JOURNAL
OF SURGICAL
RESEARCH
51,
99-105
(19%)
Impact of Exogenous Growth Hormone on Host Preservation Tumor Cell-Cycle Distribution in a Rat Sarcoma ModeF* ENG-HEN
NG, MBBS,
FRCS
and
(EDINBURGH),* CRAIG S. ROCK, M.D.,* DOUGLAS LAZARUS, PH.D.,* EVA FISCHER, M.D.,* LYLE L. MOLDAWER, PH.D.,* AND STEPHEN F. LOWRY, M.D., FACS*f3
LISA STAIANO-COICO, PH.D.,? *The
Laboratory
of Surgical Metabolism, New York Hospital-Cornell Submitted
the Department of Surgery, and tFlow Cytometry Medical Center, New York, New York 10021 for publication
Growth hormone (hGH) has been reported to improve nitrogen balances and accrue lean mass tissue in stable subjects. However, the ability of hGH to positively influence host preservation in stressed catabolic states such as cancer-induced cachexia remains unproven. Thirtyseven sham or tumor implanted Fischer 344 rats were randomized to receive either 0.5 mg/kg/day hGH or saline (SAL) subcutaneously from Days 14 to 23 postimplantation. Plasma levels of hGh and somatomedin C/ insulin-like growth factor I (IGF I) as well as IGF I bioactivity were determined at sacrifice. Gastrocnemius muscle protein content was used as a index of host lean tissue mass and the tumor response was evaluated via flow cytometry for analysis of cell-cycle distribution. Host cachexia was not attenuated by hGH as muscle protein content was similar in hGH and salinetreated groups. Despite elevated hGH levels (range, 77-222 rig/ml (GH) vs < 2 rig/ml (SAL)), IGF I levels and bioactivity were not elevated in GH-treated groups. In contrast, cancer-induced anorexia markedly decreased IGF I levels (4 U/ml vs 9 U/ml, NTB; P < 0.01) and this response remained refractory to hGH administration. While final tumor weights were similar between GH- and SAL-treated groups, hGH treatment caused a twofold increase in the proportion of aneuploid cells (P < 0.05). In conclusion, hGH failed to attenuate lean mass dissolution in the tumor bearing host and this response may be related to the failure of IGF I induction. Conversely, the altered proportion of tumor aneuploid cells suggests a direct influence on tumor cell-cycling populations. 0 1991 Academic Press, Inc.
November
1 Presented at the Annual Meeting of the Association for Academic Surgery, Houston, TX, November 14-17, 1990. 2 This study supported by grants from GM-34695 and the American Cancer Society PDT No. 347. a To whom correspondence and reprint requests should be addressed at Department of Surgery, New York Hospital-Cornell Medical Center, 525 East 68th Street, F-2016, New York, NY 10021.
20, 1990
been further substantiated through the recent use of recombinant human growth hormone (hGH) in a variety of clinical and physiological situations [4, 51. Improvements in nitrogen retention were reported in stable patients receiving hGH after gastrointestinal surgery and accrual of lean mass was also noted in highly conditioned athletes administered hGH in supraphysiological quantities. However, it is uncertain if these anabolic properties extend to stressed or cachectic states in humans since the induction of Somatomedin C insulin-like growth factor I (IGF I), which is believed to mediate many of the anabolic processes of hGH, has also been suggested to be impaired in such conditions [61. Further, possible mitogenic influences of hGH such as those observed in lymphoid malignancies [7, 81 could preclude the use of hGH in cancer-bearing hosts. Recently, hGH administered to experimental tumor models has not shown obvious enhancement of malignant activity as reflected by its effect on the size and number of tumor nodules [g-11]. Flow cytometry has been employed to evaluate tumor responses to a variety of therapies [ 12,131 and may provide additional criteria by which tumor responses can be quantified. This study was designed to characterize the effect of exogenous hGH on tumor cell-cycle distribution as well as its influence on preservation of host lean tissue. IGF I levels were also evaluated via radioimmunoassay for determination of circulating levels, and costal cartilage bioassays served as an additional index of IGF I activity. MATERIALS
INTRODUCTION
The potent anabolic properties of growth hormone demonstrated initially with pituitary extracts [l-3] have
Core Facility,
AND
METHODS
Experimental Protocol The experimental protocol was reviewed and approved by the Animal Care and Use Committee of the New York Hospital-Cornell Medical Center. Prior to utilization of hGh in the control and tumor bearing animals, the potency of recombinant human growth hormone (Genentech, South San Francisco, CA) was evaluated on hypophysectomized male Fischer 344 99
0022.4804/91 $1.50 Copyright 0 1991 by Academic Press, Inc. All rights of reproduction in any form reserved.
100
JOURNAL
OF
SURGICAL
RESEARCH:
rats (75-80 g) with once per day administration of hGH (0.5 mg/kg) subcutaneously for 12 days. Daily weights were taken at 9:00 AM. Subsequently, 37 male Fischer 344 rats (115-130 g) were either sham implanted (nontumor bearing, NTB) or inoculated with tumor fragments. Two different sizes of tumor fragments were implanted in an effort to achieve differing levels of tumor burden. The first group was inoculated with approximately 2-mm3 fragments to achieve a smaller tumor burden (STB) and the second group with 5-mm3 fragments to achieve larger tumors (LTB). These three groups (NTB, STB, and LTB) were randomized on Day 14 postimplantation by weight and tumor size to receive either hGH 0.5 mg/kg once daily subcutaneously (GH) or saline (SAL) for 10 days, after which the animals were sacrificed on post-tumor implantation Day 23. Daily weights and food intake were recorded at 9:00 AM. Tumor volume was measured serially twice a week using calipers and tumor weights were extrapolated as previously described [ 141. At sacrifice, tumors were removed and host weights were calculated as the difference between total body weight and tumor weight. Organs and gastrocnemius muscles were removed, weighed, and frozen in liquid nitrogen and stored in -70°C until analysis. Animals
and Tumors
Hypophysectomized and normal Fischer 344 rats were obtained from Charles River Breeding Laboratories (Wilmington, Massachusetts) and maintained on a 12hr light, 12-hr dark cycle in individual cages in a temperature-controlled room and were given standard rat chow and water ad libitum. Hypophysectomized rats were weighed periodically and only rats with weight gain of less than 1 g over 2 weeks and without increase in testicular size were used for experiments. Rats were anesthetized with pentobarbital 36 mg/kg ip and inoculated with methylcholanthrene (MCA) sarcoma tumor fragments in the interscapular region. Tumor tissue was obtained from a tumor-bearing donor animal which had been maintained by serial transplantation for about 1 year. This tumor is a locally invasive fibrosarcoma that seldom metastasizes and uniformly kills the animal 30-35 days following inoculation [l5]. RNA/DNA
Determinations
The RNA and DNA contents were determined by previously described techniques [ 161. Tissues were homogenized in ice-cold water and nucleic acids were precipitated with cold perchloric acid. RNA was digested to free nucleotides in alkali and the content of nucleotides was determined spectrophotometrically. DNA content was determined colorometrically [ 171. Protein
Content
Protein contents of the muscle homogenates were determined calorimetrically by methods according to Bradford [ 181.
VOL.
Flow
51, NO.
2, AUGUST
Cytometry
Analysis
1991
Only tumors from the STB groups were analyzed. Preparation of single cell suspensions. Immediately after sacrifice, a sliver of grossly viable tissue measuring approximately 3 X 3 X 0.3 cm was obtained from the periphery of the tumor, ensuring that no necrotic tissue usually located at the center was included. Specimens were immersed in Hank’s balance salt solution (Mediatech) and placed on ice. Digestion of tumor specimens into single cell suspension was obtained using a cocktail of three enzymes-collagenase D (1 mg/ml), hyaluronidase (1 mg/ml), and DNAse (50 pug/ml) (BoehringherMannheim, Indianapolis). Tumor specimens were finely minced in petri dishes and added into a flask containing 25 cc of HBSS and 5 cc each of individual enzymes and were agitated for 2-4 hr at room temperature. The suspension was then filtered through a 52-pm nylon mesh (Nytrex) and pelleted out at 1500 rpm for 10 min. The pellet was washed twice after lysis of RBC with sterile water. Finally cells were resuspended in ice-cold 70% alcohol while vortexing and stored at 4°C until analysis by flow cytometry. Cell Cycle Distribution
Analysis
Alcohol-fixed single-cell suspensions were pelleted and resuspended in PBS containing 300 Kunitz units of ribonuclease A (Sigma, St Louis, MO) for 1 hr at room temperature (to eliminate artifactual broadening of DNA content distributions which would otherwise result from fluorescence of complexes of the DNA-staining dye with double-stranded RNA) before staining with 50 pg propidium iodide for 30 min on ice. The number of fluorecscent nuclei were quantitated using a Epics 752 (Coulter Electronics Corp., Hialeh, FL) cell sorter. The proportion of doublets and dead cells were eliminated from analysis using either a pulse width, time of flight measurement, or the peak versus integrated signals for red fluorescence. The cells remaining in the analysis were then quantitated. At least 12,000 and up to 50,000 cells were analyzed for each sample. Since propidium iodide fluorescence is proportional to double-stranded nuclei [ 191, the resulting fluorescence was proportional to DNA content. Histograms of DNA content were included for analysis only if the coefficient of variation of diploid GO/G1 peak was less than 6%. Peripheral blood mononuclear cells (PBMN) were used as internal diploid controls. Specimens exhibiting nondiploid/nontetraploid DNA content were considered to have aneuploid content. Each tumor sample was analyzed using the DNAFIT software package which determines the proportion of diploid, aneuploid, and S-phase cells from a DNA histogram based on a method employing “multirectangles” [20, 211. Since the DNA histograms of all tumors contained two GO/G1 cell populations which overlapped,
NG
ET
AL.:
GROWTH
HORMONE
AND
the multirectangle analysis for the S-phase fraction could only be performed on the aneuploid population. The results of analyses are presented as the aneuploid/ diploid ratio defined as the proportion of aneuploid to diploid cells and the percentage of aneuploid S-phase, defined as the proportion of cells in this compartment over total number of cells.
CANCER
CACHEXIA
IN
A RAT
MEAN
101
MODEL
FOOD INTAKE
PER DAY
20
16 14 12
Growth Hormone Determination
and Insulin-Like
Growth
Factor
1
Radioimmunoassay. Radioimmunoassay (RIA) techniques were used to determine concentrations of hGH (Kallestadt) and IGF-1 (Nichols Institute, San Juan Capistrano, CA). Plasma (EDTA) specimens were obtained at sacrifice and fractions were separated by centrifugation and stored at -70°C until analysis. The interval between administration of hGH and collection of plasma samples did not exceed 2 hr. IGF-I concentrations are expressed as U/ml. In order to avoid interference by binding proteins, samples were analyzed after acid extraction [ 221. Costal cartilage bioassay. Somatomedin C/IGF I bioassay using the rat costal cartilage was modified after Phillips [23] and only samples from STB groups were analyzed, using NTB-SAL as controls. Costa1 cartilages from a single hypophysectomized rat were cleaned and divided into 80-100 segments. Individual pieces were then placed in 96-well plates with a final volume of 0.1 ml Krebs phosphosaline-amino acid (KPS-AA) buffer supplementation with antibiotics, glucose, and [35S] sulfate (Amersham, Illinois) with or without added serum samples. Test and standard samples were added for a final concentration of 1% (vol/vol). Test rat plasma (EDTA) was obtained at sacrifice and analyzed individually. Standard plasma was obtained from normal rats, pooled and aliquots were frozen at -70°C. After 48-hr incubation, the cartilage was soaked in boiling water, then in saturated sodium sulfate, rinsed, dried and weighed, and digested with 5 M KOH; scintillation fluid was added and disintegrations per minute were measured. Cartilage uptake was measured as counts per minute per milligram of dry cartilage and expressed as the percentage of counts obtained with samples containing standard plasma. Five repeated measurements of individual test and standard samples were performed and values are means obtained.
Statistical
Analysis
All data are presented as means _+SEM. Comparisons between groups were performed by two-tailed unpaired t test or one-way analysis of variance (ANOVA) and Newman-Keuls multiple range test where appropriate. P < 0.05 was considered statistically significant.
10 8
*
6 4
NT&GH
A A
LTB-GH LTB-SAL
*
2 0
0
L
”
I
”
”
”
”
”
”
’
9 10111213141516171819202122232425
DAYS POST
IMPLANTATION
FIG. 1. Mean food intake per day after sham tion. Food intake of smaller tumor burden groups the sake of clarity, but were not different from (NTB) groups. LTB, larger tumor burden group. OVA and Newman-Keuls’ multiple range test for STB groups.
or tumor implantaare not shown for nontumor bearing *P < 0.01, by ANLTB vs NTB and
RESULTS
Hypophysectomized Rats Twelve days administration of hGH resulted in a cumulative weight gain of 21 f 1 g compared to 0 f 1 g in saline-treated hypophysectomized rats. The rate of weight gain per day was constant and did not taper with progressive administration of hGH suggesting that antibody resistance did not develop. Control and Tumor-Bearing Rats The mean food intake per day was similar during the initial period of the experiment for all groups. By contrast, LTB animals exhibited severe anorexia from Days 19 to 23 postimplantation, secondary to increasing tumor burden (Fig. 1). Food intake was not influenced by hGH therapy regardless of tumor-bearing status. Tumor-bearing animals exhibited marked cachexia and administration of hGH did not preserve host lean mass tissue as reflected by carcass weight, gastrocnemius weights, and protein contents (Table 1). Although gastrocnemius weights and protein content were increased in both TB and NTB groups treated with hGH, the differences were not statistically significant. Total muscle RNA and DNA were significantly decreased in tumor bearers but were not affected with hGH treatment. Elevated RNA/protein ratios that reflect potential protein synthetic capacity were elevated in TB groups and are consistent with other reports of increased
102
JOURNAL
OF
SURGICAL
RESEARCH:
VOL.
TABLE Initial Group
N
NTB-GH NTB-SAL STB-GH STB-SAL LTB-GH LTB-SAL
4 8 8 8 5 4
body
and final
carcass,
123 122 122 122 123 127 Protein,
RNA,
-+ + t + + +
contents
NTB-GH NTB-SAL STB-GH STB-SAL
0.59 0.53 0.20 0.18
r t * k
and epididymal
154 150 125 125 106 95
1.27 1.23 0.98 0.96 0.87 0.76
and RNA
* r + t + +
5 5 3’ 6* 7*t 9a.f
to protein
ratios
8.5 9.0 4.7 4.3
r ?I +It
f f k 2 r +-
Epididymal fat pad/ host weight (%)
0.05 0.04 O.OZ* 0.05* 0.04*,** 0.07*,**
of gastrocnemius
RNA (md
0.06 0.02 0.02* 0.02*
1991
fat pad weights
Gastrocnemius weight (g)
Protein M
Group
muscle,
Carcass weight (g)
3 2 2 2 2 2
and DNA
2, AUGUST
1
gastrocnemius
Initial body weight (g)
51, NO.
1.13 1.25 0.80 0.92 0.53 0.59
6.1 5.8 3.8 3.5
s f f -I
RNA/Protein bg/d 0.7 0.5 0.2* 0.3*
Note. Values are means rt SEM. *P -c0.01 TB vs NTB, **P < 0.01 LTB vs STB groups by ANOVA and Newman-Keuls’ No significant differences between GH and SAL groups were observed. NTB, nontumor bearing; STB, smaller tumor burden tumor burden group; GH, growth-hormone-treated; and SAL, saline-treated control.
rates of protein synthesis in cancer-bearing hosts [24]. Epididymal fat pads, reflecting fat stores, were lighter in hGH-treated groups but differences again failed to achieve statistical significance (Table 1). Growth Hormone and IGF-1 Levels Plasma concentrations of hGH in treated groups were markedly elevated when compared to SAL groups. Values ranged from 77 to 222 rig/ml (Table 2) and variations reflect the interval between administration of hGH and collection of plasma samples. However, median concentrations of hGH levels in treated rats from this study are in the supraphysiological range when compared to rat growth hormone levels documented in female rats of the
TABLE
2
Circulating Plasma Growth Hormone Levels, Insulin-Like Growth Factor I Levels, and Bioactivity at Sacrifice Group NTB-GH NTB-SAL STB-GH STB-SAL LTB-GH LTB-SAL
GH (range) (w/ml) 89-222 12 90-156
OF SURGICAL
RESEARCH
51,
99-105
(19%)
Impact of Exogenous Growth Hormone on Host Preservation Tumor Cell-Cycle Distribution in a Rat Sarcoma ModeF* ENG-HEN
NG, MBBS,
FRCS
and
(EDINBURGH),* CRAIG S. ROCK, M.D.,* DOUGLAS LAZARUS, PH.D.,* EVA FISCHER, M.D.,* LYLE L. MOLDAWER, PH.D.,* AND STEPHEN F. LOWRY, M.D., FACS*f3
LISA STAIANO-COICO, PH.D.,? *The
Laboratory
of Surgical Metabolism, New York Hospital-Cornell Submitted
the Department of Surgery, and tFlow Cytometry Medical Center, New York, New York 10021 for publication
Growth hormone (hGH) has been reported to improve nitrogen balances and accrue lean mass tissue in stable subjects. However, the ability of hGH to positively influence host preservation in stressed catabolic states such as cancer-induced cachexia remains unproven. Thirtyseven sham or tumor implanted Fischer 344 rats were randomized to receive either 0.5 mg/kg/day hGH or saline (SAL) subcutaneously from Days 14 to 23 postimplantation. Plasma levels of hGh and somatomedin C/ insulin-like growth factor I (IGF I) as well as IGF I bioactivity were determined at sacrifice. Gastrocnemius muscle protein content was used as a index of host lean tissue mass and the tumor response was evaluated via flow cytometry for analysis of cell-cycle distribution. Host cachexia was not attenuated by hGH as muscle protein content was similar in hGH and salinetreated groups. Despite elevated hGH levels (range, 77-222 rig/ml (GH) vs < 2 rig/ml (SAL)), IGF I levels and bioactivity were not elevated in GH-treated groups. In contrast, cancer-induced anorexia markedly decreased IGF I levels (4 U/ml vs 9 U/ml, NTB; P < 0.01) and this response remained refractory to hGH administration. While final tumor weights were similar between GH- and SAL-treated groups, hGH treatment caused a twofold increase in the proportion of aneuploid cells (P < 0.05). In conclusion, hGH failed to attenuate lean mass dissolution in the tumor bearing host and this response may be related to the failure of IGF I induction. Conversely, the altered proportion of tumor aneuploid cells suggests a direct influence on tumor cell-cycling populations. 0 1991 Academic Press, Inc.
November
1 Presented at the Annual Meeting of the Association for Academic Surgery, Houston, TX, November 14-17, 1990. 2 This study supported by grants from GM-34695 and the American Cancer Society PDT No. 347. a To whom correspondence and reprint requests should be addressed at Department of Surgery, New York Hospital-Cornell Medical Center, 525 East 68th Street, F-2016, New York, NY 10021.
20, 1990
been further substantiated through the recent use of recombinant human growth hormone (hGH) in a variety of clinical and physiological situations [4, 51. Improvements in nitrogen retention were reported in stable patients receiving hGH after gastrointestinal surgery and accrual of lean mass was also noted in highly conditioned athletes administered hGH in supraphysiological quantities. However, it is uncertain if these anabolic properties extend to stressed or cachectic states in humans since the induction of Somatomedin C insulin-like growth factor I (IGF I), which is believed to mediate many of the anabolic processes of hGH, has also been suggested to be impaired in such conditions [61. Further, possible mitogenic influences of hGH such as those observed in lymphoid malignancies [7, 81 could preclude the use of hGH in cancer-bearing hosts. Recently, hGH administered to experimental tumor models has not shown obvious enhancement of malignant activity as reflected by its effect on the size and number of tumor nodules [g-11]. Flow cytometry has been employed to evaluate tumor responses to a variety of therapies [ 12,131 and may provide additional criteria by which tumor responses can be quantified. This study was designed to characterize the effect of exogenous hGH on tumor cell-cycle distribution as well as its influence on preservation of host lean tissue. IGF I levels were also evaluated via radioimmunoassay for determination of circulating levels, and costal cartilage bioassays served as an additional index of IGF I activity. MATERIALS
INTRODUCTION
The potent anabolic properties of growth hormone demonstrated initially with pituitary extracts [l-3] have
Core Facility,
AND
METHODS
Experimental Protocol The experimental protocol was reviewed and approved by the Animal Care and Use Committee of the New York Hospital-Cornell Medical Center. Prior to utilization of hGh in the control and tumor bearing animals, the potency of recombinant human growth hormone (Genentech, South San Francisco, CA) was evaluated on hypophysectomized male Fischer 344 99
0022.4804/91 $1.50 Copyright 0 1991 by Academic Press, Inc. All rights of reproduction in any form reserved.
100
JOURNAL
OF
SURGICAL
RESEARCH:
rats (75-80 g) with once per day administration of hGH (0.5 mg/kg) subcutaneously for 12 days. Daily weights were taken at 9:00 AM. Subsequently, 37 male Fischer 344 rats (115-130 g) were either sham implanted (nontumor bearing, NTB) or inoculated with tumor fragments. Two different sizes of tumor fragments were implanted in an effort to achieve differing levels of tumor burden. The first group was inoculated with approximately 2-mm3 fragments to achieve a smaller tumor burden (STB) and the second group with 5-mm3 fragments to achieve larger tumors (LTB). These three groups (NTB, STB, and LTB) were randomized on Day 14 postimplantation by weight and tumor size to receive either hGH 0.5 mg/kg once daily subcutaneously (GH) or saline (SAL) for 10 days, after which the animals were sacrificed on post-tumor implantation Day 23. Daily weights and food intake were recorded at 9:00 AM. Tumor volume was measured serially twice a week using calipers and tumor weights were extrapolated as previously described [ 141. At sacrifice, tumors were removed and host weights were calculated as the difference between total body weight and tumor weight. Organs and gastrocnemius muscles were removed, weighed, and frozen in liquid nitrogen and stored in -70°C until analysis. Animals
and Tumors
Hypophysectomized and normal Fischer 344 rats were obtained from Charles River Breeding Laboratories (Wilmington, Massachusetts) and maintained on a 12hr light, 12-hr dark cycle in individual cages in a temperature-controlled room and were given standard rat chow and water ad libitum. Hypophysectomized rats were weighed periodically and only rats with weight gain of less than 1 g over 2 weeks and without increase in testicular size were used for experiments. Rats were anesthetized with pentobarbital 36 mg/kg ip and inoculated with methylcholanthrene (MCA) sarcoma tumor fragments in the interscapular region. Tumor tissue was obtained from a tumor-bearing donor animal which had been maintained by serial transplantation for about 1 year. This tumor is a locally invasive fibrosarcoma that seldom metastasizes and uniformly kills the animal 30-35 days following inoculation [l5]. RNA/DNA
Determinations
The RNA and DNA contents were determined by previously described techniques [ 161. Tissues were homogenized in ice-cold water and nucleic acids were precipitated with cold perchloric acid. RNA was digested to free nucleotides in alkali and the content of nucleotides was determined spectrophotometrically. DNA content was determined colorometrically [ 171. Protein
Content
Protein contents of the muscle homogenates were determined calorimetrically by methods according to Bradford [ 181.
VOL.
Flow
51, NO.
2, AUGUST
Cytometry
Analysis
1991
Only tumors from the STB groups were analyzed. Preparation of single cell suspensions. Immediately after sacrifice, a sliver of grossly viable tissue measuring approximately 3 X 3 X 0.3 cm was obtained from the periphery of the tumor, ensuring that no necrotic tissue usually located at the center was included. Specimens were immersed in Hank’s balance salt solution (Mediatech) and placed on ice. Digestion of tumor specimens into single cell suspension was obtained using a cocktail of three enzymes-collagenase D (1 mg/ml), hyaluronidase (1 mg/ml), and DNAse (50 pug/ml) (BoehringherMannheim, Indianapolis). Tumor specimens were finely minced in petri dishes and added into a flask containing 25 cc of HBSS and 5 cc each of individual enzymes and were agitated for 2-4 hr at room temperature. The suspension was then filtered through a 52-pm nylon mesh (Nytrex) and pelleted out at 1500 rpm for 10 min. The pellet was washed twice after lysis of RBC with sterile water. Finally cells were resuspended in ice-cold 70% alcohol while vortexing and stored at 4°C until analysis by flow cytometry. Cell Cycle Distribution
Analysis
Alcohol-fixed single-cell suspensions were pelleted and resuspended in PBS containing 300 Kunitz units of ribonuclease A (Sigma, St Louis, MO) for 1 hr at room temperature (to eliminate artifactual broadening of DNA content distributions which would otherwise result from fluorescence of complexes of the DNA-staining dye with double-stranded RNA) before staining with 50 pg propidium iodide for 30 min on ice. The number of fluorecscent nuclei were quantitated using a Epics 752 (Coulter Electronics Corp., Hialeh, FL) cell sorter. The proportion of doublets and dead cells were eliminated from analysis using either a pulse width, time of flight measurement, or the peak versus integrated signals for red fluorescence. The cells remaining in the analysis were then quantitated. At least 12,000 and up to 50,000 cells were analyzed for each sample. Since propidium iodide fluorescence is proportional to double-stranded nuclei [ 191, the resulting fluorescence was proportional to DNA content. Histograms of DNA content were included for analysis only if the coefficient of variation of diploid GO/G1 peak was less than 6%. Peripheral blood mononuclear cells (PBMN) were used as internal diploid controls. Specimens exhibiting nondiploid/nontetraploid DNA content were considered to have aneuploid content. Each tumor sample was analyzed using the DNAFIT software package which determines the proportion of diploid, aneuploid, and S-phase cells from a DNA histogram based on a method employing “multirectangles” [20, 211. Since the DNA histograms of all tumors contained two GO/G1 cell populations which overlapped,
NG
ET
AL.:
GROWTH
HORMONE
AND
the multirectangle analysis for the S-phase fraction could only be performed on the aneuploid population. The results of analyses are presented as the aneuploid/ diploid ratio defined as the proportion of aneuploid to diploid cells and the percentage of aneuploid S-phase, defined as the proportion of cells in this compartment over total number of cells.
CANCER
CACHEXIA
IN
A RAT
MEAN
101
MODEL
FOOD INTAKE
PER DAY
20
16 14 12
Growth Hormone Determination
and Insulin-Like
Growth
Factor
1
Radioimmunoassay. Radioimmunoassay (RIA) techniques were used to determine concentrations of hGH (Kallestadt) and IGF-1 (Nichols Institute, San Juan Capistrano, CA). Plasma (EDTA) specimens were obtained at sacrifice and fractions were separated by centrifugation and stored at -70°C until analysis. The interval between administration of hGH and collection of plasma samples did not exceed 2 hr. IGF-I concentrations are expressed as U/ml. In order to avoid interference by binding proteins, samples were analyzed after acid extraction [ 221. Costal cartilage bioassay. Somatomedin C/IGF I bioassay using the rat costal cartilage was modified after Phillips [23] and only samples from STB groups were analyzed, using NTB-SAL as controls. Costa1 cartilages from a single hypophysectomized rat were cleaned and divided into 80-100 segments. Individual pieces were then placed in 96-well plates with a final volume of 0.1 ml Krebs phosphosaline-amino acid (KPS-AA) buffer supplementation with antibiotics, glucose, and [35S] sulfate (Amersham, Illinois) with or without added serum samples. Test and standard samples were added for a final concentration of 1% (vol/vol). Test rat plasma (EDTA) was obtained at sacrifice and analyzed individually. Standard plasma was obtained from normal rats, pooled and aliquots were frozen at -70°C. After 48-hr incubation, the cartilage was soaked in boiling water, then in saturated sodium sulfate, rinsed, dried and weighed, and digested with 5 M KOH; scintillation fluid was added and disintegrations per minute were measured. Cartilage uptake was measured as counts per minute per milligram of dry cartilage and expressed as the percentage of counts obtained with samples containing standard plasma. Five repeated measurements of individual test and standard samples were performed and values are means obtained.
Statistical
Analysis
All data are presented as means _+SEM. Comparisons between groups were performed by two-tailed unpaired t test or one-way analysis of variance (ANOVA) and Newman-Keuls multiple range test where appropriate. P < 0.05 was considered statistically significant.
10 8
*
6 4
NT&GH
A A
LTB-GH LTB-SAL
*
2 0
0
L
”
I
”
”
”
”
”
”
’
9 10111213141516171819202122232425
DAYS POST
IMPLANTATION
FIG. 1. Mean food intake per day after sham tion. Food intake of smaller tumor burden groups the sake of clarity, but were not different from (NTB) groups. LTB, larger tumor burden group. OVA and Newman-Keuls’ multiple range test for STB groups.
or tumor implantaare not shown for nontumor bearing *P < 0.01, by ANLTB vs NTB and
RESULTS
Hypophysectomized Rats Twelve days administration of hGH resulted in a cumulative weight gain of 21 f 1 g compared to 0 f 1 g in saline-treated hypophysectomized rats. The rate of weight gain per day was constant and did not taper with progressive administration of hGH suggesting that antibody resistance did not develop. Control and Tumor-Bearing Rats The mean food intake per day was similar during the initial period of the experiment for all groups. By contrast, LTB animals exhibited severe anorexia from Days 19 to 23 postimplantation, secondary to increasing tumor burden (Fig. 1). Food intake was not influenced by hGH therapy regardless of tumor-bearing status. Tumor-bearing animals exhibited marked cachexia and administration of hGH did not preserve host lean mass tissue as reflected by carcass weight, gastrocnemius weights, and protein contents (Table 1). Although gastrocnemius weights and protein content were increased in both TB and NTB groups treated with hGH, the differences were not statistically significant. Total muscle RNA and DNA were significantly decreased in tumor bearers but were not affected with hGH treatment. Elevated RNA/protein ratios that reflect potential protein synthetic capacity were elevated in TB groups and are consistent with other reports of increased
102
JOURNAL
OF
SURGICAL
RESEARCH:
VOL.
TABLE Initial Group
N
NTB-GH NTB-SAL STB-GH STB-SAL LTB-GH LTB-SAL
4 8 8 8 5 4
body
and final
carcass,
123 122 122 122 123 127 Protein,
RNA,
-+ + t + + +
contents
NTB-GH NTB-SAL STB-GH STB-SAL
0.59 0.53 0.20 0.18
r t * k
and epididymal
154 150 125 125 106 95
1.27 1.23 0.98 0.96 0.87 0.76
and RNA
* r + t + +
5 5 3’ 6* 7*t 9a.f
to protein
ratios
8.5 9.0 4.7 4.3
r ?I +It
f f k 2 r +-
Epididymal fat pad/ host weight (%)
0.05 0.04 O.OZ* 0.05* 0.04*,** 0.07*,**
of gastrocnemius
RNA (md
0.06 0.02 0.02* 0.02*
1991
fat pad weights
Gastrocnemius weight (g)
Protein M
Group
muscle,
Carcass weight (g)
3 2 2 2 2 2
and DNA
2, AUGUST
1
gastrocnemius
Initial body weight (g)
51, NO.
1.13 1.25 0.80 0.92 0.53 0.59
6.1 5.8 3.8 3.5
s f f -I
RNA/Protein bg/d 0.7 0.5 0.2* 0.3*
Note. Values are means rt SEM. *P -c0.01 TB vs NTB, **P < 0.01 LTB vs STB groups by ANOVA and Newman-Keuls’ No significant differences between GH and SAL groups were observed. NTB, nontumor bearing; STB, smaller tumor burden tumor burden group; GH, growth-hormone-treated; and SAL, saline-treated control.
rates of protein synthesis in cancer-bearing hosts [24]. Epididymal fat pads, reflecting fat stores, were lighter in hGH-treated groups but differences again failed to achieve statistical significance (Table 1). Growth Hormone and IGF-1 Levels Plasma concentrations of hGH in treated groups were markedly elevated when compared to SAL groups. Values ranged from 77 to 222 rig/ml (Table 2) and variations reflect the interval between administration of hGH and collection of plasma samples. However, median concentrations of hGH levels in treated rats from this study are in the supraphysiological range when compared to rat growth hormone levels documented in female rats of the
TABLE
2
Circulating Plasma Growth Hormone Levels, Insulin-Like Growth Factor I Levels, and Bioactivity at Sacrifice Group NTB-GH NTB-SAL STB-GH STB-SAL LTB-GH LTB-SAL
GH (range) (w/ml) 89-222 12 90-156
