INCREASED CONCENTRATION OF THE COMPLEMENT SPLIT PRODUCT C5a IN ACUTE PYELONEPHRITIS DURING PREGNANCY

Abstract

OBJECTIVE

Pregnant women with acute pyelonephritis develop acute respiratory distress syndrome (ARDS) more frequently than non-pregnant women. The reasons for this remain unknown. The complement system is a complex set of self-assembling proteins that have been implicated in the pathophysiology of ARDS and sepsis. The purpose of this study was to determine if activation of the complement system occurs in pregnant women with acute pyelonephritis.

METHODS

A cross-sectional study was conducted to determine the plasma concentrations of C3a, C4a and C5a (i.e., complement split products) in pregnant patients with acute pyelonephritis (n=38) and normal pregnant women (n=38). The complement split products C3a, C4a and C5a were measured using ELISA. Data were analyzed using non-parametric statistics.

RESULTS

1) The median plasma concentration of C5a in pregnant patients with acute pyelonephritis was significantly higher than that in normal pregnant women (p< 0.001); 2) there was no statistical difference in the median plasma concentration of C3a and C4a between the two groups (p> 0.05); and 3) concentrations of C3a, C4a and C5a were not different among patients with acute pyelonephritis with and without bacteremia.

CONCLUSIONS

1) Pyelonephritis in pregnant women is associated with increased plasma concentration of C5a, but not C3a and C4a; 2) an excess of C5a can predispose pregnant women to develop ARDS and multi-organ failure in pyelonephritis. This finding may have clinical implications since blocking C5a improves ARDS in experimental sepsis.

INTRODUCTION

The complement system, a major component of the innate immune response, plays an important role in host defense against invading pathogens. The complement cascade is formed by several plasma proteins with catalytic properties that react in a sequential manner, yielding active biological mediators and lytic components to clear foreign cells. In addition, the complement system is involved in the regulation of the adaptive immune response.[1]

Infection can trigger activation of the complement system directly by microbial surfaces or products, or indirectly by antigen-antibody complexes. Anaphylatoxins C3a, C4a and C5a are pro-inflammatory peptides released from the cleavage of their native components following activation of the complement cascade. These molecules induce vascular permeability,[2–4] smooth muscle contraction[3–5], the release of pro-inflammatory cytokines[6–8] and white blood cell chemotaxis.[9–11] Among anaphylatoxins, C5a is the most potent.[4] Its interaction with phagocytic cells increases oxidative burst,[12,13] induces expression of adhesion molecules[14,15] and inhibits polymorphonuclear (PMN) apoptosis.[16] C3a effects are predominantly on mast cells and eosinophils, and include chemotaxis,[10,11] granule release,[11,17] expression of adhesion molecules (β₂ integrin) and L-selectin shedding.[14] C4a, the weakest anaphylatoxins,[4] can only be generated through the classical pathway. Little is known about C4a receptor interactions and its biological functions in humans.

Even though “complement split products” participate in microbial clearance, excessive production may be harmful to the host.[18] Indeed, plasma and serum concentrations of anaphylatoxins are higher in septic patients with a poor outcome than in those with a favorable outcome.[19–21] Moreover, C5a has been implicated in the generation of acute respiratory distress syndrome (ARDS).[22,23]

Pregnancy is a unique immune state characterized by activation of the innate immune response and suppression of adaptive immunity.[24] This is thought to provide protection to the mother against infection[24] and promote tolerance of the fetal semi-allograft. Nonetheless, pregnant women are more susceptible to the effects of microbial products.[25] For example, urinary tract infections are frequently encountered during pregnancy,[26,27] and acute pyelonephritis affects 1 to 2% of pregnant women, particularly during the last two trimesters.[28] Pregnant women with acute pyelonephritis are at risk of developing sepsis and ARDS,[29–33] complications that are rare in non-pregnant women with urinary tract infections. The reasons for this remain unknown.

Previous studies have demonstrated complement activation during normal pregnancy,[34] as determined by higher plasma concentrations of anaphylatoxins (C3a, C4a and C5a) in pregnant women than in non-pregnant women.[35] However, the plasma concentrations of these complement split products in acute pyelonephritis during pregnancy has not been reported. The purpose of this study was to determine the plasma concentration of anaphylatoxins in pregnant patients with and without acute pyelonephritis.

PATIENTS AND METHODS

Study Design

A cross-sectional study was conducted by searching our clinical database and bank of biological samples. This study was designed to compare the plasma concentration of complement split products C3a, C4a and C5a among normal pregnant women (n=38) and pregnant patients with acute pyelonephritis (n=38). The control group consisted of pregnant women from 20 to 42 weeks, not in labor, and without medical or obstetrical complications. Acute pyelonephritis was diagnosed in the presence of fever (temperature ≥ 38º C), clinical signs or symptoms of upper urinary tract infection (e.g., flank pain, costovertebral angle tenderness), pyuria and a positive urine culture. Blood cultures were also performed. Both groups were matched for gestational age at venipuncture within two weeks. Eligible patients were approached at the Detroit Medical Center/Hutzel Women’s Hospital in Detroit, Michigan. All women provided written informed consent prior to the collection of samples. The collection of samples was approved by the IRBs of both the National Institute of Child Health and Human Development and Wayne State University. Many of these samples have been previously used in other studies.

Blood Collection

Samples of peripheral blood for complement determinations were collected in tubes containing EDTA (ethylene diamine tetraacetic acid). The samples were centrifuged at 1068g and stored at −70º C. Specific and sensitive enzyme-linked immunoassays (ELISA) were used to determine concentrations of complement C3a, C4a and C5a. Immunoassay systems for C3a and C4a were obtained from Assay Designs, Inc. (Ann Arbor, MI). C5a immunoassays were obtained from American Laboratory Products Company (Windham, NH). Complement C3a, C4a and C5a assays were performed according to the manufacturer’s instructions. Briefly, maternal plasma samples were incubated in duplicate wells of microtiter plates, which had been pre-coated with antigen specific (C3a, C4a or C5a) antibodies. Complement C3a, C4a or C5a present in the standards or maternal plasma samples were immobilized by their specific pre-coated antibodies (forming antigen antibody complexes) during this incubation. Repeated washing and aspiration was conducted to remove unbound materials from the assay plates. This step was followed by incubation with a specific antibody-enzyme reagent. Following a wash step to remove excess unbound materials, a substrate solution was added to the wells of the microtiter plates and color developed in proportion to the amount of antigen bound in the initial step of the individual assay. The color development was stopped with the addition of an acid solution and the intensity of color was read using a programmable microtiter plate spectrophotometer (Ceres 900 Microplate Workstation, Bio-Tek Instruments, Winooski, VT). The concentrations of complement C3a, C4a or C5a in maternal plasma samples were determined by interpolation from individual standard curves composed of purified human C3a, C4a or C5a. The calculated inter-assay coefficients of variation (CV) for C3a, C4a and C5a immunoassays in our laboratory were 5.4%, 6.1%, and 4.0%, respectively. Calculated intra-assay CV for C3a, C4a and C5a were 6.6%, 6.9%, and 2.3%, respectively. The detection limit (sensitivity) was 0.13 ng/ml for C3a, 0.32 ng/ml for C4a, and 0.06 ng/ml for C5a.

Statistical Analysis

The Shapiro-Wilk test was used to test for normal distribution of the data. Mann-Whitney U tests were utilized to compare the medians of normal pregnant women and pregnant patients with acute pyelonephritis. A chi-square test was used for comparison of proportions. The statistical package used was SPSS 12 (SPSS Inc., Chicago, IL). A probability value < 0.05 was considered statistically significant.

RESULTS

Clinical and obstetrical characteristics of women in each group are displayed in Table I. Women with acute pyelonephritis delivered neonates with lower birthweight than those in the control group (pyelonephritis median: 3175 g, range: 1080–4090g vs. normal pregnancy median: 3372 g, range 2610–4080g; p= 0.04). The other characteristics were similar between the two groups.

Table I.

Clinical characteristics of the study population and comparison of anaphylatoxin plasma concentrations

	Normal Pregnancy n = 38	Pyelonephritis n = 38	P
Maternal age (y)	24 (17–34)	22 (16–41)	0.4
Nulliparity	13 (34.2)	13 (34.2)	1
Smoking	4 (10.5)	5 (13.2)	0.7
Gestational age at venipucture (wks)	31.6 (20.5–41.7)	31.6 (20.7–42.2)	0.9
Gestational Age at delivery (wks)†	39.0 (37–41.8)	39.2 (28.5–42.7)	0.8
Birthweight (g)‡	3372 (2610–4080)	3175 (1080–4090)	0.04*
C3a (ng/ml)	2393.4 (557.9 – 6642.7)	2414.5 (769.3 – 19150)	0.4
Quartiles
25	1765.2	1913.9
75	3136.1	3328.8
C4a (ng/ml)	7827 (850.7 – 27380)	5894.6 (619.9 – 33400)	0.1
Quartiles
25	3755.4	3500
75	12925	8464.7
C5a (ng/ml)	12.3 (1.5 – 87.1)	20.1 (9.7 – 61.8)	<0.001*
Quartiles
25	7.5	16.2
75	16.2	27.9

Values expressed as median (range) or number (percent)

^*Statistically significant, P < .05

^†Pyelonephritis n= 37

^‡Pyelonephritis n= 36

Among women with acute pyelonephritis, blood cultures were positive in 11 (28.9%) patients. Microorganisms isolated from urine cultures included: Escherichia coli (31), Klebsiela (1), Proteus mirabilis (1), Pseudomonas species (1), Staphylococcus aureus (1), Lactobacili (2), Gardnerella vaginalis (2), Citrobacter koseri (1), Streptococcus species (1) and mixed flora (1). Blood cultures were positive for Escherichia coli (8), Enterobacter aerogenes (1), and Staphylococcus aureus (2).

The median plasma concentration of C5a in pregnant patients with acute pyelonephritis was significantly higher than in normal pregnant women (acute pyelonephritis median 20.1 ng/ml, range 9.7 – 61.8 ng/ml vs. normal pregnancy median 12.3 ng/ml, range: 1.5 – 87.1 ng/ml; p< 0.001; Figure 1, C). There was no significant difference in the median plasma C3a and C4a concentrations between the two groups (p = 0.4 and p = 0.1, respectively; Figures 1a and 1b) Among patients with acute pyelonephritis, no significant difference in the median plasma concentration of C5a, C4a and C3a was observed between those who had both positive urine and blood cultures and those who had only positive urine cultures (C5a, p= 0.76; C4a, p=0.34; and C3a, p= 0.22).

Figure 1.

Plasma anaphylatoxin concentration in normal pregnancy and pregnancy complicated with acute pyelonephritis. a, The median plasma C3a concentration was not significantly different between pregnant patients with acute pyelonephritis and normal pregnant women (median 2414.5 ng/ml, range 769.3 – 19150 ng/ml vs. median 2393.4 ng/ml, range 557.9 – 6642.7 ng/ml; p=0.4). b, the median plasma C4a concentration was not significantly different between pregnant women with and without acute pyelonephritis (median 5894.6 ng/ml, range 619.9 – 33400 ng/ml vs. median 7827 ng/ml, range 850.7 – 27380 ng/ml; p=0.1). c, In contrast, pregnant patients with acute pyelonephritis had a median plasma C5a concentration higher than normal pregnant women (median 20.1 ng/ml, range 9.7 – 61.8 ng/ml vs. median 12.3 ng/ml, range 1.5 – 87.1 ng/ml; p<0.001).

DISCUSSION

This study demonstrated that pregnant women with acute pyelonephritis had a higher median plasma concentration of C5a than normal pregnant women (Figure 1, C). To our knowledge, this is the first study that has determined complement anaphylatoxins in pyelonephritis in pregnancy.

C5a exerts its activities by coupling with its seven loop transmembrane receptor C5aR/CD88.[36] C5aR has been identified in a variety of cells including myeloid, lymphoid, endothelial, and cells from different tissues such as liver, lung, kidney and brain.[37–41] Recently, it has been shown that C5a-C5aR interaction participates in the production of antiviral CD8+ T cell response.[42] C5a activities on white blood cells include the release of granules containing elastase,[43] glucoronidase,[11] histamine[44] and eosinophil cationic protein (ECP),[17] stimulation of granulocyte and monocyte oxidative burst with the release of reactive oxygen species, [12,13,45,46] an increased expression of β₂-integrin adhesion molecules and shedding of L-selectin,[14] an increase in vascular permeability[2,3] and delayed neutrophil apoptosis.[16] These mechanisms of C5a have been implicated in the induction of a potent inflammatory response and prolongation of the life span of neutrophils in the presence of foreign antigens. [16]

A solid body of evidence indicates that C5a stimulates the white blood cell release of pro-inflammatory cytokines. In vitro studies have shown that C5a induces the expression of IL-1[8,47,48] and IL-8[6] by monocytes. In animal and in vitro studies, C5a augmented LPS-induced release of pro-inflammatory cytokines such as IL-6[7] from neutrophils and TNF∀ [8,49] from mononuclear cells. On the other hand, these pro-inflammatory cytokines induced the expression of anaphylatoxins in white blood cells. Indeed, IL-1 induces the expression of C3a and C5a receptors in monocytes.[50]

During normal pregnancy, maternal immunity is modified presumably to adapt to the presence of the fetus. Changes in the innate immunity involving leukocytes include increased leukocyte concentration,[51] enhanced oxidative burst and baseline intracellular reactive oxygen species,[52] increased expression of granulocyte/monocyte adhesion molecules,[52] reduced neutrophil apoptosis[53] and increased concentration of plasma granulocyte elastase.[54] Patients with pyelonephritis have marked phenotypical and metabolic changes in granulocytes and monocytes than those detected in normal pregnancy.[52] Neutrophil elastase and myeloperoxidase, as well as macrophage serine proteases, can cleave C5 directly and generate C5a.[55–58] Thus, it is possible that leukocyte activation may account for the high concentration of C5a found in the plasma of pregnant patients with acute pyelonephritis. It is unclear why C3a and C4a were not different between the groups.

There is growing evidence that excessive C5a can be harmful to the host. Indeed, patients with sepsis who had a higher concentration of C5a had lower survival rates than those who had lower C5a serum levels.[19–21] High concentrations of C5a during sepsis are associated with the loss of expression of C5aR due to granulocyte internalization of the C5a-C5aR complex[59–61] with subsequent reduction of the oxidative burst and generation of reactive oxygen species, chemotaxis and impaired bacterial killing.[60–62] In addition, excessive C5a has been linked to the pathophysiology of ARDS.[22,23] Indeed, C5a is increased in bronchoalveolar lavage[63] and plasma[22] of patients with this condition. Furthermore, the administration of anti-C5a improved the outcome of sepsis and ARDS in animal models. [23,62,64,65]

In summary, pregnancy complicated with acute pyelonephritis is characterized by increased maternal plasma C5a concentration. C5a may act on leukocytes and other host cells to potentiate the mechanisms necessary to effectively remove pathogens. On occasion, persistent or exaggerated activation of C5a could lead to detrimental effects to the mother such as ARDS.