Shorter Telomere Length in Peripheral Blood Leukocytes Is Associated with Post-Traumatic Chronic Osteomyelitis

Abstract

Background:

This study investigated the association between post-traumatic chronic osteomyelitis (COM) and peripheral leukocyte telomere length (PLTL) and explored factors associated with PLTL in COM.

Methods:

A total of 56 patients with post-traumatic COM of the extremity and 62 healthy control subjects were recruited. The PLTL was measured by real-time PCR. Binary logistic regression analysis was used to identify factors in correlation with telomere length. Sex, age, white blood cell (WBC) count, erythrocyte sedimentation rate (ESR), C-reactive protein (CRP), and infection duration were included as independent variables in the logistic regression model.

Results:

Post-traumatic COM patients had significantly shorter PLTLs (5.39 ± 0.40) than healthy control subjects (5.69 ± 0.46; p < 0.001). Binary logistic regression analysis showed that PLTL had a statistically significant association with age (B = −0.072; p = 0.013) and CRP (B = −0.061; p = 0.033). The logistic regression model was statistically significant and explained 31.4% (Nagelkerke R²) of the change in telomere length and correctly classified 69.6% of the cases.

Conclusions:

Patients with post-traumatic COM have shorter PLTLs than healthy subjects. The PLTL erosion of post-traumatic COM was partially explained by age and CRP.

Chronic osteomyelitis (COM), a chronic infectious disease that lasts months or even years, is characterized by the persistence of micro-organisms, low-grade inflammation, dead bone (sequestrum), and fistulous tracts [1]. In addition to local bone and bone marrow damage, systemic complications were found secondary to COM, such as a higher risk of end-stage renal disease [2], ischemic stroke [3], coronary artery disease [4], depression [5], and a significant increase in the long-term mortality risk among the elderly [6]. Many of the complications relate to aging; however, no study previously explored directly the relations between osteomyelitis and aging.

Telomeres are special structures that lie on the ends of eukaryotic cells' chromosomes and play an essential role in protecting chromosomes from recombination and degradation activities [7]. Telomeric sequences are lost during cell division because the cell mechanisms cannot fully replicate the 3′ end of the linear chromosome [8]. Telomerase can help to maintain the telomere length in some human cells, such as stem cells and germ cells; however, in most somatic cells, the activity of telomerase is down-regulated [7]. Peripheral leukocyte telomere length (PLTL) had been thought to be a reliable surrogate for telomere length in other somatic cells [9] and to be a marker of aging for the whole body [7].

Therefore, it is reasonable to hypothesize that COM would be associated with telomeric erosion. In the present investigation, we used a case-control study design to evaluate the association between PLTL and COM as well as with several clinical features of the disease.

Patients and Methods

Ethics approval and consent to participate

The complete details of the study design and procedures involved were in accordance with the Declaration of Helsinki. Prior to formal experience, all participants gave their written informed consent to participate. The study protocol was approved by the Ethics Committee of Nanfang Hospital, Southern Medical University.

Participants

Subjects with extremity post-traumatic COM were recruited from the in-patients in the Department of Orthopaedics and Traumatology, Nanfang Hospital, Southern Medical University, between July 2015 and April 2016. In this study, COM was defined as osteomyelitis lasting more than 10 weeks [10]. A COM diagnosis was based on either an intra-operative histopathologic test or cultures from at least two infection sites yielding the same organism or from a sinus tract connecting directly to the bone [11]. Healthy controls without current or previous osteomyelitis were collected from the Physical Examination Center of Nanfang Hospital.

Demographic characteristic information was recorded from the medical records. Those patients were excluded who had a history or presence of serious diseases probably causing telomere attrition, such as diabetic foot, diabetes mellitus, uncontrolled cardiovascular disease, lymphoma, congenital dyskeratosis, autoimmune disease, chronic infections/inflammatory disease, or severe systemic disease or having received immunosuppressive therapy.

Examination of inflammatory indicators

In our hospital, the white blood cell (WBC) count was measured by an automatic blood cell analyzer (Sysmex, XE-2100, Kobe, Japan). The upper limit of normal was 9.50 × 10⁹/L. The erythrocyte sedimentation rate (ESR) was measured using an automatic ESR analyzer (Electa Lab, XC-40B, Forli, Italy), the upper limit value of normal being 15 mm/h in males and 20 mm/h in females. The C-reactive protein (CRP) concentration was determined using an automatic biochemical analyser (Beckman AU5421/AU5431, Atlanta, GA, USA). The upper limit of normal was 5 mg/L.

DNA extraction and quality control

Peripheral blood samples (2 mL of each) were collected from all subjects using Ethylene Diamine Tetraacetic Acid (EDTA) anticoagulative tubes and quickly stored at −80°C. A commercial DNA extraction kit (Qia Amp Mini Kit) was applied to extract DNA from the thawed whole blood samples according to the manufacturer's instruction. The DNA quantity and quality were checked using a NanoDrop ND-1000 spectrophotometer (NanoDrop Technologies, Wilmington, DE, USA) and electrophoretic trace. The A260/A280 nm wavelength ratio was set in the range of 1.7–2.0 to ensure high purity of the extracted DNA. A single strip in an electrophoretic trace was regarded as satisfied integrity for each sample. After extraction, the DNA sample was stored at −20°C until further investigation.

Telomere length measurement

The relative PLTL was determined by the relative ratio of telomere repeat copy number to single copy number (T/S) using an established and validated monochrome multiplex quantitative polymerase chain reaction (PCR) protocol described by Cawthon et al. [12]. The primers were telg (ACACTAAGGTTTGGGTTTGGGTTTGGGTTTGGGTTAGTGT) and telc (TGTTAGGTATCCCTATCCCTATCCCTATCCCTATCCCTAACA). As for the single-copy reference gene, namely albumin, the primers were albu (CGGCGGCGGGCGGCGCGGGCTGGGCGGaaatgctgcacagaatccttg) and albd (GCCCGGCCCGCCGCGCCCGTCCCGCCGgaaaagcatggtcgcctgtt).

Each PCR well contained a total volume of 20 mcL. The reagent components and final amounts were 900 nM for each primer (generay, Shanghai), approximately 15 ng of genomic DNA, and 10 mcL of DNA Master SYBR Green kit (Bestar^® HRM Mastermix). Each plate held four concentrations of a reference DNA sample (the “Standard DNA”) spanning a 27-fold range of DNA concentrations to generate two standard curves. The PCRs were performed using the MyiQ Single Color Real-Time PCR Detection System (Bio-Rad) with the following settings: one cycle of 15 minutes at 95°C; two cycles of 15 seconds at 94°C and 15 seconds at 49°C; and 32 cycles of 15 seconds at 94°C, 10 seconds at 62°C, and 15 seconds at 74°C with signal acquisition, 10 seconds at 84°C, and 15 seconds at 88°C with signal acquisition. Analyses were performed automatically using CFX Manager software (Bio-Rad; Hercules, CA USA). Samples were run in triplicate. Standard curve efficiency for both primers was set at above 90%, and regression linearity R² >0.98 was accepted. Samples outside this range were excluded and run again. The acceptable standard deviation was set at 0.25 (for cycle threshold values). If at least two of the triplicates for both the single gene and telomeres differed <10%, the run was accepted, and the average of the two closest T/S values was used. Otherwise, analysis of the sample had to be repeated until satisfactory.

Statistical analysis

Data were examined for normality using the Kolmogorov Smirnov test. Parametric variables were analyzed by parametric tests; otherwise, non-parametric tests were used. Descriptive statistics, namely, mean ± standard deviation (SD), 95% confidence interval (CI), median, and inter-quartile range (IQR) were applied as appropriate. Comparisons of PLTL between patients and healthy control subjects were made using the Student t-test. Factors associated with PLTL were analyzed using binary logistic regression analysis; the association between the PLTL and the risk of COM was examined utilizing unconditional logistic regression, adjusting for age and gender, to calculate the adjusted odds ratio (OR) and 95% CI. All the statistical analyses were performed with the International Business Machines Corporation Statistical Package for the Social Sciences (IBM SPSS Statistics, New York, USA), version 21.0. The criterion for statistical significance was p < 0.05.

Results

Subject characteristics

This study recruited 56 patients with post-traumatic COM (48 males and 8 females) and 62 generally matched healthy control subjects (54 males and 8 females). Table 1 presents the characteristics of the subjects. The mean age of the patients was 41.13 ± 12.83 years and that of the control subjects was 40.95 ± 12.97 years (p = 0.942). The median duration of illness was 15 months (IQR 4, 36), and the most frequent single infection site was the tibia (42.9%).

Table 1.

Clinical Characteristics of 56 Patients with Chronic Post-Traumatic Extremity Osteomyelitis

Variable	Findings
Mean age (y) (±SD)	41.13 ± 12.83
Median illness duration (mos) (IQR）	15 (4, 36)
Infection site distribution (%)
Femur	19 (33.9)
Tibia	24 (42.9)
Foot	9 (16.1)
Other site	4 (7.1)
Inflammation index
Mean WBC ( × 10⁹/L) (±SD)	7.37 ± 1.98
Median CRP (mg/L) (IQR)	4.89 (1.39, 11.86)
Median ESR (mm/h) (IQR)	16.50 (6.25, 39.75)

CRP = C-reactive protein; ESR: = erythrocyte sedimentation rate; IQR = interquartile range; SD = standard deviation; WBC = white blood cells.

Association between PLTL and COM

Figure 1 illustrates the PLTL values in the two groups. On average, a significantly shorter PLTL was found in patients with post-traumatic COM (5.39 ± 0.40) than in healthy control subjects (5.69 ± 0.46; p < 0.001). The patients were further assigned to long PLTL and short PLTL groups according to the mean PLTL value of the healthy control subjects as a reference; a considerably higher proportion of patients was found in the short group than in the long group (82.1% vs. 17.9%).

FIG. 1.

Comparison of relative leukocyte telomere length (T/S ratio) (mean ± SD) between patients with chronic post-traumatic extremity osteomyelitis and healthy subjects.

To evaluate the association between PLTL and the risk of COM, we performed an unconditional logistic regression analysis adjusting for age and gender, and we observed a significantly higher risk of post-traumatic COM for individuals with a shorter PLTL (adjusted OR 4.386; 95% CI 1.961–9.804; p < 0.001) (Table 2).

Table 2.

Sensitivity and Specificity Using Mean Leukocyte Telomere Length of Healthy Subjects

Telomere Length Category	Cases (%)	Controls (%)	Sensitivity (%)	Specificity (%)
Long	10 (17.9)	33 (53.2)		76.7
Short	46 (82.1)	29 (46.8)	61.3

Association of PLTL with clinical characteristics

Factors attributed to PLTL in COM were revealed by binary logistic regression analysis. The dependent variable was PLTL, which was set as a binary variable with the mean value of all patients as a reference. The independent variables were sex, age, WBC count, ESR, CRP, and duration of illness, among which CRP, ESR, and duration of illness were ordinal data, sex was nominal data, and age and WBC count were scale data. In the patients, a significant association of PLTL was revealed with age (B = −0.072; p = 0.013) and CRP concentration (B = −0.061; p = 0.033), but not with gender, WBC count, ESR, or duration of illness (Table 3). The Nagelkerke R² value of the logistic regression model was 31.4%, and the overall corrected percentage of predication was 69.6%.

Table 3.

Binary Logistic Regression Results on Telomere Length and Its Influencing Factors

	B	SE	Wald	Significance	OR^a	95% CI
Sex^b	−0.956	0.950	1.013	0.314	0.384	0.060–2.473
Age	−0.072	0.029	6.140	0.013	0.930	0.879–0.985
WBC	−0.179	0.179	0.998	0.318	0.836	0.588–1.188
ESR	0.033	0.028	1.438	0.231	1.034	0.979–1.092
CRP	−0.061	0.029	4.553	0.033	0.941	0.889–0.995
Duration^c	0.002	0.020	0.007	0.935	1.002	0.963–1.041
Constant	5.932	2.546	5.428	0.020	377.065

OR: telomere length as dependent variable; binary variable: 1 = short (<5.39); 2 = long (>5.39).

Sex: binary variable: 1 = male, 2 = female.

Duration: binary variable: 1 = short (<15 mos); 2 = long (>15 mos).

B = regression coefficient; CI = confidence interval; OR = odds ratio; SE = standard error.

Discussion

For the first time, we compared the PLTL between post-traumatic COM and healthy control subjects; we found an association between post-traumatic COM and PLTL, and we determined that age and CRP concentration were risk factors for telomere erosion.

This study also revealed a significantly shorter PLTL in patients with post-traumatic COM compared with their healthy counterparts, which may be secondary to the chronic inflammation. One potential mechanism is the resultant increase in oxidative stress. Post-traumatic extremity osteomyelitis would relate to the increased amounts of oxidants and decreased concentrations of serum antioxidants [14], which may contribute to site-specific DNA damage at the telomeric GGG sequence [14] and a reduced ability to combine shelterin protein with the telomere [15], consequently accelerating telomere shortening. In addition, as an inflammation process, post-traumatic extremity osteomyelitis may indirectly affect PLTL by regulating the production of the telomere shelter in complex proteins or telomerase-associated components [16], including telomerase reverse transcriptase (TERT) and telomerase RNA component (TERC). As classic signal pathways of inflammation, NF-κB [17] and Wnt [18] have recently been found to participate in accelerating telomere attrition through regulating the expression of TERT. As a feed-forward mechanism, telomere dysfunction accelerates apoptosis, which in turn intensifies the inflammatory reaction and accelerates telomere erosion [16].

Using logistic regression analysis, we built a regression model and found that age and CRP concentration are risk factors for telomere erosion in chronic post-traumatic extremity osteomyelitis. As aging can cause PLTL erosion physiologically [19], it is logical that age showed statistical significance in this model. A statistically significant p value of CRP was found, although the influence was only slight. The model revealed that telomere length decreased as the CRP concentration increased, thus illustrating an inverse relation between telomere length and inflammation that was in accordance with previous studies [20,21]. The regression coefficient value of CRP was −0.061, a little lower than that of age (−0.072), which means the influence of CRP was lower than that of physiological aging. The logistic regression model was statistically significant but explained only 31.4% of the change in telomere length, and the overall correct percentage of predication was 69.6%. As some factors influence PLTL in osteomyelitis that were not tested in our study, more participants and more associated variables need to be examined. In this regression model, WBC count and ESR had no statistical impact on telomere length. Discrepancies regarding the association between telomere length and WBC count can be explained primarily by the variety of subset cells studied (e.g., mononuclear cells, hematopoietic cells, and lymphocytes) [20,21]), but no exact mechanism has been proved. As for ESR, it is a non-specific measure of inflammation governed by the balance between pro-sedimentation factors and factors resisting sedimentation [22]. Few studies focused on the correlation between ESR and telomere, and no correlation was found between ESR and telomere length [23].

Contrary to our expectation, the duration of COM in this model showed no statistically significant association with telomere length. As telomere erosion is a result of inflammatory accumulation [24], we expected patients with longer durations of disease to have shorter telomeres. However, further analysis showed that patients with longer duration manifested erosion only slightly most of the time, whereas those with shorter duration manifested erosion more seriously. That is to say, telomere erosion depends on the accumulation of inflammation, not only on the duration or severity of disease. As commonly accepted, females have longer telomeres than their peer males [25]. In this logistic regression model, the telomere length of females was greater than in males but with no statistical significance; this may be because of the small size of the female population.

This study has several limitations. First, only a small number of patients and controls were enrolled, which weakens our conclusions to a certain degree. Second, many inflammatory biomarkers were used to diagnose COM, such as IL-6, serum amyloid A, procalcitonin, and tumor necrosis factor, so we chose only WBC count, ESR, and CRP, partially because their analysis is more economical for patients, and partially because this combination was most accepted to diagnose COM in previous studies [26]. Third, as mentioned above, post-traumatic COM is affected by complex factors that can have a greater or lesser effect on telomere length. However, because of missing values or some other limitations, we could not record all these factors; thus, future studies should better take these into account.

Conclusions

This study revealed that patients with post-traumatic COM have shorter PLTLs than healthy controls and that the PLTL erosion of post-traumatic COM was partially explained by age and CRP.

Footnotes

Acknowledgments

This work was supported by Natural Science Foundation of China (81572219, 81871848).

Author Disclosure Statement

We hereby declare that none of the authors has any competing interests concerning this study.

References

Lew

, Waldvogel

. Osteomyelitis. Lancet, 2004; 364:369–379.

Lin

, Lin

, Tseng

, et al. Association between chronic osteomyelitis and risk of end-stage renal disease. Medicine, 2015; 94:e1141.

Tseng

, Chen

, Muo

, et al. Increased risk of ischaemic stroke amongst patients with chronic osteomyelitis: A population-based cohort study in Taiwan. Eur J Neurol, 2015; 4:633–639.

Hsiao

, Muo

, Chen

, et al. Increased risk of coronary heart disease in patients with chronic osteomyelitis: A population-based study in a cohort of 23 million. Heart, 2014; 18:1450–1454.

Tseng

, Huang

, Muo

, et al. Increased depression risk among patients with chronic osteomyelitis. J Psychosom Res, 2014; 6:535–540.

Huang

, Tsai

, Weng

, et al. Chronic osteomyelitis increases long-term mortality risk in the elderly: A nationwide population-based cohort study. BMC Geriatr, 2016; 16:72.

Shay

, Wright

. Telomeres and telomerase: Three decades of progress. Nat Rev Genet, 2019; 5:299–309.

DiLoreto

, Murphy

. The cell biology of aging. Mol Biol Cell, 2015; 26:4524–4531.

Blackburn

, Epel

, Lin

. Human telomere biology: A contributory and interactive factor in aging, disease risks, and protection. Science, 2015; 6265:1193–1198.

10.

Trampuz

, Zimmerli

. Diagnosis and treatment of infections associated with fracture-fixation devices. Injury, 2006; 37:S59–S66.

11.

Jiang

, Zhao

, Qin

, et al. Association of vitamin d receptor gene TaqI, BsmI, FokI and ApaI polymorphisms and susceptibility to extremity chronic osteomyelitis in Chinese population. Injury, 2016; 8:1655–1660.

12.

Cawthon

. Telomere length measurement by a novel monochrome multiplex quantitative PCR method. Nucleic Acids Res, 2009; 37:e21.

13.

Jyoti

, Singh

, Mukhopadhyay

, et al. Free radicals and antioxidant status in chronic osteomyelitis patients: A case control study. J Clin Diagn Res, 2015; 9(4):BC08–BC10.

14.

Kawanishi

, Oikawa

. Mechanism of telomere shortening by oxidative stress. Ann N Y Acad Sci, 2004; 1019:278–284.

15.

Sun

, Tan

, Xu

, et al. Targeted DNA damage at individual telomeres disrupts their integrity and triggers cell death. Nucleic Acids Res, 2015; 43:6334–6347.

16.

Zhang

, Rane

, Dai

, et al. Ageing and the telomere connection: an intimate relationship with inflammation. Age Res Rev, 2016; 25:55–69.

17.

Ghosh

, Saginc

, Leow

, et al. Telomerase directly regulates NF-kappaB-dependent transcription. Nat Cell Biol, 2012; 14:1270–1281.

18.

Diala

, Wagner

, Magdinier

, et al. Telomere protection and TRF2 expression are enhanced by the canonical Wnt signalling pathway. EMBO Rep, 2013; 14:356–363.

19.

Rizvi

, Raza

, Mahdi

. Telomere length variations in aging and age-related diseases. Curr Aging Sci, 2014; 7:161–167.

20.

Shin

, Shin

, Lee

. Effects of inflammation and depression on telomere length in young adults in the United States. J Clin Med, 2019; 8:E711.

21.

Flannagan

, Bowman

, Mora-Plazas

, et al. Micronutrient status and leukocyte telomere length in school-age Colombian children. Eur J Nutr, 2019; doi 10.1007/s00394-019-01968 Epub ahead of print.

22.

Harrison

. Erythrocyte sedimentation rate and C-reactive protein. Aust Prescr, 2015; 38:93–94.

23.

Svenson

, Ljungberg

, Roos

. Telomere length in peripheral blood predicts survival in clear cell renal cell carcinoma. Cancer Res, 2009; 69:2896–2901.

24.

O'Donovan

, Pantell

, Puterman

, et al. Cumulative inflammatory load is associated with short leukocyte telomere length in the health, aging and body composition study. PLoS One, 2011; 6(5):13.

25.

Gardner

, Bann

, Wiley

, et al. Gender and telomere length: Systematic review and meta-analysis. Exp Gerontol, 2014; 51:15–27.

26.

Stucken

, Olszewski

, Creevy

, et al. Preoperative diagnosis of infection in patients with nonunions. J Bone Joint Surg Am, 2013; 95:1409–1412.