Gene profiling of maternal hepatic adaptations to pregnancy

Abstract

Background: Maternal metabolic demands change dramatically during the course of gestation and must be co-ordinated with the needs of the developing placenta and fetus. The liver is critically involved in metabolism and other important functions. However, maternal hepatic adjustments to pregnancy are poorly understood.

Aim: The aim of the study was to evaluate the influences of pregnancy on the maternal liver growth and gene expression profile.

Methods: Holtzman Sprague-Dawley rats were mated and sacrificed at various stages of gestation and post-partum. The maternal livers were analysed in gravimetric response, DNA content by PicoGreen dsDNA quantitation reagent, hepatocyte ploidy by flow cytometry and hepatocyte proliferation by ki-67 immunostaining. Gene expression profiling of non-pregnant and gestation d18.5 maternal hepatic tissue was analysed using a DNA microarray approach and partially verified by northern blot or quantitative real-time PCR analysis.

Results: During pregnancy, the liver exhibited approximately an 80% increase in size, proportional to the increase in body weight of the pregnant animals. The pregnancy-induced hepatomegaly was a physiological event of liver growth manifested by increases in maternal hepatic DNA content and hepatocyte proliferation. Pregnancy did not affect hepatocyte polyploidization. Pregnancy-dependent changes in hepatic expression were noted for a number of genes, including those associated with cell proliferation, cytokine signalling, liver regeneration and metabolism.

Conclusions: The metabolic demands of pregnancy cause marked adjustments in maternal liver physiology. Central to these adjustments are an expansion in hepatic capacity and changes in hepatic gene expression. Our findings provide insights into pregnancy-dependent hepatic adaptations.

Fig. 1. Maternal liver weight responses to pregnancy and lactation in the rat

Tissues were collected from non-pregnant (np), pregnant (gestational d4.5, d8.5, d11.5, d13.5, d15.5, d18.5, and d21.5), and postpartum d10 non-lactating (nlac) and lactating (lac) rats and weighed. Maternal liver weight responses are shown in Panel A. Representative gross morphology of nonpregnant (NP) and gestation d18.5 livers is shown in Panel B. Liver weight responses were also normalized to body weight as shown in Panel C. Asterisks indicate values that are significantly different from values for non-pregnant animals (P < 0.05; n = 5 to 7).

Fig. 2. Measurement of liver DNA content from non-pregnant, pregnant, and post-partum rats

Liver tissues collected from non-pregnant (np), pregnant (gestational d11.5 and d18.5) and postpartum d10 non-lactating (nlc) and lactating (lac) rats were completely digested overnight. DNA content was measured using Quant-iT PicoGreen dsDNA Kit. DNA concentrations were determined using a standard curve of fluorescence emission intensity plotted versus DNA concentration. Asterisks indicate values that are significantly different from values for non-pregnant animals (P < 0.05; n = 5 to 7).

Fig. 3. Hepatocyte ploidy analysis

Primary rat hepatocytes were isolated from non-pregnant (np) and gestational d18.5 animals and stained with propidium iodide. Hepatocyte ploidy (2N, 4N, and 8N) was analyzed by flow cytometry using BD LSRII and FACS Diva. Histograms from the flow cytometry are shown in Panel A. Populations of hepatocytes with different DNA contents are shown in Panel B.

Fig. 4. Ki-67 immunostaining in liver tissues

Liver tissues were collected from nonpregnant (NP), pregnant (gestational d4.5, d8.5, d11.5, d13.5, d15.5, d18.5, and d21.5), and postpartum d10 non-lactating (nlac) and lactating (lac) rats. The tissues were fixed in formalin and embedded in paraffin. Liver tissue sections were prepared and Ki-67 immunostaining was performed. The nuclei of Ki67-positive cells stained dark brown. (A) Representative liver sections immunohistochemically stained for Ki-67 are shown for NP and gestation d18.5 rats. (B) Ki67-positive hepatocytes were counted (40 × optical field) and the results are shown as mean number of positive nuclei per field ± SD. Asterisks indicate values that are significantly different from values for non-pregnant animals (P < 0.05; n = 3).

Fig. 4. Ki-67 immunostaining in liver tissues

Liver tissues were collected from nonpregnant (NP), pregnant (gestational d4.5, d8.5, d11.5, d13.5, d15.5, d18.5, and d21.5), and postpartum d10 non-lactating (nlac) and lactating (lac) rats. The tissues were fixed in formalin and embedded in paraffin. Liver tissue sections were prepared and Ki-67 immunostaining was performed. The nuclei of Ki67-positive cells stained dark brown. (A) Representative liver sections immunohistochemically stained for Ki-67 are shown for NP and gestation d18.5 rats. (B) Ki67-positive hepatocytes were counted (40 × optical field) and the results are shown as mean number of positive nuclei per field ± SD. Asterisks indicate values that are significantly different from values for non-pregnant animals (P < 0.05; n = 3).

Fig. 5. Hepatic gene expression during pregnancy and lactation in the rat

Total RNA was prepared from liver tissue of non-pregnant (np), pregnant (gestational d4.5, d8.5, d11.5, d13.5, d15.5, d18.5, and d21.5), and postpartum d10 non-lactating (nlac) and lactating (lac) rats. (A) The expression of indicated genes was analyzed by northern blotting. G3PDH served as control for loading and RNA integrity. (B) Maternal hepatic mRNA levels of indicated genes were assessed by quantitative RT-PCR and are expressed as means of fold changes compared to nonpregnant controls ± SD (n = 3 for each group). *, P < 0.05.

Fig. 5. Hepatic gene expression during pregnancy and lactation in the rat

Total RNA was prepared from liver tissue of non-pregnant (np), pregnant (gestational d4.5, d8.5, d11.5, d13.5, d15.5, d18.5, and d21.5), and postpartum d10 non-lactating (nlac) and lactating (lac) rats. (A) The expression of indicated genes was analyzed by northern blotting. G3PDH served as control for loading and RNA integrity. (B) Maternal hepatic mRNA levels of indicated genes were assessed by quantitative RT-PCR and are expressed as means of fold changes compared to nonpregnant controls ± SD (n = 3 for each group). *, P < 0.05.