Relationships Between Serum α-Klotho Concentration and Inflammation-Related Cytokines in Hemodialysis Patients


 We investigated the relationship between α-Klotho and cytokines related to inflammation in HD patients. We analyzed levels of α-Klotho with ELISA and inflammatory cytokines with CBA in the serum of HD patients. There was a significant negative correlation between the concentration of serum α-Klotho and patients’ age and the serum concentration of PTH. No correlation has been found between α-Klotho and Ca or Pi. HD time, creatynine or eGFR. However, there were significant positive correlations between the concentration of α-Klotho and the serum concentration of IL-12p70, IL-10, and IL-1β. Furthermore, the concentration of IL-10 and IL-1β was significantly lower in HD patients with low α-Klotho concentrations compared with HD patients with high α-Klotho. However, in a multivariable linear regression analysis, only patients’ age was associated independently with α-Klotho level. While these results draw our attention to potential relationships between α-Klotho proteins and inflammatory markers in HD patients, our cross-sectional study could not fully explain the pathogenic link between α-Klotho and inflammation in these patients. Therefore, further studies are necessary to clarify these relationships. However, this observation aligns with previous studies that confirm a significant relationship between Klotho concentration and human aging.


Introduction
The KLOTHO gene encoding the protein of the same name discovered at the end of the 20th century is one of the genes affecting animal and human life span 1 . Its knock-out mutations in mice induced osteoporosis, vascular calci cation, muscle atrophy, hypoglycemia, and hyperphosphatemia and promoted a reduced activity, hearing impairment, generally leading to accelerated aging 2 . Many of the animal studies were later partially con rmed in humans. One of the best-known functions of Klotho, which belongs to the β-glucuronidase family, is its role in regulating calcium and phosphate metabolism, among others, by regulating pathways dependent on broblast growth factor 23 (FGF23) 3 . Therefore, one of the organs in which Klotho expression is highest are the kidneys 4 .
We now know about the existence of three Klotho isoforms, of which α-Klotho is best known. It exists either in a form associated with the cell membrane or in a secretory form (sKlotho). Secretive Klotho may arise as a result of the extracellular part being excised from the cell membrane or alternative splicing 5 . One of the essential functions of α-Klotho (both membrane and secretory) is the role of the FGF23 cofactor in increasing renal phosphate secretion 6 . Also, sKlotho with circulation goes to various tissues where it inhibits insulin-like growth factor (IGF) dependent pathways 7 .
Chronic kidney disease (CKD), currently considered a civilization disease, is associated with the loss of active nephrons occurring in the renal parenchyma and inevitably leads to end-stage renal disease (ESRD) that requires renal replacement therapy. The course of the disease is complicated by cardiovascular disease, anemia, malnutrition, acid-base, and calcium-phosphate disorders. The loss of active renal parenchyma also leads to hypocalcemia, hyperphosphataemia, and renal osteodystrophy, which are associated with a gradual decrease in serum sKlotho 8 . According to Kim et al., among patients with low α-Klotho concentrations (below the median value of 396 pg/ml), a doubling of creatinine levels within a couple of months of observation occurs in a higher percentage than in the patients with concentrations above 396 pg/ml 8 . Also, α-Klotho concentration is positively correlated with the estimated glomerular ltration rate (eGFR) and calcium level 8 .
In hemodialysis (HD) patients, α-Klotho levels are also associated with cardiovascular morbidity and mortality [9][10][11] . Otani-Takei et al. showed that HD patients with low Klotho serum concentrations (<309 pg/ml) are more likely to die from cardiovascular causes 10 . Meanwhile, HD patients with high concentrations of sKlotho have reduced the occurrence of outcomes combining cardiovascular events and cardiovascular death 9 . Therefore, serum Klotho level could is by some authors considered a potential predictor of cerebrovascular disease in HD patients 11 .
ESRD patients, especially those on hemodialysis, are characterized by increased oxidative stress and in ammation, which seem to be a central component of the uraemic phenotype. Chronic in ammation has been linked with cardiovascular mortality 12 and protein-energy wasting (PEW) in HD patients 13 .
In ammatory markers, especially interleukin (IL)-6, are also recognized as a strong predictor of poor outcome in ESRD patients 14 . The pathophysiology of chronic in ammation in CKD is a consequence of multiple factors, including the type of dialysis membrane, uraemic toxins, oxidative stress, cellular senescence, gut dysbiosis, hypoxia, uid and sodium overload 15 . It has been hypothesized that α-Klotho de ciency might contribute to increased oxidative stress and in ammation in ESRD patients. Low concentrations of α-Klotho are associated with in ammation in experimental kidney disease models 16 . Human studies show that in patients on peritoneal dialysis (PD), high concentrations of Klotho (above the median value of 329 pg/ml) are correlated with lower IL-6 levels 17 . Still, the problem of the relationship between α-Klotho and in ammation in ESRD is poorly understood, and there are only a few articles on this subject. Therefore, this work aimed to analyze relationships between the serum concentration of α-Klotho and cytokines associated with in ammation (IL12p70, TNF, IL-10, IL-6, IL-1β, IL-8) in HD patients.

Demographic, clinical, and laboratory parameters by median α-Klotho level
The primary cause of CKD or the type of DM did not in uence the serum concentration of α-Klotho. Likewise, the vitamin D supplementation did not in uence the serum concentrations of PTH, Ca, Pi, or α-Klotho.
Next, patients were divided into two subgroups according to the median Klotho value: those with α-Klotho concentrations below (34 people) or above (33 people) the median (6,96 pg/ml). The subdivision was somehow arti cial, but there are no clear ranges of serum values for α-Klotho. HD patients with low α-Klotho concentrations were characterized by higher PTH concentrations (the median value of 425 pg/ml) compared to patients with high α-Klotho (the median value of 320 pg/ml) (Tab. 4, Fig. 1A). No difference was found in calcium or phosphorus level. HD patients with low α-Klotho had lower IL-10 levels (the median value of 2,45 pg/ml) compared to patients with high α-Klotho (the median value of 3,44 pg/ml ) (Tab. 4, Fig. 1B). The concentration of IL-1β was signi cantly lower in HD patients with low α-Klotho (the median value of 1,5 pg/ml) compared with HD patients with high α-Klotho (the median value of 2,34 pg/ml) (Tab. 4, Fig. 1C). No signi cant differences were observed between the two groups in age, hemodialysis duration, eGFR, dialysis adequacy, or other serum biomarkers.

Discussion
The history of Klotho is associated with the discovery of the relationship between its expression and aging in mice 1 . Among three isoforms, the most important is α-Klotho, which occurs in the kidney in high concentrations. It is produced mainly in the distal nephron tubules and is associated with calciumphosphate metabolism 3 . CKD patients present a gradual decrease in serum sKlotho, accompanied by an increase in FGF23. In the kidney, decreased α-Klotho expression occurs as early as in stage 2 of CKD. However, even in those with ESRD, we can still detect some level of Klotho in blood. α-Klotho in these patients is secreted compensatively from other places, e.g., from choroid plexus or parathyroid glands. However, studies show that the parathyroid tissue of patients with more advanced stages of CKD has also reduced the expression of Klotho 18 . Even the vascular smooth muscle of CKD patients displays reduced Klotho content 19 .
Over the years, many publications have emerged showing signi cant correlations between the concentration of α-Klotho in the blood of CKD patients and the progression of the disease 8 , and cardiovascular complications, especially in HD patients [9][10][11] . However, we do not have data on whether there is any relationship between α-Klotho concentration and in ammatory markers in HD patients. According to some animal studies, α-Klotho exerts some antiin ammatory effects 16 .
ESRD is well characterized by increased oxidative stress and in ammation, which seem to be a central component of the uraemic phenotype and are linked with cardiovascular mortality among HD patients 12 . HD patients have high serum IL-1, IL-6, IL-8, and TNF-α 20 , produced mainly by mononuclear cells 21 . IL-6, in particular, is recognized as a strong predictor of poor outcomes in ESRD patients 14 . In our study, α-Klotho positively correlated with IL-12p70, IL-10, and IL-1β, but not with TNF or IL-6. Downregulation of renal α-Klotho expression increases kidney in ammation 16 and leads to renal brosis 22 . These processes are related to reduced production of reactive oxygen species (ROS) and reactive nitrogen species (RNS) 23 , suppressed TNF-induced NF-kappaB activation 24 , and upregulated anti-in ammatory IL-10 23 . Thus, the anti-in ammatory role of α-Klotho could explain a positive correlation with anti-in ammatory IL-10 in our study group. From human studies, such an association was only con rmed in patients with established cardiovascular disease (CVD), whose low Klotho concentrations accompanied decreased IL-10 levels 25 . We did a similar observation for our patients -those with low α-Klotho concentrations had signi cantly lower IL-10 compared with HD patients with high α-Klotho.
However, what is surprising is the positive correlation between Klotho and IL-1β and IL-12p70, which have a pro-in ammatory effect. IL-12 is a proin ammatory cytokine that induces the production of interferon-γ (IFN-γ) by T helper 1 (T h 1) cells associated with cell-mediated immunity. Studies showed that HD patients exhibit an increased percentage of T h 1 cells compared with healthy controls, and their monocytes produce high levels of IL-12 26 . IL-1β, also known as leukocytic pyrogen, is an important mediator of the in ammatory response associated with acute symptoms such as fever and hypotension in HD patients 27 . Both cytokines are associated with attenuated in ammatory responses in ESRD, aggravated by high concentrations of uremic toxins and repeated contact with the dialysis membrane 20 . Their positive correlation with a high level of α-Klotho draws attention to its potential, not yet described, role in promoting in ammation. As mentioned above, α-Klotho protects cells from oxidative stress, in ammation, and brosis 22 . Klotho depletion increases nuclear factor NFκB activation and the subsequent transcription of pro-in ammatory genes 24 . The positive correlation between α-Klotho and pro-in ammatory cytokines could be ultimately not of much importance because, in a multivariable linear regression analysis, only age was associated independently with patients' α-Klotho level.
On the other hand, HD patients with low α-Klotho concentrations had low both IL-10 and IL-1. Thus, the relationship between IL-10 and IL-1 levels is most likely the result of monocytes and macrophages responding to high levels of pro-in ammatory cytokines. This is related to the anti-in ammatory role of IL-10. In particular, macrophages produce IL-10 in a negative feedback loop to reduce the uncontrolled in ammatory cytokine production during, for example, infection 28 . By binding with its receptor on cells of innate immunity, IL-10 inhibits the release of pro-in ammatory cytokines, decreases antigen presentation and phagocytosis. Accordingly, it can be expected that high IL-1 levels will be accompanied by the higher IL-10 level.
The correlation matrix showed no relationships between serum α-Klotho and any variables associated with CKD, including creatinine, eGFR, or calcium and phosphorus concentrations, except for PTH. However, some studies show that α-Klotho concentrations strongly correlate with eGFR in different stages of CKD 8 . The exception is the study by Seiler et al. 29 , in which the authors demonstrated that serum level of Klotho in CKD patients correlates only with the patient's age. In their opinion, therefore, Klotho concentration is not related to kidney function and cannot predict adverse outcomes in CKD patients. Also, serum α-Klotho in CKD patients does not necessarily re ect Klotho de ciency at the tissue level. Besides the kidneys, Klotho is also produced in other organs, e.g., choroid plexus or parathyroid glands. Moreover, Yildirim et al. 30 have recently demonstrated that HD and PD patients have higher serum α-Klotho compared to healthy people. The presented results undermine the assumptions made so far that the level of α-Klotho decreases with CKD progress. It is also possible that in HD patients, the eGFR is already too low and the calcium-phosphate balance too disturbed to demonstrate any relationship with Klotho levels. Wei et al.
also saw no correlation between serum Klotho and Ca, Pi, or PTH in HD patients despite the larger study group 11 . Finally, in a multivariable linear regression analysis, only age was associated with patients' serum Klotho level. This observation aligns with previous studies that con rm a signi cant relationship between Klotho concentration and the aging of the human or animal organism 1,29 .
In summary, these associations, which we demonstrated in our study, draw attention to the potential relationship between α-Klotho levels and in ammation status of HD patients. However, our cross-sectional study could not fully explain the pathogenic link between α-Klotho, and in ammation. Therefore, further studies are necessary to clarify these relationships.

Patients
The study groups consisted of 67 HD patients of mean age of 67,06 ± 12,73 years (27 men and 40 women). The mean time of HD treatment was 36,47 months (minimum time was 0,5 of the month, and maximum -169 months) (Tab. 1). All patients had estimated glomerular ltration rate (eGFR) below 15 min/ml/1.73 m2 and underwent 4h sessions of hemodialysis three times a week using low-ux NIPRO PES 150DL, 170DL, 210DL or high-ux ELISIO 15H and 17H dialyzers. In addition, patients had regular dialysis adequacy assessment by measuring urea clearance using equation Kt/V ( Kurea clearance, ttime on dialysis, Vvolume of distribution); mean Kt/V was 1,52 ± 0,33.
In 13 patients, the primary cause of chronic kidney disease was glomerulonephritis (GN), in 25diabetic nephropathy (DN), in 13ischemic nephropathy (IN), in 4 -hypertensive nephropathy (HN), in 6adult polycystic kidney disease (ADPKD). In 2 patients, the primary cause of CKD was unknown. 2 patients had obstructive nephropathy, and 2 patients had granulomatosis with polyangiitis (GPA). None of the patients suffered from any infection, in ammation, malnutrition, malignancy, or blood loss during the study. 36 patients had diabetes mellitus (DM).
All participants were informed about the purpose of the tests and gave their written informed consent. The Bioethical Committee for Scienti c Research at the Medical University of Gdansk approved the study. We performed all the experiments following the relevant guidelines and regulations.
We collected 5 ml of peripheral venous blood from each patient before the HD session into anticoagulant-free tubes to collect serum to assess concentrations of cytokines and α-Klotho. We stored serum samples at -80°C.
Cytokine measurement in plasma samples

Analysis and Statistics
The concentration of α-Klotho was calculated with GraphPad Prism 9 (GraphPad Software Inc., USA). Statistica (data analysis software system), version 13 (TIBCO Software Inc. 2017), and GraphPad Prism 9 were used to perform statistical analysis. The Kolmogorov-Smirnov and Lilliefors tests were used for