Late-onset thymidine kinase 2 deficiency: a review of 18 cases

RESEARCH Open Access La e-onse hymidine kinase 2 de iciency: a e iew o 18 cases C is ina Domínguez-González 1,2,3 , Au elio He nández-Laín 4 , Eloy Ri as 5 , Ana He nández-Vo h 6 , Ja ie Sayas Ca alán 6 , Robe o Fe nández-To ón 7 , Ca men Fuiza-Luces 2,3,8 , Jo ge Ga cía Ga cía 9 , Ge mán Mo ís 10 , Mon se Oli é 11 , F ances Mi alles 12 , Jo di Díaz-Mane a 3,13 , Candela Caballe o 14 , Bosco Méndez-Fe e 15 , Ramon Ma í 3,17 , Elena Ga cía A umi 3,16 , Ma ía Ca men Badosa 17 , Jesús Es eban 1,2,3 , Cecilia Jimenez-Malleb e a 3,17 , Albe o Blazquez Encina 2,3,8 , Joaquín A enas 2,3,8 , Michio Hi ano 18 , Miguel Ángel Ma in 2,3,8 and Ca men Pa adas 19,20* Abs ac Backg ound: TK2 gene encodes o mi ochond ial hymidine kinase, which phospho yla es he py imidine nucleosides hymidine and deoxycy idine. Recessi e mu a ions in he TK2 gene a e esponsible o he ‘myopa hic o m’o he mi ochond ial deple ion/mul iple dele ions synd ome, wi h a wide spec um o se e i y. Me hods: We desc ibe 18 pa ien s wi h mi ochond ial myopa hy due o mu a ions in he TK2 gene wi h absence o clinical symp oms un il he age o 12. Resul s: The mean age o onse was 31 yea s. The i s symp om was muscle limb weakness in 10/18, eyelid p osis in 6/18, and espi a o y insu iciency in 2/18. All pa ien s de eloped a iable muscle weakness du ing he e olu ion o he disease. Hal o pa ien s p esen ed di icul y in swallowing. All pa ien s showed e idence o espi a o y muscle weakness, wi h need o non-in asi e Mechanical Ven ila ion in 12/18. Fou pa ien s had deceased, all o hem due o espi a o y insu iciency. We iden i ied common adiological ea u es in muscle magne ic esonance, whe e he mos se e ely a ec ed muscles we e he glu eus maximus, semi endinosus and sa o ius. On muscle biopsies ypical signs o mi ochond ial dys unc ion we e associa ed wi h dys ophic changes. All mu a ions iden i ied we e p e iously epo ed, being he mos equen he in- ame dele ion p.Lys202del. All cases showed mul iple m DNA dele ions bu m DNA deple ion was p esen only in wo pa ien s. Conclusions: The la e-onse is he less equen o m o p esen a ion o he TK2 de iciency and i s na u al his o y is no well known. Pa ien s wi h la e onse TK2 de iciency ha e a consis en and ecognizable clinical pheno ype and a poo p ognosis, due o he high isk o ea ly and p og essi e espi a o y insu iciency. Keywo ds: TK2 de iciency, Mi ochond ial myopa hy, Mul iple dele ions Backg ound De ec s in he main enance and epai o mi ochond ial DNA (m DNA) esul in an eme ging and he e ogeneous g oup o mi ochond ial diso de s, caused by al e a ions o he nuclea genes in ol ed in m DNA eplica ion [1–3]. This g oup includes de ec s in enzymes in ol ed in he main enance o he balanced pool o deoxynucleo ides o he mi ochond ia, which a e c ucial in he biosyn hesis o he mi ochond ial genome and ha e he apeu ic implica- ions [4,5]. The dis up ed syn hesis o m DNA esul s in quali a i e (mul iple dele ions) and/o quan i a i e (a d as ic dec ease in he numbe o copies o deple ion) de ec s o he m DNA. In pa icula , one o he ‘myopa hic o ms’o he mi ochond ial deple ion/mul iple dele ions synd omes is caused by mu a ions in he TK2 gene which encodes o mi ochond ial hymidine kinase, which phospho yla es he © The Au ho (s). 2019 Open Access This a icle is dis ibu ed unde he e ms o he C ea i e Commons A ibu ion 4.0 In e na ional License (h p://c ea i ecommons.o g/licenses/by/4.0/), which pe mi s un es ic ed use, dis ibu ion, and ep oduc ion in any medium, p o ided you gi e app op ia e c edi o he o iginal au ho (s) and he sou ce, p o ide a link o he C ea i e Commons license, and indica e i changes we e made. The C ea i e Commons Public Domain Dedica ion wai e (h p://c ea i ecommons.o g/publicdomain/ze o/1.0/) applies o he da a made a ailable in his a icle, unless o he wise s a ed. * Co espondence: [email p o ec ed] 19 Neu ology Depa men , Neu omuscula Diso de s Uni , Ins i u o de Biomedicina de Se illa, Hospi al U. Vi gen del Rocío, CSIC, Uni e sidad de Se illa, A d. Manuel Siu o s/n, 41013 Se illa, Spain 20 Biomedical Ne wo k Resea ch Cen e on Neu odegene a i e Diseases (CIBERNED), Mad id, Spain Full lis o au ho in o ma ion is a ailable a he end o he a icle Domínguez-González e al. O phane Jou nal o Ra e Diseases (2019) 14:100 h ps://doi.o g/10.1186/s13023-019-1071-z py imidine nucleosides hymidine (dT) and deoxycy idine (dC) [1,6]. Recessi e mu a ions in he TK2 gene (MIM# 609560) a e esponsible o di e se clinical p esen a ions mainly cha ac- e ized by p og essi e muscle weakness, dysphagia and e- spi a o y in ol emen wi h a wide spec um se e i y and o age o onse . TK2 de iciency was ini ially desc ibed by Saada, e al. in 2001 [6] in ou child en wi h a se e e myopa hy as- socia ed wi h deple ion o he m DNA. Since hen, a numbe o cases ha e been epo ed depic ing a he e ogeneous clin- ical p esen a ion wi h a con inuum spec um o he disease, which includes ea ly-onse ex emely se e e and apidly p o- g essi e o ms wi h su i al o less han wo yea s, o less se- e e o ms wi h la e o e y la e onse , and a a iably slowe a e o p og ession [7,8]. In 2012, Tyynismaa, e al. epo ed he i s wo cases wi h mu a ions in he TK2 gene wi h onse in he i h decade o li e, mani es ing ch onic p og essi e ex- e nal oph halmoplegia (CPEO) associa ed wi h limb muscle weakness and dysphagia [9]. A ecen publica ion ha in- cluded 92 pa ien s desc ibing he na u al his o y o his dis- o de p oposed he classi ica ion o h ee clinical o ms acco ding o ages-a -onse : in an ile (< 1 yea -old), childhood (1–12) and la e (> 12 yea s) onse . Nea ly 40% o he epo ed TK2 cases p esen ed wi h he symp oms p io o he age o 1, in ano he 41% he onse occu ed be ween he ages o one and 12, and only in 19% o pa ien s did he symp oms appea ed a e he age o 12 [7]. A subsequen e ospec i e e iew, wi h simila equencies o hose h ee subg oups, included ele en new cases o which only h ee we e classi ied as la e-onse [8]. So a , he na u al his o y o pa ien s wi h la e onse TK2 de iciency has no been de ined in de ail. He e, we epo on he clinical ea u es and assess- men s in a la ge se ies o 18 pa ien s wi h la e-onse TK2 de iciency, he less known and poo es de ined o m o his disease, o u he cha ac e ize his pa ien subg oup. Expanding he na u al his o y and p ognosis o la e-onse TK2 de iciency will acili a e ea lie diagno- sis and iden i ica ion o ea men wi h he apies unde clinical de elopmen . Me hods Pa ien s We desc ibe he pheno ypic ea u es o 16 Spanish and 2 US pa ien s wi h mi ochond ial myopa hy due o mu a- ions in he TK2 gene wi h he absence o clinical symp- oms un il he age o 12. The se ies include h ee pai s o siblings (P3-P4, P6-P10 and P14-P15). Pa ial da a om i e pa ien s ha e been p e iously published else- whe e (P1, P5, P9 [7], P3 and P12 [10]). Clinical e alua ion The elec onic eco ds we e e iewed o collec in o ma- ion abou he age o onse , ini ial symp oms, se e i y, dis ibu ion and p og ession o he muscle weakness and ex a-muscula symp oms. We ga he ed in o ma ion om he la es neu ological examina ion egis e ed in- cluding, when a ailable, he Muscle Resea ch Council (MRC) scale o assess he muscle s eng h and he 6 min walk es (6MWT) o unc ional e alua ion. Respi a o y assessmen The la es alue o he o ced i al capaci y (FVC) in sea ed and supine posi ion, maximum inspi a o y p es- su e (MIP), blood gas analysis, noc u nal en ila ion (assessed wi h noc u nal pulse oxime y and/o capno- g aphy [11] and he need o mechanical en ila ion (MV) ype and hou s o use we e eco ded. Labo a o y es s CK (c ea ine kinase) and lac a e le els we e quan i ied in se um in basal condi ions, a diagnosis. GDF-15 (g ow h/di e en ia ion ac o -15) le els we e quan i ied in plasma samples using human GDF-15 quan i a i e ELISA ki (R&D Biosys ems) acco ding o he manu ac- u e ’s ins uc ions. Muscle MRI Muscle MRI was pe o med in 8 o he 18 pa ien s. All o hem we e scanned in a 1.5 T MR scanne (Siemens). Lowe limb axial T1-weigh ed sequences we e used o mo phological analysis and sho - au in e sion eco e y (STIR) sequences we e examined o de ec muscle edema. The muscle MRI s udies we e e alua ed by he same neu ologis (R F-T) wi h wide expe ience in neu o- muscula diso de s. The e alua o was blind ega ding he clinical mani es a ions. He sco ed pel ic, high and lowe leg muscles in axial T1-sequences wi h he semi- quan i a i e Me cu i isual scale (MVS) modi ied by Fishe [12]: 0: No mal appea ance; 1: Mild in ol emen , less han 30% o indi idual muscle olume; 2: Mode a e in ol emen , 30–60% o indi idual muscle olumes; 3: Se e e in ol emen , > 60% o indi idual muscle; 4: End s age, all he muscle is se e ely a ec ed, eplaced by in- c eased densi y o connec i e issue and a , wi h only a im o ascia and neu o ascula s uc u es dis inguish- able. We compa ed median alue o muscle a y e- placemen using he Wilcoxon-Mann-Whi ney es . S a is ical analyses we e pe o med using IBM SPSS S a- is ics, V.22 (IBM, A monk, New Yo k, USA). Ae obic exe cise es ing Exe cise es ing was pe o med in 5 pa ien s on a cycle e gome e , ollowing a amp-like p o ocol (wo kload in- c eases o 1 W e e y 6 s [a e aging 10 W·min −1 ] s a ing om an ini ial load o 0 W, wi h a pedal cadence o 60– 70 pm h oughou he es ). Gas-exchange a iables we e collec ed b ea h-by-b ea h wi h an au oma ed me abolic ca (Qua k CPET, COSMED, Rome, I aly). Domínguez-González e al. O phane Jou nal o Ra e Diseases (2019) 14:100 Page 2 o 10 The peak oxygen up ake (VO 2 peak) was compu ed as he highes alue ob ained o any 10-s pe iod du ing he es s [13]. Muscle biopsy Muscle samples we e ob ained by open biopsy and p oc- essed ollowing he s anda d p ocedu es: Hema oxylin and eosin (H&E), modi ied Gomo i ich ome, ATPase (adenosine iphospha ase), NADH (nico inamide adenine dehyd ogenase), SDH (succina e dehyd ogenase), COX (cy och ome C oxidase), and COX-SDH s ains we e pe - o med in all a ailable samples. Respi a o y chain enzyme ac i i y le els we e eco ded when a ailable. Gene ic s udies Molecula diagnosis was pe o med ei he by di ec Sange sequencing o exons and in on/exon bounda ies o he TK2 gene, o by cus omized nex gene a ion sequencing (NGS) panels. Pa ien ’s skele al muscle m DNA dele ions we e in es iga ed by long- ange PCR (polyme ase chain eac ion) and/o Sou he n blo , and m DNA copy numbe was assessed by quan i a i e PCR as p e iously desc ibed [10,14]. The s udy was app o ed by he ins i u ional e iew boa d o e e y cen e and all pa ien s signed an in o med consen o he anonymous publica ion o his da a. Resul s Clinical mani es a ions (Table 1) We included 18 pa ien s (6 male, 12 emale). The mean age-a -onse was 31 yea s ( ange 12 o 60 yea s) wi h a mean age a diagnosis o 48.5 yea s ( ange 23 o 73 yea s) esul ing in an a e age o 17.4 yea s be ween he onse o he disease un il eaching a gene ic diagnosis ( ange 1 o 44 yea s). The mean du a ion o he disease was 19.8 yea s ( ange 6 o 44 yea s). Fou pa ien s om he se ies we e deceased, all o hem due o espi a o y insu i- ciency a mean o wo decades a e he onse . The i s symp om was muscle limb weakness in 10/18 (55.6%), eyelid p osis in 6/18 (33%) ( wo pa ien s also p esen ed oph halmopa esis), and espi a o y insu i- ciency in 2/18 (11.1%). All pa ien s de eloped muscle weakness du ing he e olu ion o he disease, 17/18 showing p oximal and dis al limb muscle weakness, 1/18 wi h only dis al limb weakness, and 16/18 axial in ol e- men . I is no ewo hy ha neck lexo weakness was clea ly mo e se e e han limb weakness (mean, 2.14 on he MRC scale). The ollowing muscle g oups we e he mos equen ly a ec ed, in a symme ical manne : shoulde abduc o (mean, 4 on he MRC scale), hip lexo (mean, 3.75 on he MRC scale) and hip ex enso (mean, 3.87 o bo h on he MRC scale) and inge ex enso muscles (mean, 4.14 on he MRC scale). Fou pa ien s (22%) los he abili y o walk wi hou suppo . Facial muscula u e was symme ically a ec ed in 17 pa ien s (94.4%), wi h p e- dominance o he o bicula oculis muscle. 16/18 o he pa- ien s (88.9%) also had symme ical eyelid p osis o a iable se e i y, wi h his being he i s symp om in 6 pa- ien s (33.3%). Six o hem equi ed su gical blepha o- plas y due o ision impai men . Nine pa ien s had CPEO. The majo i y (11/18) had di icul y in swallowing, which esul ed in se e e weigh loss and/o de imen o he sa e y o o al eeding in 6 cases, equi ing pe cu an- eous gas os omy ube in 5 cases (27.8%) on a e age 19.6 yea s a e he onse o he disease ( anging om 12 o 28 yea s). O he clinical mani es a ions included senso y axonal polyneu opa hy (7/18;38.9%), neu osenso y hea ing loss (3/18;16.6%) and dysphonia due o ocal co d palsy (2/ 18;11.1%). No pa ien had ca diomyopa hy. Respi a o y unc ion FVC a diagnosis om he o al coho was 55.4% ( anging om 17 o 103) wi h a mean dec ease o FVC in supine posi ion o 8% ( anging om 0 o 14), and a mean MIP o 36.8% ( anging om 20 o 101%), independen o he asso- cia ed muscle symp oms. F om a espi a o y pe spec i e, he high equency o complica ions should be no ed, wi h need o non-in asi e MV in 12/18 pa ien s (66.6%). The mean use o he MV was 11.6 h pe day ( anging om 8 o 24 h). Eigh ou o he 12 pa ien s wi h MV (66.6%) p e- sen ed wi h acu e espi a o y insu iciency ollowing a ou- ine uppe espi a o y in ec ion as he i s mani es a ion o hedisease.Noneo hesecaseshadanyp io espi a o y symp oms; howe e , once de ec ed, hey equi ed MV due o hype capnia seconda y o al eola hypo en ila ion. Al- hough limb muscle weakness and/o eyelid p osis we e al eady p esen a he onse o espi a o y insu iciency, hose neu omuscula symp oms had no p omp ed a neu - ology consul a ion in any o he eigh pa ien s. Thus, he e- spi a o y in ol emen esul ed in he diagnosis o an unde lying myopa hy in hese pa ien s; he mean FVC was o 40.8% ( ange om 28 o 58) a he ime o diagnosis. O he six pa ien s who did no needed MV, all showed e i- dence o espi a o y muscle weakness on unc ional es s, al hough only one o hem (P8) epo ed espi a o y symp- oms (o hopnoea), sugges ing diaph agma ic weakness. This pa ien displayed p osis and CPEO a he age o 50 as- socia ed wi h mode a e axial and p oximal limb muscle weakness (4 on he MRC scale). S ikingly, al hough he unc ional espi a o y es s and noc u nal pulse oxime y we e no mal (FVC sea ed 103%, FVC decubi us 100% and MIP 101%) noc u nal anscu aneous capnog aphy e ealed high mean le els o ca bon dioxide (CO 2 ,meano 48 mmHg, wi h a maximum peak o 54 mmHg). Fou pa ien s died o espi a o y insu iciency a mean age o 56 yea s ( anging om 40 o 68), and a mean o Domínguez-González e al. O phane Jou nal o Ra e Diseases (2019) 14:100 Page 3 o 10 Table 1 Clinical mani es a ions summa y ID Age A Onse Gende Cu en Age Clinical Assessmen s Respi a o y Assessmen s O he mani es a ions P osis CPEO Facial Weakness Dysphagia Neck lexo weakness Limb Weakness Wheelchai - bound 6MWT (Me e s) PEG (AT AGE) O hopnea BMI Si ing FVC (%) FVC Si ing/ Supine (%) Hou s on MV (AT AGE) 1 12 F 32 + –+ Yes +++ + No 530 No No 13.86 46 −8 8 (32) – 220 F 33 ––+ Yes +++ + No 390 Yes (32) Yes 17.9 26 −22 8–9 (28) Hepa opa hy 3 30 F 61 + –+ Yes ++ + No 386 No Yes 27.12 71 −14 8 (61) PNP-S 4 60 F 73 + + ++ Yes ++ + No 345 No Yes 26 69 NA 8 (73) Hea ing loss PNP-S 5 50 M 50 ++ –+ No + + No 475 No Yes 27.6 47 NA 12 (50) PNP-S 6 14 F 58 ++ ++ ++ No +++ + No 450 No Yes 24.5 51 −28 8 (56) – 7 40 M Dea h a 68 ++ ++ + Yes +++ ++ Yes NA Yes (68) Yes 23.7 48 NA 10–12 (49) Hea ing loss PNP-S Vocal co d palsy 8 50 F 63 + + ++ No + + No 417 No Yes 28.7 103 −3–PNP-S 9 23 F Dea h a 40 ––+ Yes +++ ++ Yes NA Yes (39) Yes 15 28 NA 22 (30) – 10 14 M Dea h a 49 +–– Yes ++ +++ Yes NA Yes (42) Yes 17 32 NA 24 (42) – 11 40 M 51 + –+No–+ No 600 No No 26.3 70 −10 –– 12 30 M 60 ++ –+ Yes NA + No NA No No 23 75 NA –Vocal co d palsy 13 48 F Dea h a 67 + + + No +++ +++ Yes NA No Yes NA 43 NA 12 (58) – 14 15 M 26 + + + Yes + + No 413 No No NA 72 −10 –Hea ing loss PNP-S 15 12 F 31 + + + No + + No 428 No No NA 75 −12 –PNP-S 16 30 F 43 ++ + + No + + No 425 No Yes NA 69 2 8 (43) – 17 45 F 51 ++ + ++ Yes + + No NA No No NA NA NA –NA 18 25 F 58 ++ ++ ++ Yes ++ ++ No 228 Yes (39) Yes NA 17 NA 11 (59) NA NA No a ailable. P osis: ++ His o y o eyelid su ge y, + mild, −no p osis. CPEO ch onic p og essi e ex e nal oph halmoplegia: ++ comple e, + pa ial, −no ocula weakness. Facial weakness: ++ se e e, + mild, −no acial weakness. Neck lexo weakness; +++ unable o li head in supine posi ion, ++ 3 on Medical Resea ch Council (MRC) scale, + 4 on MRC scale, −no neck lexo weakness. Limb weakness: +++ 1–2 on MRC scale, ++ 3 on MRC scale, + 4 on MRC scale, −no weakness. 6MWT six-minu e walking es , FVC o ced i al capaci y, MV Mechanical en ila ion, PNP-S senso y polyneu opa hy, BMI body mass index Domínguez-González e al. O phane Jou nal o Ra e Diseases (2019) 14:100 Page 4 o 10 24 yea s a e he onse o hei ini ial symp oms ( an- ging om 17 o 35). CK and lac a e le els (Table 2) 94.4% o pa ien s had inc eased a iable se um CK le els anging om 190 o 2435 UI/l (no mal le els < 170 UI/ l)), and 16.7% showed le els 10- old abo e he uppe no mal limi . Se um lac a e le els we e measu ed in basal condi ions in 12 o he 18 cases. O hese, only h ee (25%) displayed sligh ly inc eased le els (1.4-2x abo e he uppe no mal limi ). GDF-15 le els GDF-15, a bioma ke iden i ied in he analysis o ansc ip- omic p o iling o TK2 de icien human skele al muscle [15], has been p o en use ul in he diagnosis o mi ochond ial myopa hies [16], being especially inc eased in pa ien s wi h mi ochond ial TK2 de iciency [17]. Se um le els o GDF-15 we e inc eased in 5 ou o 5 cases analysed (100%), anging om 1529 o 2438 pg/mL (2113 pg/mL ± 462, mean ± s anda d de ia ion, uppe limi o no mal =550 pg/mL) [16]. Muscle MRI indings I was pe o med in 8 pa ien s. Mean age a muscle MRI was 46.4 yea s old ( ange: 23–73). Mean disease du a ion a he ime o he scan was 18 yea s ( ange 10–31). The mos se e ely a ec ed muscles in axial T1-weigh ed se- quences we e he glu eus maximus, semi endinosus, sa - o ius and gas ocnemius medialis (median MVS: 3). O hese, only he glu eus maximus and sa o ius we e a - ec ed in all pa ien s. Apa om he la e , glu eus med- ius, adduc o magnus and semi endinosus we e also mode a ely a ec ed in he highs and gas ocnemius la e alis in he legs (median MVS: 2). No muscle a in- il a ion was obse ed in ob u a o , quad a us emo is, ex enso is digi o um and ibialis pos e io (Fig. 1). The Table 2 Biochemical and molecula cha ac e is ics ID Mu a ion Muscle Biopsy Mul iple Dele ions Residual m DNA (%) Respi a o y Chain Enzyme Ac i i y CK (UI/l) GDF-15 (pg/mL) Lac a e (mmol/l) Allele 1 Allele 2 1 c.323C > T (p.Th 108Me ) c.323C > T (p.Th 108Me ) Yes Yes 17 CI, CIII and CIV de ici 2435 2423 1.95 2 c.323C > T (p.Th 108Me ) c.323C > T (p.Th 108Me ) Yes Yes 39 No mal 303 2439 2.3 3 c.604–606 AAGdel (p.Lys202del) c.604–606 AAGdel (p.Lys202del) Yes Yes 60 CIII de ici 294 1695 2.6 4 c.604–606 AAGdel (p.Lys202del) c.604–606 AAGdel (p.Lys202del) ND ND NA ND 647 2483 2.2 5 c.604–606 AAGdel (p.Lys202del) c.604–606 AAGdel (p.Lys202del) Yes Yes 66 No mal 357 1529 1.5 6 c.323C > T (p.Th 108Me ) c.323C > T (p.Th 108Me ) Yes Yes 19 No mal 425 NA 1.6 7 c.604–606 AAGdel (p.Lys202del) c.604–606 AAGdel (p.Lys202del) Yes Yes 33 NA 568 NA 2.6 8 c.604–606 AAGdel (p.Lys202del) c.604–606 AAGdel (p.Lys202del) Yes Yes NA NA 405 NA 2.4 9 c.323C > T (p.Th 108Me ) c.323C > T (p.Th 108Me ) Yes Yes 35 CI, CIII and CIV de ici 190 NA NA 10 c.323C > T (p.Th 108Me ) c.323C > T (p.Th 108Me ) Yes NA NA No mal 405 NA 3 11 c.604–606 AAGdel (p.Lys202del) c.604–606 AAGdel (p.Lys202del) Yes Yes NA ND 266 NA NA 12 c.604–606 AAGdel (p.Lys202del) c.604–606 AAGdel (p.Lys202del) Yes Yes NA ND 350 NA NA 13 c.388C > T (p.A g130T p) c.415G > A (p.Ala139Th ) Yes Yes NA CI, CIII and CIV de ici 170 NA 4.14 14 c.623A > G (p.Ty 208Cys) c.623A > G (p.Ty 208Cys) Yes Yes NA CI, CIII and CIV de ici 1739 NA NA 15 c.623A > G (p.Ty 208Cys) c.623A > G (p.Ty 208Cys) ND NA NA ND 381 NA NA 16 c.323C > T (p.Th 108Me ) c.323C > T (p.Th 108Me ) Yes Yes 53 No mal 233 NA 1.77 17 c.469_470insTGGG (p.Asp157Val s*11) c.156 + 6 T > G Yes Yes 50 NA 537 NA NA 18 c.604–606 AAGdel (p.Lys202del) c.604–606 AAGdel (p.Lys202del) NA NA NA NA 1348 NA NA NA no a ailable, ND no done, CK c ea ine kinase, GDF-15 G ow h di e en ia ion ac o -15 Domínguez-González e al. O phane Jou nal o Ra e Diseases (2019) 14:100 Page 5 o 10 Fig. 1 Axial T1 muscle MRI and ba cha s wi h Me cu i Visual Scale (MVS) dis ibu ion o 7 pa ien s and pe ana omical egion. a, Axial T1 muscle MRI in pel is: These wo consecu i e slices om di e en pa ien s a e showing ha he glu eus maximus (ma ked wi h as e isk) is he mos a ec ed muscle. Tenso ascia la ae is a ec ed while ob u a o and qua a us emo is a e less a ec ed. b, Ba cha MVS a eplacemen in pel is: MVS (0: no a eplacemen , 4: he muscle is comple ely eplaced) o all pa ien s. Glu eus maximus is he mos a ec ed muscle, ollowed by enso ascia la ae. c, Axial T1 muscle MRI in highs: These wo slices om wo di e en pa ien s a e showing he a eplacemen o sa o ius (wide whi e a ow) and as us la e alis ( hin whi e a ow). O he muscles like semi endinosus, semimemb anosus and g acilis a e also mode a ely a ec ed. d, Ba cha MVS a eplacemen in highs: MVS o all pa ien s. Sa o ius, semimemb anosus, semi endinosus, g acilis and as us la e alis a e he mos a ec ed muscles. Sa o ius and g acilis a e a ec ed in all pa ien s. e, Axial T1 muscle MRI in legs: These wo slices om wo di e en pa ien s a e showing he a eplacemen o gas ocnemius medialis (whi e a ow head). Gas ocnemius la e alis and soleus a e also mode a ely a ec ed. Tibialis an e io and ibialis pos e io a e he leas a ec ed. , Ba cha MVS a eplacemen in legs: MVS o all pa ien s. Gas ocnemius medialis and la e alis a e he mos a ec ed muscles in legs. Tibialis an e io , ex enso is digi o um and ibialis pos e io a e he leas a ec ed muscles Domínguez-González e al. O phane Jou nal o Ra e Diseases (2019) 14:100 Page 6 o 10 a eplacemen ollowed a di use pa e n and no ocal a eas o a in il a ion we e de ec ed. We did no ob- se e s a is ical di e ences ega ding asymme ic in- ol emen . STIR sequence was no mal in all pa ien s. Ae obic exe cise es ing In addi ion o weakness, one o he mos equen clin- ical mani es a ions in he mi ochond ial myopa hies is poo exe cise capaci y [18].The la e is e lec ed by low le els o VO 2 peak o by poo muscle-oxygen ex ac ion (as assessed wi h nea -in a ed spec oscopy) du ing g aded cycle-e gome e / eadmill es ing [19]. Ae obic exe cise es ing was pe o med on a cycle e gome e in i e pa ien s. The mean ± SD VO 2 peak ob ained was 14.8 ± 3.2 mL/kg −1 /min −1 , wi h no mal consump ion alues o 40.0 ± 9.5 mL/kg −1 /min −1 [20]. Muscle biopsies Muscle biopsies we e pe o med in 16 pa ien s, 11 we e a ailable o e-analysis. The mo phological s udy e- ealed nume ous agged- ed ibe s in 100% o he biop- sies, which we e hype - eac i e wi h SDH eac ion and usually COX-de icien . COX-de icien ibe s accoun ed o app oxima ely 5–15% o all ibe s. F equen ly hese muscles also showed dys ophic ea u es wi h equen nec o ic ibe s, some wi h phagocy osis, and inc eased endomysial connec i e issue (p esen in 7 ou o 11 bi- opsies e ised). Ma ked ype I ibe p edominance was also obse ed in 2 pa ien s (Fig. 2). These indings di e om he usual pa e n displayed in o he mi ochond ial myopa hies, whe e he ypical signs o mi ochond ial p oli e a ion and dys unc ion a e no associa ed wi h o he ele an changes in muscle his ology s uc u e [21]. We ha e esul s o he analysis o he enzyma ic ac- i i y o espi a o y chain complexes o 10 pa ien s. Only in hal o hem a educ ion in he ac i i y o one o mo e enzyma ic complexes we e iden i ied (Table 2). Gene ic s udies All pa ien s ha bo ed biallelic mu a ions in he TK2 gene (Re .Seq. NM_004614.4) (Table 2). Mos pa ien s (16/18;88.9%) we e homozygous. All mu a ions we e p e iously epo ed [7,8], wi h he in- ame dele ion p.Lys202del (c.604_606AAGdel) being he mos e- quen (16/36 alleles; 44.4%), ollowed by he missense mu a ion p.Th 108Me (c.323C > T) (12/36;27.8%). Addi ionally, h ee missense mu a ions we e iden i ied in 3 pa ien s: p.A g130T p (c.388C > T), p.Ala139Th (c.415G > A), and p.Ty 208Cys (c.623A > G). Finally, one pa ien ha bo ed a ameshi mu a ion p.Asp157 Val s*11(c.469-470insTGGG) in compound he e ozy- gosis wi h a splice si e mu a ion c.156 + 6 T > G. Gene ic da a om pa ien s P1, P2, P5, P9 and P12, we e p e iously epo ed [7,10]. Muscle m DNA copy-numbe was s udied in 9 pa ien s and se e e m DNA deple ion was de ec ed in only wo (17% o esidual m DNA in P1 and 19% o esidual m DNA in P6). Fou een ou o 14 pa ien s (100%) showed hep esenceo mul iplem DNAdele ionsinmuscle. Discussion The la e-onse p esen a ion o TK2 de iciency is he leas equen clinical mode o p esen a ion known. These pa- ien s a e conside ed o ha e a milde p esen a ion han hose wi h in ancy and childhood onse disease, howe e , ew cases ha e been desc ibed o da e and hose e- po ed we e no ex ensi ely explo ed. So a , 17 pa ien s wi h la e-onse we e epo ed o ha bou TK2 biallelic mu a ions [7–10,22]. Howe e clinical de ails we e sca ce, he e ogeneous, and epo s did no clea ly de ine he pheno ype o a e o p og ession o he disease. In some cases, clinical p esen a ion is simila o ha de- sc ibed in he childhood onse pa ien s, wi h p og essi e limb, acial, ex aocula , o opha yngeal and espi a o y muscle weakness, bu wi h a slowe p og ession, whe eas in o he cases, CPEO is he main mani es a ion [9]. Re- spi a o y insu iciency has been men ioned as a po en ial cause o dea h al hough comp ehensi e da a abou he espi a o y in ol emen is no a ailable o all he p e i- ously published pa ien s: se e e espi a o y insu iciency is desc ibed in 41% o he epo ed cases bu in he emaining 59% his da a is una ailable o supe icially desc ibed [7,8,22]. We iden i ied 16 Spanish and wo No h Ame ican pa- ien s, om 13 di e en amilies, wi h TK2 mu a ions and a la e-onse p esen a ion. Exhaus i e clinical desc ip ion is he e p o ided o acili a e ea lie and accu a e diagnosis and o imp o e he knowledge o he na u al his o y o his a e, and p obably unde diagnosed diso de . The clinical ea u es and esul s o he diagnos ic es s desc ibed in ou se ies show a homogeneous pheno ypic pa e n in la e-onse TK2 de iciency consis ing o p o- g essi e p oximal limb, axial neck lexo and acial muscle weakness equen ly associa ed wi h p osis, oph halmopa esis and bulba weakness, along wi h an ea ly and se e e, al hough un ecognized, espi a o y in- ol emen . Diaph agma ic weakness is e y cha ac e is- ic, occu ing in all o ou cases, showing an ea ly onse bu slow p og ession; 12/18 (66.6%) equi ed MV du ing he e olu ion o he disease and in 8/18 (44.4%) was he cause o he i s medical consul a ion. This pa e n o espi a o y in ol emen was ound e en in pa ien s who only had an appa en ly isola ed CPEO pheno ype. The e o e, i is c i ical o iden i y signs o noc u nal hypo en ila ion du ing clinical e alua ion o hese pa- ien s, ega dless o he se e i y o he skele al myopa hy. This disc epancy be ween diaph agma ic and limb weak- ness was also e lec ed in some pa ien s wi h i ually Domínguez-González e al. O phane Jou nal o Ra e Diseases (2019) 14:100 Page 7 o 10 no mal 6MWT esul s, despi e using MV (see Table 1). In ou se ies, he capnog aphy was he mos sensi i e es o de ec ing he espi a o y dys unc ion, since i was abno mal e en be o e basal FVC and MIP e ealed al e a ions. Muscle biopsies showed he ypical indings o mi o- chond ial dys unc ion desc ibed in mos mi ochond ial myopa hies. Howe e , as in o he TK2 de iciency o ms, hey also e ealed dys ophic ea u es which a e dis inc om he majo i y o o he mi ochond ial myopa hies. Thus, ou da a suppo ha he associa ion o bo h mi ochond ial and dys ophic pa e n s ongly sugges mu a ions in he TK2 gene as he unde lying cause. All p e iously published la e-onse pa ien s showed mul iple m DNA dele ions, while m DNA deple ion was ound only in one o he i e cases in whom he m DNA copy-numbe was quan i ied. Ou indings co obo a e he p e ious esul s indica ing he p esence o mul iple m DNA dele ions is mo e equen han m DNA deple- ion in he la e-onse TK2 de icien pa ien s. P e ious e- po s showed ha m DNA deple ion is ound in he mos o ea ly onse pa ien s [7], bu ou da a suppo ha i canno be conside ed a alid p ognos ic ma ke since i can also be ound in la e-onse cases. In muscle MRI he a muscle eplacemen was di use, esembling many muscula dys ophies and congeni al myopa hies. Muscle degene a ion in MRI was desc ibed in i e MERRF pa ien s wi h he m.8344A > G mu a ion [23], and mo e ecen ly a y in il a ion has been com- munica ed in pa ien s wi h single, la ge-scale dele ions o mi ochond ial DNA [24]. Howe e , no ex ensi e s ud- ies ha e been published ying o de ine muscle MRI pa e ns in di e en mi ochond ial myopa hies. So, he e is no speci ic MRI pa e n o any mi ochond ial myop- a hy desc ibed so a . In ou se ies o TK2 pa ien s al- hough no clea pa e n o a in il a ion was de ec ed, we ha e iden i ied some adiological common ea u es, as he in ol emen o he sa o ius muscle in all cases. This muscle is usually spa ed un il la e s ages in many gene ic muscle diseases (is only a ec ed ea ly in some myo ib illa myopa hies, in he Laing dis al myopa hy and in RYR1- ela ed myopa hies (encodes o yanodine ecep o 1 p o ein) [12,25–27]), so his inding could be help ul o di e en ial diagnosis. Se um GDF-15 le els ha e ecen ly been e ealed as a sensi i e and speci ic bioma ke o he diagnosis o mi ochond ial myopa hies [16,17]. In ou se ies, i p o ed o be e y high in all analysed cases, so i could o ien a e he molecula diagnosis in a p ope clinical con ex , be o e he muscle biopsy was pe o med. As in o he mi ochond ial myopa hies [19], in ou se ies he ca diopulmona y exe cise es ing iden i ied a Fig. 2 Mo phological al e a ions in muscle biopsies om pa ien s P1 (a, ,k,p), P5 (b,g,l,q), P9 (c,h,m, ), P14 (D, i,n,s) and P16 (e,j,o, ). a-e H&E shows dys ophic ea u es in all cases wi h mild endomysal ib osis, adipose issue eplacemen , a ophy and nec o ic ibe s. Ragged- ed ibe s a e equen ly iden i ied in all muscle samples (a ows). -j Gomo i ich ome showed he cha ac e is ic agged- ed ibe s in all he biopsies. k-o Succina e dehyd ogenase (SDH) e eals an inc ease o he oxida i e s aining in nume ous ibe s. p- F equen cy och ome C oxidase (COX) de icien ibe s a e p esen in a iable p opo ion in he di e en cases (pand , COX s aining; o,sand , COX-SDH combined s aining). Scale ba =100 μm Domínguez-González e al. O phane Jou nal o Ra e Diseases (2019) 14:100 Page 8 o 10 e y educed consump ion o oxygen, e en in pa ien s wi h CPEO as a p edominan clinical mani es a ion (P8). This indica es ha , al hough he weakness may no be se e e in la e-onse TK2 de iciency pa ien s, he exe cise capaci y is abno mally low, ul ima ely impai ing physical ac i i y. No iceably, he p.Lys202del was he mos equen mu a ion in he TK2 gene in ou se ies o la e-onse pa- ien s, which is consis en wi h he inding ha his mu- a ion appea s o be es ic ed o adul -onse cases, since i has no been epo ed in any in an ile-onse pa ien s who ha e no e en ha bou ing his mu a ion in a single allele [8]. Ne e heless, i was epo ed in one pa ien wi h childhood-onse , who was compound he e ozygous o his mu a ion and a ameshi mu a ion, and began showing symp oms a 2.5 yea s bu su i ed un il 8.5 yea s-old [28]. The eigh cases wi h his mu a ion in ou se ies we e all homozygous suppo ing he idea ha his mu a ion is associa ed wi h a milde e ec (age a onse anging om 25 o 60 yea s). In e es ingly, his mu a ion has only been iden i ied in 13 un ela ed Spanish pa ien s ((11, 13, 26, 27, and his s udy), 2 ela ed pa ien s om Hispanic e hnic backg ound [10], and one pa ien om Venezuela ( his s udy) sugges ing ha i could be a p i- a e mu a ion and ha Spanish/Hispanic candida e pa- ien s may be amenable o a apid gene ic sc eening o his mu a ion. Howe e , haplo ype analysis would be e- qui ed o con i m he possible ounde e ec o his mu- a ion. The p.Th 108Me mu a ion was he second mos common mu a ion in his s udy, howe e i has been ound in in an ile and childhood onse cases [6,7] o di - e en geog aphic o igin. TK2 de iciency is a se e e diso de causing p ema u e dea h. In ecen p e-clinical s udies, i has been demon- s a ed ha ea men wi h py imidine nucleosides (dC + dT) in he H126N knock-in mouse model o TK2 de i- ciency, leads o a p olonged li e span in he animals and a es o ed m DNA copy numbe , wi hou signi ican ox- ici y [4]. This opens he doo o a po en ial he apeu ic in e en ion in humans wi h his me abolic he edi a y diso de , making i necessa y o de ine sensi i e and ob- jec i e ou comes o assess an e en ual esponse o ea - men . Ou indings sugges ha unc ional espi a o y es s, se um GDF-15 le el and he s ess cyclome e e alua ion a e po en ially good candida es o moni o - ing he p og ession o disease. Conclusion In summa y, ou s udy shows ha la e-onse pa ien s wi h mi ochond ial TK2 de iciency ha e a consis en and ecognizable clinical pheno ype, cha ac e ized by a p o- g essi e myopa hy wi h p edominan acial and axial neck lexo weakness, and espi a o y in ol emen , o en asso- cia ed o CPEO. Thei p ognosis is poo , due o he high isk o ea ly and p og essi e espi a o y insu iciency. Ye , some pa ien s may p esen wi h a se e e acu e espi a o y ailu e. Ea ly de ec ion o espi a o y in ol emen equi es an ac i e sea ch in he clinics, e en in asymp oma ic pa- ien s. A small numbe o a ionally designed ea men s a e being de eloped o mi ochond ial diso de s [29], in- cluding nucleoside subs a e enhancemen he apy de- signed speci ically o TK2 de iciency [4]. The e o e, ea ly diagnosis o TK2 de iciency is impo an as pa ien s could bene i om he exis ence o a po en ial he apy. Abb e ia ions 6MWT: 6-min walking es ; ATPase: Adenosine iphospha ase; BMI: Body mass index; CK: C ea ine kinase; CO 2 : Ca bon dioxide; COX: Cy och ome C oxidase; CPEO: Ch onic p og essi e ex e nal oph halmoplegia; dC: Deoxycy idine; dT: Thymidine; FVC: Fo ced i al capaci y; GDF-15: G ow h di e en ia ion ac o 15; H&E: Hema oxylin and eosin; MIP: Maximal inspi a o y p essu e; MRC: Muscle Resea ch Council; MRI: Magne ic esonance imaging; m DNA: Mi ochond ial DNA; MV: Mechanical en ila ion; MVS: Me cu y isual scale; NADH: Nico inamide adenine dehyd ogenase; NGS: Nex gene a ion sequencing; PCR: Polyme ase chain eac ion; SDH: Succina e dehyd ogenase; STIR: Sho au in e sion eco e y; VO 2 peak: Peak oxygen up ake Acknowledgemen s No applicable. Funding This wo k was suppo ed by esea ch g an s o Plan Nacional de I + D + I and Ins i u o de Salud Ca los III (ISCIII), Subdi ección Gene al de E aluación y Fomen o de la In es igación Sani a ia”, p ojec PI16–01843 (CP), PI16/00579 and CP09/00011 o CJM and he Eu opean Regional De elopmen Fund (FEDER a way o achie e Eu ope). MAM has ecei ed unding om he Spanish ISCIII (g an PI 15/00431). A mul icen ic g an unded by he ISCIII (PMP15/00025 o MAM, RM, MO, CP). A ailabili y o da a and ma e ials The da ase s used and/o analysed du ing he cu en s udy a e a ailable om he co esponding au ho on easonable eques . Au ho s’con ibu ions CDG, AHV, JSC, JG, GM, MO, FM, JDM, CC, JBM, JE, MH and CP handled pa ien s and ecollec ed clinical da a o he manusc ip . RT collec ed and analysed he MRI da a o he pa ien s. EG and ABE pe o med molecula analysis; CB and CJ p o ided GDF-15 analysis. AH and ER collec ed and e iew muscle biopsy da a. CD and CP coo dina ed all he s udy. CD, MAM and CP w o e he ini ial manu- sc ip . RM, JA, MAM, CP, and MH p o ided c i ical discussion o he esea ch. All au ho s con ibu ed o he inal e sion o he manusc ip . All au ho s ead and app o ed he inal manusc ip . E hics app o al and consen o pa icipa e W i en in o med consen was ob ained om pa ien s o publica ion o anonymised clinical da a. Consen o publica ion No applicable. Compe ing in e es s MH and RM a e co-in en o s on pa en applica ions iled by Columbia Uni e - si y Medical Cen e (CUMC) o deoxynucleoside he apy o mi ochond ial DNA deple ion synd omes including TK2 de iciency. The pa en applica ions and o he in ellec ual p ope y ha e been licensed by CUMC o Me es Pha maceu i- cals, Inc. CUMC may be eligible o ecei e paymen s ela ed o he de elop- men and comme cializa ion o he echnology. Any po en ial licensing ees ea ned will be paid o CUMC and a e sha ed wi h in en o s h ough CUMC dis- ibu ion policy. MH and RM a e paid consul an s o Me es Pha maceu ical, Inc. The es o au ho s decla e ha hey ha e no con lic o in e es . Domínguez-González e al. O phane Jou nal o Ra e Diseases (2019) 14:100 Page 9 o 10