Low dietary fiber intake increases Collinsella abundance in the gut microbiota of overweight and obese pregnant women

Figures & data

Table 1. Fasting biochemical and dietary characteristics of pregnant women at 16 weeks gestation according to maternal BMI.

	Pregnancy-specific reference range	Overweight (n = 57)	Obese (n = 73)	P value
Maternal characteristics
Age	NA	32.0 (29.0 – 34.0)	30.5 (28.0 – 34.0)	0.299
BMI (kg/m^2)	NA	27.9 (27.0 – 29.1)	34.3 (31.8 – 41.3)	< 0.0001
Gestational age (days)	NA	112 (105 – 112)	105 (98 – 112)	0.008
Insulin (μU/L)	NA	4.9 (3.4 – 7.6)	7.8 (5.7 – 10.75)	< 0.0001
C-peptide (pg/mL)	NA	0.5 (0.4 – 0.6)	0.7 (0.5 – 0.8)	< 0.0001
Glucose (mmol/L)	4.2-4.4	4.3 (4.1 – 4.5)	4.4 (4.1 – 4.6)	0.125
HbA1C (%)	4.0-6.0	4.8 (4.7 – 5.0)	4.9 (4.7 – 5.1)	0.038
HbA1C (mmol/mol)	20-42	29 (28 – 31)	30 (28 – 32)	0.038
HOMA-IR	NA	0.7 (0.5 – 0.9)	1.0 (0.7 – 1.3)	< 0.0001
Total cholesterol (mmol/L)	4.6-7.7	5.3 (4.7 – 5.8)	5.3 (4.6 – 6.2)	0.704
HDL (mmol/L)	1.4-2.3	1.9 (1.6 – 2.1)	1.6 (1.4 – 1.9)	< 0.0001
LDL (mmol/L)	2.0-4.8	2.7 (2.3 – 3.3)	3.0 (2.6 – 3.7)	0.058
VLDL (mmol/L)	0.3-0.6	0.6 (0.4 – 0.7)	0.6 (0.5 – 0.8)	0.011
Triglycerides (mmol/L)	0.9-4.3	1.3 (0.8 – 1.6)	1.4 (1.1 – 1.7)	0.020
Dietary intake
Carbohydrate (g/d)		152.4 (121.0 – 201.4)	155.9 (132.3 – 180.3)	0.902
Protein (g/d)		70.6 (56.9 – 98.1)	77.6 (60.7 – 88.1)	0.661
Fiber (g/d)		17.7 (13.8 – 25.1)	17.4 (13.7 – 21.1)	0.358
Fat (g/day)		59.9 (47.6 – 83.7)	59.1 (49.9 – 82.4)	0.885
Saturated fat (g/d)		25.0 (18.6 – 35.8)	27.0 (19.3 – 36.5)	0.756
Polyunsaturated fat (g/d)		8.3 (5.6 – 12.4)	7.1 (6.0 – 9.9)	0.260
Monounsaturated fat (g/d)		21.5 (17.4 – 29.3)	22.3 (17.9 – 28.9)	0.759
Starch (g/d)		81.9 (56.7 – 112.3)	85.0 (61.1 – 105.8)	0.846
Total energy (Kj/d)		6360 (4663 – 7837)	6098 (5080 – 7671)	0.806
Pregnancy outcomes
Gestational diabetes mellitus (%)		0 (0)	0 (0)	ND
Hypertensive disorders of pregnancy (%)		3 (5.5)	13 (17.8)	0.06
Gestational age at delivery		281 (274-287)	285 (282-289)	0.15
Cesarean section (%)		14 (25)	26 (36)	0.25
Infant gender (F/M)^$		26/23	37/28	0.70
Infant birth weight		3570 (3230-3944)	3615 (3273-3831)	0.96

Data is presented as median (IQR) when not indicated otherwise; Differences determined by Mann-Whitney U comparisons for continuous variables and Fisher's exact tests for proportions; NA, not applicable/no specific pregnancy range;

* Ranges presented for the second trimester of pregnancy for women, not specific for BMI categories⁶³;

^# Women who developed gestational diabetes mellitus were excluded from this sub-study; $ infant gender not known for 8 infants of overweight and 7 obese mothers.

Table 2. Metabolic parameters and their correlation with genus Collinsella abundance at 16 weeks gestation.

	Collinsella abundance
Metabolic parameters	rho	P value
BMI	0.161	0.070
Insulin	0.362	< 0.001
C-peptide	0.263	0.003
Fasting glucose	0.073	0.411
HOMA-IR	0.331	0.0002
Hb1Ac	0.132	0.138
Total cholesterol	−0.097	0.273
HDL	−0.079	0.370
LDL	−0.059	0.505
VLDL	0.032	0.715
Triglycerides	0.037	0.671

Figure 1. Collinsella relative abundance and maternal circulating insulin. (A) Positive correlation between genus Collinsella and insulin levels in overweight (black) and obese pregnant women (grey). (B) Collinsella abundance in overweight (black) and obese (grey) pregnant women.

Figure 2. Connections between Collinsella relative abundance, circulating insulin and dietary fiber intake. (A) Negative correlation between dietary fiber intake and Collinsella relative abundance. (B) Comparison of genus Collinsella between women with low and high dietary fiber intake. (C) Negative correlation between dietary fiber intake and insulin levels in overweight and obese pregnant women.

Figure 3. Differences in microbiota composition according to high and low fiber intake. (A) Network analysis based on high (n = 13) and low (n = 13) dietary fiber intake. Genera are represented as nodes. Green nodes represent genera overrepresented in the low-fiber group. Red nodes represent genera overrepresented in the high-fiber group. Color intensity explains the strength of the associations with fiber intake. (B) Principal component analysis showing the clustering of selected samples based on high (red) and low (green) fiber intake. (C) Identification of significant associations between low and high dietary fiber groups by Anosim test.

Figure 4. Network analysis after correction of dietary fiber by total energy intake. (A) Network analysis based on high (n = 13) and low (n = 13) dietary fiber intake. Genera are represented as nodes. Green nodes represent genera overrepresented in the low-fiber group. Red nodes represent genera overrepresented in the high-fiber group. Color intensity explains the strength of the associations with fiber intake. (B) Principal component analysis showing the clustering of selected samples based on high (red) and low (green) fiber intake. (C) Identification of significant associations between low and high dietary fiber groups by Anosim test.

Figure 5. Predictive function analysis of gut microbiota profiles of women with low or high dietary fiber intake. Pathways predicted to be upregulated in women with low fiber intake in green bars, those upregulated in women with high fiber intake in red bars.

Abbassi-Ghanavati M, Greer LG, Cunningham FG. Pregnancy and laboratory studies: a reference table for clinicians. Obstet Gynecol. 2009;114:1326–31. doi:10.1097/AOG.0b013e3181c2bde8. PMID:19935037.

 PubMed Web of Science ®Google Scholar