Discussion
In this study, we measured plasma FAA and carnitine levels in HM patients and healthy pregnant women as a control group. The original finding of the present study was that three amino acids (i.e., alanine, arginine, and valine) were significantly decreased, and three carnitines (i.e., C8DC, C16:1, and C18) were significantly increased in the pregnant women with HM.
The mean plasma total carnitine concentration in non-pregnant women has been reported to be approximately 40 µmol/L [16]. It has been found that plasma carnitine concentrations decreased during pregnancy and this decrease was mostly in the first half of pregnancy [17]. The reasons for these low plasma carnitine concentrations in pregnancy are currently unknown, but several hypotheses have been asserted: Ringseis et al. speculated that plasma carnitine concentrations may be low in pregnant women due to reduced carnitine synthesis [18]. However, their study had a clear limitation because the number of subjects was very small. Cho and Cha [17] demonstrated that low carnitine levels were caused by increased urinary loss of carnitine in Korean pregnant women. Grube et al. [8] showed that increased expression of carnitine/organic cation transporter 2 (OCTN2), a carnitine-transporting protein in human placenta, can play an important role in carnitine supply of growing fetus especially in early gestation. Grube et al. [8] detected that this protein located in the membrane of syncytiotrophoblast, had a dominant role for carnitine uptake from maternal circulation. However, they have not simultanously measured the carnitine level changes in maternal plasma. Bai et al. [19] have also investigated the reason of carnitine reduction and concluded that OCTN2 mediated L-Car transfer across placenta played a major role in maternal plasma L-Car reduction during pregnancy. We found higher carnitine levels in plasma of pregnant women with HM compared to the healthy pregnant women. This finding suggests a defect in carnitine uptake by syncytiotrophoblasts due to the absence of a healthy pregnancy in HM patients or decreased expression of OCTN2 protein in HM cases. Although we did not study OCTN2 protein in pathology specimens of HM patients, this speculation may be a hypothesis for further studies.
Studies on amino acids have been previously published in several cancers and obstetric diseases such as gestational diabetes [20-23]. Neoplastic cells need some amino acids, such as glutamine, glycine, aspartic acid and serine, for the new vessel formation, DNA synthesis, duplication of protein, and synthesis of hormones. In an in vitro study on mice, Roux et al. [12] demonstrated that the glutamine amino acid uptake of pancreatic adenocarcinoma cells increased compared to non-pancreatic tumor cells, therefore, the plasma glutamine levels decreased in the environment. Additionally, Saglik et al. [24] found that glutamine amount was higher in pterygium tissue than in normal conjunctival tissue. Both studies were designed at the tissue level, unlike ours. In our study, circulating plasma glutamine levels were found similar in both groups, however, alanine, arginine, and valine levels were lower in HM group than in healthy pregnants. This discrepancy may be due to different histopathogenesis of the diseases, differences in method used, small sizes of our study, female gender or younger age of our participants.
Liu and colleagues [25] pointed out changes in FAA concentrations in both plasma and gastric fluid, including threonine, serine, alanine, valine, methionine, isoleucine, leucine, tyrosine, phenylalanine, lysine, and arginine in the diagnosis of gastric cancer. They found an increase in gastric fluid FAA levels and a decrease in plasma FAA concentrations in gasric cancer patients compared to nongastric cancer patients. The inconsistency of these metabolic phenotypes between plasma and gastric fluid in gastric cancer patients (opposite FAA levels) was explained as abnormal accumulation of several metabolites in tumor microenvironment. Camelo et al. [13] showed that high FAA levels in the intervillous space were compatible with syncytiotrophoblast activity in healthy pregnant women. They thought that these high levels occurred due to an asymmetric influx or active transport from the trophoblast cells to the blood in the intervillous space. However, there is no study researching the association between the plasma FAAs and HM. As the first study in the literature, we found decreased concentrations in several maternal plasma FAAs in patients with HM. This finding may be caused by impaired active transport from the intervillous space to the blood in patients with HM, or due to excessive accumulation of FAAs in trophoblastic microenvironment, because trophoblastic cells are highly proliferative and are similar to cells in cancer tissue. Although molar pregnancy is not malignant tumor, it has a risk for gestational trophoblastic neoplasia. However, we think that if the tissue-supported studies are performed, it will contribute to our results.
Miyagi et al. [20] suggested that plasma FAA profiling has an important role for cancer screening and diagnosis in patients with asymptomatic early-stage of the disease. Morever, they have demonstrated that the differences in FAA metabolism did not releated with impaired nutritional support of cancer patients. In contrast, some authors have suggested that low plasma FAA levels were associated with anorexia, malnutrition, and progressive weight loss in cancer patients. Thus, the factors affecting plasma FAA levels in these subjects still remain controversial and unclear [10,11]. Our HM patients had impaired nutritional support due to hyperemesis, but there was no correlation between the plasma FAA and urine ketone levels.
Several factors, such as obesity and food intake, may affect the serum FAA and carnitine levels. Rigamonti et al. [26] showed that whey consumption in obese female subjects increased some circulating amino acids (alanine, arginine, asparagine, citrulline, glutamine, hydroxyproline, isoleucine, histidine, leucine, lysine, methionine, ornithine, phenylalanine, proline, serine, threonine, tyrosine and valine). They reported that of these amino acids, isoleucine, leucine, lysine, methionine, phenylalanine, proline, tyrosine, and valine correlatated negatively with starvation and positively with satiety. Additionally, Yamada et al. [27] speculated that deficiencies of certain dietary substrats, such as folic acid, proteins, and vitamin B-12, which are essential for nucleic acid formation in decidual cells, may affect the risk of abortion releated with chromosomal disorders. Although, we did not measure other circulating parameters such as folic acid and vitamins, we showed that HM patients had different plasma FAA and carnitine levels compared to the healthy pregnants. We excluded obese pregnants, smokers, and women who use food supplements from the study to avoid changes in plasma FAA and carnitine levels that may occur due to these factors. However, further epidemiological and biochemical studies are needed to obtain a more precise definition of specific dietary correlation.
The limitations of this case-control study were that it was designed only between pregnant women and only by measuring the maternal plasma. However, this study is the first in the literature to investigate FAA and carnitine profiling in HM. Although more research is needed to support our results, including larger sample sizes and investigating tissue samples, we think current results may be considered as references for subsequent studies on HM.