Gestational diabetes mellitus (GDM), defined as any degree of glucose intolerance with onset or first recognition during pregnancy (1), affects a considerable number of pregnant women worldwide (2). GDM is associated with many short- and long-term adverse health outcomes for mothers and their offspring. Mothers with GDM are more likely to develop preeclampsia during pregnancy and type 2 diabetes mellitus (T2DM) 5-10 years after delivery (3, 4). The offspring of mothers with GDM have a higher incidence of macrosomia and a higher risk of diabetes and other cardiometabolic diseases later in life (5–7). Both genetic susceptibility and lifestyle risk factors have been shown to play a role in the etiology of GDM, yet the impact of environmental exposure remains unclear (8–10). Recently, the effects of environmental metal exposure on GDM risk have attracted increasing attention worldwide. Different metals exhibited protective (11–13) or deleterious (14–17) effects on the development of GDM.

Rare earth elements (REEs) have gained worldwide attention due to the rapid development of industrial, agricultural and medical technologies in the last decades. China has the world’s richest reserves of REEs, sufficient to meet the global demand (18, 19). REEs are beneficial elements for plants and have been used in fertilizers in Chinese agriculture for about 40 years (20). REEs are also used in medicine. For example, rare earth-doped nanoparticles have the potential to target and treat glioblastoma (21). With the exploitation and application of REEs in multiple fields, there is an increasing possibility for humans to absorb REEs from the workplace, environment and food (22, 23). The health effects of REEs are controversial (24). While some studies found that REEs impair human health (25–33), other studies showed that higher than background concentrations of REEs in human tissues did not have significant health effects (34, 35).

Oxidative stress plays a key role in the development and progression of diabetes. The vulnerability to oxidative stress during pregnancy suggests that oxidative stress may also be an important factor in triggering GDM (35, 36). It is known that low doses of REEs have antioxidant effects, while high concentrations of REEs contribute to oxidative stress (37). This unique redox property makes rare earth oxalates show great potential for the treatment of oxidative stress-induced diseases, such as diabetes, in several animal studies (38–41). In 2011, Pourkhalili N et al. first reported the beneficial effect of the combination of cerium oxide nanoparticles/sodium selenite in diabetes treatment (42). Treatment of rats with this combination resulted in a significant reduction in blood glucose. Hence, a small intake of REEs might reduce the risk of developing GDM. However, to our knowledge, we did not find epidemiological evidence of an association between REEs exposure and GDM.

Therefore, this study aimed to investigate whether early pregnancy exposure to REEs (single metal elements or a metal mixture) is associated with the risk of GDM.

In this study, we investigated the association between serum REE concentrations and GDM risk among pregnant women in Beijing, China. We didn’t observe a significant association between REEs and GDM by logistic regression analysis. However, the WQS model indicated that exposure to lower levels of REEs mixtures was associated with an increased risk of GDM, and three elements, Nd, Pr, and La, were important contributors to the REE mixture.

Blood concentrations of REEs in unexposed populations vary considerably by race, region, age, gender, and the different analytical methods used (28). As shown in Table S1 , previous studies (28, 31, 46–49) showed that the range of the median values was 0.01-0.854 μg/L for La, 0.02-2.546 μg/L for Ce, 0.01-0.132 μg/L for Pr, 0.01-0.839 μg/L for Nd, 0.01-0.132 μg/L for Sm, 0.01-0.034 μg/L for Eu, 0.049-0.059 μg/L for Gd, 0.001-0.011 μg/L for Tb, 0.01-0.047 μg/L for Dy, 0.001-0.01 μg/L for Lu, 0.01-0.529 μg/L for Yb among the non-occupationally exposed population in China, Spain, Romania, and Sub-Saharan Africa. The concentrations of all the serum REEs in our study fell in these ranges. The high detection rates in blood for most REEs in humans are remarkable, considering that to date there are very few references in the scientific literature dealing with the effects of chronic exposure to REEs on human health (24–28). Therefore, it remains unclear whether the exposure to REEs during pregnancy is related to GDM or not.

In the present study, we didn’t observe a significant association between individual REEs and GDM risk. When we classified the concentrations of the REEs according to quartiles, we found no statistically significant association. Considering the absence of clear toxicological criteria, some authors analyzed the combined effects of these elements, as they were often used in combination in the manufacture of multiple devices (50). The WQS model can not only examine the effects of the REE mixture but also identify those elements most strongly associated with GDM (45). In WQS analysis, lower levels of REE mixture were significantly associated with a higher risk of GDM, and Nd, Pr and La were likely to be the most important contributors. The joint effect of the LREEs on GDM was similar to the results of WQS analysis for all REEs. A possible explanation for the similar results is that the three most important contributors (Nd, Pr, and La) belong to LREEs. To the best of our knowledge, there is no published report in the literature on REEs in maternal serum and risk for GDM. A previous study conducted in Shanxi province of China reported that higher concentrations of REEs were associated with an increased risk of hypertension (24). Two studies conducted in Shanxi and Hebei provinces of China found that high concentrations of REEs were associated with increased risks of fetal neural tube defects (NTDs) (25, 26). Besides, Liu et al. found that increased maternal urinary REEs (Ce and Yb) were associated with decreased neonatal thyroid-stimulating hormone (TSH) levels (27). Another study suggested that REEs might play a role in the development of anemia (28). Liu et al. showed that prenatal exposure to REEs and REE mixtures were associated with an increased risk of premature rupture of membranes (29). Liu et al. found that in utero exposure to REE mixtures increased the risk of orofacial clefts (30). According to these studies, exposure to higher levels of REEs was a risk factor for some diseases, while in our study, exposure to higher levels of REEs might be a protective factor for GDM.

The possible underlying mechanisms of this inconsistency are as follows. REEs may act either as antioxidants or pro-oxidants, depending on the environment, the nature of the bonding in their compounds, and their concentration in the tissues (51). Results of previous studies might be explained by oxidative stress induced by exposure to REEs. However, some REEs have been reported to have antioxidant effects in clinical applications (52). For example, cerium oxide nanoparticles could exhibit superoxide dismutase (SOD) activity and act as a free radical scavenger (42). GDM is usually accompanied by increased levels of oxidative stress and inadequate antioxidant defense responses (36). Therefore, REEs may reduce oxidative stress associated with GDM. Moreover, individual REEs and REE mixtures have been shown to have hormetic effects on several health endpoints. In other words, REEs can increase or improve biological events (e.g. growth) at low concentrations and exhibit inhibitory or toxic effects at increasing doses/concentrations (35). Another possible explanation is related to the relatively low concentrations of REEs in our study. Low concentrations of REEs might stimulate pancreatic beta (β) cells to secrete insulin to regulate glucose metabolism and maintain blood glucose stability (53–55). Within a certain range of concentrations, exposure to lower concentrations of REEs might be a risk factor for GDM. Further studies are needed to validate the role of REEs in the pathophysiology of GDM.

To the best of our knowledge, this study is the first to evaluate the association between early pregnancy exposure to REEs and the risk of GDM. This is one of the strengths of our study. Second, comprehensive information regarding potential confounders was prospectively collected in the birth cohort and controlled in our statistical analyses. Third, the sample size of this prospective cohort study was relatively large, and we collected blood samples in early pregnancy to evaluate metal exposure, which meets the criterion of temporality for causal inference. Finally, the information in our questionnaire was collected using standardized interview procedures by trained local health workers at the hospital. However, there were some limitations in our study. First, causality cannot be made based on the findings of the current study because of the observational nature of the study design. Second, everyone is exposed to multiple toxic substances in real life, and it is difficult to fully rule out the effects of residual confounders. Thirdly, the data of dietary intake were retrospectively collected and had recall bias, causing underestimation of energy intake in this study. However, we think that the underestimation of energy intake may not affect the main results, because it was a covariate in our study. More validation studies are needed in the future.

In conclusion, we observed that early pregnancy exposure to lower levels of REE mixtures increased the risk of GDM. Nd, Pr, and La showed the greatest contribution to the metal mixture index related to GDM. Our findings suggested that REEs are more likely to play a role through oxidative stress or other pathways. Animal studies are needed to uncover the possible mechanisms underlying the association between the mixture of REEs and GDM risk. The results of this study warrant further epidemiological investigation in other populations.

The original contributions presented in the study are included in the article/ Supplementary Material . Further inquiries can be directed to the corresponding authors.

The studies involving human participants were reviewed and approved by the institutional review boards at Peking University (IRB00001052-18003). The patients/participants provided their written informed consent to participate in this study.

Conceptualization, XX, YW, LY, XM, and H-JW. Methodology, XX, YW, SL, ZL, and HB. Software, XX and XY. Validation, XY and H-JW. Formal Analysis, XX, YW, and YJ. Data Curation, NH, XY, JL, CJ, LL, and SZ. Writing - Original Draft Preparation, XX and YW. Writing–Review and Editing, LY, HW, and BW. Supervision, LY, XM, and H-JW. Project Administration, XX. Funding Acquisition, H-JW. All authors have read and agreed to the published version of the manuscript.

This study was funded by the National Key Research and Development Program (No. 2016YFC1000300 and No. 2016YFC1000307) and National Natural Science Foundation of China (81973053 and 81703240).

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.