• 2.a. The FlexiScope Trial
  
• 2.b. Data collection
  
• 2.c. DNA Extraction
  
• 2.d. NAT1 Genotype Analysis
  
• 2.e. NAT2 Genotype Analysis
  
• 2.f. Collaboration with other centres
  
• 2.g. Statistical analysis
  
• 2.h. Cleaning the FFQ data
  

The basis of all data collected for this work was the FlexiScope Trial.  Ethics approval was obtained locally for both the FlexiScope Trial protocol and the enrolment of volunteers in to this study on N-acetyltransferase genotype, diet and adenomatous polyps (Appendix 2: Ethics Committee Approval).

2.a.i. Trial Centres

The initial trial phase was performed in Welwyn Garden City and then in Leicester²⁶⁶ where a total of 4422 people were screened.  Subsequently, 12 centres were set up across the country in Birmingham, Glasgow, Harrow, Leeds, Liverpool, Manchester, Newcastle, Newport, Cambridge / Norwich, Oxford, Portsmouth and Swansea.

The data presented herein is the data collected from the Leeds Trial centre with collaborative data with the Portsmouth and Cambridge Trial Centres.

2.a.ii. General Practice Recruitment

Central to volunteer recruitment for the trial was the recruitment of G.P. practices.  The general practitioner gave his or her permission for the F.H.S.A to produce a list of all persons aged between 55 and 64 within the practice.  The general practitioner then screened the list to ensure that it was appropriate to approach their patients as potential volunteers.  Exclusion criteria are listed in Table 15.

2.a.iii. Volunteer recruitment

The households for potential volunteers were randomised to whether they would be approached in order that there would not arise the situation where one member of a household was approached for screening and another was not as it was felt this would produce anxiety between the family members.

The recruitment was in 2 stages in order to increase compliance by asking a first round "Interest in Bowel Screening" question (Appendix 3: Initial Questionnaire) before randomisation (Figure 7).

It was estimated that to recruit a total of 3000 volunteers then 30,000 would need to be approached.

Those in the study arm were sent out an initial letter giving them an appointment time and date (Appendix 4: Invitation Letters).  Those who accepted were given a reminder of this appointment when the enema and instructions (Appendix 5: Guidelines for using the enema), map and directions were sent out.  Those who did not respond were sent a repeat appointment letter if they failed to respond to this then there was no further contact (Figure 7)

2.a.iv. Bowel Preparation

The initial trial centres in Welwyn Garden City and Leicester investigated the use of either Picolax oral bowel preparation or a self-administered phosphate enema as bowel preparation prior to flexible sigmoidoscopy in a single blind randomised trial²⁶⁷.  They concluded that the phosphate enema gave better bowel preparation, was associated with greater compliance with fewer side effects in comparison to Picolax.

Therefore, a single phosphate enema (Fletchers' long tube version, Pharmax, Bexley, Kent) was self-administered in most cases an hour before appointment time according to the guidelines sent to the volunteer (Appendix 5: Guidelines for using the enema).

Volunteers who felt they could not comply with self-administration were offered the enema in the department given by a trained nurse.

A proportion of volunteers had inadequate bowel preparation and were given a second enema in the endoscopy department.

2.a.v. Pre-screening medical questionnaire

A pre-screening medical questionnaire  (Appendix 6: Pre-Screening Medical Form) was completed by all participants.  This enabled the endoscopist to identify any comorbidity that might be pertinent to the examination such as giving volunteers with artificial heart valves prophylactic antibiotics and not performing polypectomy in patients on warfarin.

2.a.vi. Flexible Sigmoidoscopy

Flexible sigmoidoscopy in the Leeds centre was performed by a single endoscopist in the endoscopy department at The General Infirmary at Leeds using 60cm Olympus endoscopes supplied by KeyMed.  Sedation was not routinely employed.  The examination was deemed complete if the sigmoid / descending colon junction was assessed to have been reached.  The screening findings were recorded as per Appendix 7: Screening Findings.

Polypectomy was performed using standard techniques so long as bowel preparation allowed.  If a polyp was greater than 1cm and thus an absolute indication for colonoscopy then the polypectomy was carried out at the time of colonoscopy.  Colonoscopy screening results were recorded on the form in Appendix 9: Colonoscopy Findings.

2.a.vii. Histopathology

All polyps were analysed histopathologically by a single histopathologist, Professor P. Quirke in the Leeds Trial Centre.  Each polyp was assessed according to the pathology report pro forma (Appendix 8: Endoscopic Pathology Report) detailing the histological diagnosis (with % villousness as appropriate), grade of dysplasia and shape.

2.a.viii. Post-screening Outcome

It was the intention of the trial to screen and discharge the majority (~95%) of volunteers attending.  Occasionally, a repeat flexible sigmoidoscopy would be indicated if the bowel preparation was not ideal but the volunteer could not stay for the duration of another enema and repeat examination or if the endoscopist felt there was an incomplete excision of a "low risk" adenoma.

The possible outcomes following screening are listed in Table 16.

Colonoscopy and surgical referral in the Leeds centre was to Mr. Paul Finan, Consultant Coloproctologist at the General Infirmary at Leeds.

The FlexiScope trial itself gave the means for recording the data concerning the volunteers and their polyps as detailed above using the "Epi Info 6" epidemiology database (provided by the Centres for Disease Control and Prevention www.cdc.gov ).

2.b.i. Dietary data

All volunteers attending for screening were asked to complete a pre-screening medical form which included data on smoking (Appendix 6: Pre-Screening Medical Form) and a dietary questionnaire (Appendix 12: Food Frequency Questionnaire).  This was filled in on arrival in the endoscopy department prior to actual screening in order to minimise recall bias caused by the volunteer knowing whether they had polyps or not.

2.b.ii. Cases and controls

Cases and controls were defined as follows:

Cases

Any Caucasian volunteer with a histologically confirmed adenomatous polyp in the left colon found at screening flexible sigmoidoscopy

Controls

A Caucasian volunteer with a normal flexible sigmoidoscopy complete to the sigmoid - descending junction or more proximally in the presence of adequate bowel preparation matched to a case by age, sex and general practice.  Age matching was within 2 age categories either 55 - 59 or 60 - 64 years of a case.

Matching for general practice was used in an effort to minimise any effect for social status as catchment areas for the different practices recruited varied in terms of affluence.

2.b.iii. Blood for DNA analysis

Following flexible sigmoidoscopy, volunteers identified as potential cases (i.e. any polyp awaiting histological confirmation) or controls were asked if they would participate in this research by donating blood for DNA analysis.  The consent form and the information sheets are listed in Appendix 10: Information and Consent Forms for Blood Taking.

Dependent on the timing of the flexible sigmoidoscopy screening list and access to the laboratory, two methods for DNA extraction were used dependent on whether the DNA was extracted immediately from fresh blood or if the blood was stored frozen at -20^oC prior to extraction.

2.c.i. DNA Extraction from Frozen Blood (Phenol-Chloroform Method)

3ml of blood was added to 9ml of R.C.L.₁ solution.

Placed on rotator to lyse for 10 minutes.

Centrifuged at 3000 rpm for 10 minutes and the supernatant discarded.

3ml ice cold R.C.L. solution was added and the tube inverted.

Centrifuged at 3000 rpm for 10 minutes and the supernatant discarded.

To the remaining pellet, 100ml Proteinase K and 3ml 1 x S.D.S. / S.T.E.₅ solution was added.

Sample was left to incubate at 37^oC overnight.

3ml saturated TRIS buffered Phenol was added.

Centrifuged at 3000 rpm for 10 minutes.

The two phases were separated by pipetting the top phase into a clean tube,

To the top phase was added 3ml chloroform.

Centrifuged at 3000 rpm for 10 minutes.

Again, the two phases were separated by pipetting the top phase into a clean tube,

To the top phase was added 7.5ml 100% Ethanol.

The tube was inverted several times until any DNA precipitates out of solution.

Centrifuged at 3000 rpm for 10minutes.

Supernatant was discarded leaving the DNA pellet on the bottom of the tube,

Tube was placed in a fume cupboard to dry.

When all the ethanol had evaporated 110ml of 1 x T.E.₄ solution was added.

Following dissolution the DNA was transferred to a clean 1.5ml DNA tube.

10ml of the DNA solution was used to measure the concentration by optical density ratio.

2.c.ii. DNA Extraction From Fresh Blood (Salt-Precipitation Method)

3ml of blood was added to 9ml of R.C.L. solution.

Sample was left on rotator to lyse.

Centrifuged at 3000 rpm for 10minutes.

Supernatant discarded, leaving a buff white cell pellet.

3ml ice cold R.C.L. solution was gently washed over pellet and discarded.

3ml W.C.L.₂ solution pipetted up and down to break up and dislodge the pellet.

Sample was incubated at 37^oC for at least 30 minutes.

Allow sample to cool to room temperature.

1ml of protein precipitation solution₃ was added.

Mixture was vortexed for at least 20seconds.

Centrifuged at 3000 rpm for 10 minutes.

Supernatant pipetted in to a fresh tube.

3ml of isopropanol was added and gently mixed to precipitate the DNA

Centrifuged at 3000 rpm for 10minutes.

Supernatant was discarded leaving the DNA pellet at the bottom of the tube.

3ml of 70% ethanol was added.

Centrifuged at 3000 rpm for 10minutes.

Supernatant was discarded leaving the DNA pellet at the bottom of the tube.

3ml of 70% ethanol was added.

Centrifuged at 3000 rpm for 10minutes.

Supernatant was discarded leaving the DNA pellet at the bottom of the tube.

Tube was placed in a fume cupboard to dry.

When all the ethanol had evaporated 110ml of 1 x T.E. solution was added.

Following dissolution the DNA was transferred to a clean 1.5ml DNA tube.

10ml of the DNA solution was used to measure the concentration by optical density ratio.

2.c.iii. DNA Extraction Solutions

1.  R.C.L. Solution

     8.3g Ammonium Chloride, 155mM

     1g Potassium Chloride, 10mM

     0.37g EDTA, 1mM

     make up to 1 litre with distilled water and autoclaved

2.  W.C.L. Solution

     9.3g EDTA, 25mM

     20g of 10% SDS stock solution, 2%

     make up to 1 litre with distilled water and autoclaved

3.  Protein Precipitation Solution

     385.4g Ammonium Acetate, 10M

     make up to 500ml with distilled water

4.  1 x T.E.

     1.21g TRIS, 10mM

     0.37g EDTA, 1mM

     make up to 1 litre with distilled water and autoclave

5.  1 x S.D.S. / S.T.E.

     5.88g Sodium Chloride, 10mM

     1.21g TRIS, 10mM

     0.37g EDTA, 1mM

     100ml of 10% S.D.S. stock solution

     made up to 1 litre with distilled water

Six alleles of the NAT1 gene (1*3, 1*4, 1*10, 1*11, 1*14 and 1*15) were investigated using a combination of Polymerase Chain Reaction (PCR) and Restriction Fragment Length Polymorphism (RFLP) with the enzyme MBOII, allele-specific PCR and heteroduplex analysis.

In light of the discovery of further NAT1 alleles the following interpretations are made for the method described below:

• This method will pick up NAT1*11A and NAT1*11C correctly as NAT1*11 variants but will miss NAT1*11B which does not have the C1095A substitution (see Appendix 1: Arylamine N-Acetyltransferase Nomenclature).  It does not distinguish between NAT1*11A and NAT1*11C
• Heteroduplex analysis for NAT1*14 was only performed on those with the C1095A substitution and therefore will only pick up NAT1*14A alleles as NAT1*14B does not have this substitution (see Appendix 1: Arylamine N-Acetyltransferase Nomenclature).
• Further reference to these alleles herein will be as NAT1*11 and NAT1*14.

The initial PCR amplifies the end of the NAT1 gene and the beginning of the non-coding sequence.  The choice of reverse primer was made to introduce an MboII enzyme restriction site which is lost in the presence of the C1095A mutation which is common to NAT1*3, NAT 1*10, NAT1*11  and NAT1*14 alleles.

All NAT1*11 alleles also have a 9 base pair deletion between 1065bps and 1090 bps and so can be distinguished from NAT1*3, NAT1*10 and NAT1*14 alleles by having a slightly smaller largest fragment.

2.d.i. NAT1 initial PCR

A 50 ml PCR reaction was performed using the following primers:

NAT1 F (forward primer) 5' TAA AAC AAT CTT GTC TAT TTG 3'

NAT1 R (reverse primer)  5' ACA GGC CAT CTT TAG AA 3'

The PCR mix was as follows:

           dH2O                                            25.5 ml

           10 x PCR buffer                             5.0 ml

           MgCl₂ (25 mM)                                7.0 ml

           D.M.S.O.                                         0.5 ml

           dNTPs (20mM) - A                         1.0 ml

dNTPs (20mM) - C                        1.0 ml

dNTPs (20mM) - G                        1.0 ml

dNTPs (20mM) - T                         1.0 ml

NAT1 F (150 ng/ml)                        1.0 ml

NAT1 R (150 ng/ml)                      1.0 ml

Taq ¹ (1 unit/ml)                             1.0 ml

           DNA extract (~50-100 ng/ml)           5.0 ml

2.d.ii. NAT1 PCR conditions

The PCR was performed on the Gene Amp 9600 system under the following conditions:

           94^oC   4 minutes               1 cycle

           50^oC   30 seconds

           72^oC   45 seconds            35 cycles

           94^oC   30 seconds

           72^oC   5 minutes               1 cycle

which generated a single product of 230bp.

2.d.iii. NAT1 PCR product confirmation

The PCR product was confirmed on a 8% acrylamide gel prior to enzyme digestion against a DNA ladder size standard ².

2.d.iv. NAT1 restriction analysis

Each PCR product is digested with MboII overnight at 37^oC using the following mix:

              PCR product                       20 ml

              10 x react buffer                2.5 ml

              enzyme (5/6 units)             0.5 ml

              dH2O                               2.0 ml

The resultant products were analysed on 12% acrylamide gels.

2.d.v. NAT1 Genemap and mutations

                  start codon

                      5' atggacattgaagcatatcttgaaagaattggctataaga

41   agtctaggaacaaattggacttggaaacattaactgacattcttcaacaccagatccgag

101  ctgttccctttgagaaccttaacatccattgtggggatgccatggacttaggcttagagg

161  ccatttttgatcaagttgtgagaagaaatcggggtggatggtgtctccaggtcaatcatc

221  ttctgtactgggctctgaccactattggttttgagaccacgatgttgggagggtatgttt

281  acagcactccagccaaaaaatacagcactggcatgattcaccttctcctgcaggtgacca

341  ttgatggcaggaactacattgtcgatgctgggtttggacgctcataccagatgtggcagc

401  ctctggagttaatttctgggaaggatcagcctcaggtgccttgtgtcttccgtttgacgg

461  aagagaatggattctggtatctagaccaaatcagaagggaacagtacattccaaatgaag

521  aatttcttcattctgatctcctagaagacagcaaataccgaaaaatctactcctttactc

*14  _______________________________________a____________________

*15  ______________________________________t_____________________

581  ttaagcctcgaacaattgaagattttgagtctatgaatacatacctgcagacatctccat

*11  ___________________________________________________________g

641  catctgtgtttactagtaaatcattttgttccttgcagaccccagatggggttcactgtt

701  tggtgggcttcaccctcacccataggagattcaattataaggacaatacagatctaatag

761  agttcaagactctgagtgaggaagaaatagaaaaagtgctgaaaaatatatttaatattt

                                                   stop codon

821  ccttgcagagaaagcttgtgcccaaacatggtgatagattttttactatttagaataagg

NAT1 F -------------------->

881  agtaaaacaatcttgtctatttgtcatccagctcaccagttatcaactgacgacctatca

941  tgtatcttctgtacccttaccttattttgaagaaaatcctagacatcaaatcatttcacc

1001 tataaaaatgtcatcatatataattaaacagctttttaaagaaacataaccacaaacctt 

                                        <---------------- NAT1 R

                                        ttctaaagatggcctgt        

1061 ttcaaataataataataataataataataaatgtcttttaaagatggcctgtggttatct

*3   __________________________________a_________________________

*10  ___________________________a______a_________________________

*11A and *11C__________________________a_________________________

*14A ___________________________a______a_________________________

MboII Recognition Sequence: gaagannnnnnnn^¯

cttctnnnnnnn^¯n

2.d.vi. NAT1 RFLP banding patterns

Figure 8. NAT1 banding patterns for homozygote *4, homozygote C1095A mutation and homozygote *11

Figure 9. Photograph of gel showing NAT1 banding pattern following MboII digest.

The homozygotes are labelled.

2.d.vii. NAT1 allele specific PCR

In order to differentiate between NAT1*3 alleles from NAT1*10 or NAT 1*14 alleles an allele specific PCR was used.  The samples that were identified as having the C1095A mutation (either heterozygous or homozygous) underwent a separate PCR reaction for the presence or absence of the T1088A mutation.

The forward primer NAT1 F was the same as in the initial PCR and the 2 reverse primers were selected to distinguish the T1088A mutation (Figure 10. NAT1 allele specific PCR).

NAT1 F (forward primer) 5' TAA AAC AAT CTT GTC TAT TTG 3'

Either NAT1 R*3            5' GCC ATC TTT AAA ATA CAT TTA   3'

Or NAT1 R*10                  5' GCC ATC TTT AAA ATA CAT TTT 3'

Figure 10. NAT1 allele specific PCR

To act as a control to ensure that the PCR reaction had worked, forward and reverse primers were used that amplified exon 3 of the epoxide hydrolase gene²⁶⁸.

The PCR mix was as follows:

           dH2O                                            28.0 ml

           10 x PCR buffer                             5.0 ml

           MgCl2 (25 mM)                             3.0 ml

           dNTPs (20mM)₂ - A                     1.0 ml

dNTPs (20mM)₂ - C                       1.0 ml

dNTPs (20mM)₂ - G                       1.0 ml

dNTPs (20mM)₂ - T                     1.0 ml

NAT1 F (150 ng/ml)                        1.0 ml

NAT1 R (150 ng/ml)                      1.0 ml

Epo3 F    (50 ng/ml)                       1.0 ml

Epo3 R   (50 ng/ml)                       1.0 ml

Taq (1 unit/ml)                               1.0 ml

           DNA extract                                  5.0 ml

2.d.viii. NAT1 allele specific PCR conditions

The PCR was performed on the Gene Amp 9600 system under the following conditions:

           94^oC   4 minutes               1 cycle

           52^oC   30 seconds

           72^oC   30 seconds            34 cycles

           94^oC   30 seconds

           72^oC   5 minutes               1 cycle

2.d.ix. NAT1 allele specific PCR banding patterns

The resultant products were analysed on a 3% agarose gel ³.

The leading band following PCR is that of epoxide hydrolase exon 3 and therefore the presence or absence of a more proximal band indicates the presence or absence of the specific allele (Figure 11).

Figure 11.  *3 allele specific PCR.

2.d.x. Heteroduplex analysis

Heteroduplex analysis was performed by GenoVar Diagnostics Ltd in order to distinguish the NAT1*14 allele from the NAT1*10 allele and to identify NAT1*15 C559T mutations.  The method used was as described by Hubbard et al.²¹³.

2.d.xi. NAT1 Materials

All materials used for the NAT1 analysis were provided by GenoVar Diagnostics Ltd.

1.      The Taq enzyme used was ABgene Thermoprime Plus DNA polymerase (AB-0301).

2.      The DNA ladder used was DNA micromarker E840 from Anachem (Catalogue No. E840-100UG-F)

Table 17: DNA ladder  - HaeIII digest of pUC18)

3.      Anachem general purpose agarose I (catalogue No. 0710-100G-R).

At the time of investigation there were six common alleles which had been identified at the N-acetyltransferase 2 (NAT2) gene locus in Caucasian populations.  Five of these (S1k, S1kd, S1d, S2 and S3) (these have been renamed 2*5A, 2*5B, 2*5C, 2*6A and 2*7B according to the revised nomenclature for NAT2 alleles²²⁴ (Appendix 1: Arylamine N-Acetyltransferase Nomenclature) are recessive and lead to inheritance of a "slow acetylator" phenotype with compromised acetylation capacity.  The remaining allele, F1 (2*4), is dominant i.e. only one F1 allele is required for full NAT2 activity though some studies have suggested that the F1/F1 homozygote exhibits slightly faster acetylation than the heterozygote^218,219.  The presence of each of the allelic variants can be identified by comparison of the restriction patterns obtained following Dde I₁, Kpn I₂, BamH I₃ and Taq I₄ digestion of PCR products obtained from amplification of a 547bp fragment of the NAT2 gene²¹⁸.

2.e.i. NAT2 PCR

A 100 ml PCR reaction was performed using the following primers:

NAT2 F (forward primer) 5' GCT GGG TCT GGA AGC TCC TC 3'

NAT2 R (reverse primer)  5' TTG GGT GAT ACA TAC ACA AGG G 3'

The PCR mix was as follows:

           dH2O                                            67.5 ml

           10 x PCR buffer₁                            10.0 ml

           MgCl2 (25 mM)₁                              6.0 ml

           D.M.S.O.                                         5.0 ml

           dNTPs (20mM)₂ - A                     1.0 ml

dNTPs (20mM)₂ - C                       1.0 ml

dNTPs (20mM)₂ - G                       1.0 ml

dNTPs (20mM)₂ - T                     1.0 ml

NAT2 F (250 ng/ml)  ₃                     1.0 ml

NAT2 R (250 ng/ml)₃                       1.0 ml

AmpliTaq Gold (5 units/ml)₁           0.5 ml

           DNA extract (~50-100 ng/ml)           5.0 ml

2.e.ii. NAT2 PCR conditions

The PCR was performed on the Gene Amp 9600 system under the following conditions:

           94oC     4 minutes               1 cycle

           59oC     30 seconds

           72oC     45 seconds            34 cycles

           94oC     30 seconds

           59oC     5 minutes               1 cycle

           72oC     5 minutes               1 cycle

which generated a single product of 547bp.

2.e.iii. NAT2 PCR product confirmation

The PCR product was confirmed on a 0.8% agarose gel prior to enzyme digestion against a 50bp DNA ladder size standard₄.

2.e.iv. NAT2 restriction analysis

Each PCR product is digested with BamH I₅, Dde I₆ and Kpn I₇ overnight at 37oC and with Taq I₈ overnight at 65oC using the following mix:

              PCR product                       20 ml

              10 x react buffer                2.5 ml

              enzyme (5/6 units)             0.5 ml

              dH2O                               2.0 ml

The resultant products were analysed on 3% MetaPhor₉ agarose gels.

2.e.v. NAT2 Genemap and RFLP sites

Gene map of wildtype (F1) allele with mutations and their RFLP sites:

                                                               Start codon

                             5' atggacattgaagcatattttgaaagaattggctataagaact

ctaggaacaaattggacttggaaacattaactgacattcttgagcaccagatccgggctgttccctttgagaacc

ttaacatgcattgtgggcaagccatggagttgggcttagaggctatttttgatcacattgtaagaagaaaccggg

gtgggtggtgtctccaggtcaatcaacttctgtactgggctctgaccacaatcggttttcagaccacaatgttag

gagggtatttttacatccctccagttaacaaatacagcactggcatggttcaccttctcctgcaggtgaccattg

acggcaggaattacattgtcgat 3'

                                                               NAT2 F primer

   Start of 547bp     367  ------------------>                      418
amplified sequence   5' gctgggtctggaagctcctcccagatgtggcagcctctagaattaatttctg

            DdeI                                            KpnI       493
ggaaggatcagcc^¯tcaggtgccttgcattttctgcttgacagaagagagaggaatctggtac^¯ctggaccaaatca

                                            481 c-t loss of KpnI site                                                              568
ggagagagcagtatattacaaacaaagaatttcttaattctcatctcctgccaaagaagaaacaccaaaaaatat

                   TaqI                                            643
acttatttacgcttgaacct^¯cgaacaattgaagattttgagtctatgaatacatacctgcagacgtctccaacat

          590 g-a loss of TaqI site

                                                                    718
cttcatttataaccacatcattttgttccttgcagaccccagaaggggtttactgtttggtgggcttcatcctca

                                       TaqI              DdeI          793
cctatagaaaattcaattataaagacaatacagatctggt^¯cgagtttaaaactctcac^¯tgaggaagaggttgaag

                                                              BamHI     868
aagtgctgaaaaatatatttaagatttccttggggagaaatctcgtgcccaaacctggtgatg^¯gatcccttacta

a-g creation of DdeI site                    857 g-a loss of BamHI site
                                            913                             
tttagaataaggaacaaaataaacccttgtgtatgtatcacccaa 3'
Stop codon                <------------------

                              NAT2 R primer

Enzyme                         Recognition Sequence
BamHI                               g^¯gatcc           

DdeI                                c^¯tnag        

KpnI                               ggtac^¯c           

TaqI                                 t^¯cga 

2.e.vi. NAT2 RFLP banding patterns

Figure 12. NAT2 RFLP banding pattern and examples

 (Letters in capitals indicate wildtype allele and small letters the mutant allele.)

2.e.vii. Determination of NAT2 genotype

The combinations of the RFLP banding patterns determine the definitive NAT2 genotype  .  The only genotypes that are indistinguishable by this method are F1/S1kd (2*4/2*5B) and S1k/S1d (2*5A/2*5B) as the number of mutations are the same but the phase is different i.e. on which chromosome the mutation occurs.  The alleles S1k and S1d are rare and therefore the genotype F1/S1kd is assumed on balance of probabilities.

Figure 13. NAT2 assay genotypes

2.e.vi. NAT2 Materials

1.  AmpliTaq Gold (catalogue no. N808-0244), 10x PCR buffer and 25mM MgCl₂ solution (catalogue no. N808-0245) was obtained from Perkin Elmer.

2.  Ultrapure dNTP set (each nucleotide supplied as a 100mM solution in water) obtained from Amersham Pharmacia Biotech catalogue no. 27-2035-01.

3.  NAT2 forward and reverse primers were synthesised by the Imperial Cancer Research Fund oligonucleotide synthesis service, Clare Hall Laboratories.

4.  50 bp DNA ladder was obtained from GibcoBRL catalogue no. 10416-014 and used at a concentration of 0.3mg/lane.

5.  BamH I restriction enzyme (10 u/ml) was obtained from MBI Fermentas catalogue no. KR4431 and used with Buffer BamH I+ (10mM Tris-HCl, 5mM MgCl2, 100mM KCl, 0.02% Triton X-100 and 0.1mg/ml BSA).

Recognition sequence:       5'...G^¯GATC  C...3'

                                        3'...C  CTAG_­G...5'

6.  Dde I restriction enzyme (12 u/ml) was obtained from Promega catalogue no. R6295, and used with restriction enzyme buffer D (6mM Tris-HCl, 150mM NaCl, 6mM MgCl2 and 1mM DTT).

Recognition sequence:     5'...C^¯TNA  G...3'

                                      3'...G  ANT_­C...5'

7.  Kpn I restriction enzyme (12 u/ml) was obtained from Promega catalogue no. R6341, and used with restriction enzyme buffer J (10mM Tris-HCl, 50 mM KCl, 7mM MgCl2 and 1mM DTT).

Recognition sequence:     5'...G GTAC^¯C...3'

                                      3'...C_­CATG G...5'

8.  Taq I restriction enzyme (10 u/ml) was obtained from Promega catalogue no. R6151, and used with restriction buffer E (6mM Tris-HCl, 100mM NaCl, 6mM MgCl2 and 1mM DTT).

Recognition sequence:     5'...T^¯CG A...3'

                                      3'...A GC_­T...5'

9.  MetaPhor agarose was obtained from Flowgen catalogue no.

In order to increase the numbers of cases and controls to improve the power of the studies being performed, it was agreed that data and DNA was to be pooled from three of the participating centres - Leeds, Norwich and Southampton.

Power calculations for Leeds assuming a screened population size of 3000 and adenoma prevalence of 9% giving 270 cases and controls:

• 94% power to detect a relative risk of 2.0 for a homozygote frequency of 25% (assumed for NAT1).

• 97% power to detect a relative risk of 2.0 for a homozygote frequency of 40% (assumed for NAT2).

DNA was extracted at each centre and aliquots were sent to the other centres for gene analysis.

All trial data from the centres was held centrally at the Imperial Cancer Research Fund in St. Mark's Hospital, Harrow, London.  Data analysis on the collaborative dataset was coordinated centrally by Rob Edwards, medical statistician, under the guidance of Dr. Wendy Atkin.

Statistical analysis was performed using the Stata software package (Stata Corp 1999. Stata Statistical Software: Release 6.0. College Station, TX: Stata Corporation).

Tests of association between categorical variables were assessed by chi-squared (c²) statistics and tests of ordered categorical variables were assessed by chi-squared tests for trend.

Whenever possible analysis was performed on matched cases and controls both on the Leeds dataset and the collaborative dataset as a whole.

Case-control comparisons were performed using conditional logistic regression (or unconditional logistic regression for unmatched data) for qualitative (binary) outcomes. Tests for interaction between genotype and environmental exposure were performed by comparing models with and without an interaction term using likelihood ratio (LR) tests.

As part of the data cleaning exercise the following amendments were agreed upon in interpreting the food frequency questionnaire (when the data record was incorrectly or inadequately completed).

Whole record exclusion

1.      More than 2 double entries per food item.

2.      Where any of the following were stated to be consumed 6+ times per day:  broccoli, brussel sprouts, cabbage/cauliflower, peas, fish (all types), beef, beef burgers, pork. lamb, chicken, bacon, corned beef, sausages, savoury pies. (We decided that consuming leeks, onions or garlic, or meat (all types) 6+ a day were acceptable!).

3.      People who any of the following 30 or more times each week : broccoli, brussel sprouts, cabbage/cauliflower, peas.

Keep in the following:

1.      People who tick the same frequency for many food items

2.      People who eat meat, all vegetable types etc more than 30 times/week (since they mean they consume it a lot)

Converting null values to 'never'

We agreed that we would convert all nulls to 'never' unless the item with a missing value follows or precedes an item with a double entry, in which case both would be coded as 'missing'.