Original Papers

Anaesth Intensive Care 2015 | 43:2

RYR1 and CACNA1S analysis of Australian MH families

Original Papers Analysis of the entire ryanodine receptor type 1 and alpha 1 subunit of the dihydropyridine receptor (CACNA1S) coding regions for variants associated with malignant hyperthermia in Australian families R. L. Gillies*, A. R. Bjorksten†, D. Du Sart‡, B. M. Hockey§

Summary

Defects in the genes coding for the skeletal muscle ryanodine receptor (RYR1) and alpha 1 subunit of the dihydropyridine receptor (CACNA1S) have been identified as causative for malignant hyperthermia (MH). Sixty-two MH susceptible individuals presenting to the same diagnostic centre had copy deoxyribonucleic acid, derived from muscle ribonucleic acid, sequenced to identify variants with the potential to be responsible for the MH phenotype in both RYR1 and CACNA1S. These genetic findings were combined with clinical episode details and in vitro contracture test results to improve our understanding of the Australian MH cohort. Twelve novel variants were identified in RYR1 and six in CACNA1S. Known RYR1 causative mutations were identified in six persons and novel variants in RYR1 and CACNA1S in a further 17 persons. Trends indicated higher mutation identification in those with more definitive clinical episodes and stronger in vitro contracture test responses.

Key Words: malignant hyperthermia, ryanodine receptor calcium release channel Malignant hyperthermia (MH) is an autosomal dominant pharmacogenetic condition of disordered skeletal muscle calcium homeostasis triggered by exposure to volatile anaesthetic agents or suxamethonium1, which can result in death, even in modern times2. Susceptibility to MH has traditionally been assessed with the in vitro contracture test (IVCT)3,4 in Australia and Europe (or the caffeine halothane contracture test5 in the United States) but this requires an open muscle biopsy with anaesthesia. Both IVCT and the caffeine halothane contracture test require specific equipment and expertise. This restricts availability to relatively few sites, and thus not all suspected MH patients can access this testing. The association between mutations in the ryanodine receptor (RYR1) gene and MH has long been known6 but is complicated by the very large number of potentially

* MBBS(Hons), FANZCA, Head of Malignant Hyperthermia Diagnostic Unit, Department of Anaesthesia and Pain Management, Royal Melbourne Hospital, Parkville, Victoria † PhD, Senior Scientist, Malignant Hyperthermia Diagnostic Unit, Department of Anaesthesia and Pain Management, Royal Melbourne Hospital, Parkville, Victoria ‡ PhD, FHGSA, FFSc(RCPA), Research Affiliate/Head Molecular Genetics Lab, Victorian Clinical Genetics Services, Parkville, Victoria § BSc(Hons), MBChB, FANZCA, Staff Anaesthetist, Malignant Hyperthermia Diagnostic Unit, Department of Anaesthesia and Pain Management, Royal Melbourne Hospital, Parkville, Victoria Address for correspondence: Dr Robyn Gillies. Email: [email protected] Accepted for publication on December 15, 2014

causative variants (rare DNA changes resulting in an amino acid substitution which have not yet been functionally characterised), most of which are of vanishingly small incidence7 even within the MH population. More recently, mutations (DNA changes functionally characterised and known to cause MH) in the dihydropyridine receptor (CACNA1S) gene have also been shown to be associated with MH in a few families8. Other loci have been suggested as potential sites for mutations which could be responsible for MH9 and other proteins have been shown to be involved in calcium release from the sarcoplasmic reticulum10 but, as yet, no mutations associated with these potential sites have been reported. Genetic testing for MH is simpler, cheaper and less inconvenient for the patient but is dependent on prior elucidation of the causative mutation in each MH family. A previous study of the genetics of MH in the Australian population11 using sequencing of the hot-spot regions of RYR1 (exons 1 to 20, 38 to 47, 85 to 87 and 98 to 104) from genomic DNA (gDNA) found mutations in nine of 38 families (24%) and variants in a further nine families (for a total of 47%). The large size of the RYR1 gene (15,117 base pairs of coding region in 106 exons) makes sequencing of the entire coding region of RYR1 gene a daunting task from gDNA. Consequently, attention has shifted to copy DNA (cDNA) sequencing in order to facilitate complete coverage of the

157


R. L. Gillies et al

protein coding sequence. This has been used to increase the likelihood of finding mutations segregating with MH with variable results—a yield as high as 70% in the North American population12, 58% in a large cohort from the United Kingdom13 and 52% in a recent European study14. Accordingly, the aim of this study was to sequence the whole of RYR1 and CACNA1S from cDNA derived from ribonucleic acid (RNA) extracted from muscle samples taken at IVCT to maximise the likelihood of identification of putative mutations in the Australian population.

Materials and methods Vastus lateralis muscle and peripheral blood samples were collected from 62 consecutive, unrelated probands undergoing IVCT and yielding an MHShc (malignant hyperthermia susceptible with a threshold response to both caffeine and halothane), MHSh (malignant hyperthermia susceptible with a threshold response to halothane only) or MHSc (malignant hyperthermia susceptible with a threshold response to caffeine only) result according to the European Malignant Hyperthermia Group (EMHG) IVCT protocol4,15 at the Malignant Hyperthermia Diagnostic Unit in the Department of Anaesthesia and Pain Management at the Royal Melbourne Hospital, Victoria. The IVCT laboratory is accredited by the EMHG. Samples for genetic analysis were collected with patient consent and in accordance with Melbourne Health Human Research Ethics Committee approval (Approval No. 2006.223). Muscle specimens were immediately dissected and placed in RNA®later solution (Qiagen, Limburg, the Netherlands) and blood was collected into an ethylenediaminetetraacetic acid blood collection tube (Sarstedt, Nümbrecht, Germany) according to the manufacturer’s instructions. The Clinical Grading Scale (CGS)16 was determined, where possible, from information recorded at the time of the suspected MH reaction in the proband or nearest relative in the pedigree. RNA was extracted from the muscle specimen using TRIzol reagent (Invitrogen, Thermo Fisher Scientific, Tarren Point, New South Wales) and immediately reverse transcribed to cDNA with Superscript III Supermix (Invitrogen), both according to the manufacturer's instructions. gDNA was extracted from the blood specimen according to the standard protocol of the Victorian Clinical Genetics Service Molecular Genetics Laboratory using a robotic method with Chemagen chemistry (Perkin Elmer Inc., Waltham, MA, USA). The cDNA was amplified twice in 28 overlapping fragments for RYR1 and 10 overlapping fragments for CACNA1S of 400 to 700 bp using HotstarTaq polymerase (Qiagen) and a nested approach. The internal amplification primers were used for sequencing in each direction on an ABI 3730 (Applied Biosystems, Thermo Fisher Scientific, Tarren Point, New South Wales) automated sequencer with Big Dye chemistry (Applied Biosystems). The quality of the sequencing was sufficient to

158

read each fragment in both forward and reverse directions. Exons 70 and 83 were examined using gDNA as they are alternately spliced. All cDNA sequence variants resulting in an amino acid substitution were confirmed by sequencing of the relevant exon in gDNA extracted from the blood sample from the same proband. Nucleotides were numbered according to the reference sequences NM_000540.2 for RYR1 and NM_000069.2 for CACNA1S. The minor allele frequency (MAF) of each observed variant in the Exome Variant Server (EVS)17 was used to differentiate between polymorphisms (common DNA changes causing an amino acid substitution but not segregating with MH in the population) and variants rare enough to have a possible association with MH, with 1% MAF used as the cut-off point. Polymorphism Phenotyping v2 (PolyPhen2), PhastCons and genomic evolutionary rate profiling (GERP++) scores predicting functionality and conservation across species were generated by the SeattleSeq Annotation Server 138 for each of the observed variants18. PolyPhen2 is a tool which predicts the possible impact of an amino acid substitution on the structure and function of a human protein using physical and comparative considerations, giving a score between 0 (benign) and 1 (probably damaging)19. GERP++ score is a measure of evolutionary constraint derived by aligning multiple mammalian genomes20. Negative scores imply a relative lack of conservation and the more positive the score, the stronger the conservation across species, implying a region of the protein which is probably functionally important. PhastCons scores are based on conservation of amino acids in multiple alignments, given a phylogenetic tree, with scores between 0 (little conservation) and 1 (complete conservation)21. The results from IVCT were classified into not achieving threshold (NT), minimal (MIN), moderate (MOD) and maximal (MAX) on the basis of the strength of the response to either halothane or caffeine. A response of less than 0.2 g at 2% halothane or 2 mM caffeine was graded as NT, 0.2 g to 0.29 g as MIN, 0.3 g to 0.99 g as MOD and 1.0 g or greater as MAX. These are arbitrary classifications allowing us to differentiate between borderline MHS (MIN) and strong responders (MAX).

Results IVCT data from 62 consecutive, unrelated probands included three MHSc, five MHSh and 54 MHShc. Table 1 provides a summary of the IVCT findings, CGS and non-synonymous nucleotide changes observed. Of note, there were six persons with four known pathogenic EMHG mutations22 (c.1840C >T, c.6502G >A, c.6617C >T [3], c.7300G> A) and all of these had maximum IVCT responses to caffeine and halothane. A further seventeen persons had at least one variant identified in either RYR1 or CACNA1S.

Table 1 In vitro contracture test results, Clinical Grading Scores versus mutations, variants and polymorphisms observed in each proband Family

IVCTc

IVCTh

CGS

Gene

EMHG

Variants

Polymorphisms

A

MOD

MOD

15

CACNA1S

c.4615C> T

B

MOD

MIN

10

CACNA1S

c.1373T> A

C

MIN

NT

EIR

D

MOD

NT

10

CACNA1S

c.1373T> A

E

MOD

MOD

35

CACNA1S

c.1373T> A, c.5399T> C

F

MOD

MOD

EIR

CACNA1S

G

MOD

MOD

25

CACNA1S

c.3091G>A**

H

MAX

MAX

FHx

RYR1 CACNA1S

c.6548G>A**

I

MAX

MAX

30

RYR1 CACNA1S

c.11266C> G c.773G> A c.1373T> A c.4973G> A

J

MOD

MOD

FHx

CACNA1S

c.773G>A c.1373T>A c.1817G>A

K

MAX

MAX

35

RYR1

L

MOD

MOD

18

CACNA1S

c.4615C> T

M

MOD

MAX

15

CACNA1S

c.1373T> A

N

MOD

MAX

31

CACNA1S

c.1373T> A

O

MIN

MIN

15

P

MAX

MAX

28

RYR1

Q

MOD

MOD

33

CACNA1S

c.1373T> A c.206C> G c.1373T> A c.1373T>A, c.4973G>A

c.4711A> G** c.8654C> G** c.10043G> A c.10097G> A** c.11798A> G**

c.6617C> T c.773G> A c.206C> G c.1373T> A c.1817G> A c.5399T> C c.5360C> T c.6178G> T

RYR1 R

MAX

MAX

48

RYR1

c.6617C> T

S

MOD

MOD

30

CACNA1S

T

MOD

MAX

38

U

MAX

MOD

FHx

CACNA1S RYR1

V

NT

MIN

EIR

CACNA1S

c.1373T> A

W

MOD

MOD

FHx

CACNA1S

c.1373T> A c.4615C> T

X

MAX

MAX

40

RYR1

Y

MOD

MOD

58

CACNA1S

Z

MOD

MOD

10

RYR1

c.1373T> A c.4060A> T c.4178A> G

c.5399T> C

c.7300G> A c.206C> G c.1373T> A c.13513G> C**

CACNA1S AA

MOD

MOD

28

RYR1

AB

MAX

MAX

33

CACNA1S

AC

MIN

MOD

60

CACNA1S RYR1

c.2275C> T**

AD

MOD

MOD

15

RYR1

c.4405C> T** c.6302T> A**

c.1373T> A c.5399T> C

c.4024A> G c.4055C> G c.8360C> G c.206C> G c.1373T> A c.5360C> T c.6178G> T

Table 1 (Cont.) In vitro contracture test results, Clinical Grading Scores versus mutations, variants and polymorphisms observed in each proband Family

IVCTc

IVCTh

CGS

Gene

EMHG

Variants

Polymorphisms

AE

NT

MOD

20

CACNA1S

c.1373T> A

AF

MOD

MOD

45

CACNA1S

c.1373T> A c.4973G> A

AG

MOD

MIN

30

CACNA1S

c.4615C> T

AH

MOD

MOD

30

CACNA1S

AI

NT

MIN

FHx

CACNA1S

AJ

MOD

MOD

EIR

AK

MOD

MAX

43+

CACNA1S

c.4615C> T

AL

MOD

MOD

FHx

CACNA1S

c.1373T> A (homozygous)

AM

MIN

MOD

FHx

RYR1 CACNA1S

AN

MOD

MOD

FHx

CACNA1S

c.1373T> A c.4615C> T

AO

MOD

MOD

FHx

AP

MOD

MAX

30

CACNA1S

c.4615C> T

AQ

NT

MOD

15

CACNA1S

AR

MOD

MOD

FHx

CACNA1S RYR1

c.2767G> A**

c.5399T> C c.11266C> G

AS

MAX

MAX

FHx

CACNA1S

c.1493G> T**

c.1373T> A c.4615C> T

AT

MOD

MOD

48

CACNA1S

c.1373T> A c.2099C> T** c.2440G> A** c.4178A> G

RYR1

c.[12823G>A(+)12824C>A]

c.4973G> A

c.206C> G c.1373T> A

c.1373T> A

c.4615C> T c.4973G> A c.6178G> T

RYR1 AU

MOD

MOD

52

RYR1

AV

MOD

MIN

10

CACNA1S

c.1201C> T

AW

MOD

MOD

15

CACNA1S

AX

MAX

MAX

30

RYR1

AY

NT

MOD

33+

CACNA1S

AZ

MAX

MAX

45+

RYR1 CACNA1S

c.12533G> T**

BA

MOD

MOD

30

RYR1

c.5132A> G**

BB

MAX

MAX

50+

CACNA1S

c.5399T> C

BC

MOD

MOD

30

BD

MIN

NT

EIR

CACNA1S

c.773G> A c.1373T> A c.4615C> T

BE

MAX

MAX

33

RYR1

BF

NT

MIN

FHx

CACNA1S

BG

MOD

MOD

>45

CACNA1S

BH

MOD

MAX

50+

RYR1

BI

MAX

MAX

48

RYR1

BJ

MIN

MOD

15

CACNA1S

c.1373T> A (homozygous) c.4060A> T

c.1373T> A c.5399T> C

c.1840C> T c.4615C> T c.4973G> A c.5360C> T

RYR1

c.4615C> T

c.6502G> A c.206C> G c.1373T> A c.1373T> A c.10237A> T** c.6617C> T c.1373T> A

** Variants revealed in this study. Previously revealed variants and polymorphisms are referenced the first time they appear in the table. IVCTc=strongest response to 2 mmol caffeine, IVCTh=strongest response to 2% halothane, RYR1=ryanodine receptor, CACNA1S=dihydropyridine receptor, EIR=exercise-induced rhabdomyolysis, FHx=family history. 160



Table 2 RYR1 variants and EMHG mutations along with their amino acid change, minor allele frequency in a large exome population, PolyPhen2 predicted pathogenicity and GERP++ and PhastCons conservation scores. Nucleotide Change

Exon

Amino acid change

MAF (%)

PolyPhen2

GERP++

PhastCons

c.1201C> T

12

Arg401Cys

Not detected

1.0

3.08

1.0

c.1840C> T**

17

Arg614Cys

0.012

1.0

3.76

1.0

c.4024A> G

28

Ser1342Gly

0.062

0.002

4.03

0.983

c.4055C> G

28

Ala1352Gly

Not detected

0.094

1.45

0.27

c.4178A> G

29

Lys1393Arg

0.58

0.394

4.6

1.0

c.4405C> T

30

Arg1469Trp

0.012

1.0

5.41

1.0

c.4711A> G

33

Ile1571Val

0.128

0.991

3.99

1.0

c.5132A> G

34

Tyr1711Cys

Not detected

1.0

3.98

1.0

c.6302T> A

39

Met2101Lys

Not detected

0.005

1.48

0.994

c.6502G> A**

39

Val2168Met

Not detected

1.0

4.63

1.0

c.6548G> A

39

Gly2183Glu

Not detected

1.0

4.42

1.0

c.6617C> T**

40

Thr2206Met

Not detected

1.0

4.59

0.946

c.7300G> A**

45

Gly2434Arg

Not detected

1.0

3.99

1.0

c.8360C> G

53

Thr2787Ser

0.047

0.009

4.14

1.0

c.8654C> G

56

Thr2885Arg

Not detected

1.0

3.14

0.992

c.10043G> A

67

Arg3348His

Not detected

0.997

3.54

0.99

c.10097G> A

67

Arg3366His

0.12

0.954

3.54

1.0

c.10237A> T

67

Ile3413Phe

Not detected

0.982

3.71

1.0

c.11798A> G

86

Tyr3933Cys

0.12

1.0

4.11

1.0

c.12533G> T

90

Gly4178Val

Not detected

1.0

3.67

0.998

c.[12823G> A(+)12824C> A]

91

Ala4275Lys

Not detected

0.01

-0.169

0.003

c.13513G> C

92

Asp4505His

0.43

1.0

3.93

0.949

** EMHG known causative mutation. RYR1=ryanodine receptor, EMHG=European Malignant Hyperthermia Group, PolyPhen2= Polymorphism Phenotyping v2, GERP++= genomic evolutionary rate profiling, MAF=minor allele frequency. Table 3 CACNA1S variants, along with their amino acid change, minor allele frequency in a large exome population, PolyPhen2 predicted pathogenicity and GERP++ and PhastCons conservation scores. Nucleotide Change

Exon

Amino acid change

MAF (%)

PolyPhen2

GERP++

PhastCons

c.1493G>T

11

Arg498Leu

0.127

1.0

4.78

1.0

c.2099C>T

15

Thr700Met

Not detected

0.001

-1.38

0

c.2275C>T

17

Arg759Cys

Not detected

1.0

3.25

0.262

c.2440G>A

18

Ala814Thr

0.21

0.008

2.67

0.364

c.2767G>A

22

Val923Met

Not detected

0.999

4.41

1.0

c.3091G>A

25

Val1032Met

Not detected

0

-1.34

0.01

c.4060A>T

33

Thr1354Ser

0.51

0.09

4.43

1.0

CACNA1S=dihydropydridine receptor, PolyPhen2= Polymorphism Phenotyping v2, GERP++= genomic evolutionary rate profiling, MAF=minor allele frequency.

Tables 2 and 3 show the EMHG mutations and the observed variants, MAF in the EVS and the amino acid change associated with the base pair alteration. In 39 of the 62 individuals, there were no non-synonymous changes or polymorphisms only. In RYR1 there were 18 observed variants, five of which have been previously described (c.1201C> T, c.4024A> G,

c.4055C> G, c.4178A> G, c.8360C> G). The variant c.4178A> G appeared in two unrelated individuals and the variant c.12533G> T was also identified in the same location as the variant c.12532G> A p.Gly4178Ser reported in one individual in a large European cohort, but resulting in a different amino acid substitution14.

161


R. L. Gillies et al

25

40

32%

29%

35

20 30

15

25

14% 78%

10

Subjects

Subjects

63%

20

10

5

5

0

3

5 4 Clinical Grading Score Category

6

Figure 1: Proportion of patients with and without a variant observed in RYR1 versus CGS category, where known. Solid bar=EMHG mutation, grey bar=RYR1 variant, open bar=no RYR1 variant observed. Combined data from cDNA (this study) and gDNA subjects from our Unit. Percentages shown above the columns indicate the proportion of subjects with either a mutation or variant in RYR1. RYR1=ryanodine receptor, CGS=Clinical Grading Scale, EMHG=European Malignant Hyperthermia Group, cDNA=copy deoxyribonucleic acid, gDNA=genomic deoxyribonucleic acid.

In CACNA1S, there were seven variants identified. One variant has been previously described (c.4060A> T) and appeared in two unrelated individuals. Three variants were found in only one MHSh individual (family AI). No other variants or EMHG mutations were found in individuals with threshold responses to either caffeine or halothane but not both. There were four families with more than one mutation or variant, and of these, there was one individual with five RYR1 variants whose histology showed central cores but had no manifestations of clinical disease. There was also one individual with five polymorphisms in CACNA1S. Of the 22 RYR1 variants and EMHG mutations revealed in the individuals described in this paper, only eight would have been identified using traditional hotspot targeted genetic analysis. Figures 1 and 2 show further analysis of the relationship between presence of RYR1 variants or EMHG mutations and CGSs in combination with our previous study11and the observation of RYR1 or CACNA1S variants or EMHG mutations and strength of IVCT response from this study, respectively. Figure 3 presents the association between the presence or absence of RYR1 or CACNA1S variants and the magnitude of the IVCT responses to caffeine and halothane.

Discussion The number of families where a variant or known mutation has been identified remains low in this Australian cohort. The yield of individuals with one or more variants

162

77%

15

0

17%

33%

0%

MHSc

MHSh

MIN IVCT Response Level

MOD

MAX

Figure 2: Proportion of patients with and without a variant in either RYR1 or CACNA1S versus IVCT result. The subjects were categorised on the basis of the lesser of their IVCT responses to caffeine or halothane, such that MHSc includes all subjects where halothane failed to reach threshold and MHSh includes all subjects where caffeine failed to reach threshold for contracture up to MAX which includes all patients with MAX responses to both caffeine and halothane. Solid bar=EMHG mutation, grey bar=RYR1 or CACNA1S variant, open bar=no variant. IVCT=in vivo contracture test, RYR1=ryanodine receptor, CACNA1S=dihydropydridine receptor, MHSc=malignant hyperthermia susceptible with a threshold response to caffeine only, MHSh= malignant hyperthermia susceptible with a threshold response to halothane only, MIN=minimal, MOD=moderate, MAX=maximal.

or EMHG mutations from this study of the entire coding sequence of RYR1 and CACNA1S was only 37%. Combined with data from our previous study11, only 41% of Australian MH families studied displayed variants or EMHG mutations. If we limit the numbers to RYR1 only then 18/62 (29%) RNA/ cDNA alone or 31/88 (35%) combined with gDNA hotspot analysis showed variants or mutations. Despite the low yield, 12 new variants were identified in RYR1 and six new CACNA1S variants were identified, outlining the importance of this more comprehensive genetic testing. We used 1% MAF in the European American sample of the EVS17 as a cut-off to eliminate variants too common to be considered in the context of MH. We firmly believe that it is important to not discount any rare variants until further information about their functionality in vivo, or their concordance with IVCT phenotype in a large pedigree, is known. The genetic incidence of MH has been estimated in the French population by Monnier et al23 to be at least 1:3000 and, if we use the same logic regarding dual mutation families as this French group (we have six families currently with more than one likely variant or mutation in a cohort of over 100 families and an estimated catchment population of ten million people), the genetic incidence of potentially damaging variants in our population could be even higher. We were also persuaded by the previous work of Pirone and colleagues’ analysis of CACNA1S variant c.4060A> T. This appears in EVS at a MAF of 0.51% and yet



segregated with MHS family members, was not present in any of 278 MH negative subjects and has been functionally tested to have an impact on calcium influx in myotubes, albeit in a non-isolated model24 . We have reported PolyPhen2, GERP++ and PhastCons scores for all the variants observed in this study as an indication of the likelihood that they might be MH causative. However, the size, complexity and overall conservation of RYR1 and the heterogeneity of causative mutations for MH (where a substitution can be completely benign until an exposure to a specific triggering drug precipitates an MH crisis) cautions us to be very circumspect in ruling out a variant based on these scores25. While all of the EMHG mutations have high PolyPhen2 scores, five of the seven known polymorphisms in RYR17 also have scores predicting damage; as do two of the six known CACNA1S polymorphisms8. Similarly, although the majority of the 31 known causative EMHG mutations have GERP++ and PhastCons scores consistent with high conservation, this is not the case for all and five each of the known RYR1 and CACNA1S polymorphisms occur at highly conserved residues. Thus it would appear that the only realistic ways to classify variants for MH would be presence of concordance with IVCT phenotype in a large pedigree and functional testing of a putative mutation in a sarcoplasmic reticulum calcium release assay. Access to sufficient individuals in a pedigree is limited to a large extent by the invasive nature of the muscle biopsy procedure for IVCT and compounded by the elective nature of this test and location 3.5

IVCT Force at Threshold Concentration (g)

3

2.5

2

1.5

1

0.5

0 RYR1

CACNA1S Caffeine

None

RYR1

CACNA1S

None

Halothane

Figure 3: Force of contracture at the IVCT threshold concentration (2 mM for caffeine, 2% for halothane) for each IVCT agent versus presence of a variant. Subjects with variants in both RYR1 and CACNA1S were excluded. Box=interquartile range, line across box=median, whiskers=range. IVCT=in vivo contracture test, RYR1=ryanodine receptor, CACNA1S=dihydropydridine receptor.

of testing facilities. Functional testing of individual variants is clearly the way forward, although this will not be a trivial exercise24,26–29. This study suggests that targeted area testing (in regions of RYR1 with a potentially greater influence on channel activity—the so called hot-spot regions) is not extensive enough for future variant identification in our population. Only 28% (9 out of 32) of the variants and EMHG mutations reported in this study would have been identified by this method. The possible exception to this is in those with very high clinical likelihood (i.e. a high CGS) and maximal responses on IVCT. While the Leeds group13 showed a trend towards association of different RYR1 mutations with clinical onset of MH, our observations show a trend towards a much higher frequency of RYR1 variants in those with more severe MH presentations (Figure 1), although the limited numbers in our study preclude meaningful statistical analysis. This information is useful to us because it helps us to direct our diagnostic testing for MH. Whereas in the past, we have insisted on an IVCT to prove MH diagnosis, the data from our studies suggests that a high percentage (80% using cDNA, 50% using gDNA hotspots in our cohort) of those with a CGS of 5 or 6 will have a variant or mutation identified. Because access to IVCT is limited to those who live near to or can travel to an MH testing centre and to those more than 10 years old, it means that some of our likely MH episodes have not been investigated with IVCT. These figures represent a convincing argument to go straight to RYR1 gDNA hotspot11 analysis, or even next-generation sequencing25,30 rather than wait for IVCT for children under the age of 10, and to consider this strategy in future for those whose access to IVCT is limited when a high CGS is recorded following an MH-like crisis during anaesthesia. As genetic testing becomes standard in the investigation of MH, the number of tests increases, as does the expense. The expense of RYR1 testing is likely then to fall to the family and not just as part of research projects. The figures we have in relation to the strength of IVCT contraction and likelihood of mutation detection will help us to advise families on the likelihood of discovery of a potentially causative variant, so that they can decide whether the expense is warranted for them. Figure 2 represents the data presented in this manuscript combined with data from our prior hotspot study11 where there were 27 individuals with both a maximum response to caffeine and halothane. Of these 27, RYR1 variants were found in 22 cases (13 of which were EMHG). If there is a maximal response to halothane and caffeine, then from the Unit’s genetics analysis so far, we could realistically expect up to an 81% chance of finding an RYR1 variant and a 48% chance of finding an EMHG mutation. All of the EMHG mutations were found in those whose IVCT responses

163

R. L. Gillies et al

were maximal to caffeine and halothane. The yield of RYR1 variants is high enough to warrant RYR1 testing of those with maximal responses to caffeine or halothane, but genetic testing for those with a lesser response, or who are MHSc or MHSh is harder to justify. As part of informed consent to a potentially expensive, time-consuming and complex test, this information is important to pass on to the patient. The caveat being that not all IVCT muscle samples are equal and the quality of the specimen may need to be taken into account. Interestingly, our data shows CACNA1S variants trending towards lower responses to both caffeine and halothane (Figure 3) and a CACNA1S variant was the only variant identified in any MHSh or MHSc subject. This may indicate that those with non-RYR1 defects have a less aggressive mechanism to produce an MH episode. CACNA1S variants may have a less direct mechanism of altering RYR1 channel activity than changes in RYR1, as the alpha 1 subunit of the dihydropyridine receptor is a protein which interacts with RYR1 rather than being a protein directly responsible for calcium movement itself31. It is, therefore, possible that CACNA1S variants, on the whole, represent a less malignant form of MH as measured by the IVCT and Eltit et al have provided evidence to support this32. Unsurprisingly, many of the study subjects had common polymorphisms. What is possibly significant is that four of our subjects had three or more polymorphisms with no other variants in RYR1 or CACNA1S that could explain the MH phenotype. The presence of three or more polymorphisms in the same gene infers that at least two must be on the same allele. Each variant (including a polymorphism) has the potential to change the conformation of the encoded protein at that position which may confer gain of function, loss of function or no change on its own. A variant or polymorphism may not produce enough change to be MH causative in isolation but in the presence of other RYR1 substitutions there may be interactions, depending on their relative positions in the three-dimensional protein structure which could confer MH phenotype, or change the severity of the phenotype. There is also the possibility that variants in other genes could interact with RYR1 polymorphisms33,34. Even if not responsible for MH alone, the modifying effect of more than one polymorphism on an allele may yet prove to be significant, as was alluded to with RYR1 polymorphisms and patients with centronuclear myopathy35. Much work studying family pedigrees and tracking changes across multiple generations is needed to clarify whether these ideas are another plausible explanation for MH. Over half of our MHShc and seven out of eight of the MHSh and MHSc subjects had no variant identified. The IVCT protocol used to phenotype our subjects carries with it a small, inherent risk of false positive phenotyping4,36 and

164


the study was not aided by the fact that the overwhelming majority of our cases were sole index cases and no further family investigation has been possible to date to confirm the presence of MH in these families. In addition, Sanger sequencing occasionally misses variants simply because one cannot be sure that both alleles are being amplified equally25,30. It is also likely that we have examined a genetically distinct population from those studied in Europe and Northern America12–14. For those without a variant, it may be more appropriate to move to whole exome sequencing as the next step in identification of potential causative mutations in other genes and to keep in mind the potential for combinations of variants and polymorphisms to cause disease. This is a potential application for next generation sequencing which is beginning to be applied to mutation detection in MH25,30. One individual in our study displayed five different variants. The RYR1 variants c.4711A> G, c.8654C> G, c.10043G> A and c.10097G> A all lie outside of the cytosolic N-terminal and central domains whereas c.11798A> G lies around the C-terminal membrane37 coding region which has been associated with central core disease. This subject's histology revealed the presence of central cores, although no muscle symptoms or signs were evident. The possibility that more than one variant on an allele may alter the receptor’s function to a greater or lesser degree is again pertinent. One subject displayed a previously reported, potentially causative variant in both RYR1 and CACNA1S. The probability of interaction between these variants would be small as this would require the two variants to exist such that the changes in the respective three-dimensional protein structures interact in some way, although the necessary threedimensional structures are not yet available to support this assertion. More plausibly, the MH susceptibility sensitivity would be equal to the more ‘sensitive’ of the two variants but MH could be triggered as a result of either variant. Perhaps each variant requires a different set of conditions to trigger. Unfortunately, the clinical information on this family is limited to a distant report of MH in the now deceased mother of the IVCT positive individual and no other clinical details. In conclusion, we have identified further variants in RYR1 and CACNA1S that have the potential to be responsible for the MH phenotype in our cohort. We have shown trends towards differing phenotypes in different variants and these trends combined with clinical information will help us rationalise diagnostic testing for MH in the medium term. We have reinforced that RYR1 and CACNA1S are not likely to be the only proteins involved in MH in our cohort while demonstrating the usefulness of genetic testing for MH patients and their families. Each mutation identified has the potential to change the MH diagnostic experience for many family members from an invasive muscle biopsy to a simple blood test—an important goal not to be undervalued.


References 1. Denborough MA. Malignant hyperthermia. Anesthesiology 2008; 108:156-157. 2. Larach MG, Brandom BW, Allen GC, Gronert GA, Lehman MS. Cardiac Arrests and Deaths Associated with Malignant Hyperthermia in North America from 1987 to 2006. Anesthesiology 2008; 108:603-611. 3. A protocol for the investigation of malignant hyperpyrexia (MH) susceptibility. The European Malignant Hyperpyrexia Group. Br J Anaesth 1984; 56:1267-1269. 4. Ording H, Brancadoro V, Cozzolino S, Ellis FR, Glauber V, Gonano EF et al. In vitro contracture test for diagnosis of malignant hyperthermia following the protocol of the European MH Group: results of testing patients surviving fulminant MH and unrelated low-risk subjects. The European Malignant Hyperthermia Group. Acta Anaesthesiol Scand 1997; 41:955-966. 5. Larach MG. Standardization of the caffeine halothane muscle contracture test. North American Malignant Hyperthermia Group. Anesth Analg 1989; 69:511-515. 6. Gillard EF, Otsu K, Fujii J, Khanna VK, de Leon S, Derdemezi J et al. A substitution of cysteine for arginine 614 in the ryanodine receptor is potentially causative of human malignant hyperthermia. Genomics 1991; 11:751-755. 7. Robinson R, Carpenter D, Shaw M-A, Halsall J, Hopkins P. Mutations in RYR1 in malignant hyperthermia and central core disease. Human Mutation 2006; 27:977-989. 8. Carpenter D, Ringrose C, Leo V, Morris A, Robinson RL, Halsall PJ et al. The role of CACNA1S in predisposition to malignant hyperthermia. BMC Med Genet 2009; 10:104. 9. Loke J, MacLennan DH. Malignant hyperthermia and central core disease: disorders of Ca2+ release channels. Am J Med 1998; 104:470-486. 10. Protasi F, Paolini C, Dainese M. Calsequestrin-1: a new candidate gene for malignant hyperthermia and exertional/environmental heat stroke. J Physiol 2009; 587:3095-100. 11. Gillies R, Bjorksten A, Davis M, Du Sart D. Identification of genetic mutations in Australian malignant hyperthermia families using sequencing of RYR1 hotspots. Anaesth Intensive Care 2008; 36:391-403. 12. Sambuughin N, Holley H, Muldoon S, Brandom BW, de Bantel AM, Tobin JR et al. Screening of the entire ryanodine receptor type 1 coding region for sequence variants associated with malignant hyperthermia susceptibility in the North American population. Anesthesiology 2005; 102:515-521. 13. Carpenter D, Robinson RL, Quinnell RJ, Ringrose C, Hogg M, Casson F et al. Genetic variation in RYR1 and malignant hyperthermia phenotypes. Br J Anaesth 2009; 103:538-548. 14. Klingler W, Heiderich S, Girard T, Gravino E, Heffron JJ, Johannsen S et al. Functional and genetic characterization of clinical malignant hyperthermia crises: a multi-centre study. Orphanet J Rare Dis 2014; 9:8. 15. Hopkins PMRH, Snoeck MM, Girard T, Glahn KPE, Ellis FR, Müller CR et al. The European Malignant Hyperthermia Group guidelines for the investigation of malignant hyperthermia susceptibility. Br J Anaesth 2014; In Press:. 16. Larach MG, Localio AR, Allen GC, Denborough MA, Ellis FR, Gronert GA et al. A Clinical Grading Scale to Predict Malignant Hyperthermia Susceptibilty. Anesthesiology 1994; 80:771-779.


17. Exome Variant Server, NHLBI GO Exome Sequencing Project (ESP), Seattle, WA. From http://evs.gs.washington.edu/EVS/. Accessed December 2013. 18. SeattleSeq Annotation Server 138, NHLBI, Seattle, WA. From http://snp.gs.washington.edu/SeattleSeqAnnotation138/. Accessed April 2014. 19. Adzhubei IA, Schmidt S, Peshkin L, Ramensky VE, Gerasimova A, Bork P et al. A method and server for predicting damaging missense mutations. Nature Methods 2010; 7:248-249. 20. Davydov EV, Goode DL, Sirota M, Cooper GM, Sidow A, Batzoglou S. Identifying a high fraction of the human genome to be under selective constraint using GERP++. PLoS Comput Biol 2010; 6:e1001025. 21. Siepel A, Bejerano G, Pedersen JS, Hinrichs AS, Hou M, Rosenbloom K et al. Evolutionarily conserved elements in vertebrate, insect, worm, and yeast genomes. Genome Research 2005; 15:1034-1050. 22. European Malignant Hyperthermia Group. List of all causative mutations. 2013. From https://emhg.org/genetics/mutations-inryr1/. Accessed December 2013. 23. Monnier N, Krivosic-Horber R, Payen JF, Kozak-Ribbens G, Nivoche Y, Adnet P, Reyford H et al. Presence of two different genetic traits in malignant hyperthermia families: implication for genetic analysis, diagnosis, and incidence of malignant hyperthermia susceptibility. Anesthesiology 2002; 97:1067-1074. 24. Pirone A, Schredelseker J, Tuluc P, Gravino E, Fortunato G, Flucher BE et al. Identification and functional characterization of malignant hyperthermia mutation T1354S in the outer pore of the Cavalpha1S-subunit. Am J Physiol Cell Physiol 2010; 299:C1345-1354. 25. Kim JH, Jarvik GP, Browning BL, Rajagopalan R, Gordon AS, Rieder MJ et al. Exome sequencing reveals novel rare variants in the ryanodine receptor and calcium channel genes in malignant hyperthermia families. Anesthesiology 2013; 119:1054-1065. 26. Sato K, Pollock N, Stowell K. Functional studies of RYR1 mutations in the skeletal muscle ryanodine receptor using human RYR1 complementary DNA. Anesthesiology 2010; 112:1350-1354. 27. Sato K, Roesl C, Pollock N, Stowell KM. Skeletal Muscle Ryanodine Receptor Mutations Associated with Malignant Hyperthermia Showed Enhanced Intensity and Sensitivity to Triggering Drugs when Expressed in Human Embryonic Kidney Cells. Anesthesiology 2013; 119:111-118. 28. Vukcevic M, Broman M, Islander G, Bodelsson M, RanklevTwetman E, Müller CR et al. Functional properties of RYR1 mutations identified in Swedish patients with malignant hyperthermia and central core disease. Anesth Analgesia 2010; 111:185-190. 29. Yang T, Ta TA, Pessah IN, Allen PD. Functional defects in six ryanodine receptor isoform-1 (RyR1) mutations associated with malignant hyperthermia and their impact on skeletal excitationcontraction coupling. The Journal Of Biological Chemistry 2003; 278:25722-25730. 30. Schiemann A, Durholt E, Pollock N, Stowell K. Sequence capture and massively parallel sequencing to detect mutations associated with malignant hyperthermia. Br J Anaesth 2013; 110:122-127. 31. Rios E, Brum G. Involvement of dihydropyridine receptors in excitation-contraction coupling in skeletal muscle. Nature 1987; 325:717-720. 32. Eltit JM, Bannister RA, Moua O, Altamirano F, Hopkins PM,

165

R. L. Gillies et al

Pessah IN et al. Malignant hyperthermia susceptibility arising from altered resting coupling between the skeletal muscle L-type Ca2+ channel and the type 1 ryanodine receptor. Proceedings of the National Academy of Sciences of the United States of America 2012; 109:7923-7928. 33. Robinson R, Curran JL, Ellis FR, Halsall PJ, Hall WJ, Hopkins PM et al. Multiple interacting gene products may influence susceptibility to malignant hyperthermia. Annals of Human Genetics 2000; 64:307-320. 34. Robinson R, Hopkins P, Carsana A, Gilly H, Halsall J, Heytens L et al. Several interacting genes influence the malignant hyperthermia phenotype. Human genetics 2003; 112:217-218. 35. Yamamoto L, Maia L, Ayubguerrieri D. G.P.13.10 High frequency of polymorphisms in the RYR1 gene in Brazilian patients with centronuclear myopathy. Neuromuscular Disorders 2008; 18:810. 36. Allen GC, Larach MG, Kunselman AR. The sensitivity and specificity of the caffeine-halothane contracture test: a report from the North American Malignant Hyperthermia Registry. The North American Malignant Hyperthermia Registry of MHAUS. Anesthesiology 1998; 88:579-588. 37. Amador F, Stathopulos P, Enomoto M, Ikura M. Ryanodine receptor calcium release channels: lessons from structure-function studies. The FEBS journal 2013; 280:5456-5470. 38. Brown RL, Pollock AN, Couchman KG, Hodges M, Hutchinson DO, Waaka R et al. A novel ryanodine receptor mutation and genotype-phenotype correlation in a large malignant hyperthermia New Zealand Maori pedigree. Hum Mol Genet 2000; 9:1515-1524. 39. Gillard E, Otsu K, Fujii J, Duff C, de Leon S, Khanna VK et al. Polymorphisms and deduced amino acid substitutions in the coding sequence of the ryanodine receptor (RYR1) gene in individuals with malignant hyperthermia. Genomics 1992; 13:1247-1254. 40. Levano S, Vukcevic M, Singer M, Matter A, Treves S, Urwyler A. Increasing the number of diagnostic mutations in malignant hyperthermia. Hum Mutat 2009; 30:590-598. 41. Guis S, Figarella-Branger D, Monnier N, Bendahan D, KozakRibbens G, Mattei JP et al. Multiminicore disease in a family susceptible to malignant hyperthermia: histology, in vitro contracture tests, and genetic characterization. Arch Neurol 2004; 61:106-113. 42. Davis M, Brown R, Dickson A, Horton H, James D, Laing N et al. Malignant hyperthermia associated with exercise-induced rhabdomyolysis or congenital abnormalities and a novel RYR1 mutation in New Zealand and Australian pedigrees. Br J Anaesth 2002; 88:508-515.

166
