Main

During the early stages of rat gestation, tissue anabolism is manifested by increased accretion of adipose tissue(1), liver glycogen(2), and lean tissue mass(3). In the final days of a 3-wk pregnancy, insulin resistance emerges, and exaggerated tissue catabolism supervenes, particularly during periods of fasting. Adipose tissue turnover is accelerated(1), liver glycogen content falls(2), and lean body mass is markedly reduced(3).

There is a physiologic purpose for the onset of a catabolic phase in this prepartum period, because mobilization of stored nutrients guarantees their flux to rapidly growing fetuses whose nutrient demands rise near term. Hyperphagia and nitrogen retention contribute to the positive nitrogen balance common in gestation(4). At late gestation, protein turnover accelerates whereas amino acid catabolism decreases, both processes contributing to an enhanced amino acid availability to the growing fetus(5). These events are remarkably similar to what is observed during the course of normal human pregnancy(6). The shift from an early anabolic period to a late catabolic stage is believed to be caused primarily by the changing hormonal status of advancing pregnancy.

To some extent, the fetus can be compared with a fast growing tumor because their growth rates are exponential, both being highly dependent on glucose and amino acids. The Yoshida AH-130 ascites hepatoma is a rapidly growing tumor with a volume doubling time of 1 d, and its implantation to normal rats severely alters protein turnover toward a clear catabolic state(7). Consequently, the implantation of this tumor to pregnant rats may alter maternal protein turnover as well as fetal growth. Previous studies with pregnant rats have shown a normal fetal growth coexisting with a tumor growth that was not related to either an enhanced fetal uptake of amino acids or an increased blood flow to the fetus(8). In fact, we have recently described, using reconstituted placental membrane vesicles, important changes in placental transport as a result of tumor growth, affecting the kinetics of both alanine and leucine transport(9).

Because tumor burden does not seem to affect fetal growth in our rat model, the aim of the present investigation was to determine whether the presence of the tumor modifies maternal protein turnover to support both the malignant(tumor) and nonmalignant (fetus) growth.

METHODS

Animals. All animals (female Wistar rats) were fed ad libitum on a chow diet consisting (by weight) of a 54% carbohydrate, 17% protein, and 5% fat (the residue was nondigestible material), with free access to drinking water, and were maintained at an ambient temperature of 22± 2°C with a 12-h light/12-h dark cycle (lights on from 0800 h). They were mated at 7 wk of age (200 g of body weight, approximately). Day 0 of pregnancy was considered when the presence of spermatozoa in vaginal smears was detected. Rats pregnant at 13, 17, and 20 d were used in the experiments.

Biochemicals and radioactive compounds. They were all reagent grade and were obtained either from Boehringer Mannheim S.A. (Barcelona, Spain) or from Sigma Chemical Co. (St. Louis, MO). [14C]Bicarbonate(specific activity, 0.1 mCi/mmol) was obtained from Amersham Corp. (Amersham, Bucks., England).

Tumor implantation. A Yoshida AH-130 ascites hepatoma cell suspension (approximately 120 × 106 cells in 2 mL) was injected intraperitoneally; control rats were injected with 2 mL of 0.9% (w/v) NaCl solution. Total cell content was estimated using trypan blue staining. All the experiments took place at 0, 4, or 7 d after tumor transplantation, days at which the cells were growing exponentially (d 4) or had reached the stationary phase (d 7)(7, 10). To carry out the experiments with 17- and 20-d pregnant rats, they were inoculated previously with the tumor at d 13 of gestation.

Estimation of fractional protein turnover rates. Animals were injected intraperitoneally with 400 μCi/kg of body weight with NaH14CO3 dissolved in 0.15 M NaCl, 24 h before tumor inoculation. At 0, 4, and 7 d after tumor inoculation, the animals corresponding to each group were killed. Liver and gastrocnemius muscle were rapidly weighed and homogenized. Total protein content was determined using the Lowry et al.(11) method. Trichloroacetic acid-insoluble proteins were processed for lipid extraction, extensively hydrolyzed, and counted in a liquid scintillation counter. Total protein radioactivity and specific protein radioactivity (per mg of protein) were evaluated for gastrocnemius muscle and liver. Fractional rates of protein synthesis (Ks), degradation (Kd), and accumulation (Ka) were calculated by the followingequations: as previously reported(12, 13), and expressed as percent/d. The procedure adopted in studying protein radioactivity decay, with zero time corresponding to 24 h after administration of the label, takes into account only the long life pool of tissue proteins, which represents in muscle the large majority of the protein mass.

RNA isolation and Northern blot analysis. Total RNA from soleus muscle was extracted using the acid guanidinium isothiocyanate/phenol/chloroform method(14). RNA samples (20 μg/mL) were denaturated, subjected to electrophoresis in 1.2% agarose gels containing 6.3% formaldehyde, and transferred to Hybond N membranes (Amersham Corp.). RNA was fixed to membrane by UV cross-linking. The RNA in gels and in filters was visualized by ethidium bromide dying in a UV transilluminator to ensure the integrity of RNA, to check the loading of equivalent amounts, and to confirm proper transfer. RNA was transferred in 20× standard saline citrate (SSC; 0.15 M NaCl and 15 mM sodium citrate, pH 7.0). Hybridization with denatured fluorescein-labeled probes was done at 65°C overnight in the hybridization buffer (0.25 M Na2HPO4/7% SDS/1 mM EDTA/1% BSA/10% dextran sulfate). Labeled probes were prepared by the random primer method (Amersham). The ubiquitin probe used was a cDNA clone containing 12 pairs of the second ubiquitin coding sequence plus a complete third and fourth ubiquitin coding sequence and 120 bp of the 3′-untranslated region of the chicken polyubiquitin gene UBI(15). Filters were exposed to Hyperfilm-MP (Amersham) at room temperature for 5 min.

Plasma hormones and blood metabolites. Whole blood glucose was determined by the method of Slein(16). Lactate was measured by the method of Hohorst et al.(17). Plasma insulin was determined by RIA using a rat insulin standard(18). Circulating corticosterone was estimated using a specific rat immunoassay kit (IDS, Boldon, England).

Amino acid analysis. Blood samples were sonicated for 5 min and then diluted 1:1 with distilled water. After mixing, the mixture was deproteinized using 10% trifluoroacetic acid. After centrifugation at 12 000× g for 30 min, the clear supernatants were filtered with a Millipore filter (10 000 molecular weight cut-off; Millipore Corp., Bedford, MA) and then submitted to vacuum evaporation to eliminate the acid. After resuspension of the deproteinized samples in distilled water, amino acid analysis was carried out by means of an Alpha Plus amino acid analyzer(Pharmacia LKB Biotechnology, Uppsala, Sweden), using ninhydrine as the amino group reagent.

Statistical analysis of the results. Comparison between groups were made by means of a t test; p<0.05 was considered statistically significant.

RESULTS AND DISCUSSION

Maternal body and fetoplacental unit weights and tumor volume. As clearly shown in Figure 1A, important changes took place during late gestation concerning body weight (excluding both the fetoplacental unit and the tumor). As previously described(3), a clear shift was observed between d 13 and 17 of gestation, being the maternal final body weight at d 20 lower than that corresponding to virgin rats. The implantation of the ascites hepatoma during the last third of gestation altered maternal body weight, the effects being already clear at d 17 of gestation (4 d after tumor inoculation). Despite these changes on the mother, the growth of the fetoplacental unit (Fig. 1B) and the tumor cell number (Fig. 1C) were similar to those previously reported(8, 9).

Figure 1
figure 1

(A). Maternal body weight (excluding the fetoplacental unit and tumor) during late gestation is expressed as a percentage of the initial body weight and referred to virgin rats. Data are means ± SEM (n=5-7). Virgin, closed circles; pregnant, closed triangles; pregnant + tumor, closed squares. Statistical significance of the results (tumor-bearers vs pregnant): **p<0.01. (B) Increase in the fetoplacental unit weight in both groups of pregnant rats, expressed as a percentage of the initial maternal body weight. (C) Total cell number during tumor growth in pregnant rats.

Full size image

Tissue weights and protein content. Tissue wet weight(Table 1) and protein (Table 2) are referred to 100 g of initial body weight for the purpose of standardization(7, 13). Pregnancy inflicted changes in most of the maternal tissue weights studied (liver, gastrocnemius, soleus, and white adipose tissue). Previous studies have reported that maternal tissues develop both hypertrophy and hyperplasia during gestation, being the liver the tissue most clearly affected(19).

Table 1 Maternal tissue weights
Full size table
Table 2 Total protein content in gastrocnemius muscle and liver
Full size table

Concerning liver, a clear hypertrophy was present over all the period studied in control pregnant rats parallel to an increase in protein content(Table 2), as previously reported(20). On the other hand, it is well established that liver undergoes a biphasic behavior due to the tumor presence in virgin rats(7), characterized by an initial hypertrophy (associated with decreased protein degradation) followed by a huge decrease in its weight. Surprisingly, these alterations were not evident in pregnant tumor-bearing rats neither in wet weight nor in protein content.

Gastrocnemius muscle wet weight decreased at the end of gestation, thus revealing the catabolic state present in the mother at this stage; this process was accelerated by the presence of the tumor, although no differences were found between pregnant groups at d 7 of tumor growth. A slight decrease in gastrocnemius protein content was found at d 20 of gestation, whereas this was detected earlier in the tumor group (Table 2). These results agree with those observed in virgin tumor-bearing animals, although a much lesser decrease is found in pregnant rats. Similar effects were observed in soleus muscle, whereas heart weight was not altered during all the period.

Although gestation did not alter interscapular brown adipose tissue weight in any group, white adipose tissue weight increased at the end of pregnancy. Physiologically, these lipidic reserves would contribute to the maternal nutrient supply during late gestation and, probably, the initial lactation. This increase in white adipose tissue weight was also present, to the same extent, in the tumor-bearing group. So, despite tumor growth, no decrease in adipose tissue weight was found in pregnant rats, in contrast with what was expected(21). From all these observations, a protective action on adipose tissue weight during late gestation can be suggested; this takes place even in the presence of the tumor.

Fetal tissue weight and protein content. During the last third of gestation, the fetus grows exponentially as can be seen in Table 3. Since no differences were observed in the fetoplacental unit weight during tumor burden, and to discard possible differences due to changes in amniotic fluid volume, both fetal and placental weights and protein contents were assessed. In fact, both groups of fetuses grew at the same rate at d 7 (d 20 of gestation; 16-fold with respect to d 13), and neither the fetus nor the placenta wet weights were altered by the tumor. In agreement with these observations, protein content was maintained in both tissues. These results also support a protective action of pregnancy on fetal growth, in spite of tumor burden.

Table 3 Fetal tissue weight and protein content
Full size table

Fractional tissue protein turnover rates. The effects of pregnancy and tumor growth on liver protein turnover rates are shown in Table 4. No differences were found between virgin and pregnant control animals in any of the fractional turnover rates assessed, although, as usually seen in virgin tumor-bearing rats(22), tumor growth effected increasing Kd. Nevertheless, it is interesting to point out that these changes were already present at the early 0-4-d interval in virgin rats, whereas in pregnant rats they appeared later (at the 4-7-d interval). This evidence is in favor of the idea that pregnancy protects against the cachexia induced by tumor or, at least, retards its appearance. Little or no changes were observed in liver protein synthesis due to tumor burden, being only slightly decreased at the 4-7-d interval in the pregnant group. These changes were more evident in Ka during the second interval studied, where the pregnant control group behaved as virgin, whereas a 5-fold decrease in the pregnant tumor group was found. However, compared with virgin tumor-bearing rats, this decrease in Ka was more similar to the 0-4- than the 4-7-d interval value(22).

Table 4 Fractional tissue protein turnover rates
Full size table

Concerning skeletal muscle, gastrocnemius protein turnover was assessed. As a result of pregnancy, there was an increase in Kd at the end of gestation (4-7-d interval), where, in spite of an increase in Ks, a clearly negative protein accumulation was observed. As a result of these changes, the global behavior during the 0-7-d interval reflected the increase in both Kd and Ks, and also the negative Ka. All these changes in muscle protein turnover would favor the release of amino acids from skeletal muscle. This, added to the described retention of nitrogen observed in the mother(4), together with the low activity of amino acid catabolic enzymes(5), suggests that the fetus benefits from the utilization of amino acids for both energetic and anabolic fates.

Tumor growth resulted in an accelerated gastrocnemius protein degradation, because great changes were already present at the 0-4-d interval. Previous studies in our laboratory revealed an activation of the ubiquitin-dependent proteolytic pathway in different catabolic situations such as tumor growth or sepsis(23). Bearing this in mind, we studied the expression of the ubiquitin gene in soleus muscles from these animals. The results indicated that the increased proteolytic rates were accompanied by an augmented expression of the ubiquitin-dependent proteolytic system in tumor-bearing pregnant rats (Fig. 2), similar to that found in tumor-bearing virgin rats(24). Surprisingly, as a result of a great reduction in both Kd and Ks during the second interval studied (4-7-d), the global appreciation of these changes was similar to the control pregnant group, and only little differences in Kd and Ks resulted in a lesser Ka (only 1.7-fold compared with 4.6-fold normally found in virgin tumor-bearing rats)(22).

Figure 2
figure 2

Expression of ubiquitin mRNAs in soleus muscles of pregnant (P) and tumor-bearing pregnant rats (PT) at d 4 of tumor growth; a representative Northern blot is shown.

Full size image

Effect of tumor growth on plasma hormones and circulating amino acids. Total amino acids were decreased 7 d after tumor inoculation(Table 5), as previously reported(8). There was a significant increase in plasma insulin in the pregnant group, possibly as a result of the well known insulin resistance present during gestation(6). In addition, there was a decrease in insulin levels in response to tumor growth(Table 5), as previously seen in virgin animals bearing this tumor(22).

Table 5 Plasma hormones and total amino acids
Full size table

An interference between glucocorticoids and insulin action at the receptor and/or postreceptor level has been described(25, 26). Consequently, high levels of glucocorticoids may explain the resistance to insulin present during pregnancy. Under our conditions, corticosterone levels remained unchanged between virgin and pregnant rats(Table 5), and surprisingly, no increase was found in the tumor group as a consequence of tumor growth, in contrast to what was expected(22). However, a possible role of glucocorticoids modulating protein turnover cannot be discarded, because the lack of increase in the tumor-bearing group at d 7 could normalize protein turnover rates to normal levels in pregnancy.