The influence of obesity and associated fatty acids on placental inflammation

Abstract

Purpose:

Maternal obesity, affecting nearly 1 in 4 pregnancies, is associated with increased circulating saturated fatty acids, such as palmitate. These fatty acids are implicated in placental inflammation, which may in turn exacerbate both maternal-fetal tolerance and responses to pathogens like Group B Streptococcus. In this review, we address the question “how do obesity and associated fatty acids influence placental inflammation?”

Methods:

In this narrative review, we used queries of PubMed and Google Scholar using combinations of “placental inflammation” or “pregnancy” and “lipids”, “fatty acids”, “obesity”, “palmitate” or other closely related search terms. We also used references found within these papers that may have been absent from our original search queries. We analyzed methods and key results of these papers to compare and contrast their findings, which were occasionally at odds with each other.

Findings:

While obesity can be studied as a whole, complex phenomena with in vivo mouse models and human samples from patients with obesity, in vitro modeling often relies on the treatment of cells or tissues with one or more fatty acids and occasionally other compounds (e.g. glucose and insulin). We have found that palmitate, most commonly used in vitro to recreate hallmarks of obesity, induces apoptosis, oxidative stress, mitochondrial dysfunction, autophagy defects, and inflammasome activation in many placental cell types. We compare this to in vivo models of obesity wherever possible. We find that obesity as a whole may have more complex regulation of these phenomena (apoptosis, oxidative stress, mitochondrial dysfunction, autophagy defects, and inflammasome activation) compared to in vitro models of fatty acid treatment (primarily palmitate) due to the presence of unsaturated fatty acids (i.e. oleate) which may have anti-inflammatory effects.

Implications:

The interaction of unsaturated fatty acids with saturated fatty acids may ameliorate many inflammatory effects of saturated fatty acids alone, which complicates interpretation of in vitro studies focusing on a particular fatty acid in isolation. This may explain why certain studies of obesity in vivo show differing outcomes from studies of specific fatty acids in vitro.

Introduction

Both pregnancy and obesity modulate the immune responses at baseline and to infectious agents^1–3. The interplay of obesity and pregnancy is an emerging area of interest, as pre-pregnancy obesity affects 1 in 4 pregnancies in the United States⁴, and may be a factor in 1 in 4 major pregnancy complications⁵. Obesity during pregnancy is a risk factor for chorioamnionitis^6–10, preterm pre-labor rupture of membranes (PPROM)¹¹, stillbirth^12–14, and neonatal sepsis^{15, 16}. Numerous types of infection are more common during pregnancies complicated with obesity, including surgical site infection during C-section¹⁷, rectovaginal colonization with Group B Streptococcus (which is a leading cause of PPROM, chorioamnionitis, neonatal sepsis)^18–20, increased severity of influenza infection²¹, and increased complications associated with the severe acute respiratory syndrome coronavirus type 2 (SARS CoV-2), the cause of the coronavirus disease 2019 (COVID-19)^22–28. In this review, we examine studies investigating the connection between obesity and placental inflammation and dysfunction, with a special focus on fatty acids.

Obesity has been associated with excess circulating fatty acids in the bloodstream. Fatty acids are categorized into three major classifications: poly-unsaturated fatty acids (PUFAs), saturated fatty acids (SFAs), and monounsaturated fatty acids (MUFAs) (for chemical representations of relevant fatty acids included in this review, see Figure 1). SFAs lack double bonds and are saturated with hydrogen atoms. MUFAs have a single-double bond, while PUFAs have two or more double bonds within the molecule. Fatty acids are further classified in the delta naming system according to the number of carbon atoms (C) in the chain and the number of double bonds (D) present in the chain, expressed as a ratio of C:D. For instance, the SFA palmitate has 16 carbons and no double bonds and is expressed as 16:0, while the unsaturated fatty acid oleate has 18 carbons and 1 double bond and is expressed as 18:1. Unsaturated fatty acids can further be classified by an omega (ω) nomenclature, which describes the location of the double bond closest to the methyl (ω) end of the fatty acid²⁹. In this system, the first bond in the double carbon bond that is closest to the w end is counted, resulting in an w number (often 3, 6 or 9; ω−3 fatty acids are heavily studied in nutrition). This review will primarily focus on SFAs (primarily palmitate), as these are most commonly used by in vitro studies modeling obesity in the placenta, but will also include MUFAs (primarily oleate, as it is the most common MUFA used in in vitro studies modeling obesity in the placenta), and some discussion of the anti-inflammatory effects of PUFAs and the effect of obesity as a whole.

Figure 1. Chemical representations of fatty acids included in this review.

(A) The saturated fatty acid palmitate features heavily into models of obesity, and stearate is occasionally used as a comparison. (B) The mono-unsaturated fat oleate can often act as a counter to the inflammatory properties of palmitate⁴⁸; the monounsaturated fatty acid palmitoleic acid and polyunsaturated fatty acids linoleic acid are used as comparisons. The omega-3 fatty acid docosahexaenoic acid (DHA, or one of its major sources, fish oil) has been frequently observed to be anti-inflammatory⁸⁶, and arachidonic acid plays a major role in fetal development³¹. (C) Some speculation exists that palmitate acts as a ligand for Toll-like receptor 4 (TLR4) because it resembles portions of bacterial lipopolysaccharide (LPS)^{119, 124}, the canonical agonist for TLR4, although this remains controversial¹²².

During pregnancy, circulating fatty acid levels rise progressively to increase the sources of metabolic and storage energy, reduce inflammation, and aid in the synthesis of prostaglandins³⁰. PUFAs such as ω−3 fatty acids (i.e. docosahexaenoic acid or DHA) and ω−6 fatty acids (i.e. arachidonic acid or AA) play a major role in fetal neural and retinal development, especially in the third trimester³¹. Recent studies have indicated that obesity and excessive gestational weight gain in pregnant patients is associated with higher levels of circulating SFAs, MUFAs, and n-6 long chain PUFAs, as well as imbalances of ω−3 to ω−6 fatty acid ratios compared to patients with normal weight^32–35. Typical total plasma palmitate concentrations have been measured at 147–155 μM³⁶, but can range higher (to the 400 μM range) in obese individuals even without accompanying metabolic syndromes³⁷. Interestingly, during pregnancy, obesity has been associated with lower plasma levels of DHA and AA, indicating specific subtypes of fatty acids are disproportionately affected by obesity³⁴, and DHA has been shown to reduce inflammation in pregnant patients with obesity³⁸, although not to as great an extent as in normal weight pregnant patients³⁹.

Lipotoxicity

Lipotoxicity includes the effects of mitochondrial dysfunction, endoplasmic reticulum stress, insulin resistance, and inflammation brought about by the accumulation of lipids in non-adipose tissue due to increased free fatty acids in circulation⁴⁰. Many of the phenomena discussed in this review fall under the heading of “lipotoxicity.”

Excess free fatty acids can be stored as lipid droplets, which is a way to prevent mitochondrial damage and cellular stress, since the oxidative products of free fatty acids can generate reactive oxygen species (ROS) and may act as ligands for pattern recognition receptors^{41, 42}. Lipid droplet formation can be a byproduct of autophagy (a cell component recycling program detailed below) during periods of prolonged starvation⁴³, which can also be affected by lipotoxicity and obesity^44–46. Distinct fatty acids can result in greater or lesser lipid droplet formation; for instance, the MUFA oleate (18:1; ω−9) induces lipid droplets, but the SFA palmitate does not⁴⁷. The addition of oleate to palmitate is sufficient to restore lipid droplet formation in many cell types throughout the body (i.e. hepatocytes⁴⁷) including in both the syncytiotrophoblasts⁴⁸ and the trophoblast⁴⁹. However, high glucose in circulation (such as that found in people with diabetes) can impair lipid droplet formation and stability, particularly in human trophoblasts⁴⁹. Notably, supplementation with the omega-3 fatty acid DHA was shown to decrease total lipid content in placentas of patients with obesity^{50, 51}. Thus the phenotype that emerges from palmitate treatment alone may not reflect the in vivo obesity phenotype, which has other circulating fatty acids and often excess glucose. We will now delve into the many forms of lipotoxicity associated with pregnancies complicated by obesity.

Endoplasmic Reticulum Stress

The endoplasmic reticulum (ER) is the site of protein synthesis (with the aid of accompanying ribosomes), folding (with the aid of ER-resident chaperone proteins), and post-translational modifications for the cell⁵². As such, its ionic and electronic state is tightly controlled and tailored to protein folding activities^{53, 54}. Proteins that are incorrectly folded must be removed and degraded (primarily through transport to the cytoplasm followed by ubiquitylation and proteasomal degradation) to ensure proper functioning of the ER⁵⁵. When there is a backlog of proteins to be folded, and a buildup of incorrectly folded proteins (the cellular response to which is termed the unfolded protein response (UPR)), the ER is under stress. Many pathways can lead to ER stress and the UPR, including hypoxia (an accompaniment to obesity in the placenta⁵⁶), mutations to proteins, and loss of calcium (thus changing the ionic and electronic composition of the ER and altering protein folding dynamics)⁵⁷. The UPR evolved as a pathway to slow protein translation, enlarge the ER, and expedite the removal and degradation of misfolded proteins to the cytoplasm. There are many sensors for ER stress that can help initiate the UPR, which then either results in relief of the unfolded protein backlog, or activation of alternate UPR signaling which can result in cell death through apoptosis⁵⁸.

Obesity is thought to induce ER stress through multiple mechanisms (reviewed in⁵⁹): increases in bile acid and their associated compounds⁶⁰, increased free SFAs^61–64, and increased palmitoylation (the attachment of palmitate to a substrate) of ER quality control protein calnexin and insulin responsive amino peptidase^65–67. Even in the absence of ER stress, increases in ER membrane lipids can induce the UPR^68–72. During pregnancy, alterations to bile acids in the placenta can have effects on offspring metabolic health⁷³ that may be dependent upon ER stress. Placentas from mothers with obesity showed decreases in ER stress-associated proteins Atf4 and Xbp1, as well as the molecular chaperone calnexin⁵⁶. Palmitate in conjunction with the proinflammatory cytokine tumor necrosis factor alpha (TNFα) induced UPR gene expression⁷⁴. Palmitate induces ER stress and the UPR in syncytiotrophoblasts⁴⁸ as measured by increased levels of misfolded protein chaperone BiP/GRP78/HSPA5 and increased levels of CHOP protein, but oleate alone did not, and the addition of oleate to palmitate rescued palmitate-induced ER stress. In a model of palmitate-induced ER stress, stress-inducible Sestrin-2 was upregulated in first-trimester trophoblasts and rescued some of the effects of palmitate-induced ER stress⁷⁵. Thus, modeling obesity strictly with palmitate may exaggerate some effects of obesity on ER stress, given that there are many circulating fatty acids in obesity.

Apoptosis

One common outcome of ER stress is apoptosis. Apoptosis is programmed cell death that results in controlled cell death without the release of inflammatory compounds from the dying cell; it is accomplished via cellular proteases called caspases and results in DNA fragmentation, calcium release from the mitochondria, destruction of proteins, and cell surface ligands to induce phagocytosis of the dead cell⁷⁶. Apoptosis can be induced by cytokine/Fas receptor interactions, an ER stress and/or calcium-driven pathways, and a mitochondrial-mediated mechanism⁷⁷. It is a way of limiting inflammation during cell death and is increased in many tissues during obesity^78–80. However, in human placentas, one study found a surprising inverse relationship between body mass index (BMI) and apoptotic markers⁸¹, and another found increased levels of the cellular inhibitors of apoptosis 1 and 2 proteins (cIAP 1 and 2) in placentas of patients with obesity compared to placentas from lean patients⁸², although the focus of the study was induction of inflammation due to cIAP 1 and 2, and not apoptosis. In vitro studies on the impact of specific fatty acids on placental cells yield different results. In the syncytiotrophoblast, palmitate induced apoptosis via cleaved Caspase-3 (an indicator of apoptosis)⁴⁸, but the addition of oleate treatment alone did not, and the addition of oleate at equimolar concentrations to palmitate prevented apoptosis. The absence of apoptosis in trophoblasts treated with a 1:1 ratio of palmitate and oleate was also shown earlier by Pathmaperuma, et al.⁴⁹. Lee, et al. 2020 also found that palmitate induced ER stress and apoptosis in Sw.71 cells (a 1^st trimester cytotrophoblast cell line⁸³) via increases in the cleaved form of Caspase-3⁷⁵. This study further found that apoptosis in palmitate-treated cells is more drastic when the stress-inducible protein Sestrin2 is knocked down⁸³. The apoptotic effects of palmitate were further shown in placental macrophages, where apoptotic cell death (by propidium iodide and Annexin V staining) was induced by 0.4 mM palmitate⁸⁴. While palmitate induces trophoblast (and macrophage) apoptosis and oleate can suppress placental palmitate-induced apoptosis, different fatty acids have distinct effects on proliferation and apoptosis at the maternal-fetal interface. Outside the context of obesity, Klingler, et al. in 2006 showed that supplementation of pregnant women’s diets with fish oil (containing high amounts of the ω−3 PUFA DHA) resulted in increased proliferation from placental trophoblasts, but no differences in apoptosis⁸⁵. Wietrak, et al. in 2014⁸⁶, using a similar model testing women who received fish oil containing 300mg of DHA, found that administration of DHA did inhibit apoptosis induced during homeostasis, although this was not investigated in the placenta but rather umbilical cord blood. In the context of obesity, administration of DHA in conjunction with another ω−3 PUFA (Eicosapentaenoic acid (EPA)) resulted in less reduction in inflammation in patients with obesity relative to normal weight pregnant patients, although it is unclear if this affected apoptosis³⁹. Taken together, these studies suggest that fatty acids, depending on their chemical properties, can alter cellular apoptosis and/or proliferation in the placenta and other maternal-fetal interface regions (i.e. cord blood⁸⁶), but that in vivo conditions of obesity as a whole (not just isolated palmitate treatment) may not affect placental apoptosis due to complex interactions of fatty acids, sugars, and mediators.

Mitochondrial function

Obesity is associated with chronic inflammation and mitochondrial dysfunction in many tissues^87–89, and is often tied to oxidative stress and ROS^{90, 91}, mitochondria being the main source (and eliminators) of ROS. The consequences of mitochondrial dysfunction are decreased adenosine triphosphate (ATP) levels (the cell’s energy substrate) and, as a byproduct, ROS production through the respiratory chain⁸⁹. Mele, et al. in 2014 found 14-fold higher ROS production in placental villous tissue from patients with obesity⁹². They then used placental ATP levels as a proxy for mitochondrial function and found that ATP was twice as high in placentas from normal weight patients versus placentas from patients with obesity⁹². They further demonstrated that mitochondrial DNA copy number were decreased in placentas from pregnancies complicated by obesity (also found by Hastie, et al. in 2014⁹³); mitochondrial DNA is replicated during the pre-S phase of the cell cycle⁹⁴ and a decrease in mitochondrial DNA is associated with apoptosis^{95, 96} while an increase can induce cell proliferation⁹⁷. Mele, et al also found that the ability to use alternate sugar sources (i.e. galactose instead of glucose), an indicator of normal mitochondrial function, was hindered in placentas from patients with obesity⁹².

In sharp contrast to the two papers showing decreased mitochondrial DNA levels in placentas from pregnancies complicated by obesity, Mando, et al. in a 2015 abstract demonstrated the opposite with respect to mitochondrial DNA: placentas from pregnancies associated with obesity had greater mitochondrial DNA levels than placentas associated with normal weight⁹⁸ and followed this up with a full study demonstrating higher mitochondrial DNA levels in placentas from pregnancies associated with obesity⁹⁹. Further, on the maternal side, Anelli, et al. in 2017 found mitochondrial DNA to be increased in the blood and plasma of pregnancies associated with obesity¹⁰⁰ along with levels of hepcidin, a positive regulator of inflammation. The reason for this discrepancy between Mando, et al. and the findings of Mele, et al. and Hastie, et al. is unclear; it may have to do with the genes assayed by qPCR to determine mitochondrial DNA: Mando et al. used cytochrome-β and normalized to RNase-P⁹⁹, while Mele, et al. used mitochondrial 16S rRNA and normalized to β₂-microglobulin⁹² and Hastie, et al. used mitochondrial cytochrome C oxidase and β-actin to normalize⁹³. Mitochondrial copy number may not be the best readout for mitochondrial dysfunction, and alterations to oxidative stress and ROS may be more indicative of mitochondrial function. We were unable to find any studies using an in vitro model of the effects of obesity on mitochondrial function, thus it is unclear the contributions of individual fatty acids to mitochondrial function during conditions of obesity.

Autophagy

Autophagy is the recycling of cellular components such as organelles, proteins, and lipids, through incorporation and degradation in double-membrane vesicles (autophagosomes) that fuse with lysosomes¹⁰¹ after which, cargo in the new autolysosome is degraded by lysosomal hydrolases. The contents are then shuttled back to the cytosol for reuse. Autophagy is associated with longer-lived proteins in the cytoplasm¹⁰², and certain structures can also be degraded by autophagy, including inflammasomes (discussed later in this review)¹⁰³. While autophagy is a basal process that is virtually always active, cellular stressors can induce autophagy to higher levels, these stressors can include inflammation¹⁰³, hypoxia, and the unfolded protein response¹⁰⁴. The reciprocal of this is also true in some cases: autophagy defects can cause inflammation¹⁰¹.

In a recent study by Cohen, et al., maternal obesity was shown to reduce placental autophagy through quantification of autophagy-related proteins LC3B and p62¹⁰⁵ from placentas from pregnancies complicated by obesity compared to normal weight counterparts. However, hypoxia, a known inducer of autophagy^106–108 was found to be increased in placentas from patients with obesity and was accompanied by a concomitant increase in placental markers of autophagy in Gohir, et al. 2019⁵⁶, albeit by mRNA expression and not protein expression as shown in Cohen, et al¹⁰⁵. The genes used as autophagy markers were also different between the two studies, with Vsp15 and Vps34 used in Gohir, et al.⁵⁶ and LC3B and p62 in Cohen, et al.¹⁰⁵, which may account for differences between the two studies. Furthering the case for obesity inducing defects in autophagy, Hong et al. found that addition of palmitate to human extravillous trophoblasts induced protein aggregates and also increased the association of p62 with these protein aggregates and increased LC3B levels in the cell¹⁰⁹. However, palmitate decreased the ability of autophagosomes to fuse with lysosomes by examining co-localization of autophagosome marker LC3B with lysosome marker LAMP1¹⁰⁹, suggesting that increases in autophagy-related proteins does not necessarily correspond to increased autophagy function. Of note, addition of the unsaturated fatty acid oleate to palmitate during stimulation of extravillous trophoblasts rescued the palmitate-induced autophagy defects at both the protein aggregate accumulation and autophagosome-lysosome fusion. Outside of the trophoblast, palmitate can also negatively impact decidualization of endometrial stromal cells¹¹⁰, and the mechanism at play may be a reduction in autophagy. Complicating the picture further, obesity-induced placental autophagy defects may be sex-specific¹¹¹. In a unique model of obesity using TNFα stimulation of primary placental trophoblasts (due to its elevation in pregnancies complicated by obesity^{112, 113}), trophoblasts from pregnancies with female fetuses only showed upregulation of the negative autophagy regulator Rubicon, suggesting a downregulation of autophagy¹¹⁴. Autophagy defects can lead to inflammation through backups in the UPR¹⁰¹, particularly if there may be apoptosis defects as well. In the case of autophagy, some of the discrepancy between in vitro palmitate studies and in vivo obesity studies may be the outcome analyzed (i.e. autophagy protein abundance versus function).

TLR4

Toll-like receptors (TLRs) are pattern recognition receptors expressed on immune cells and a variety of other cells of the body. TLR4 is primarily a receptor for lipopolysaccharide (LPS), a major component of gram-negative cell walls. Early research showed that fatty acids (e.g. palmitate, laurate, stearate, and palmitoleic acid) appear to be ligands of TLR4^115–118 through speculation that the lipid portion of LPS is structurally similar to fatty acids^{119, 120}, and more work characterized the binding of palmitate to TLR4¹²¹. However, according to a recent study, palmitate is not a TLR4 ligand¹²² but instead causes lipid-induced inflammation by reprogramming macrophage metabolism. Yet more studies suggest a role for TLR4 in obesity-induced inflammation¹²³, as TLR4 knock-out and knock-down studies show distinct differences in cytokine profiles of immune and non-immune cells during obesity and stimulation with palmitate¹²⁴. For the purposes of this review, we will accept that TLR4 is somehow involved in inflammation induction by palmitate and other fatty acids (e.g. laurate, stearate, and palmitoleic acid), and obesity in general, and look specifically at examples of this in the placenta.

Yang, et al. showed in primary human trophoblasts that palmitate induced proinflammatory cytokines TNFα, IL-6, and IL-8 while treatment with oleate did not¹²⁵. Investigating chain length and saturation further, the SFA palmitic acid (C16:0) and stearic acid (C18:0) both induced proinflammatory cytokines while unsaturated C:16 and C:18 fatty acids palmitoleic acid (C16:1; w-7), oleic acid (C18:1; w-9), and linoleic acid (C18:2; w-6) did not. This was attributed to TLR4 signaling through NFkB, as TLR4 inhibition and NFκB inhibition both reduced the induction of proinflammatory cytokines¹²⁵. Yang and colleagues went on to show in 2016 that TLR4, IL-6, and IL-8 were increased in human placentas complicated by obesity compared to their normal weight counterparts¹²⁶. In studies performed in fetal vascular smooth muscle cells (for background on vascular smooth muscle cells, see¹²⁷) obtained from rats instead of cytotrophoblasts, Thompson, et al.¹²⁸ reported that palmitate treatment induced TLR signaling through MyD88 and activated other inflammatory pathways including JNK and ERK, induced downstream COX-2 activation, and upregulated proinflammatory cytokines at the mRNA level (TNFα, IL-6, IL-1β). Further, in vivo studies in rats fed a high fat diet showed increased TLR2 and TLR4 expression in placenta, which correlated to increased TNFα and IL-6. In studies conducted in placental cotyledonary tissue of the sheep¹²⁹, maternal obesity was also shown to upregulate TLR2 and TLR4 expression, activate JNK and NFκB signaling, and induce a similar panel of proinflammatory cytokines: TNFα, IL-6, IL-8, and IL-18. These studies show that saturated fatty acids can both upregulate TLRs and signal through them to induce proinflammatory cytokine production. However, other studies have shown that there is no difference in placental TLR4 levels between low BMI and high BMI mothers¹³⁰. Despite finding no difference in TLR4 levels between low and high BMI mothers, Lager and colleagues showed an effect downstream of TLR4 signaling stimulated by fatty acids (in this case, the unsaturated oleic acid (18:1; w-9)): fetal overgrowth, mediated by a TLR4-dependent upregulation of amino acid transporters¹³⁰. They also found that oleic acid induced JNK and STAT3 signaling, but not the NFκB, mTOR, or p38 MAPK pathways¹³⁰. Lager, et al. went on to show in a later paper in 2014 the differences between palmitate, oleic acid, and DHA signaling through TLR4¹³¹. Only oleic acid induced ERK signaling, and in this work did not induce JNK signaling, contrary to their previous findings and many other published works^{128, 129}. This may be due to differences in time points- Yang, et al. 2015 used very early timepoints (2 hours after treatment)¹²⁵ for their protein signaling experiments, while Lager used 24 hour lysates^{130, 131}, which may be a late enough time point to miss key early signaling events. While induction of TLR4 expression in placentas from pregnancies complicated by obesity is unclear (some studies show increases^{126, 129}, while others do not¹³⁰), there is broad agreement that TLR4 signaling is induced by obesity and fatty acids^{125, 126, 128–131}, whether directly or indirectly.

NLRP3 inflammasome

The inflammasome is a multiprotein complex responsible for the production of IL-1β from pro-IL-1β and the cleavage of Caspase-1 from Pro-caspase-1, which is downstream of pattern recognition receptors such as TLR4. However, there are also reports of fatty acids directly activating the NRLP3 inflammasome due to crystallization of SFAs like palmitate¹³²; interestingly, formation of these crystals was inhibited in the presence of the MUFA oleate, providing a potential mechanism for the anti-inflammatory effects of oleate upon palmitate stimulation. Using the human Sw.71 trophoblast cell line treated with palmitate (400 μM, but not 200 μM, and 800 μM was toxic)¹³³, Shirasuna, et al. found that palmitic acid stimulated Caspase-1 activation and IL-1β secretion, and also Caspase-3 activation, and IL-6 and IL-8 secretion (which were dependent upon reactive oxygen species production) as detailed above¹³³. As noted above, typical plasma palmitate concentrations have been measured at 147–155 μM³⁶, but can range higher (to the 400 μM range) in obese individuals even without accompanying metabolic syndromes³⁷. Rogers, et al. also demonstrated the induction of NLRP3 inflammasome in cells treated with palmitate⁸⁴ and the accompanying induction of IL-1β. Whether by direct activation or activation through other pattern recognition receptors like TLR4, SFAs activate inflammasome activation in multiple cell types of the placenta, although this has only been modeled with palmitate and not examined in an in vivo mouse model of obesity or human samples, which may have more complex inflammasome regulation.

Conclusions

Circulating fatty acids in the bloodstream can impact inflammation, either augmenting it (in the case of saturated fatty acids like palmitate) or dampening it (in the case of unsaturated fatty acids such as oleate and DHA). Alterations to the types and amounts of circulating fatty acids affects lipid droplet formation in the placenta^47–49, which in turn further affects the amount of circulating fatty acids. Co-morbidities of obesity in the placenta, including bile acids^{60, 73}, palmitoylation^65–67, and hypoxia⁵⁶ can activate ER stress pathways, which in turn affect apoptosis and autophagy. Obesity-associated apoptosis in many cell types associated with the placenta is a complex phenomenon; treatment with SFAs alone induced apoptosis⁴⁸, but addition of MUFAs rescued cell survival^{48, 49, 83}, suggesting that pure treatment with SFAs like palmitate may be too reductive an approach to truly mimic the fatty acid milieu in obesity. While oxidative stress induced by fatty acids was largely consistent, mitochondrial function as assessed by mitochondrial DNA was less consistent, with studies finding conflicting results (possibly explained by different mitochondrial DNA replication-associated mRNAs assessed)^{93, 98, 134}. Autophagy was similarly conflicted, with some studies finding increases in autophagy markers⁵⁶ while others found defects in markers or function^{105, 109}, which may be explained by studies quantifying different autophagy-associated mRNAs. The study that showed autophagosome-lysosome fusion defects found increases in autophagy-associated mRNAs while also showing autophagy defects¹⁰⁹, which may provide the best answer to this question: autophagy may be increased when querying the mRNA level, but functional studies can still ultimately show persistent defects in autophagy¹⁰⁹. Further controversy exists concerning palmitate binding to TLR4: many knockout studies show that TLR4 contributes to palmitate-induced inflammation^115–121, while other studies suggest that palmitate cannot be a ligand for TLR4¹²². The mechanism through which palmitate may act through TLR4 is still unclear, but involvement of TLR4 in obesity-associated inflammation in the placenta is well-documented^{123–125, 128–131}. NLRP3 inflammasome involvement is less ambiguous at this point, with palmitate inducing caspase-1 activation and IL-1β secretion consistently in multiple cell types of the placenta^{84, 133}, although the mechanism behind palmitate activation of the NLRP3 inflammasome is unclear. Furthermore, each of these processes are interconnected (Figure 2). Apoptosis can be induced by mitochondrial dysfunction, ER stress, and autophagy defects, among other pathways. Autophagy affects the NRLP3 inflammasome, while TLR4 signaling can also prime the NLRP3 inflammasome. The precise point(s) where fatty acids directly and mechanistically induce inflammation remain unclear, but much important work has been done to provide an initial outline of the obesity-induced inflammation phenomenon.

Figure 2. The lipid milieu in circulation during obesity features multiple saturated and unsaturated fatty acids, which can activate or suppress multiple inflammatory or inflammation-adjacent cellular pathways.

Palmitate can decrease the anti-inflammatory lipid droplet formation, while oleate can counter the effect of palmitate on lipid droplet formation in syncytiotrophoblasts and trophoblasts^47–49. Excess palmitate can increase key autophagosome gene expression, but suppresses fusion of the autophagosome and lysosome¹⁰⁹, thus causing autophagy defects, the buildup of these products can induce ER stress. Mitochondrial dysfunction and reactive oxygen species formation can also activate the ER stress pathway. Unmitigated ER stress can trigger apoptotic cell death pathways, as can mitochondrial dysfunction and inflammasome activation. Diagram created using BioRender.

We conclude that modeling obesity using reductive approaches such as treatment with SFAs provides insight into how these biological molecules might alter reproductive immunology, but may be incomplete, because in reality a circulating milieu of MUFAs and PUFAs may mitigate many effects of SFAs. However, SFAs acting alone induce inflammation through multiple pathways, including through TLR4 and NLRP3 (either directly or indirectly), defects in autophagy, and lipotoxicity and its associated effects. Supplementation with omega-3 fatty acids such as DHA can mitigate many effects of obesity-induced inflammation. The interplay of multiple fatty acids will be of utmost importance in determining the true effects of obesity on placental inflammation and function.