Single Nucleotide Polymorphisms in Osteogenic Genes in Atrophic Delayed Fracture-Healing: A Preliminary Investigation

Single Nucleotide Polymorphisms in OsteogenicGenes in Atrophic Delayed Fracture-Healing

A Preliminary Investigation

Vikram Sathyendra, MD, Henry J. Donahue, PhD, Kent E. Vrana, PhD, Arthur Berg, PhD,David Fryzel, MD, Jonathan Gandhi, MD, and J. Spence Reid, MD

Investigation performed at the Departments of Orthopaedics and Rehabilitation, Pharmacology, and Public Health Sciences,Pennsylvania State University College of Medicine, Milton S. Hershey Medical Center, Hershey, Pennsylvania

Background: We propose that fracture-healing potential is affected by the patient’s genome. This genotype is thenphenotypically expressed by the patient at the time of injury. We examined the hypothesis that patients who exhibitdelayed or impaired fracture-healing may have one or more single nucleotide polymorphisms (SNPs) within a series ofgenes related to bone formation.

Methods: We performed a population-based, case-controlled study of delayed fracture-healing. Sixty-two adults with along-bone fracture were identified from a surgical database. Thirty-three patients had an atrophic nonunion (delayedhealing), and twenty-nine displayed normal fracture-healing. These patients underwent buccal mucosal cell harvesting.SNP genotyping was performed with use of bead array technology. One hundred and forty-four SNPs (selected fromHapMap) within thirty genes associated with fracture-healing were investigated. Three SNPs did not segregate in thepopulation and were excluded from the analysis. Eight of the remaining SNPs failed the test for Hardy-Weinberg equilibrium(p value smaller than the Bonferroni-corrected level of 0.05/141 = 0.000355) and were excluded.

Results: Five SNPs on four genes were found to have a p value of <0.05 in the additive genetic model. Of these five significantSNPs, three had an odds ratio (OR) of >1, indicating that the presence of the allele increased the risk of nonunion. Thers2853550 SNP, which had the largest effect (OR = 5.9, p = 0.034), was on the IL1B gene, which codes for interleukin 1 beta.The rs2297514 SNP (OR = 3.98, p = 0.015) and the rs2248814 SNP (OR = 2.27, p = 0.038) were on the NOS2 gene codingfor nitric oxide synthase. The remaining two SNPs had an OR of <1, indicating that the presence of the allele may be protectiveagainst nonunion. The rs3819089 SNP (OR = 0.26, p = 0.026) was on the MMP13 gene for matrix metallopeptidase 13, andthe rs270393 SNP (OR = 0.30, p = 0.015) was on the BMP6 gene for bone morphogenetic protein 6.

Conclusions: Variations in the IL1B and NOS2 genes may contribute to delayed fracture-healing and warrant furtherinvestigation.

Clinical Relevance: Impaired fracture union may have genetic contributions.

Disclosure: One or more of the authors received payments or services,either directly or indirectly (i.e., via his or her institution), from a thirdparty in support of an aspect of this work. None of the authors, or theirinstitution(s), have had any financial relationship, in the thirty-sixmonths prior to submission of this work, with any entity in the bio-medical arena that could be perceived to influence or have the po-tential to influence what is written in this work. In addition, no authorhas had any other relationships, or has engaged in any other activities,that could be perceived to influence or have the potential to influencewhat is written in this work. The complete Disclosures of PotentialConflicts of Interest submitted by authors are always provided withthe online version of the article.

A commentary by Thomas A. Einhorn, MD,is linked to the online version of this articleat jbjs.org.

Peer Review: This article was reviewed by the Editor-in-Chief and one Deputy Editor, and it underwent blinded review by two or more outside experts. The Deputy Editorreviewed each revision of the article, and it underwent a final review by the Editor-in-Chief prior to publication. Final corrections and clarifications occurred during one ormore exchanges between the author(s) and copyeditors.

1242

COPYRIGHT � 2014 BY THE JOURNAL OF BONE AND JOINT SURGERY, INCORPORATED

J Bone Joint Surg Am. 2014;96:1242-8 d http://dx.doi.org/10.2106/JBJS.M.00453

Bone regeneration is a complex process that involves mul-tiple interacting biological mechanisms and is a criticalcomponent of many aspects of musculoskeletal care,

including fracture-healing, spinal fusion, and osseointegrationof implants. It is estimated that nearly 10% of the approximately7.9 million fractures that occur in the United States each year arecomplicated by impaired healing1. The cost of treating a de-layed union or nonunion can be as high as $25,000 per patient(2007 data)2, and a tibial nonunion has a health impact com-parable to that of end-stage hip arthrosis and is more debili-tating than congestive heart failure3. Identifying patients at riskfor impaired fracture-healing and modifying their initial treat-ment strategy might prevent nonunions.

Local and systemic patient factors, as well as surgeon-controlled variables, can affect the union rate of a fracture.Diabetes, smoking, corticosteroid or NSAID (nonsteroidalanti-inflammatory drug) use, vitamin-D deficiency, and othercomorbidities have all been associated with delayed fracture-

healing4-7. We propose that a patient has a fracture-healingpotential that is based, in part, on their genome and that isphenotypically expressed at the time of injury. We examinedthe hypothesis that patients who exhibit delayed or impairedfracture-healing may have one or more single nucleotidepolymorphisms (SNPs) within a set of bone-related genes.These SNPs may affect fracture-healing directly, or they mayinteract with other epigenetic host or environmental factorsto result in delayed fracture union8. Additionally, these SNPsmay be potential biomarkers for associated genetic defectswithin the coding regions of these genes.

Identification of a patient as having a genetic risk ofdelayed or impaired fracture-healing at the time of a fracturemay justify more aggressive initial treatment of the fracture. Ifthe fracture would typically be treated nonsurgically, knowl-edge of the existence of a genetic defect may either tip the scalesin favor of surgical treatment or justify the immediate use ofphysical modalities such as ultrasound or electric field therapy9-12

TABLE I Genes Selected

Category Gene Symbol Official Name Chromosome

Cytokines CSF1 Colony stimulating factor 1 (macrophage) 1p21-p13

IL1B Interleukin 1, beta 2q14

IL6 Interleukin 6 (interferon, beta 2) 7p21

IL11 Interleukin 11 19q13.3-q13.4

TNFSF11 Tumor necrosis factor (ligand) superfamily, member 11 13q14

TNFRSF11B Tumor necrosis factor receptor superfamily, member 11b 8q24

IFN1a Interferon, type 1, cluster 9p22

TNF Tumor necrosis factor 6p21.3

Extracellular matrix COL2A1 Collagen, type II, alpha 1 12q13.11-q13.2

COL1A1 Collagen, type I, alpha 1 17q21.33

Morphogens TGFB1 Transforming growth factor, beta 1 19q13.1

TGFB2 Transforming growth factor, beta 2 1q41

TGFB3 Transforming growth factor, beta 3 14q24

BMP2 Bone morphogenetic protein 2 20p12

BMP4 Bone morphogenetic protein 4 14q22-q23

BMP5 Bone morphogenetic protein 5 6p12.1

BMP6 Bone morphogenetic protein 6 6p24-p23

BMP7 Bone morphogenetic protein 7 20q13

BMP8A Bone morphogenetic protein 8a 1p34.2

MSTN Myostatin 2q32.2

GDF10 Growth and differentiation factor 10 10q11.22

Proteases MMP9 Matrix metallopeptidase 9 20q11.2-q13.1

MMP13 Matrix metallopeptidase 13 11q22.3

Angiogenic VEGFA Vascular endothelial growth factor A 6p12

VEGFC Vascular endothelial growth factor C 4q34.1-q34.3

ANGPT1 Angiopoietin 1 8q22.3-q23

PTN Pleiotrophin 7q33-q34

Others NOS2 Nitric oxide synthase 2A 17q11.2-q12

ADRB2 Beta adrenergic receptor 2 5q

PTGS2 Prostaglandin-endoperoxide synthase 2 (cyclooxygenase 2 [COX2]) 1q25.2-q25.3

1243

TH E J O U R N A L O F B O N E & JO I N T SU R G E RY d J B J S . O R G

VO LU M E 96-A d NU M B E R 15 d AU G U S T 6, 2014SI N G L E NU C L E O T I D E PO LY M O R P H I S M S I N OS T E O G E N I C

GE N E S I N AT R O P H I C DE L AY E D F R AC T U R E -HE A L I N G

or the short-term use of recombinant human parathyroid hor-mone (rhPTH 1-34) supplementation13,14. If the fracture wouldtypically be treated surgically, knowledge of the existence of agenetic defect may justify early bone-grafting or the use of acommercially available bone morphogenetic protein (BMP)product15-19, in conjunction with the nonsurgical modalitiesdescribed above. Such early modulation of treatment may pre-vent a nonunion and the associated delay in return to activities. Agenetic marker for nonunion may also identify patients who aremore likely to experience problems with spinal fusion or os-seointegration of an ingrowth prosthesis. Identification of SNPsassociated with impaired fracture-healing may also provide fur-ther insights into the molecular genetics of bone-healing. In thefuture, this information may yield relatively inexpensive screen-ing studies and contribute to the development of gene-basedtherapy for patients with a particular genetic marker20,21. In thepresent study, we identified five SNPs in four genes that may beassociated with altered fracture-healing.

Materials and Methods

This was a population-based, case-controlled study of delayed fracture-healing. We used our surgical database to select sixty-two adult patients

(age, eighteen to seventy-nine years) who sustained a fracture involving thefemur, tibia, humerus, or ulna from 2005 to 2010. Thirty-three patients with anatrophic nonunion (the nonunion group) and twenty-nine patients with un-eventful fracture union (the healing group) were identified. An atrophic non-union was defined as a fracture that had minimal callus formation six monthsafter injury and required additional surgery to obtain union. In every case,the additional surgery required the use of autogenous bone graft or anotherosteoinductive agent (BMP2 or 722) to augment the defective biology. In somecases, this additional surgery also required revision of internal fixation im-plants. Patients with a Gustilo-Anderson

23grade-II or III open fracture or a

positive bone culture at the time of the nonunion surgery were excluded fromthe study, as these factors likely predispose to nonunion regardless of the geneticpotential for fracture-healing. Patients who had a pathologic or insufficiencyfracture or were undergoing chemotherapy for cancer were also excluded.

Normal healing was defined as radiographic and clinical evidence thatthe fracture had healed by six months without secondary intervention. Sincethe inclusion criteria for both groups was essentially radiographic and clinical,and were therefore somewhat subjective, the radiographs and operative notesfor each patient were reviewed by the senior author (J.S.R.), who is an expe-rienced fracture surgeon, to verify the suitability of the patient to the healingor the nonunion group.

After approval from the institutional review board, suitable patientswere contacted by telephone. Interested patients then presented to the ortho-

paedic clinic to complete the study questionnaire, provide written informedconsent, and undergo harvesting of buccal mucosal cells from the oral cavity.

Isolation of genomic DNA was performed with use of an XIT GenomicDNA from Buccal Cells kit (G-Biosciences). Cells from the oral cavity wereharvested with use of cytology brushes and suspended in XIT Lysis Buffer. Eachvial was initially placed on ice until the end of the clinic session and then storedat 280�C. Genomic DNA was extracted and purified and was then stored at280�C. At the conclusion of the collection period, DNA from the entire cohortwas processed en masse in the Functional Genomics Core Facility at our in-stitution with use of BeadXpress VeraCode (Illumina) SNP analysis

24,25.

A PubMed literature search was performed to identify thirty genes thathad a known or suspected association with fracture-healing (Table I). Theseselected genes can be divided into six groups on the basis of their function. Thecytokine group contains eight genes coding for interleukins and tumor necrosisfactors. The extracellular matrix group contains two genes coding for type-Iand II collagen. The morphogen group contains eleven genes, three of whichcode for transforming growth factor beta proteins and six of which code forBMPs. The protease group contains two genes coding for matrix metal-lopeptidases (MMPs) 9 and 13. The angiogenic group contains four genescoding for vascular endothelial growth factors A and C as well as angiopoietinand pleiotrophin. The final group contains three miscellaneous genes codingfor inducible nitric oxide synthase (NOS2), beta adrenergic receptor 2, andprostaglandin-endoperoxide synthase 2 (coding for cyclooxygenase 2 [COX2]).

A Single Nucleotide Polymorphism (SNP) data file for each gene wasdownloaded from HapMap

26-28. The data were then analyzed with use of

Haploview software (Broad Institute of MIT and Harvard)29

, and 144 SNPsfrom these thirty genes were selected to be genotyped (see Appendix). Thenumber of SNPs examined per gene ranged from two to seven. The SNPswere selected on the basis of their known or suspected association withfracture-healing or bone formation. SNP genotyping was performed with useof Illumina GoldenGate bead array technology

30-32. Binary logistic regression

was used to test the association of patient age and sex with the developmentof a fracture nonunion

33.

Source of FundingThis study was funded by a research grant from the Orthopaedic TraumaAssociation.

Results

Thirty-three patients (fourteen men and nineteen women witha mean age of 48.6 years) developed an atrophic long-bone

TABLE II Patient Groups

Healing Nonunion P Value

No. 29 33

Mean age (yr) 47.3 48.6 0.75

Female 11 (38%) 19 (58%) 0.2

Smoker 0.241 0.348 0.4

Femur 10 (34%) 13 (39%) 0.79

Tibia 15 (52%) 18 (55%) 1

Humerus 4 0

Ulna 0 2

TABLE III SNPs That Failed Hardy-Weinberg Equilibrium andWere Excluded

Genotype (no.)

Gene SNP AA AB BB P Value*

BMP7 rs162316 5 26 31 3.43 · 10215

MSTN rs16832285 0 2 60 3.43 · 10215

BMP8 rs1066945 54 7 0 3.43 · 10215

COL1A1 rs1061970 48 14 0 5.71 · 10215

BMP7 rs1998190 16 33 13 3.2 · 10212

ADRB2 rs1800888 0 1 61 6.9 · 1026

TNF rs3093671 0 4 58 0.000177

IL6 rs1524107 2 4 56 0.000247

*For Hardy-Weinberg equilibrium.

1244




nonunion. The control group consisted of twenty-nine patients(eighteen men and eleven women with a mean age of 47.3 years)who had uneventful fracture union. Simple statistical analyseswith use of the exact binomial test and t test were performed onpatient parameters, including age, sex, smoking, and the longbone fractured (Table II).

SNPs were screened for quality with use of GenCall andGenTrain software (Illumina) to measure the reliability of SNPdetection on the basis of the distribution of genotypic classes34.Only SNPs with a GenCall score of >0.25 and a GenTrain scoreof >0.25 were retained. At this stage, three SNPs (rs3758853 inMMP13, rs1143641 in IL1B, and rs2075554 in COL1A1) didnot segregate in the population and were therefore excludedfrom the analysis. Finally, SNPs were tested for Hardy-Weinbergequilibrium, and eight were excluded because the Hardy-Weinbergp value was smaller than the Bonferroni-corrected level of0.05/141 = 0.00035535-41. In some cases, the exclusion resultedfrom a small minor-allele frequency (Table III). This left 133SNPs for analysis. Logistic regression was used in the additivegenetic model, and five SNPs on four genes were found to besignificant at the 0.05 level (Table IV). An odds ratio (OR) of>1 indicated that the presence of the allele predisposed pa-tients toward developing a nonunion, whereas an OR of <1indicated that the presence of the allele protected patientsfrom developing a nonunion.

Three SNPs were found to have an OR of >1 (Table IV).The presence of the C/T genotype or T allele at SNP rs2853550on the IL1B gene yielded the largest OR, 5.9 (p = 0.034). Thus,

the presence of the T allele at this SNP increased the risk ofnonunion approximately sixfold. Two SNPs on the NOS2 genealso increased the risk of nonunion; the presence of the C/Tgenotype or T allele at rs2297514 was associated with an OR of3.98 (p = 0.015), and the presence of the A/G genotype or Gallele at rs2248814 was associated with a lower OR of 2.27 (p =0.038). It is interesting to note that two of the five tested SNPson the NOS2 gene (see Appendix) were found to have signif-icance in this genetic model.

Two SNPs were found to have an OR of <1. The presenceof the G allele at rs3819089 on the MMP13 gene was found tobe protective from nonunion (OR = 0.26, p = 0.026). Thepresence of the G allele at rs270393 on the BMP6 gene was alsofound to be protective (OR = 0.30, p = 0.015).

Discussion

Fracture-healing is a multifactorial process that can be influ-enced by a variety of variables. Epigenetic patient factors such

as preexisting disease states, medication use, nutritional impair-ment, and endocrine abnormalities all affect fracture-healing.The magnitude and location of the injury itself may be causativeof impaired fracture-healing. In our experience, fracture non-union frequently occurs in high-energy trauma, which causesextensive soft-tissue injury, or in association with patient-relatedfactors such as smoking, the use of NSAIDs, or vitamin-D defi-ciency. Nevertheless, we have occasionally seen patients withsimple closed fractures and no apparent systemic risk factorsdevelop an atrophic nonunion. Occasionally, patients develop

TABLE IV Significant Findings

Gene SNP Genotype No. with Healing/Nonunion Significant Genotype P Value OR

IL1B rs2853550 T allele increases risk 0.034 5.9

C/C 0/0

C/T 8/2

T/T 21/31

NOS2 rs2297514 T allele increases risk 0.015 3.98

C/C 15/7

C/T 14/26

T/T 0/0

NOS2 rs2248814 G allele increases risk 0.038 2.27

A/A 7/1

A/G 11/14

G/G 11/18

MMP13 rs3819089 G allele protective 0.026 0.26

A/A 0/2

A/G 4/11

G/G 25/20

BMP6 rs270393 G allele protective 0.015 0.30

C/C 14/25

C/G 12/8

G/G 3/0

1245




nonunions in multiple extremities after fractures. Thus, it ap-pears likely that there are factors at the genetic level that pre-dispose certain patients toward the development of a fracturenonunion even in the absence of any known epigenetic factors.Alternatively, there may be a genetic predisposition towarddelayed fracture-healing in a given patient that is then phe-notypically expressed when one or more epigenetic factors arepresent. For example, it may be that smoking in associationwith a certain SNP or combination of SNPs is particularlydeleterious to fracture-healing.

This concept is fairly novel. To our knowledge, only twopreviously published studies have investigated the possibleexistence of a genetic predisposition toward the development ofa fracture nonunion. Dimitriou et al. investigated fifteen SNPsspread across four genes in 109 patients, sixty-two with non-union and forty-seven controls42. They found two specificgenotypes (the G/G genotype at the rs1372857 SNP located onthe NOGGIN gene and the T/T genotype at the rs2053423 SNPlocated on the SMAD6 gene) that may predispose patientstoward development of fracture nonunion. Both of the genescode for proteins involved in the BMP pathway; the product ofthe NOGGIN gene serves as an antagonist, and that of SMAD6is an intracellular inhibitor of BMP expression. That study didnot take into account Hardy-Weinberg equilibrium, lack ofwhich disqualified eight SNPs in the present study from furtheranalysis.

In a more recent study, Zeckey et al. investigated twentySNPs for possible associations with fracture nonunion. Theyfound only a trend toward the development of nonunion inpatients with a particular genotype at the rs2252070 SNP in theMMP13 gene. The authors went further to perform a haplotypestudy on the corresponding proteins, and they found a significantassociation with one particular haplotype of PDGF (platelet-derived growth factor) A involving three different SNPs43.

A related study by Mitchell et al. examined the geneticpredictors of heterotopic ossification in 1095 consecutive traumapatients. Heterotopic ossification is a clinical problem that is at theother end of the biological spectrum from nonunion, but it mayinvolve similar genetic loci. That group examined sixty-one SNPsin a variety of genes that may affect formation of heterotopic bone.They identified three SNPs (on the ADRB2, toll-like receptor 4,and complement factor H genes) that were associated with anincreased or decreased frequency of heterotopic ossification44.

We found five SNPs on four genes that were associatedwith a nonunion in an additive genetic model at the 0.05 levelof significance. Three of these SNPs (rs2853550 on IL1B andrs2297514 and rs2248814 on NOS2) had an OR of >1 and werethus associated with impaired fracture-healing. The other twoSNPs (rs3819089 on MMP13 and rs270393 on BMP6) wereassociated with protection from nonunion.

The SNP with the largest odds ratio (5.9) was found onthe IL1B gene. Interleukin 1 beta is a proinflammatory cyto-kine that is a dose-dependent stimulator of human osteoblastdifferentiation45, and it is usually present in the early phasesof bone-healing46. It has been shown to recruit osteoblasts toinjured sites and protect osteoblasts from apoptosis47,48.

Expression of NOS2 occurs during the early stages offracture-healing. Corbett et al. showed that osteoblasts andchondrocytes express NOS2 in a rat tibial fracture model49. Zhuet al. showed that NOS2 mRNA (messenger RNA) expressionpeaked approximately four days after fracture in a rat femurundergoing intramembranous bone formation; by three weeks,there was no detection of NOS2 mRNA. In enchondral boneformation, NOS2 mRNA expression peaked on day seven andwas virtually undetectable by twenty-one days after fracture50.Furthermore, Baldik et al. showed that deletion of the NOS2gene impairs mouse fracture-healing51. We found two SNPs inthe NOS2 gene (rs2248814 and rs2297514) that may predis-pose patients toward development of a nonunion. Patients withthe T allele at the rs2297514 SNP were found to be 3.98 (95%confidence interval [CI], 1.22 to 8.7) times more likely to de-velop a nonunion, and patients with the presence of the G alleleat the rs2248814 SNP were found to be 2.27 (95% CI, 1.1 to4.27) times more likely to develop a nonunion after fracture.

MMP13 is expressed by hypertrophic chondrocytes andosteoblasts in fracture callus. MMP13 is involved in normalremodeling of bone and cartilage during adult skeletal repair,and it may act directly in the initial stages of degradation of theextracellular matrix in these tissues prior to invasion of bloodvessels and osteoclasts52,53. Kosaki et al. showed that fracture-healing in an MMP13-deficient rat model was significantlydelayed and was characterized by retarded cartilage resorptionin the fracture callus. They suggested that MMP13 is crucial tothe process of angiogenesis during fracture-healing, especiallyin the cartilage resorption process and endochondral ossifica-tion54. In the present study, the presence of the G allele at SNPrs3819089 on the MMP13 gene was found to be protectivefrom nonunion (OR = 0.26, p = 0.026). Zeckey et al. found asimilar result in their earlier study43.

The expression of BMPs has been shown in numerousstudies to affect fracture-healing. In a study by Ishidou et al.,BMP7, BMP2, and BMP4 were expressed in the periosteumand extracellular matrix during intramembranous ossification.BMP7 expression peaked on day seven after fracture and de-creased to negligible levels by day fourteen. BMP2 and BMP4were expressed continuously until the very late stages of fracture-healing55. Mizrahi et al. found that recombinant BMP6 inducedmore efficient osteogenic differentiation of mesenchymal stemcells in vitro compared with recombinant BMP256. The presentstudy indicated that the presence of the G allele at the rs270393SNP on the BMP6 gene may be protective against nonunion(OR = 0.030).

Although the present study revealed five SNPs that mayaffect fracture union in either a positive or negative way, threecaveats suggest that caution should be used in interpreting ourdata. First, we began by investigating 144 SNPs and analyzed 133after eliminating the eight SNPs that failed Hardy-Weinbergequilibrium and three that failed quality measures; thus, the strictBonferroni level of significance would be p = 0.05/133 = 0.000376.Consequently, it is likely that using a p < 0.05 level of signifi-cance will have yielded a type-I error and generated false-positive findings. However, strict adherence to the Bonferroni

1246




criteria for the type of multiple comparisons employed in thisstudy might well have yielded a type-II error, excluding SNPsthat are actually associated with impaired fracture-healing.This statistical conundrum is the reason that we consider thepresent study to be a preliminary one. We believe that theresults of this study will allow us to focus on a smaller numberof SNPs in a much larger patient group and thus demonstratesignificance at the Bonferroni-corrected level.

Second, we chose SNPs from HapMap that are known orsuspected to be associated with fracture-healing. There mayexist other, not yet identified, SNPs or SNP combinations that(alone or in the presence of certain epigenetic conditions) maybe associated with impaired fracture-healing. Identification ofthese SNPs would require a genome-wide analysis. Third, ourpatients did not receive screening for endocrine abnormalities,vitamin deficiencies, protein malnutrition, or the use of NSAIDsor other medications affecting bone-healing.

In conclusion, this study provides preliminary datashowing that the technique of SNP genotyping applied to theproblem of defective fracture-healing has merit and is worthyof further investigation. A large genome-wide associationstudy may yield novel SNP-nonunion associations outsideof those genes that are currently understood to be involvedin fracture-healing57. Information from such a study may

also direct further basic-science investigations into the pre-cise mechanisms of osseous healing and osseointegration ofimplants58.

AppendixA table listing the 144 SNPs selected from HapMap and thecorresponding genes is available with the online version of

this article as a data supplement at jbjs.org. n

Vikram Sathyendra, MDHenry J. Donahue, PhDKent E. Vrana, PhDArthur Berg, PhDDavid Fryzel, MDJonathan Gandhi, MDJ. Spence Reid, MDDepartments of Orthopaedics andRehabilitation (V.S., H.J.D., D.F., J.G., and J.S.R.),Pharmacology (K.E.V.), andPublic Health Sciences (A.B.),Penn State Hershey College of Medicine,500 University Drive,Hershey, PA 17033.E-mail address for J.S. Reid: [email protected]

References

1. Busse JW, Bhandari M, Sprague S, Johnson-Masotti AP, Gafni A. An economicanalysis of management strategies for closed and open grade I tibial shaft fractures.Acta Orthop. 2005 Oct;76(5):705-12.2. Antonova E, Le TK, Burge R, Mershon J. Tibial shaft fractures: costly burden ofnonunions. BMC Musculoskeletal Disorders. 2013;14:42.3. Brinker MR, Hanus BD, Sen M, O’Connor DP. The devastating effects of tibialnonunion on health-related quality of life. J Bone Joint Surg Am. 2013 Dec18;95(24):2170-6.4. Moghaddam A, Zimmermann G, Hammer K, Bruckner T, Grutzner PA, von RecumJ. Cigarette smoking influences the clinical and occupational outcome of patientswith tibial shaft fractures. Injury. 2011 Dec;42(12):1435-42. Epub 2011 Jun 12.5. Brinker MR, O’Connor DP, Monla YT, Earthman TP. Metabolic and endocrineabnormalities in patients with nonunions. J Orthop Trauma. 2007 Sep;21(8):557-70.6. Haining SA, Atkins RM, Guilland-Cumming DF, Sharrard WJ, Russell RG, Kanis JA.Vitamin D metabolites in patients with established non-union of fracture. BoneMiner. 1986 Jun;1(3):205-9.7. Spiro AS, Beil FT, Baranowsky A, Barvencik F, Schilling AF, Nguyen K, Khadem S,Seitz S, Rueger JM, Schinke T, Amling M. BMP-7-induced ectopic bone formationand fracture healing is impaired by systemic NSAID application in C57BL/6-mice.J Orthop Res. 2010 Jun;28(6):785-91.8. Hung HC, Chuang J, Chien YC, Chern HD, Chiang CP, Kuo YS, Hildesheim A,Chen CJ. Genetic polymorphisms of CYP2E1, GSTM1, and GSTT1; environmentalfactors and risk of oral cancer. Cancer Epidemiol Biomarkers Prev. 1997 Nov;6(11):901-5.9. Simonis RB, Parnell EJ, Ray PS, Peacock JL. Electrical treatment of tibial non-union: a prospective, randomised, double-blind trial. Injury. 2003 May;34(5):357-62.10. Nolte PA, van der Krans A, Patka P, Janssen IM, Ryaby JP, Albers GH. Low-intensity pulsed ultrasound in the treatment of nonunions. J Trauma. 2001Oct;51(4):693-702; discussion 702–3.11. Nelson FRT, Brighton CT, Ryaby J, Simon BJ, Nielson JH, Lorich DG, Bolander M,Seelig J. Use of physical forces in bone healing. J Am Acad Orthop Surg. 2003 Sep-Oct;11(5):344-54.12. Jingushi S, Mizuno K, Matsushita T, Itoman M. Low-intensity pulsed ultrasoundtreatment for postoperative delayed union or nonunion of long bone fractures.J Orthop Sci. 2007 Jan;12(1):35-41. Epub 2007 Jan 31.13. Paridis D, Karachalios T. Atrophic femoral bone nonunion treated with 1-84 PTH.J Musculoskelet Neuronal Interact. 2011 Dec;11(4):320-2; quiz 323.

14. Della Rocca GJ, Crist BD, Murtha YM. Parathyroid hormone: is there a role infracture healing? J Orthop Trauma. 2010 Mar;24(Suppl 1):S31-5.15. Friedlaender GE, Perry CR, Cole JD, Cook SD, Cierny G, Muschler GF, Zych GA,Calhoun JH, LaForte AJ, Yin S. Osteogenic protein-1 (bone morphogenetic protein-7)in the treatment of tibial nonunions. J Bone Joint Surg Am. 2001;83 Suppl 1(Pt 2 ):S151-8.16. Friedlaender GE. Osteogenic protein-1 in treatment of tibial nonunions: currentstatus. Surg Technol Int. 2004;13:249-52.17. Lissenberg-Thunnissen SN, de Gorter DJJ, Sier CFM, Schipper IB. Use andefficacy of bone morphogenetic proteins in fracture healing. Int Orthop. 2011Sep;35(9):1271-80. Epub 2011 Jun 23.18. Moghaddam A, Elleser C, Biglari B, Wentzensen A, Zimmermann G. Clinicalapplication of BMP 7 in long bone non-unions. Arch Orthop Trauma Surg. 2010Jan;130(1):71-6. Epub 2009 Nov 3.19. den Boer FC, Bramer JAM, Blokhuis TJ, Van Soest EJ, Jenner JMGT, Patka P,Bakker FC, Burger EH, Haarman HJ. Effect of recombinant human osteogenic protein-1 on the healing of a freshly closed diaphyseal fracture. Bone. 2002 Jul;31(1):158-64.20. Kimelman-Bleich N, Pelled G, Zilberman Y, Kallai I, Mizrahi O, Tawackoli W, GazitZ, Gazit D. Targeted gene-and-host progenitor cell therapy for nonunion bone fracturerepair. Mol Ther. 2011 Jan;19(1):53-9. Epub 2010 Sep 21.21. Baltzer AW, Lattermann C, Whalen JD, Wooley P, Weiss K, Grimm M, GhivizzaniSC, Robbins PD, Evans CH. Genetic enhancement of fracture repair: healing of anexperimental segmental defect by adenoviral transfer of the BMP-2 gene. Gene Ther.2000 May;7(9):734-9.22. Bostrom MP. Expression of bone morphogenetic proteins in fracture healing.Clin Orthop Relat Res. 1998 Oct;(355 Suppl):S116-23.23. Gustilo RB, Anderson JT. Prevention of infection in the treatment of one thou-sand and twenty-five open fractures of long bones: retrospective and prospectiveanalyses. J Bone Joint Surg Am. 1976 Jun;58(4):453-8.24. Shen R, Fan JB, Campbell D, Chang W, Chen J, Doucet D, Yeakley J, Bibikova M,Wickham Garcia E, McBride C, Steemers F, Garcia F, Kermani BG, Gunderson K,Oliphant A. High-throughput SNP genotyping on universal bead arrays. Mutat Res.2005 Jun 3;573(1-2):70-82.25. Tsuchihashi Z, Dracopoli NC. Progress in high throughput SNP genotypingmethods. Pharmacogenomics J. 2002;2(2):103-10.26. International HapMap Consortium. A haplotype map of the human genome.Nature. 2005 Oct 27;437(7063):1299-320.

1247




27. International HapMap Consortium, Frazer KA, Ballinger DG, Cox DR, Hinds DA,Stuve LL, Gibbs RA, Belmont JW, Boudreau A, Hardenbol P, Leal SM, Pasternak S,Wheeler DA, Willis TD, Yu F, Yang H, Zeng C, Gao Y, Hu H, Hu W, Li C, Lin W, Liu S,Pan H, Tang X, Wang J, Wang W, Yu J, Zhang B, Zhang Q, Zhao H, Zhao H, Zhou J,Gabriel SB, Barry R, Blumenstiel B, Camargo A, Defelice M, Faggart M, Goyette M,Gupta S, Moore J, Nguyen H, Onofrio RC, Parkin M, Roy J, Stahl E, Winchester E,Ziaugra L, Altshuler D, Shen Y, Yao Z, Huang W, Chu X, He Y, Jin L, Liu Y, Shen Y,Sun W, Wang H, Wang Y, Wang Y, Xiong X, Xu L, Waye MM, Tsui SK, Xue H, Wong JT,Galver LM, Fan JB, Gunderson K, Murray SS, Oliphant AR, Chee MS, Montpetit A,Chagnon F, Ferretti V, Leboeuf M, Olivier JF, Phillips MS, Roumy S, Sallee C, Verner A,Hudson TJ, Kwok PY, Cai D, Koboldt DC, Miller RD, Pawlikowska L, Taillon-Miller P,Xiao M, Tsui LC, Mak W, Song YQ, Tam PK, Nakamura Y, Kawaguchi T, Kitamoto T,Morizono T, Nagashima A, Ohnishi Y, Sekine A, Tanaka T, Tsunoda T, Deloukas P,Bird CP, Delgado M, Dermitzakis ET, Gwilliam R, Hunt S, Morrison J, Powell D,Stranger BE, Whittaker P, Bentley DR, Daly MJ, de Bakker PI, Barrett J, Chretien YR,Maller J, McCarroll S, Patterson N, Pe’er I, Price A, Purcell S, Richter DJ, Sabeti P,Saxena R, Schaffner SF, Sham PC, Varilly P, Altshuler D, Stein LD, Krishnan L, SmithAV, Tello-Ruiz MK, Thorisson GA, Chakravarti A, Chen PE, Cutler DJ, Kashuk CS, LinS, Abecasis GR, Guan W, Li Y, Munro HM, Qin ZS, Thomas DJ, McVean G, Auton A,Bottolo L, Cardin N, Eyheramendy S, Freeman C, Marchini J, Myers S, Spencer C,Stephens M, Donnelly P, Cardon LR, Clarke G, Evans DM, Morris AP, Weir BS,Tsunoda T, Mullikin JC, Sherry ST, Feolo M, Skol A, Zhang H, Zeng C, Zhao H,Matsuda I, Fukushima Y, Macer DR, Suda E, Rotimi CN, Adebamowo CA, Ajayi I,Aniagwu T, Marshall PA, Nkwodimmah C, Royal CD, Leppert MF, Dixon M, Peiffer A,Qiu R, Kent A, Kato K, Niikawa N, Adewole IF, Knoppers BM, Foster MW, Clayton EW,Watkin J, Gibbs RA, Belmont JW, Muzny D, Nazareth L, Sodergren E, Weinstock GM,Wheeler DA, Yakub I, Gabriel SB, Onofrio RC, Richter DJ, Ziaugra L, Birren BW, DalyMJ, Altshuler D, Wilson RK, Fulton LL, Rogers J, Burton J, Carter NP, Clee CM,Griffiths M, Jones MC, McLay K, Plumb RW, Ross MT, Sims SK, Willey DL, Chen Z,Han H, Kang L, Godbout M, Wallenburg JC, L’Archeveque P, Bellemare G, Saeki K,Wang H, An D, Fu H, Li Q, Wang Z, Wang R, Holden AL, Brooks LD, McEwen JE, GuyerMS, Wang VO, Peterson JL, Shi M, Spiegel J, Sung LM, Zacharia LF, Collins FS,Kennedy K, Jamieson R, Stewart J. A second generation human haplotype map ofover 3.1 million SNPs. Nature. 2007 Oct 18;449(7164):851-61.28. Olivier M. A haplotype map of the human genome. Physiol Genomics. 2003 Mar18;13(1):3-9. Epub 2003 Mar 18.29. Barrett JC, Fry B, Maller J, Daly MJ. Haploview: analysis and visualization of LDand haplotype maps. Bioinformatics. 2005 Jan 15;21(2):263-5. Epub 2004 Aug 5.30. Gerry NP, Witowski NE, Day J, Hammer RP, Barany G, Barany F. Universal DNAmicroarray method for multiplex detection of low abundance point mutations. J MolBiol. 1999 Sep 17;292(2):251-62.31. Lin CH, Yeakley JM, McDaniel TK, Shen R. Medium- to high-throughput SNPgenotyping using VeraCode microbeads. Methods Mol Biol. 2009;496:129-42.32. Mefford HC, Cooper GM, Zerr T, Smith JD, Baker C, Shafer N, Thorland EC,Skinner C, Schwartz CE, Nickerson DA, Eichler EE. A method for rapid, targeted CNVgenotyping identifies rare variants associated with neurocognitive disease. GenomeRes. 2009 Sep;19(9):1579-85. Epub 2009 Jun 8.33. R Foundation for Statistical Computing. R project. A language and environmentfor statistical computing. http://www.r-project.org/. Accessed 2014 Feb 19.34. Carlson CS, Eberle MA, Rieder MJ, Yi Q, Kruglyak L, Nickerson DA. Selecting amaximally informative set of single-nucleotide polymorphisms for association anal-yses using linkage disequilibrium. Am J Hum Genet. 2004 Jan;74(1):106-20. Epub2003 Dec 15.35. Zhang H, Zheng G, Li Z. Statistical analysis for haplotype-based matched case-control studies. Biometrics. 2006 Dec;62(4):1124-31.36. Boehringer S, Epplen JT, Krawczak M. Genetic association studies of bronchialasthma—a need for Bonferroni correction? Hum Genet. 2000 Aug;107(2):197.37. Tanner MS, Sharrard MJ, Rigby AS. Gene polymorphisms and the use of theBonferroni correction factor: when and when not to apply? Arch Dis Child. 1997Apr;76(4):386.38. Gao G, Kang G, Wang J, Chen W, Qin H, Jiang B, Li Q, Sun C, Liu N, Archer KJ,Allison DB. A generalized sequential Bonferroni procedure using smoothed weightsfor genome-wide association studies incorporating information on Hardy-Weinbergdisequilibrium among cases. Hum Hered. 2012;73(1):1-13. Epub 2011 Dec 30.

39. Gao X, Becker LC, Becker DM, Starmer JD, Province MA. Avoiding the highBonferroni penalty in genome-wide association studies. Genet Epidemiol. 2010Jan;34(1):100-5.40. Fu YX. Probability of a segregating pattern in a sample of DNA sequences. TheorPopul Biol. 1998 Aug;54(1):1-10.41. Dupuis J, Albers K, Allen-Brady K, Cho K, Elston RC, Kappen HJ, Tang H, ThomasA, Thomson G, Tsung E, Yang Q, Zhang W, Zhao K, Zheng G, Ziegler JT. Effect oflinkage disequilibrium between markers in linkage and association analyses. GenetEpidemiol. 2007;31(Suppl 1):S139-48.42. Dimitriou R, Carr IM, West RM, Markham AF, Giannoudis PV. Genetic predis-position to fracture non-union: a case control study of a preliminary single nucleotidepolymorphisms analysis of the BMP pathway. BMC Musculoskelet Disord.2011;12:44. Epub 2011 Feb 10.43. Zeckey C, Hildebrand F, Glaubitz LM, Jurgens S, Ludwig T, Andruszkow H, HufnerT, Krettek C, Stuhrmann M. Are polymorphisms of molecules involved in bonehealing correlated to aseptic femoral and tibial shaft non-unions? J Orthop Res.2011 Nov;29(11):1724-31. Epub 2011 Apr 28.44. Mitchell EJ, Canter J, Norris P, Jenkins J, Morris J. The genetics of heterotopicossification: insight into the bone remodeling pathway. J Orthop Trauma. 2010Sep;24(9):530-3.45. Frost A, Jonsson KB, Nilsson O, Ljunggren O. Inflammatory cytokines regulateproliferation of cultured human osteoblasts. Acta Orthop Scand. 1997 Apr;68(2):91-6.46. Szczesny G. Molecular aspects of bone healing and remodeling. Pol J Pathol.2002;53(3):145-53.47. Olmedo ML, Landry PS, Sadasivan KK, Albright JA, Meek WD, Routh R, MarinoAA. Regulation of osteoblast levels during bone healing. J Orthop Trauma. 1999 Jun-Jul;13(5):356-62.48. Kon T, Cho TJ, Aizawa T, Yamazaki M, Nooh N, Graves D, Gerstenfeld LC,Einhorn TA. Expression of osteoprotegerin, receptor activator of NF-kappaB ligand(osteoprotegerin ligand) and related proinflammatory cytokines during fracturehealing. J Bone Miner Res. 2001 Jun;16(6):1004-14.49. Corbett SA, Hukkanen M, Batten J, McCarthy ID, Polak JM, Hughes SP. Nitricoxide in fracture repair. Differential localisation, expression and activity of nitricoxide synthases. J Bone Joint Surg Br. 1999 May;81(3):531-7.50. Zhu W, Diwan AD, Lin JH, Murrell GA. Nitric oxide synthase isoforms duringfracture healing. J Bone Miner Res. 2001 Mar;16(3):535-40.51. Baldik Y, Diwan AD, Appleyard RC, Fang ZM, Wang Y, Murrell GAC.Deletion of iNOS gene impairs mouse fracture healing. Bone. 2005 Jul;37(1):32-6.52. Behonick DJ, Xing Z, Lieu S, Buckley JM, Lotz JC, Marcucio RS, Werb Z, Miclau T,Colnot C. Role of matrix metalloproteinase 13 in both endochondral and intra-membranous ossification during skeletal regeneration. PLoS One. 2007 Nov7;2(11):e1150.53. Yamagiwa H, Tokunaga K, Hayami T, Hatano H, Uchida M, Endo N, TakahashiHE. Expression of metalloproteinase-13 (collagenase-3) is induced during fracturehealing in mice. Bone. 1999 Aug;25(2):197-203.54. Kosaki N, Takaishi H, Kamekura S, Kimura T, Okada Y, Minqi L, Amizuka N,Chung UI, Nakamura K, Kawaguchi H, Toyama Y, D’Armiento J. Impaired bonefracture healing in matrix metalloproteinase-13 deficient mice. Biochem BiophysRes Commun. 2007 Mar 23;354(4):846-51. Epub 2007 Jan 19.55. Ishidou Y, Kitajima I, Obama H, Maruyama I, Murata F, Imamura T, Yamada N,ten Dijke P, Miyazono K, Sakou T. Enhanced expression of type I receptors for bonemorphogenetic proteins during bone formation. J Bone Miner Res. 1995 Nov;10(11):1651-9.56. Mizrahi O, Sheyn D, Tawackoli W, Kallai I, Oh A, Su S, Da X, Zarrini P, Cook-WiensG, Gazit D, Gazit Z. BMP-6 is more efficient in bone formation than BMP-2 whenoverexpressed in mesenchymal stem cells. Gene Ther. 2013 Apr;20(4):370-7.Epub 2012 Jun 21.57. Kruglyak L. Prospects for whole-genome linkage disequilibrium mapping ofcommon disease genes. Nat Genet. 1999 Jun;22(2):139-44.58. Dimitriou R, Kanakaris N, Soucacos PN, Giannoudis PV. Genetic predispositionto non-union: evidence today. Injury. 2013 Jan;44 Suppl 1:S50-3.

UpdateThis article was updated on September 10, 2014, because of a previous error. On page 1242, in the byline, and on page 1247, in theauthor addresses, the academic degree for Henry J. Donahue had previously read ‘‘MD.’’ The degree now reads ‘‘PhD.’’

1248




Documents

Single Nucleotide Polymorphisms in Osteogenic Genes in Atrophic Delayed Fracture-Healing: A Preliminary Investigation