (260 mmol K/kg). Graded K-deficiencies were induced by maintaining the animals on distilled water and K-deficient fodder supplemented with KCI to various concentrations between 1 and 260 mmol K/kg. Pairfeeding experiments were performed both during K-depletion and K-repletion. In the first instance, K-deficiency was induced by maintaining the animals on K-deficient fodder. The daily food consumption by the K-depleted animals was measured each morning. The pair-fed control animals were each given the same amount of fodder as a percentage of body weight, as that consumed in the K-depleted group. In addition, an ad libitum fed control group with free access to fodder was included. All three groups were kept on K-free fodder and the K-depleted group received distilled water, while pair-fed and ad libitum fed control animals were given access to a KC1 solution (250 mmol K/L). In the second pair-feeding experiment, 15 animals were maintained for 1 week on K-deficient fodder (1 mmol K/kg) and distilled water. After this period, one group continued on the K-deficient fodder. Another group received the same amount of food as a percentage of body weight as the K-depleted rats, but supplemented with K (260 mmolikg) (pair-fed repleted), whereas the third group was ad libitum repleted by giving them free access to K-deficient fodder supplemented with K (260 mmol&g).
Growth Parameters In all experiments, body weight was recorded daily. and in one experiment, the weight of soleus muscles, tail length, and tibia length were measured.
ET AL
Blood Samples Blood samples were collected in the nonfasting state from the retrobulbar plexus through heparinized capillary tubes into plastic tubes and allowed to clot. This was done under ether anesthesia. except in the growth hormone-releasing factor (GRF)-test. where animals were anesthetized with sodium barbital (see below). The blood samples were centrifuged and serum was frozen and kept at -20°C until further analyses.
fGF-I Radioimmunoassay IGF-I antihody UB 286 (raised by L.E. Underwood and J.J. van Wyk, Pediatric Endocrinolog, University of Carolina. Chapel Hill. NC) was donated by the US National Hormone and Pituitary Program. For standards (0.5 to IO kg/L) and iodination. a full amino acid sequence IGF-I analogue (Amgen Biologicals, Thousand Oaks. CA) was used, purchased from Amersham International. Amersham. Bucks, UK. IGF-I was measured in rat serum (1:400) after previous extraction in methanol/acetic acid.“’ Intraassay coefficient of variance (CV) on duplicates was 5% and interassay CV 17% (mean, 1.220 kg/L; n = 10). The IGF-I antibody has O-5?, cross-reactivity with IGF-II and cross-reacts minimally with insulin at 1OY mol. GRF- Test and GH-Assay Under sodium barbital anesthesia (20 mg/kg body weight), blood was collected from the rrtrobulbar plexus at t = 0.5. 10,20. and 30 minutes. Independent of body weight, 1 kg GH-releasing factor 40 (GRF 40) was injected intravenously in K-depleted, pair-fed, and ad libitum fed control animals after the first blood sample. Rat GH (rGH) was determined by radioimmunoassay using wick-chromatography.” rGH and rabbit anti-rGH were donated by the National Hormone and Pituitary Program. GRF 40 was obtained from Bachem. Bubendorf, Switzerland.
Insulin Assqv Serum insulin (s-insulin) previously described.”
K-Content
of Muscle
was measured
by radioimmunoassay
as
and Serum
Muscles for K-determination were excised immediately after decapitation and snap-frozen in liquid nitrogen. The frozen samples were weighed and homogenized in 5% trichloroacetic acid (TCA) using an Ultra Turrax Tissue homogenizer model TD 18110 (Janke-Kunkel GmbH, Stauten. Germany). After centrifugation at 2,000 rpm, samples of the supernatant were taken for determination of K. This procedure has previously been shown to give the same value as measurements on nitric acid digests of tissue samples.” K-concentrations of serum and TCA extracts were determined by flame photometry using a Radiometer (Copenhagen, Denmark) FLM 3 flame photometer and lithium as internal standard. CV for muscle K-determinations was less than 0.4%.
Statistics All results are given as mean values 2 SEM. The significance of difference was assessed by two-tailed I test for groups of nonpaired observations, paired t test and one-way ANOVA. ANOVA was performed in all experiments where more than two groups of animals were compared. In those instances where the ANOVA showed statistical significance, a two-tailed t test was subsequently performed to establish the significance of difference between single groups. Linear regression analysis of unweighted values was performed by the method of the least squares. Pvalues less than .05 were considered statistically significant.
GH, IGF-I, AND GROWTH RETARDATION
771
IN K-DEFICIENCY
In the K-deficient group, s-IGF-I decreased over the lo-day period (p = -03, paired t test), and amounted on day 10 to only 54% of the value seen in the pair-fed control group. However, it is noteworthy that the increase in s-IGF-I in pair-fed animals amounted to only 50% of that seen in ad libitum fed controls, demonstrating an effect of the food restriction per se on s-IGF-I. In contrast, s-insulin was unchanged in the pair-fed control and in the K-depleted group. Finally, a strong correlation between body weight and s-IGF-I could be established (r = .80, P < ,001) when including the values of all animals at day 10. The effects of K-depletion and subsequent K-repletion were evaluated in two different experiments. Figure 1 shows the time course of the changes in body weight, s-K, and s-IGF-I in a group of 6-week-old male rats maintained on K-deficient fodder (1 mmol K/kg) for 9 days and then repleted by changing to the same fodder supplemented with KC1 to a content of 260 mmol K/kg. Age-matched control animals were maintained on the same KCl-enriched fodder. The animals on K-deficient fodder showed an immediate growth retardation, and weight gain during 9 days amounted to only 5.1 2 1.3 g, whereas the control animals showed a weight gain of 42.7 + 2.9 g (P < .OOl) (Fig la). When the K-depleted animals were repleted by giving free access to K-enriched fodder, they showed a weight gain of 24.2 + 1.5 g during the first 2 days compared with that of 5.8 * 1.3 g in the control animals (P < .OOl). K-depletion for 9 days caused a 53% decrease in s-K compared with values in controls (Fig lb). Following K-repletion, s-K reached normal levels within 24 hours. In the control animals, s-IGF-I (Fig tc) showed the
RESULTS
The effects of caloric and K-intake on growth parameters (body weight, tail length, tibia length, and muscle weight), s-K, s-IGF-I, and s-insulin were analyzed in a pair-feeding experiment where 4%-week-old female rats were studied for 10 days. One group of animals had free access to the K-deficient diet and distilled water (K-depleted), another group of animals had free access the K-deficient diet and K-supplemented water (ad libitum fed controls). A third group received the same amount of K-deficient fodder as a percentage of body weight as consumed in the K-depleted group the previous day and in addition K-supplemented water (pair-fed controls). Table 1 shows that pair-feeding caused a reduction in body weight gain (61%), tail length, tibia length, and soleus weight when compared with ad libitum feeding, while s-K was unchanged. In the same experiment, the K-depleted group showed a more pronounced reduction in weight gain when compared with both the ad libitum fed (99%) and the pair-fed controls (97%), while s-K had decreased as much as 54% as compared with the pair-fed controls. In addition, tail length, tibia length, and soleus weight were all significantly reduced when compared with the pair-fed control animals. Body weight on day 10 was significantly correlated to tail length (r = .71, P < .OOl), tibia length (r = .85, P < .OOl), and muscle weight of soleus (r = .85, P < ,001) when including all animals, demonstrating that measurements of body weight in these young animals is a dependable indicator for longitudinal growth status. In both the ad libitum fed and the pair-fed control groups, the age-dependent surge in s-IGF-I was observed.
Table 1. Effect of K-Depletion on Growth Parameters, s-K, s-IGF-I, and s-Insulin Group
1
Ad Libitum
Fed
Controls (n= 8)
Group
2
Pair-fed
Controls (n= 8)
Group
P lv2
3
K-depleted
P
(n = 8)
2~3
Body weight (g) 73.4 + 0.8
73.0 + 1.2
NS
72.6 k 1.6
NS
117.3 2 1.5
90.3 + 1.0
VOL 40, NO 8
Evidence That Potassium Reduced Circulating Levels
Deficiency Induces of Growth Hormone
Allan Flyvbjerg,
lnge DWp,
1991
Growth Retardation Through and Insulin-like Growth Factor
I
Maria E. Everts, and Hans Orskov
Growth retardation and impaired protein synthesis are major characteristics of potassium (K)-deficiency in animals and man. We have evaluated the effect of K-deficiency on growth, serum growth hormone (s-GH), insulin-like growth factor I (s-IGF-I), and insulin (s-insulin) in young rats. After 10 days on K-deficient fodder, 4’/z-week-old rats showed a 54% reduction in serum potassium (s-K) and a weight gain that was reduced by 97%. compared with pair-fed controls. In addition, tail length, tibia length, and muscle weight of soleus in K-depleted animals were all significantly reduced compared with pair-fed controls. The growth retardation was accompanied by a 46% reduction in s-IGF-I, while s-insulin showed no decrease. K-repletion in animals depleted for 7 days showed complete normalization of s-K within 24 hours, in addition to a significant increase in both s-IGF-I and weight. In 4-week-old rats maintained on K-deficient fodder with variable K-content (1 to 260 mmol/kg) for 1 week, a strong correlation between the K-content of fodder and s-IGF-I could be established (r = 68.P < .OOl), as well as between s-IGF-I and weight gain (I = .90, P < .OOl). Furthermore, a stepwise reduction in basal s-GH was seen with the graded reduction of dietary K-content. The peak in s-GH after stimulation with growth hormone-releasing factor (GRF 40) was 255 ? 25 pg/mL in 4’/2-week-old rats that had been K-depleted for 10 days, while age-matched pair-fed controls showed a peak of 478 ? 50 pg/L (P < .OOl). In conclusion, K-depletion leads to growth retardation that is accompanied by reduced GH response to GRF and reduced circulating IGF-I, but unchanged insulin levels. The reduction in s-GH and IGF-I could not be attributed to reduced caloric intake, but seems to be a more specific effect of nutritional deficiency of K. It is suggested that the growth retardation seen in even mildly K-depleted animals may be mediated through suppressed levels of circulating GH and IGF-I. Copyright 0 1997 by W.B. Saunders Company
OTASSIUM is essential for growth in animals’~Z and in man, as it appears in children with Bartters syndrome.’ Furthermore, part of the growth retardation seen in children with protein-caloric malnutrition may be caused by accompanying potassium (K)-deficiency.4,5 In rats, nutritional deficiency of K leads to a progressive and marked reduction in weight gain and protein synthesis, which is reversible following K-repletion.‘,’ Since protein synthesis even in cell-free systems seems to depend on the presence of K,6 it would appear plausible that the inhibition seen in the intact K-deficient animal is secondary to a decreased intracellular concentration of K. However, earlier studies performed in this laboratory have shown that the incorporation of [3H]leucine into muscle protein, as well as weight gain, was significantly reduced in rats maintained on moderately K-deficient fodder, despite almost undetectable reductions in muscle K-content.’ Furthermore, growth retardation was evident already within the first days on K-deficient fodder, before any significant cellular K-deficiency had developed.’ Therefore, the impairment of growth cannot readily be accounted for as a result of reduced availability of K in the major pools of protein synthesizing cells. This suggests that insufficient supplies of K may influence
P
Metabolism,
Vol40, No
8
(August), 1991: pp 769-775
growth via endocrine regulatory systems. Growth hormone (GH), insulin-like growth factor I (IGF-I), and insulin are hormones that are interlinked at many levels and of great importance for normal growth. The somatogenic actions of GH on the peripheral tissues is thought to be mediated in part by circulating levels of IGF-I,7 60% of which is estimated to be of hepatic origin.8 The present study was performed to measure the serum levels of GH, IGF-I, and insulin during dietary K-deficiency From the Medical Department M (Diabetes and Endocrinologyl, Aarhus Kommunehospital, Institute of Physiology, Universily of Aarhus; and the Institute of Experimental Clinical Research, Universityof Aarhus, Aarhus, Denmark. Supported by grants from DANIDA, the Danish Diabetes Association, the Danish Medical Research Council (Grant No. 128461) the Ruth Kanig Petersen Foundation, the Aa. Louis-Hansen Memorial Foundation and the Nordic Insulin Foundation. Inge D&up held a research fellowshipfrom Alfred Benzon’s Foundation. Address reprint requests to Allan Flyvbjerg, MD, Medical Depatiment M (Diabetes and Endocrinology), Aarhus Kommunehospital, DK-8000Aarhus C, Denmark. Copyright 0 1991 by U?B. Saunders Company 0026-0495/9114008-0001$03.00/0 769
FLYVBJERG
770
in rats and compare K and growth.
these parameters
MATERIALS
to changes
in serum
AND METHODS
Animals Since the growth-retarding effect of K-depletion is particularly pronounced during the rapid growth phase, all experiments were performed using young male or female Wistar rats, 3 to 6 weeks old at the start of the experiment. However, it is known that serum IGF-I levels increase considerably with age, in the range 3 to 10 weeks,” and therefore the experimental and control animals in each experiment were carefully matched with respect to age and body weight. The animals were kept at constant temperature (23°C). humidity (53%). and a 12:12 hour artificial light cycle (6:00 AM to 6:00 PM). Stainless steel frames were positioned in the bottom of the cages to prevent the animals from having access to urine or feces. Until used for experiments, all rats were maintained on a standard fodder containing 260 mmol Wkg (Altromin. Lage, Germany).
K-deficiency was induced by maintaining the animals on distilled water and a semisynthetic K-free fodder with the same caloric content as control fodder (13.8 kJ/g) (Altromin). The K-content of all batches of the fodder was analyzed and found to be on average 1.15 + 0.17 mmolikg (mean + SEM). With the exception of the two pair-feeding experiments (see below), control animals were maintained on distilled water and K-free fodder supplemented with KCI to a final concentration of 260 mmol K/kg fodder. Three days before the start of the experiments, the rats were accustomed to the semisynthetic fodder by feeding the K-supplemented fodder. After this adjustment period, they were randomly divided into the experimental groups. K-repletion after a depletion period was performed by giving the K-depleted animals free or restricted access (see below) to K-enriched fodder (260 mmol K/kg). Graded K-deficiencies were induced by maintaining the animals on distilled water and K-deficient fodder supplemented with KCI to various concentrations between 1 and 260 mmol K/kg. Pairfeeding experiments were performed both during K-depletion and K-repletion. In the first instance, K-deficiency was induced by maintaining the animals on K-deficient fodder. The daily food consumption by the K-depleted animals was measured each morning. The pair-fed control animals were each given the same amount of fodder as a percentage of body weight, as that consumed in the K-depleted group. In addition, an ad libitum fed control group with free access to fodder was included. All three groups were kept on K-free fodder and the K-depleted group received distilled water, while pair-fed and ad libitum fed control animals were given access to a KC1 solution (250 mmol K/L). In the second pair-feeding experiment, 15 animals were maintained for 1 week on K-deficient fodder (1 mmol K/kg) and distilled water. After this period, one group continued on the K-deficient fodder. Another group received the same amount of food as a percentage of body weight as the K-depleted rats, but supplemented with K (260 mmolikg) (pair-fed repleted), whereas the third group was ad libitum repleted by giving them free access to K-deficient fodder supplemented with K (260 mmol&g).
Growth Parameters In all experiments, body weight was recorded daily. and in one experiment, the weight of soleus muscles, tail length, and tibia length were measured.
ET AL
Blood Samples Blood samples were collected in the nonfasting state from the retrobulbar plexus through heparinized capillary tubes into plastic tubes and allowed to clot. This was done under ether anesthesia. except in the growth hormone-releasing factor (GRF)-test. where animals were anesthetized with sodium barbital (see below). The blood samples were centrifuged and serum was frozen and kept at -20°C until further analyses.
fGF-I Radioimmunoassay IGF-I antihody UB 286 (raised by L.E. Underwood and J.J. van Wyk, Pediatric Endocrinolog, University of Carolina. Chapel Hill. NC) was donated by the US National Hormone and Pituitary Program. For standards (0.5 to IO kg/L) and iodination. a full amino acid sequence IGF-I analogue (Amgen Biologicals, Thousand Oaks. CA) was used, purchased from Amersham International. Amersham. Bucks, UK. IGF-I was measured in rat serum (1:400) after previous extraction in methanol/acetic acid.“’ Intraassay coefficient of variance (CV) on duplicates was 5% and interassay CV 17% (mean, 1.220 kg/L; n = 10). The IGF-I antibody has O-5?, cross-reactivity with IGF-II and cross-reacts minimally with insulin at 1OY mol. GRF- Test and GH-Assay Under sodium barbital anesthesia (20 mg/kg body weight), blood was collected from the rrtrobulbar plexus at t = 0.5. 10,20. and 30 minutes. Independent of body weight, 1 kg GH-releasing factor 40 (GRF 40) was injected intravenously in K-depleted, pair-fed, and ad libitum fed control animals after the first blood sample. Rat GH (rGH) was determined by radioimmunoassay using wick-chromatography.” rGH and rabbit anti-rGH were donated by the National Hormone and Pituitary Program. GRF 40 was obtained from Bachem. Bubendorf, Switzerland.
Insulin Assqv Serum insulin (s-insulin) previously described.”
K-Content
of Muscle
was measured
by radioimmunoassay
as
and Serum
Muscles for K-determination were excised immediately after decapitation and snap-frozen in liquid nitrogen. The frozen samples were weighed and homogenized in 5% trichloroacetic acid (TCA) using an Ultra Turrax Tissue homogenizer model TD 18110 (Janke-Kunkel GmbH, Stauten. Germany). After centrifugation at 2,000 rpm, samples of the supernatant were taken for determination of K. This procedure has previously been shown to give the same value as measurements on nitric acid digests of tissue samples.” K-concentrations of serum and TCA extracts were determined by flame photometry using a Radiometer (Copenhagen, Denmark) FLM 3 flame photometer and lithium as internal standard. CV for muscle K-determinations was less than 0.4%.
Statistics All results are given as mean values 2 SEM. The significance of difference was assessed by two-tailed I test for groups of nonpaired observations, paired t test and one-way ANOVA. ANOVA was performed in all experiments where more than two groups of animals were compared. In those instances where the ANOVA showed statistical significance, a two-tailed t test was subsequently performed to establish the significance of difference between single groups. Linear regression analysis of unweighted values was performed by the method of the least squares. Pvalues less than .05 were considered statistically significant.
GH, IGF-I, AND GROWTH RETARDATION
771
IN K-DEFICIENCY
In the K-deficient group, s-IGF-I decreased over the lo-day period (p = -03, paired t test), and amounted on day 10 to only 54% of the value seen in the pair-fed control group. However, it is noteworthy that the increase in s-IGF-I in pair-fed animals amounted to only 50% of that seen in ad libitum fed controls, demonstrating an effect of the food restriction per se on s-IGF-I. In contrast, s-insulin was unchanged in the pair-fed control and in the K-depleted group. Finally, a strong correlation between body weight and s-IGF-I could be established (r = .80, P < ,001) when including the values of all animals at day 10. The effects of K-depletion and subsequent K-repletion were evaluated in two different experiments. Figure 1 shows the time course of the changes in body weight, s-K, and s-IGF-I in a group of 6-week-old male rats maintained on K-deficient fodder (1 mmol K/kg) for 9 days and then repleted by changing to the same fodder supplemented with KC1 to a content of 260 mmol K/kg. Age-matched control animals were maintained on the same KCl-enriched fodder. The animals on K-deficient fodder showed an immediate growth retardation, and weight gain during 9 days amounted to only 5.1 2 1.3 g, whereas the control animals showed a weight gain of 42.7 + 2.9 g (P < .OOl) (Fig la). When the K-depleted animals were repleted by giving free access to K-enriched fodder, they showed a weight gain of 24.2 + 1.5 g during the first 2 days compared with that of 5.8 * 1.3 g in the control animals (P < .OOl). K-depletion for 9 days caused a 53% decrease in s-K compared with values in controls (Fig lb). Following K-repletion, s-K reached normal levels within 24 hours. In the control animals, s-IGF-I (Fig tc) showed the
RESULTS
The effects of caloric and K-intake on growth parameters (body weight, tail length, tibia length, and muscle weight), s-K, s-IGF-I, and s-insulin were analyzed in a pair-feeding experiment where 4%-week-old female rats were studied for 10 days. One group of animals had free access to the K-deficient diet and distilled water (K-depleted), another group of animals had free access the K-deficient diet and K-supplemented water (ad libitum fed controls). A third group received the same amount of K-deficient fodder as a percentage of body weight as consumed in the K-depleted group the previous day and in addition K-supplemented water (pair-fed controls). Table 1 shows that pair-feeding caused a reduction in body weight gain (61%), tail length, tibia length, and soleus weight when compared with ad libitum feeding, while s-K was unchanged. In the same experiment, the K-depleted group showed a more pronounced reduction in weight gain when compared with both the ad libitum fed (99%) and the pair-fed controls (97%), while s-K had decreased as much as 54% as compared with the pair-fed controls. In addition, tail length, tibia length, and soleus weight were all significantly reduced when compared with the pair-fed control animals. Body weight on day 10 was significantly correlated to tail length (r = .71, P < .OOl), tibia length (r = .85, P < .OOl), and muscle weight of soleus (r = .85, P < ,001) when including all animals, demonstrating that measurements of body weight in these young animals is a dependable indicator for longitudinal growth status. In both the ad libitum fed and the pair-fed control groups, the age-dependent surge in s-IGF-I was observed.
Table 1. Effect of K-Depletion on Growth Parameters, s-K, s-IGF-I, and s-Insulin Group
1
Ad Libitum
Fed
Controls (n= 8)
Group
2
Pair-fed
Controls (n= 8)
Group
P lv2
3
K-depleted
P
(n = 8)
2~3
Body weight (g) 73.4 + 0.8
73.0 + 1.2
NS
72.6 k 1.6
NS
117.3 2 1.5
90.3 + 1.0
