Effect of Prepartum Adiposity and Lipolysis on Gestational and Postnatal Adipose Tissue Inflammation and Immune Cell Infiltration

Overweight and obese women are at a higher risk for developing gestational diabetes as these patients exhibit high lipolytic rates and adipose tissue (AT) inflammation that impair insulin signaling. Similarly, high adiposity in gestating dairy cows is associated with a surge in AT lipolysis and inflammatory responses leading to insulin resistance. Lipolysis induces a remodeling process that is characterized by an inflammatory response with immune cell migration, proliferation of cellular components of the stromal vascular fraction (SVF), and changes in the extracellular matrix. This study evaluated the effect of adiposity and lipolytic rate on markers of AT remodeling in gestating and lactating cows. Blood and subcutaneous AT samples were collected from multiparous Holstein cows with high (HA; n=12) or low (LA; n=9) adiposity scores at −27±7 (FO) and −10±5 (CU) d prepartum and at 8±3 (PP) d postpartum. Expression of genes related to AT remodeling were analyzed by RT‐PCR and immune cell trafficking in subcutaneous SVF was quantified by flow cytometry. Lipolysis increased at CU and reached its peak at PP compared to FO as demonstrated by serum FFA concentrations (0.27±0.05, 0.39±0.05 and 0.99±0.05 mEq/L for FO, CU and PP, respectively). Serum FFA were higher in HA (0.63±0.02 mEq/L) compared to LA (0.47±0.02 mEq/L) cows reflecting a significant effect of adiposity on lipolysis rate during gestation and after parturition. Immune cell trafficking analysis showed that HA had fewer SVF cells expressing T cell markers CD8, CD4, and CD3, and B cells, compared to LA. HA had no effect on the expression of macrophage markers CD14, CD16, and CD163. SVF expressing antigen presenting cell markers CD172a and MHCII and B cell marker were reduced after parturition compared to FO and CU. Gene expression analysis indicated that PP transcription of ostepontin ( SPP1 ), an inflammatory cytokine that triggers macrophage infiltration during AT remodeling, its receptor CD44, and signal regulatory protein α ( SIRPA ), which is a marker of mononuclear immune cells, were higher at PP compared with FO and CU. Serum FFA concentrations were positively correlated with AT expression of SIRPA (r=0.59), and negatively correlated with AT expression of IL10 (r=−0.54) , an anti‐inflammatory cytokine. This indicates that enhanced lipolytic rate during late gestation and postnatal periods impairs the anti‐inflammatory response performed by cytokines. Collectively these data indicate that exacerbated lipolysis around parturition enhances immune cell infiltration into AT and alters the inflammatory response, independently of prepartum adiposity. Further studies on how adiposity and AT lipolysis affect AT remodeling processes and maternal health during critical periods of pregnancy and in the postpartum are required. Support or Funding Information This project was supported by USDA‐National Institute of Food and Agriculture (Washington, DC) grants 2014–68004–21972 and 2015–67015–23207, and Michigan Alliance for Animal Agriculture. 1 The effect of adiposity and sampling timepoint on immune cell trafficking in subcutaneous adipose tissue of multiparous dairy cows ( n = 13) with high (HA) or low (LA) adiposity during gestational and postnatal periods. Subcutaneous adipose tissue was collected at (means ± SD): far off (−26 ± 7 d), close‐up (−8 ± 5 d) and postpartum (10 ± 2 d).Adiposity SEM P value Sampling SEM P valueSurface Marker 1 Low High Far off (FO) Close‐up (CU) Postpartum (PP)CD8 0.91 0.43 0.20 0.01 0.81 0.74 0.55 0.19 NSCD4 1.02 0.35 0.20 <0.01 0.88 0.71 0.59 0.19 NSCD3 1.80 1.06 0.36 0.06 1.76 1.44 1.30 0.35 NSB CELL 1.43 0.88 0.28 0.04 1.56 a 1.24 ab 0.83 b 0.26 0.08MHCII 0.45 0.43 0.24 NS 2 0.81 a 0.45 b 0.31 b 0.23 0.06CD172A 5.63 4.08 1.34 NS 6.75 a 5.35 ab 4.03 b 1.28 0.06CD163 1.40 0.91 0.49 NS 1.04 1.50 1.66 0.47 NSCD16 7.29 5.65 1.38 NS 6.90 6.22 7.29 1.31 NSCD14 1.36 0.99 0.32 NS 1.39 1.18 1.14 0.3 NSMHCII: Major histocompatibility complex class II; CD172a: Signal regulatory protein α; a–c Different letters within the same row indicate a significant difference (P ≤ 0.05). 1 Values are least squares means of percentages of stromal vascular fraction cells expressing phenotype surface markers ± SEM as measured by flow cytometry using specific bovine monoclonal antibodies. 2 NS: non‐significant effect (P > 0.05).2 Gene expression fold change of genes related to inflammation, immune cell infiltration, and extracellular matrix in the adipose tissue of multiparous cows ( n = 21) with high (HA) or low (LA) adiposity through gestational and postnatal periods. Subcutaneous adipose tissue was collected at (means ± SD): far off (−26 ± 7 d), close‐up (−8 ± 5 d) and postpartum (10 ± 2 d). Statistical analysis was performed using Δct values 1.Fold Change relative to LA FO sample 1, 3 Fold Change relative to FO sample 2, 3Adiposity P value Sampling P valueGene Low (LA) High (HA) Far off (FO) Close‐up (CU) Postpartum (PP)ARG1 0.70 1.49 0.02 1.00 0.69 0.68 NSCD44 1.16 1.67 NS 4 1.00 b 0.90 b 2.50 a <0.01COL6A2 0.69 0.74 NS 1.00 a 0.42 b 1.00 a 0.04IL10 1.07 2.00 0.09 1.00 0.93 1.41 NSSIRPA 0.78 10.13 <0.01 1.00 b 0.26 c 4.23 a <0.01SPP1 1.68 2.27 NS 1.00 b 0.99 b 4.59 a <0.01ARG1: Arginase 1; CD44: CD44 antigen; COL6A2: Collagen alpha‐2(VI) chain; IL10: Interleukine‐10; SIRPA: Signal regulatory protein α; SPP1: Osteopontin‐1. a–c Different letters within the same row indicate a significant difference (P ≤ 0.05). 1 Gene expression fold changes were calculated from least squares mean differences of the ΔCt values (ΔΔCt) normalized to the mean of EIF3K, B2M, and RPS9 housekeeping genes. 2 The FO subcutaneous adipose tissue sample of LA cows was the calibrator for the calculation of ΔΔct values transformed to fold change. Fold Change = 2 (−ΔΔCt); ΔΔCt = ΔCt LA‐FO sample − ΔCt target sample.3 The FO subcutaneous adipose tissue sample was the calibrator for the calculation of ΔΔct values transformed to fold change. Fold Change = 2 (−ΔΔCt); ΔΔCt = ΔCt FO sample − ΔCt target sample.4 NS: non‐significant effect (P > 0.05).