Introduction

Hip fracture in older people is a common event that implies a high morbidity and mortality1, and would also involve a high socioeconomic healthcare cost2. Diverse risk factors regarding mortality of hip fracture patients, both early and long-term mortality, have already been described3,4,5.

The factors influencing the outcome of older hip fracture patients could be divided in two main groups: extrinsic factors, such as those related to the care provided, and patient-intrinsic factors, such as patient’s age or comorbidity. Among the intrinsic risk factors would also be genetic determinants.

Some studies have analyzed the relationship of certain genetic determinants with the risk of suffering a fragility fracture6. Indeed, many different genetic polymorphisms have been analyzed for frailty assessment7, and some genetic polymorphisms have been related to the bone density and turnover8,9,10. Many genetic factors have been related not only with hip fracture risk, but also on the skeletal architecture11. However, to our knowledge, there are no previous studies analyzing the influence of genetic factors on the functional status of hip fracture patients.

Autophagy is a highly regulated physiological process involved in the turnover of the cellular components, also involved in bone homeostasis and metabolism12,13. It has also been related to bone disease14 and it has been associated to osteoporosis onset15. Many aging-related diseases would be determined by the autophagy activity16, as the stimulation of autophagy would have antiaging effects17,18. Autophagy-related genes (ATG) regulates the cellular catabolic process responsible for the degradation and elimination of damaged organelles, cytoplasmic proteins, and protein aggregates19.

Several studies have linked single nucleotide polymorphisms (SNPs) on ATG genes to some pathologies20,21, but the relationship between autophagy and hip fracture has not been yet explored.

We have selected polymorphisms from ATG genes involved in autophagosome generation, that have been previously reported in the literature. Initially, we selected non-synonym polymorphisms with a population frequency of the minor allele higher than 10% in Caucasian population and that were located in sequences highly conserved throughout the evolution. ATG2B rs3759601, ATG16L1 rs2241880 and ATG10 rs1864183 polymorphisms, which are missense mutations.

Older patients who suffer a hip fracture usually present a compromised clinical situation associated with significant comorbidity. One of the aspects that can influence the prognosis of older patients with hip fracture is their previous functional status. There seems to be a relationship between worse functional status and the worst prognosis3,22. The functional status of older adult patients with hip fracture is usually established by means of the so-called “geriatric scores”, which are part of the initial evaluation strategy of these patients as a part of the comprehensive geriatric assesment. The most commonly used geriatric scores are the Barthel index (BI), Katz index (KI), Lawton-Brody index (LBI), and Physical Red Cross Scale (PRCS).

The aim of the present study was to analyze the relationship of three SNPs on ATG2B, ATG10 and ATG16L1 with the functional status of older hip fracture patients established through the analysis of geriatric scores.

Materials and methods

Design and population

We designed a prospective observational study including 87 hip fracture patients admitted to the Orthogeriatric Unit of the University Hospital of Salamanca. Inclusion criteria included patients aged 80-year-old or older, presenting an osteoporotic hip fracture treated surgically, and signing an informed consent. Exclusion criteria excluded patients presenting a pathologic fracture of the hip caused by neoplasic activity, a high-energy-trauma-caused hip fracture, or antecedent of bone marrow transplant. Patients under 80 years of age, not surgically treated or who did not sign the informed consent were also excluded from the study.

The sample size calculation was done through an a priori power analysis, with the aim of estimating the optimal sample size for determining significant clinical effects. Kendall Correlations test was used as it is the main contrast on which this research is based. A Tau-c value of 0.3 was considered (for being a moderate effect size, which represents a conservative threshold to detect a significant association), alpha was set on 0.05 and beta at 0.20, for a power level of 0.80. It resulted in an optimal sample size calculation of 84 subjects.

The whole study was conducted following the principles of the Declaration of Helsinki, and IRB Comité Ético del Área de Salud de Salamanca previously approved the study (ref. code: PI 2020 01 418).

Study protocol

On hospital admission, a clinical and radiological examination of the patient is carried out, checking inclusion/exclusion criteria. On admission to the ward, an initial blood test is taken, and the comprehensive geriatric assessment (CGA) and anesthetic evaluation are carried out.

During the hospital stay, surgical treatment is carried out. The usual hemacytometric controls will be used to extract an additional 5 ml of blood for the genetic analysis of the patient samples.

Variables

Based on the study of the clinical records and patients’ samples, the following groups of variables will be analyzed: Biodemographic variables are including, but not limited to gender, age, and place of residence of the patient; Clinical variables are those derived from the analysis of the clinical records of the patients, regarding comorbidities; Radiological variables describe the type of fracture; Functional status variables are derived from the comprehensive geriatric assessment performed during the in-hospital process, and are including, Katz index, Barthel index, Lawton-Brody index, and Physical Red Cross Scale; Genetic variables are addressing the studied SNPs (Table 1).

Table 1 Autophagy polymorphisms analyzed in the study.
Full size table

DNA isolation and polymorphism genotyping

Genomic DNA was extracted from peripheral blood by standard phenol/chloroform procedure. Genotyping of the three SNPs on ATG2B, ATG10 and ATG16L1 (Table 1) was performed using TaqMan® 5′-exonuclease allelic discrimination assays (Assay IDs: C___9690166_10, C__11953871_20, C___9095577_20). PCR reactions were carried out using TaqMan® universal PCR Maxter Mix following instructions in a Step-One Plus Real-time PCR system (ThermoFisher Scientific). To assess reproducibility, a random selected 5% of the samples were re-genotyped, all of these genotypes matched with genotypes initially designated.

Comprehensive geriatric asessment (CGA)

Geriatric scores

Four geriatric scores (Barthel index, Katz index, Lawton-Brody index and Physical Red Cros scale) are used as part of the comprehensive geriatric assessment of older patients admitted for hip fracture in our center. The characteristics and description of the geriatric scores have been described in a previous work of our research group22.

The Barthel index (BI) analyzes the ability to perform activities of daily living. It can collect a score between 0 (total dependence) and 100 (independence), classifying patients in the following categories: total dependence, severe dependence, moderate dependence, low dependence, and independent patients. The Katz index (KI) also evaluates the ability to perform activities of daily living in 6 aspects and classifies patients into seven categories labeled with letters from A to G, where A reflects a situation of functional independence and G that of total dependence. The Lawton-Brody index (LBI) assesses the ability to perform instrumental activities of daily living. Some of these activities are conditioned by cultural factors, for which this index has been criticized. The Physical Red Cross scale (PRCS) is a score developed in Spain that evaluates the mobility capacity of patients. In our center, we have removed the cognitive function component. It establishes 5 levels of ambulatory capacity, where 0 is a total capacity for autonomous mobility and 5 is a total disability.

Statistical analysis

The statistical analysis of the results was performed using IBM® SPSS® Statistics 23. The sample size calculation was conducted using GPower Software, version 3.1.9.7.

Genotypes distribution followed by Hardly Weinberg equilibrium (p > 0.05). Quantitative variables were expressed by mean and SD and were analyzed by non-parametric test. Qualitative variables were analyzed by Kendall’s Tau-C test for ordinal variables. In all cases, a p-value < 0.05 was considered statistically significant.

Results

Of the 87 patients initially included in the study, sufficient data could only be obtained in 84 patients who were enrolled in the study (69.0% women). The mean age of the population was 87.6 ± 4.2 years. The mean length of stay for the studied patients was 8.6 ± 3.3 days and the mean time to surgery was 2.7 ± 1.7 days. A descriptive resume of biodemographic, radiological and functional variables of the population is exposed in Table 2.

Table 2 Characteristics of the population.
Full size table

Attending to the genotyping of patients, the distribution of studied genotypes and allelotypes are resumed in Table 3.

Table 3 Genotyping distribution on the study population.
Full size table

ATG2B rs3759601 C > G

We analyzed the geriatric scores as ordinal scales for measuring the functional status of patients. There are statistically significant differences in the distribution of the functional status of patients and ATG2B genotypes. The GG genotype would present a better functional status, compared to those patients presenting the C allelotype, which is associated with values in the geriatric scores that indicate a worse functional status of the patients.

When analyzing the ATG2B rs3759601 genotype in relation to the functional situation determined by BI (Barthel index), we found a clear functional superiority of patients with GG genotype (Table 4, Kendall’s Tau-c p-value on BI: 0.008).

Table 4 Genotyping distribution of the study population regarding the patient’s Barthel index.
Full size table

When analyzing this functional superiority by allelotype, however, we found that the significant difference would be caused by the presence of the C allele, which would determine a worse functional situation, despite not showing extreme values of dependence (Kendall’s Tau-c p-value for C allelotype on BI: 0.002).

When analyzing the genotype of the ATG2B rs3759601 in relation to the functional status determined by the KI (Katz index), we again found a clear functional superiority of patients with GG genotype (Table 5, Kendall’s Tau-c p-value on KI: 0.002). This functional superiority would again be caused by the presence of the C allele, which would determine a worse functional situation (Kendall’s Tau-c p-value for C allelotype on KI: 0.001).

Table 5 Genotyping distribution of the study population regarding the patient’s Katz index.
Full size table

The functional inferiority of the CG genotype is exposed even more clearly when the results are biased by the previously determined inflexion point between the A/B categories of the IK versus the rest23, finding frequencies of good functional status of 62.9% in patients with GG genotype, compared to 31.7% of patients with good functional status and CG genotype (Chi-square p-value: 0.022).

When analyzing the genotype of the ATG2B rs3759601 in relation to the functional status determined by the LBI (Lawton–Brody Index), the previous results are corroborated. We observed that CG and CC genotypes, present a worse functional situation (lower IL categories) than patients with GG genotype (Table 6, Kendall’s Tau-c p-value on LBI: 0.001 / Kendall’s Tau-c p-value for C allelotype on LBI: 0.001).

Table 6 Genotyping distribution of the study population regarding the patient’s Lawton–Brody index.
Full size table

When analyzing the genotype of the ATG2B rs3759601 in relation to functional status in terms of mobility as determined by PRCS (Red Cross scale), we again find similar results. We observe that the CG genotype presents the worst mobility (Table 7, Kendall’s Tau-c p-value on PRCS: <0.001/Kendall’s Tau-c p-value for C allelotype on PRCS: 0.016).

Table 7 Genotyping distribution of the study population regarding the patient’s physical red cross scale (PRCS).
Full size table

ATG10 rs1864183 C > T

When analyzing the genotype of the ATG10 rs1864183 in relation to the functional situation determined by the BI, we found a clear functional inferiority of patients with CC genotype (Table 4, Kendall’s Tau-c p-value on BI: 0.011).

When analyzing this functional inferiority by allelotype of the ATG10 rs1864183, we found that the significant difference would be caused by the presence of the T allele, which would determine a better functional situation, avoiding the development of extreme values of dependence (Kendall’s Tau-c p-value for T allelotype on BI: 0.041).

When analyzing the genotype of the ATG10 rs1864183 in relation to the functional situation determined by the KI, we again found a clear functional inferiority of patients with CC genotype (Table 5, Kendall’s Tau-c p-value on KI: 0.017). This functional inferiority would be justified by the absence of the T allele, which would determine a better functional situation (Kendall’s Tau-c p-value for T allelotype on KI: 0.022).

When analyzing the genotype of the ATG10 rs1864183 in relation to the functional situation determined by the LBI, we again found statistically significant differences in the distribution of genotypes (Table 6, Kendall’s Tau-c p-value on LBI: 0.025), but not by allelotype (p > 0.05). Patients with CC genotype would present a higher incidence of severe dependence for instrumental activities of daily living (lower LBI categories). The result of the analysis of the functional status by means of Lawton’s index can be modified by cultural issues, since, for example, it analyzes the performance of activities that in some environments are more frequently developed by a specific gender.

When analyzing the genotype of the ATG10 rs1864183 in relation to the functional status in terms of mobility as determined by PRCS, we again find similar results. We observe that the CC genotype presents the worst mobility, while patients with T allelotype present better mobility (Table 7, Kendall’s Tau-c p-value on PRCS: 0.009 / Kendall’s Tau-c p-value for T allelotype on PRCS: 0.013).

ATG16L1 rs2241880 A > G

In the genotypic analysis of ATG16L1 rs2241880 in relation to functional status as determined by geriatric scores measured in the CGA, we found no statistically significant difference in the distribution of these SNPs or in the analysis by allelotype.

Discusion

The most important finding of this work is that it shows the existence of a relationship between some autophagy genes polymorphisms and the functional status of older patients presenting a hip fracture. Hip fracture in the older population remains one of the most important health problems in terms of morbi-mortality24. Studies on the relationship between genetic factors and hip fracture have focused mainly on aspects linked to the risk of suffering a fracture, bone quality and fragility6,25,26. It is known how there are genetic factors that predispose to hip fracture; accordingly, we already know that there is an elevated risk of hip fracture in women whose mothers also suffered it. Few studies have related certain genetic factors to the prognosis of older old patients with hip fracture.

The processes known as autophagy encompass various intracellular processes that involve the elimination of cellular debris. Several genes are known that regulate these processes and whose polymorphisms have been related to some pathologies. It has been considered that autophagy processes could have a protective effect against some pathologies such as infection, cancer, neurodegenerative pathologies, as well as against aging16,19. Until now and to our knowledge, these genes and their variations had not been related to the situation and therefore the prognosis of patients with hip fracture. We have analyzed three genes as they are the ones for which more detailed information is available in the literature for the purpose of this work.

Genetic studies using yeasts, worms, flies and mice demonstrate the ATG genes are required for lifespan extension requirement in caloric restriction, loss-of-function insulin signaling and other conserved longevity paradigms (reviewed in27. Systemic autophagy induction exerts anti-aging effects in worms, flies and mice, and genetically engineered mice with constitutively increased autophagy have extended lifespan and improved healthspan28. Interestingly, the offspring of people with exceptional longevity have enhanced activation-induced T cell autophagy and immune function compared to age-matched controls29. Autophagy may prevent aging through improved organellar quality control and homeostasis (e.g. via selective autophagy pathways such as mitophagy, lipophagy, lysophagy, aggrephagy), enhanced insulin sensitivity, maintenance of stemness and promotion of genomic stability. Interestingly, tissue-specific autophagy in certain tissues (e.g. in the muscle, intestine and brain) may also exert favorable effects on longevity, potentially by modulating a range of inter-tissue interactions27. Moreover, it is possible that ATG genes may have autophagy-independent effects that promote longevity; for example, their roles in secretion and exocytosis might contribute to inter-tissue effects.

Autophagy gene expression and lysosomal function decline with aging in different tissues in worms, flies and mammals (including human brains), resulted in an age-related decline in autophagic capacity27. This age-related decline contributes both to the aging process itself as well as the development of age-related diseases.

Atg2B is essential for autophagosome formation. Atg2B extracts phospholipids from the membrane source and transfers them to Atg9 (ATG9A or ATG9B) for membrane expansion. ATG2B rs3759601 reflects a C > G transversion in exon 5 that produces a glutamine-to-glutamic acid change in position 1383 (p.Gln1383Glu) and has been associated to pharyngeal cancer and glioblastoma30,31. Moreover, it has been reported that monocytes carrying the G allele show better response to BCG inmunotherapy increasing the levels of IL1b and TNFa32. Our results, on the other hand, showed a worse functional status of patients with the C allele; however, it would be necessary to further investigate the biological mechanisms underlying this SNP in order to definitively establish a relationship between the activation of the immune response and the presence of the C allele.

Atg10 is an essential E2-like enzyme that mediates the formation of Atg12-Atg5 conjugate. Nonsynonymous ATG10 variant rs1864183 encodes a threonine-to-methionine change at codon 212 (p.Thr212Met). It has been predicted to be located at exonic splicing enhancers (ESEs) and has been proposed to lead to the catalytic change of the protein, causing a dysregulation of autophagosome formation. Previous study of this polymorphism in patients with Paget disease of bone showed that being a carrier of the T allele of ATG10 rs1864183 polymorphism decreased the risk of suffering this disease33. Considering previous studies and our results, the presence of the T allele, associated with a better functional state, would imply a lower functionality of the Atg10 protein. This is somewhat controversial, since we could expect that a higher degree of autophagy activation would allow a better handling of waste products. Disabling one of the steps in the autophagic pathway would not have led us to believe that it would allow a higher functional state. This small biological controversy should be studied further in order to determine with greater accuracy the consequences of the presence of this polymorphism at the organic-functional level.

Atg16L1 plays an essential role in autophagy through interactions with Atg12-Atg5 to mediate the lipidation to Atg8 family proteins. It has been demonstrated that ATG16L1 rs2241880 increased Atg16L1 sensitivity to caspase-3-mediated processing, resulting in diminished of autophagy. Also, in a previous study in patients with Paget disease of bone, the homozygous CC genotype was associated with an increased risk of developing the disease33.

Despite, in addition to the clinical situation, the functional status of hip fracture patients is relevant to their prognosis. The functional status is usually determined by using the so-called geriatric scores. In our center their collection is a part of the so-called comprehensive geriatric assessment (CGA). We have previously described how there is a relationship between these scores and the outcome of surgical treatment of this type of lesion and how there are also inflection points in these scores that more accurately determine this prognosis. Geriatric scores are a widely used and validated assessment tool. These indexes analyze each individual’s ability to perform daily activities as well as the ability to move independently. These are factors closely related to patients’ quality of life. This is especially relevant when we consider patients as old as those in the sample.

The findings of this work seem to indicate that there is a certain genetic conditioning with respect to functional status which, as we have pointed out, is related to the treatment outcome of older hip fracture patients. Particularly interesting are the findings that show the relationship of some polymorphisms with the scores that indicate a worse functional status and consequently a worse prognosis. ATG10 rs1864183 T allele would be associated to a better functional status and prognosis of hip fracture patients, while CC genotype would be associated to the worse functional status of the patient. On the other hand, ATG2B rs3759601 C allele would determine a worse functional situation of hip fracture patients. Both conclusions are in agreement with the cited previous works, so the genetic polymorphisms here studied could affect the prognosis of studied patients.

The present work has some limitations, such as the analysis of a relatively small sample of patients and a limited number of genes, but it should be remembered that it is an initial study on the relationship between autophagy processes and hip fracture.

We do not currently know which mechanisms explain the relationship between autophagy genes and the functional status of older patients, although it is thought to be related to the relationship between autophagy processes and the state of fragility. Dysregulated autophagy, impairing removal of dysfunctional proteins and organelles, is suggested as one of the underlying mechanisms that some authors have pointed out about aging, both the process of autophagosome formation and autophagosome-lysosome fusion would be affected in frail older population34.

Conclusion

ATG10 rs1864183 and ATG2B rs3759601 are significantly related to the functional status of older hip fracture patients. The autophagy process is related to the functional status of older hip fracture patients and consequently to their prognosis and would be needed more wide studies for further elucidating the complete underlying mechanism of this association.