- Email: [email protected]
Characterization of Deletions of the HBA and HBB Loci by Array Comparative Genomic Hybridization
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62
The Journal of Molecular Diagnostics, Vol. -, No. -, - 2015
jmd.amjpathol.org
Characterization of Deletions of the HBA and HBB Loci by Array Comparative Genomic Hybridization Q11 Q1
Daniel E. Sabath,* M.A. Bender,yz Vijay G. Sankaran,x Esther Vamos,{ Alex Kentsis,k Hye-Son Yi,* and Harvey A. Greisman* From the Departments of Laboratory Medicine* and Pediatrics,y University of Washington, Seattle, Washington; the Clinical Research Division,z Fred Hutchinson Cancer Research Center, Seattle, Washington; the Division of Hematology/Oncology,x Boston Children’s Hospital and Department of Pediatric Oncology, Dana-Farber Cancer Institute, Harvard Medical School, Boston, Massachusetts; the Departments of Pediatrics and Genetics,{ Université Libre de Bruxelles, Brussels, Belgium; and the Department of Pediatrics,k Memorial Sloan Kettering Cancer Center, New York, New York Accepted for publication July 28, 2015.
Q4
Q5
Address correspondence to Daniel E. Sabath, M.D., Ph.D., and Harvey A. Greisman, M.D., Ph.D., Department of Laboratory Medicine, University of Washington, Box 359743, Harborview Medical Center, 3NJ345.1, 325 Ninth Ave., Seattle, WA 98104 E-mail: dsabath@ uw.edu or hagreisman@gmail. com.
Thalassemia is among the most common genetic diseases worldwide. a-Thalassemia is usually caused by deletion of one or more of the duplicated HBA genes on chromosome 16. In contrast, most b-thalassemia results from point mutations that decrease or eliminate expression of the HBB gene on chromosome 11. Deletions within the HBB locus result in thalassemia or hereditary persistence of fetal Hb. Although routine diagnostic testing cannot distinguish thalassemia deletions from point mutations, deletional hereditary persistence of fetal Hb is notable for having an elevated HbF level with a normal mean corpuscular volume. A small number of deletions accounts for most a-thalassemias; in contrast, there are no predominant HBB deletions causing b-thalassemia. To facilitate the identification and characterization of deletions of the HBA and HBB globin loci, we performed array-based comparative genomic hybridization using a custom oligonucleotide microarray. We accurately mapped the breakpoints of known and previously uncharacterized HBB deletions defining previously uncharacterized deletion breakpoints by PCR amplification and sequencing. The array also successfully identified the common HBA deletions –SEA and –FIL. In summary, comparative genomic hybridization can be used to characterize deletions of the HBA and HBB loci, allowing high-resolution characterization of novel deletions that are not readily detected by PCR-based methods. (J Mol Diagn 2015, -: 1e8; http://dx.doi.org/10.1016/j.jmoldx.2015.07.011)
Thalassemia is one of the most common genetic diseases and is responsible for significant morbidity worldwide. a-Thalassemia is most commonly caused by deletion of one or more of the duplicated HBA genes on chromosome 16. In contrast, most b-thalassemia results from point mutations that cripple or completely inactivate expression from the HBB gene on chromosome 11. These mutations commonly result in a decrease in HBB and a relative increase in HBD protein expression, resulting in an elevated HbA2 level. There is frequently a relative increase in HBG1/2 expression as well, with a corresponding increase in HbF. Deletions within the HBB locus occur and can eliminate HBB expression. If the HBD and HBG genes are deleted along with the HBB gene, the characteristic increases in HbA2 and HbF will not be seen, making it impossible to distinguish from a-thalassemia on the basis of HbA2 quantification alone. Deletions resulting in levels of HBD and HBG polypeptide production insufficient to balance that
from the HBA locus result in db- or gdb-thalassemia with a decreased mean corpuscular volume (MCV). In contrast, some deletions result in elevated levels of HBG sufficient to balance that from the HBA locus, resulting in hereditary persistence of fetal Hb (HPFH), the hallmark of which is a normal MCV. In both of these cases, an HBB locus deletion is assumed. HBB deletions causing thalassemia may be more common than generally thought because they may not be recognized during routine thalassemia diagnostic procedures, especially in the rare deletions that do not result in HbF elevation. Several well-characterized deletions account for most a-thalassemia; thus, gap-PCRebased methods are routinely used to identify HBA gene deletions.1 In contrast, HBB Supported by NIH grant DK44746 (M.A.B.) and funds from the Q2 Department of Laboratory Medicine, University of Washington. Q3 Disclosures: None declared.
Copyright ª 2015 American Society for Investigative Pathology and the Association for Molecular Pathology. Published by Elsevier Inc. All rights reserved. http://dx.doi.org/10.1016/j.jmoldx.2015.07.011
FLA 5.4.0 DTD JMDI458_proof 20 November 2015 8:03 pm EO: JMD15_0091
63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124
125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186
Sabath et al Table 1
PCR and Sequencing Primers Used in This Study
Subject no. Forward PCR primer 4 5 7*
8 9
0
Reverse PCR primer 0
Additional sequencing primers (if used)
0
0
5 -TAGATCCCTTTGCCATTATG-3 50 -AGCCTCATGGTAGCAGAATC-30 50 -CCTTTTTCTTGTGGTAAATGCTT-30
5 -TTGGGTTTCTGATAGGCACTG-3 50 -TGGTATCTGCAGCAGTTGCC-30 50 -TTTCCTTGTGTTTGAAAGTGCT-30
50 -ATGCCAGTGCTCTCCACAAT-30
50 -ATAATAAGCCTGCGCCCTTC-30
50 -ACCCTTTGAGTAATAGTTTCCTGA-30 (forward) 50 -GCAAATAAGCACACATATATTCCAA-30 (reverse) 50 -AGTTTCTTGTTTACTCTGGA-30 (reverse) 50 -GGCCATATAGGGTAACTTTCTGAC-30 (reverse) 50 -AAAAGTGTGCCATGGTTTTAATG-30 (reverse)
50 -CATCCACCACTTTCTGATAGG-30 50 -TAGCATGCATGAGCAAATTAAGA-30 50 -GCCACATGGTATGGGAGGTA-30 50 -TGTTACTTGTCTGGTGTGGCTAA-30 *Upper line shows sequences for telomeric breakpoint; lower line shows sequences for centromeric breakpoint.
deletions, when they occur, are diverse, so routine PCRbased methods are not efficient or routinely available. This is especially true in the United States, where no set of mutations predominates because of the ethnic diversity of this population. To provide a rapid inexpensive and efficient platform to identify common, previously described, and unique deletions of either the HBA or HBB loci, we developed a microarray-based comparative genomic hybridization (CGH) assay. This method can be used to genotype patients with deletional forms of a- or b-thalassemia, HPFH, or complex mixed HBA and HBB globin deletions, whose genotype is not apparent from protein-based studies, such as electrophoresis and high-performance liquid chromatography (HPLC). We found the use of this array to characterize both previously characterized and novel deletions of the HBB locus. This array has been used in patients with apparent thalassemia trait but no evidence of a-thalassemia and no increase in HbA2, providing diagnostic information that allows accurate genetic counseling.
Materials and Methods Patient Specimens and Diagnostic Studies
Q6
Q7
Patient specimens were sent to the University of Washington Red Cell Disorders laboratory for routine clinical testing. Clinical testing at the University of Washington, Helena, included qualitative Hb analysis by isoelectric focusing and HPLC (Bio-Rad Variant II). HbF and HbA2 were quantified by HPLC. Detection of HbH inclusions was performed by brilliant cresyl blue staining as previously described.2 Iron deficiency was assessed by determining the zinc protoporphyrin/heme ratio3; serum iron, ferritin, total iron binding capacity, and transferrin saturation were determined in a subset of cases. For samples collected at the University of Washington, DNA was isolated using a Corbett automated DNA extractor. PCR to detect HBA deletions was performed
2
as previously described.4 Additional specimens were obtained from Boston Children’s Hospital. All studies were approved by the University of Washington Human Subjects Division.
Microarray Design Custom arrays were designed essentially as described, including the software used for probe design and repetitive sequence masking.5 Genome coordinates used throughout this article are from the human hg38 assembly (December 2013). For chromosome 11, probes were tiled at a median density of 60 bp across an approximately 450-kbp region surrounding the HBB locus (coordinates chr11:5,022,271-5,472,133), and at a median density of 57 to 62 kbp on either side of this region for the remainder of chromosome 11 (coordinates chr11:218,3645,022,271 and chr11:5,472,133-134,998,453). For chromosome 16, probes were tiled at a median density of 64 bp across an approximately 400-kbp region surrounding the HBA loci (coordinates chr16:11,177-414,936) and a median density of 56,261 bp for the remainder of chromosome 16 (chr16:429,413-90,081,925). An additional 883 probes representing the other chromosomes (median spacing of 1.8 Mbp) were included for the purpose of signal normalization. The array design is accessible in the National Center for Biotechnology Information Gene Expression Omnibus (http://www. ncbi.nlm.nih.gov/geo; accession number GPL20136).
Array Hybridization, Scanning, and Data Analysis Subject genomic DNA was labeled with Cy5-dUTP, and normal human genomic DNA (Promega, Madison, WI) was labeled with Cy3-dUTP using the Agilent Genomic DNA Labeling kit PLUS (Agilent). Array hybridization, Q8 scanning, and data analysis were performed as described previously.5 The raw data are accessible at the National Center for Biotechnology Information Gene Expression
jmd.amjpathol.org
-
The Journal of Molecular Diagnostics
FLA 5.4.0 DTD JMDI458_proof 20 November 2015 8:03 pm EO: JMD15_0091
187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248
249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310
Array CGH for Globin Deletions Table 2 Q9
Clinical Data for Patients with HBB Deletions
Subject no.
Age
Sex
MCV
HbA2
HbF
Hb
1 2 3 4 5 6 7 8 9 10 11
2 months 20 months 26 years 25 years 25 years 36 years 30 years 29 years 12 years 4.5 years 2 months
F F F F F M F F M F F
93 76 82 69 79 83 65 72 60.5 69.1 66
0.9 3 16.5 ND 4.3 1.9 8.2 7 3 2.6 2.1
91.1 34.1 11.7 44.2 20.6 40.8 1.2 9.5 ND 16.8 26.3
11.3 11.6 13.1
16.6 10.2 11.6 12 7.4
Hct
35 35 50.6 34
36.8
Electrophoresis
Ethnicity
HbF, HbS HbS, HbF HbA, Hb Kenya HbE, HbF HbA, HbF, HbA2 HbA, HbF, HbA2
African African African Cambodian Vietnamese Sri Lankan Kurdish Irish/Romanian Belgian Sri Lankan Caucasian
HbA, HbA, HbA, HbA,
HbF, HbA2 HbA2 HbF, HbA2 HbF, HbA2
F, female; Hct, hematocrit; M, male; MCV, mean corpuscular volume; ND, not determined.
Omnibus (http://www.ncbi.nlm.nih.gov/geo; number GSE68587).
accession
Confirmation of Deletion Breakpoints PCR primers flanking breakpoints identified by CGH were designed using Primer 3 software version 1.1.0 (http://bioinfo. ut.ee/primer3, last accessed June 1, 2009). DNA sequencing was performed using standard BigDye Terminator reagents (Applied Biosystems, Foster City, CA) using the same PCR ½T1 primers (Table 1).
Results Deletions within the HBB Locus Subjects at the University of Washington were suspected of having a deletional form of b-thalassemia if they were microcytic without evidence of iron deficiency, had normal levels of HbA2, and had no evidence of a-thalassemia trait (eg, negative HbH inclusion body assay). Other subjects were identified because of unexplained elevations of HbF or unusually high HbA2. Hematologic findings are detailed in ½T2 Table 2. For array CGH, Cy5-labeled subject DNA was combined with Cy3-labeled reference DNA, and the labeled DNA was hybridized to a microarray containing oligonucleotide probes spanning the HBA and HBB loci. The density of probes varied (see Materials and Methods for genomic coordinates), with a high density of probes spanning structural genes of the loci and their upstream regulatory elements (approximately one probe per 60 to 64 bp) and a low density of probes in the flanking regions of the loci (approximately one probe per 56 to 57 kbp). Additional probes from the other chromosomes were included as normalization controls. Figure 1 shows the results of array CGH for four DNA ½F1 specimens from patients (subjects 1 to 4) with previously characterized HBB locus deletions (a generous gift of Dr. David H.K. Chui, Boston University, Boston, MA), including HPFH1 (subject 1) (Figure 1A), HPFH-2 (subject 2) (Figure 1B), and Hb Kenya (subject 3) (Figure 1C). The HPFH-1 deletion is
The Journal of Molecular Diagnostics
-
shown as a region of Cy5:Cy3 ratio of log2 1 extending from chromosome 11 coordinate chr11:5,153,221 within a cluster of olfactory receptor genes at the telomeric end of the locus to coordinate chr11:5,238,138 between the HBD and HBG1 genes (Figure 1A), resulting in deletion of the HBB and HBD genes. This deletion is most commonly found in populations of African ancestry and results in very high HbF levels (25% to 30%) in heterozygotes.6 The HPFH-2 deletion is very similar to the HFPH-1 deletion, with breakpoints just centromeric to the HPFH-1 breakpoints at both ends (telomeric breakpoint at chr11:5,158,438, centromeric breakpoint at chr11:5,242,749). The HPFH-2 deletion is also common in individuals of African ancestry, with HbF levels similar to those seen with HPFH-1.6 Hb Kenya (Figure 1C) is a deletion that excises sequences between the HBB and HBG2 genes (telomeric breakpoint chr11:5,226,631 to centromeric breakpoint at chr11:5,249,422), resulting in a HBG1-HBB fusion gene and protein.7 The abnormal fusion protein
Figure 1 Examples of array comparative genomic hybridization showing previously characterized HBB deletions. Deletions identified include hereditary persistence of fetal Hb (HPFH)-1 (subject 1, A), HPFH-2 (subject 2, B), and Hb Kenya (subject 3, C). Numbers at the bottom of the figure represent distance from the chromosome 11p telomere. The ratio of Cy5 to Cy3 is 1 in the absence of copy number change, reflected in a log2 of 0. Heterozygous deletions are identified by a 50% loss (log2 change of 1) of Cy5 signal in a group of contiguous probes. Homozygous deletions result in a log2 change of N.
jmd.amjpathol.org
FLA 5.4.0 DTD JMDI458_proof 20 November 2015 8:03 pm EO: JMD15_0091
3
Q10
311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372
Sabath et al 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395 396 397 398 399 400 401 402 403 404 405 406 Figure 2 Array comparative genomic hybridization used to characterize deletions in patient samples that had not been defined molecularly. These samples 407 include a deletional hereditary persistence of fetal Hb co-inherited with HbE trait in a Cambodian subject (subject 4, A), an HBB deletion in a Vietnamese 408 subject (subject 5, B), a deletion causing HPFH in a Sri Lankan individual that is identical to HPFH-3 (subject 6, C), a noncontiguous deletion removing the 409 HBBP1 and HBB genes while sparing the intervening HBD gene in a Kurdish individual (subject 7, D, expanded in inset), a novel deletion causing 410 db-thalassemia in an individual of Irish/Romanian ancestry (subject 8, E, expanded in inset), a deletion causing b-thalassemia that includes the b-globin locus control region and the HBE1, HBG1/2, and HBD genes that spares the linked HBB gene (subject 9, F), a deletion causing db-thalassemia in a Sri Lankan 411 individual (subject 10, G), and a very large (>2 Mbp) deletion causing b0-thalassemia with an inset showing an overview of the 11-Mb region containing the 412 deletion (subject 11, H). Numbers at the bottom of the figure represent distance from the chromosome 11p telomere. Note that the inset for H shows a 413 sequence span of approximately 10 Mb. 414 415 416 co-elutes with HbA2 on HPLC, explaining the elevated adult Cambodian patient with HbE, markedly elevated HbF, 417 level of HbA in subject 3; the intact HBG2 gene in cis is and no HbA, suggesting compound heterozygosity for HbE 418 2 419 expressed at an elevated level as well. and a HBB gene deletion (subject 4). The lack of anemia, 420 ½F2 Figure 2 shows the results obtained from subjects with very high HbF level, and MCV consistent with HbE trait 421 HPFH or deletional b-thalassemia whose HBB locus suggested a deletional HPFH rather than b0-thalassemia trait 422 deletions had not been characterized molecularly (subjects 4 on the non-E allele. This patient was found to have a 12.6-kbp 423 to 11). All these deletion breakpoints were sequenced by deletion from telomeric breakpoint chr11:5,220,623 to 424 designing PCR primers that flanked the breakpoints and centromeric breakpoint chr11:5,233,234, resulting in deletion 425 determining the sequence of the resulting PCR products of the HBB gene and a small segment of the 30 end of the HBD 426 using Sanger sequencing other than subject 11, whose gene. This deletion has been termed the Thai/Vietnamese 427 breakpoints occurred in sequences tiled at low density. (db)0 deletion.8 Figure 2B shows a 27.4-kbp deletion (telo428 Breakpoint sequence data for deletions not shown elsewhere meric breakpoint chr11:5,201,647 to centromeric breakpoint 429 can be found in Supplemental Figure S1 and Supplemental chr11:5,229,059) in a Vietnamese patient with an elevated 430 431 Table S1. A diagram showing the extent of the deletions HbF level, slightly elevated HbA2 level, and mild microcytosis 432 with respect to the globin genes on chromosome 11 is (subject 5) having the SEA HPFH deletion.9 The HBB gene is 433 ½F3 shown in Figure 3. Figure 2A is a DNA specimen from an deleted, but the HBD and HBG genes are intact, resulting in a 434
4
jmd.amjpathol.org
-
The Journal of Molecular Diagnostics
FLA 5.4.0 DTD JMDI458_proof 20 November 2015 8:03 pm EO: JMD15_0091
435 436 437 438 439 440 441 442 443 444 445 446 447 448 449 450 451 452 453 454 455 456 457 458 459 460 461 462 463 464 465 466 467 468 469 470 471 472 473 474 475 476 477 478 479 480 481 482 483 484 485 486 487 488 489 490 491 492 493 494 495 496
497 498 499 500 501 502 503 504 505 506 507 508 509 510 511 512 513 514 515 516 517 518 519 520 521 522 523 524 525 526 527 528 529 530 531 532 533 534 535 536 537 538 539 540 541 542 543 544 545 546 547 548 549 550 551 552 553 554 555 556 557 558
Array CGH for Globin Deletions
relative increase in both HBD and HBG expression. Figure 2C shows a 49.8-kbp HBB-HBD deletion (telomeric breakpoint chr11:5,194,453 to centromeric breakpoint chr11:5,244,225) consistent with HPFH-310,11 in a patient from Sri Lanka (subject 6). Figure 2D shows a complex deletion in a Kurdish family that was previously described but only partially characterized12 (subject 7). This is a discontinuous deletion that removes the HBB (deleted nucleotides chr11:5,223,9605,231,570) and HBBP1 (deleted nucleotides chr11:5,238,7595,246,317) genes but spares the HBD gene and results in microcytosis consistent with b-thalassemia trait but with more elevated HbA2 and normal HbF levels. Figure 2E shows the results from a patient of Irish/Romanian ancestry (subject 8) with a previously unreported 1518-bp deletion of the HBB promoter region (telomeric breakpoint chr11:5,225,894 to centromeric breakpoint chr11:5,227,412), resulting in a b-thalassemic phenotype with relatively high levels of both HBG and HBD. Figure 2F shows a 59.9-kbp deletion (telomeric breakpoint chr11:5,238,245 to centromeric breakpoint chr11:5,298,126) in a Belgian male with microcytosis and normal HbF level (subject 9). This deletion spares the HBB gene but removes the remaining globin genes and the locus control region. Figure 2G shows a 38.2-kbp deletion (telomeric breakpoint chr11:5,194,461 to centromeric breakpoint chr11:5,232,686) in a Sri Lankan patient, causing db-thalassemia with increased HbF (subject 10). Of note, this deletion shares a common breakpoint downstream of the locus with the HPFH-3 deletion but differs in the breakpoint within the locus, located immediately telomeric to the HBD gene and has been termed Sri Lankan (db)0 thalassemia.10 Finally, Figure 2H shows a deletion of >2 Mbp in a 2-month-old girl with microcytosis (subject 11). Because the resolution of the array is lower at large distances from the structural HBB genes, we were not able to precisely map the breakpoints of this deletion. However, the telomeric breakpoint is between chr11:4,109,643-4,188,151, and the centromeric breakpoint is between chr11:6,384,408-6,422,730. Of interest, this patient had early neurologic developmental abnormalities and white matter changes, raising the possibility that this large deletion is responsible for more than the hematologic changes observed.
The Journal of Molecular Diagnostics
-
559 560 561 562 563 564 Figure 3 Comparison of deletions for subjects with HBB locus deletions. Coordinates of chromo565 some 11 are shown at the bottom. Deletions not 566 previously described in the literature are indicated 567 by asterisks. 568 569 570 571 572 573 574 Deletions Causing a-Thalassemia 575 576 To determine the ability to characterize HBA deletions by 577 array CGH, DNA samples that had been characterized 578 previously by gap-PCR were studied using the globin array. 579 Figure 4 shows results from patients with cis (two-gene, ½F4 580 a-thalassemia-1) deletions. A sample heterozygous for the 581 582 Filipino deletion (aa/–FIL) yielded a 50% decrease in signal 583 spanning nucleotides chr16:150,821-182,671, including the 584 HBZ, the two HBA, and the HBQ1 genes (Figure 4A), 585 consistent with accurate mapping of the mutation. A 586 Southeast Asian deletion (aa/–SEA) sample similarly resul587 ted in the expected 50% decrease in signal from nucleotides 588 589 590 591 592 593 594 595 596 597 598 599 600 601 602 603 604 605 606 607 608 609 610 611 612 Figure 4 Array comparative genomic hybridization characterization of 613 HBA two-gene deletions in cis-causing a-thalassemia. These deletions 614 include heterozygosity for the common Filipino aa/–FIL deletion (A), 615 SEA heterozygosity for the common Southeast Asian deletion aa/– (B), and 616 SEA homozygosity for the – deletion (C). The arrows in B and C indicate the 617 SEA inter-z hypervariable region, which is shorter in individuals with the – 618 deletion than in individuals without the deletion. Numbers at the bottom 619 of the figure represent distance from the chromosome 16p telomere. 620
jmd.amjpathol.org
FLA 5.4.0 DTD JMDI458_proof 20 November 2015 8:03 pm EO: JMD15_0091
5
621 622 623 624 625 626 627 628 629 630 631 632 633 634 635 636 637 638 639 640 641 642 643 644 645 646 647 648 649 650 651 652 653 654 655 656 657 658 659 660 661 662 663 664 665 666 667 668 669 670 671 672 673 674 675 676 677 678 679 680 681 682
Sabath et al chr16:165,401-184,701, sparing the HBZ gene, and removing the HBM, the two HBA, and the HBQ genes (Figure 4B). Finally, a case of a homozygous –SEA deletion, associated with Hb Bart hydrops fetalis, reveals a complete loss of signal in the deleted region (Figure 4C). Of interest, the –SEA specimens (heterozygous and homozygous) had a shorter a-globin inter-z hypervariable region (HVR)13 when compared with the reference DNA, telomeric to the large deletion. This finding appears to be in linkage disequilibrium with the –SEA deletion because it is present in all –SEA specimens examined for this study. Similarly, the –FIL specimen appears to have a shorter a-globin 30 HVR14 than the reference DNA, seen as a decrease in signal just centromeric to the deleted region (Figure 4A) and the –SEA homozygous specimen apparently has a longer 30 HVR as evidenced by increased signal (Figure 4C). We also investigated the utility of this array to detect the common single HBA gene deletions. Interpretation of the plots is difficult due to the sequence homology between the two HBA genes because DNA from the remaining (nondeleted) HBA gene in cis hybridizes to the probes spanning the deleted HBA gene, as well as its own probes. Although it is not possible to determine the extent of these deletions from array CGH a priori, the patterns of signal loss for each deletion are reproducible. Examples are shown in Supplemental Figure S2.
CGH produces reproducible patterns for deletional athalassemia. Two other groups have reported using similar technology to identify HBA and HBB deletions,15,16 but sequencing of breakpoints was not reliable. In contrast, the higher resolution of our array has facilitated one-step PCR and sequencing when breakpoints occur in the high-density region of the array.
b-Globin Deletions
Molecular diagnosis of thalassemia is a useful adjunct to the more common techniques of Hb electrophoresis and HPLC when genetic counseling or prenatal diagnosis is a concern. It is especially important when standard methods cannot be used for a definitive diagnosis, for example, in the setting of b-thalassemia or HbE traits, where a-thalassemia can be masked, or in the setting of deletional b-thalassemia, where the characteristic increase in HbA2 is often not present. In this study, we found that array CGH can be accurately used to diagnose patients as having deletions of the HBA and HBB loci. We designed a custom oligonucleotide array representing 400- to 450-kbp genomic segments surrounding the HBA and HBB loci with a median probe spacing of 60 to 64 bp, enabling high-resolution mapping of deletions involving these loci. As a demonstration of the ability of this technique to map known deletions, we accurately map the 50 and 30 breakpoints of the previously characterized HPFH-1, HPFH-2, and Hb Kenya deletions (Figure 1). We then reveal that deletions that were mapped previously using lower-resolution DNA blots can be mapped at high resolution using the CGH array, allowing a definitive molecular characterization after the design of a single set of PCR primers and a single Sanger sequencing run. This technique allows the rapid identification of novel mutations and facilitates their rapid sequencing. Finally, we reveal that
Diagnosis of deletional b-thalassemia can be challenging. Most b-thalassemias are due to point mutations of the HBB structural or regulatory sequences and are routinely characterized by DNA sequencing.17 In populations where specific mutations are very common, methods are available to detect these mutations using allele-specific reagents.18e21 Similarly, gap-PCR methods, in which primers are used that flank specific deletion breakpoints, may be used to confirm specific mutations when suspected from less specific studies or family history.22 To detect unknown deletions, DNA blot analysis has been the standard approach but is not commonly used in clinical laboratories because it is labor intensive, low resolution, and slow. Quantitative PCR using primer sets spaced throughout the HBB locus followed by capillary electrophoresis has been reported for characterizing HBB deletions.23 More recently, multiplex ligation-dependent probe amplification has been used to characterize deletions of both the HBA and HBB loci.24e27 These latter techniques, although useful to reveal the presence of deletions, do not have the resolution available with the CGH method presented in this study. Our CGH approach was able to map previously uncharacterized HBB locus deletions with high enough resolution to design PCR primers that could amplify the DNA sequences flanking the deletion breakpoints and obtain definitive sequence analysis in a single run. This finding resulted in the identification and characterization of novel HPFH and (g) (d)b0-thalassemia deletions as well as definitive sequencing of prior mutations that were only imprecisely mapped. Subject 4, an adult Cambodian woman, was a compound heterozygote for HbE trait and bthalassemia with 44.2% HbF. CGH array analysis revealed chromosome 11 breakpoints that correspond to previously described Vietnamese/Laotian/Thai HPFH deletions.8,28,29 In a previously reported case,28 the HbF level was 21.3%. In our case, the HbF level is roughly twice that in previous reports. This can be accounted for by the fact that expression of bE is approximately half that of bA. Both subjects 6 and 10, of Sri Lankan ethnicity, had elevated HbF levels but with quite different phenotypes. Subject 6, a 36-year-old man, has a very high HbF level of 40.8% with low-normal MCV. CGH revealed the 49,772-bp HPFH-3 deletion.10 Subject 10, a 4.5-year-old girl, was microcytic with a moderately elevated HbF of 16.8% and a novel 38,225-bp deletion we define as Sri Lankan (db)0-thalassemia.10 The telomeric breakpoints were almost
6
jmd.amjpathol.org
Discussion
-
The Journal of Molecular Diagnostics
FLA 5.4.0 DTD JMDI458_proof 20 November 2015 8:03 pm EO: JMD15_0091
683 684 685 686 687 688 689 690 691 692 693 694 695 696 697 698 699 700 701 702 703 704 705 706 707 708 709 710 711 712 713 714 715 716 717 718 719 720 721 722 723 724 725 726 727 728 729 730 731 732 733 734 735 736 737 738 739 740 741 742 743 744
745 746 747 748 749 750 751 752 753 754 755 756 757 758 759 760 761 762 763 764 765 766 767 768 769 770 771 772 773 774 775 776 777 778 779 780 781 782 783 784 785 786 787 788 789 790 791 792 793 794 795 796 797 798 799 800 801 802 803 804 805 806
Array CGH for Globin Deletions identical for these two patients (chr11:5,194,453 for subject 6 versus chr11:5,194,461 for subject 10), whereas the centromeric breakpoints were 11,539 bp apart (chr11:5,244,225 for subject 6 versus chr11:5,232,686 for subject 10). The region between chr11:5,244,225 and chr11:5,232,686, which is immediately centromeric to the HBD gene, contains sequences that repress HBG expression based on the large difference in HbF levels between these two subjects, and chromatin immunoprecipitation has reveasled a binding site for the key HbF silencing factor BCL11A in this interval,30 along with binding sites for corepressors and repressive histone and chromatin marks.10 This site has also been associated with regulation of HBG expression in genome-wide association studies.31,32 Subject 7, a 30-year-old Kurdish woman, has b-thalassemia with typical microcytosis and a significantly elevated HbA2 level but no significant increase in HbF. CGH analysis revealed two noncontiguous deletions removing the HBB and HBBP1 genes but sparing the HBD gene. The retained sequences include this BCL11A binding site.10 Other mechanisms must exist for elevated HBG expression, however, because the deletions in subjects 4 and 5 do not remove the BCL11A binding site centromeric to the HBD gene, yet these subjects have elevated HbF levels of 44.2% and 20.6%, respectively. Perhaps the increased HbF in these subjects is due to a telomeric enhancer brought into proximity of the HBG genes. One other interesting feature of subject 10’s Sri Lankan (db)0 thalassemia deletion is that the HBD gene is intact, yet there is no increase in HbA2. The breakpoint occurs approximately 200 bp 30 (telomeric) to the HBD structural sequences and 30 untranslated region. This result suggests the presence of an HBD-positive regulatory element located somewhere telomeric to this breakpoint.
of >0.5 (Supplemental Figure S2). Standard gap-PCR is adequate for detecting the common HBA deletions leading to a-thalassemia,1 but the array CGH method still has potential utility to identify unusual deletions not tested for by gap-PCR and novel deletions of the HBA locus. Although we did not have samples to test, array CGH should also be able to identify and characterize triplicated HBA genes.
Conclusions We found that CGH using a high-density custom oligonucleotide array can be used to rapidly and efficiently define both previously described and unique deletions of the HBA and HBB loci. This method will be a useful adjunct to current diagnostic methods for patients suspected of having thalassemia where standard diagnostic techniques are inconclusive. The typical patient would be microcytic with normal iron studies and have a normal level of HbA2. If gap-PCR to detect the common HBA deletions is negative, an HBB or uncommon HBA deletion is likely. The lack of HbA2 elevation makes an HBB point mutation unlikely, so Sanger sequencing of HBB would not likely be diagnostic, and thus proceeding directly to array CGH would be the most efficient diagnostic pathway. The other clinical scenario where HBB deletion analysis would be useful is HPFH, where there is a large variety of deletions that are not amenable to gap-PCR, particularly in an ethnically diverse population. Next-generation sequencing should eventually replace CGH and most other genetic technologies for clinical testing, although currently CGH represents a less labor- and data-intensive way to characterize deletions, such as those in the globin loci, and thus could remain more accessible to laboratories with limited resources.
a-Globin Deletions
Acknowledgment
We also report the utility of array CGH for detecting HBA gene deletions. For subjects heterozygous for large deletions of the HBA locus, such as the –SEA and –FIL deletions, the expected 50% decrease in signal intensity is observed throughout the deleted region. CGH is also able to characterize some of the polymorphic repeat regions in the HBA locus, including the inter-z HVR, a polymorphic region on chromosome 16 that lies between the HBZ and HBZP1 genes13 and the 30 HVR centromeric to the HBQ gene.14 It appears that the inter-z HVR is smaller and the 30 HVR is larger for the –SEA allele than for the corresponding regions of the reference DNA used for the CGH study (Figure 4, A and B). The –FIL allele appears to have a shorter 30 HVR (Figure 4A). Identification of alleles lacking one of two HBA genes, such as the -a3.7 and -a4.2 deletions, is more difficult because the two HBA genes are highly homologous, and when only one is deleted, labeled DNA fragments from the nondeleted HBA genes (and their homologous pseudogenes) cross-hybridize to probes covering the deleted regions, resulting in a signal ratio of subject to reference DNA
HBB locus deletions were a generous gift of Dr. David H.K. Chui (Boston University, Boston, MA).
The Journal of Molecular Diagnostics
-
Supplemental Data Supplemental material for this article can be found at http://dx.doi.org/10.1016/j.jmoldx.2015.07.011.
References 1. Chong SS, Boehm CD, Higgs DR, Cutting GR: Single-tube multiplexPCR screen for common deletional determinants of a-thalassemia. Blood 2000, 95:360e362 2. Sabath DE, Cross ST, Mamiya LY: An improved method for detecting red cells with hemoglobin H inclusions that does not require glass capillary tubes. Clin Lab Haematol 2003, 25:87e91 3. Labbe RF, Rettmer RL: Zinc protoporphyrin: a product of irondeficient erythropoiesis. Semin Hematol 1989, 26:40e46 4. Chong SS, Boehm CD, Cutting GR, Higgs DR: Simplified multiplexPCR diagnosis of common Southeast Asian deletional determinants of a-thalassemia. Clin Chem 2000, 46:1692e1695
jmd.amjpathol.org
FLA 5.4.0 DTD JMDI458_proof 20 November 2015 8:03 pm EO: JMD15_0091
7
807 808 809 810 811 812 813 814 815 816 817 818 819 820 821 822 823 824 825 826 827 828 829 830 831 832 833 834 835 836 837 838 839 840 841 842 843 844 845 846 847 848 849 850 851 852 853 854 855 856 857 858 859 860 861 862 863 864 865 866 867 868
869 870 871 872 873 874 875 876 877 878 879 880 881 882 883 884 885 886 887 888 889 890 891 892 893 894 895 896 897 898 899 900 901 902 903 904 905 906 907 908 909 910 911 912 913 914 915 916 917 918 919 920 921 922 923 924 925 926 927 928 929 930
Sabath et al 5. Greisman HA, Hoffman NG, Yi HS: Rapid high-resolution mapping of balanced chromosomal rearrangements on tiling CGH arrays. J Mol Diagn 2011, 13:621e633 6. Tuan D, Feingold E, Newman M, Weissman SM, Forget BG: Different 30 end points of deletions causing db-thalassemia and hereditary persistence of fetal hemoglobin: implications for the control of gglobin gene expression in man. Proc Natl Acad Sci U S A 1983, 80: 6937e6941 7. Huisman TH, Wrightstone RN, Wilson JB, Schroeder WA, Kendall AG: Hemoglobin Kenya, the product of fusion of g and b polypeptide chains. Arch Biochem Biophys 1972, 153:850e853 8. Chalaow N, Thein SL, Viprakasit V: The 12.6 kb-deletion in the bglobin gene cluster is the known Thai/Vietnamese (db)0-thalassemia commonly found in Southeast Asia. Haematologica 2013, 98: e117ee118 9. Xu XM, Li ZQ, Liu ZY, Zhong XL, Zhao YZ, Mo QH: Molecular characterization and PCR detection of a deletional HPFH: application to rapid prenatal diagnosis for compound heterozygotes of this defect with b-thalassemia in a Chinese family. Am J Hematol 2000, 65: 183e188 10. Sankaran VG, Xu J, Byron R, Greisman HA, Fisher C, Weatherall DJ, Sabath DE, Groudine M, Orkin SH, Premawardhena A, Bender MA: A functional element necessary for fetal hemoglobin silencing. N Engl J Med 2011, 365:807e814 11. Henthorn PS, Mager DL, Huisman TH, Smithies O: A gene deletion ending within a complex array of repeated sequences 30 to the human b-globin gene cluster. Proc Natl Acad Sci U S A 1986, 83:5194e5198 12. Oner C, Oner R, Gurgey A, Altay C: A new Turkish type of b-thalassaemia major with homozygosity for two non-consecutive 7.6 kb deletions of the jb and b genes and an intact d gene. Br J Haematol 1995, 89:306e312 13. Goodbourn SE, Higgs DR, Clegg JB, Weatherall DJ: Molecular basis of length polymorphism in the human z-globin gene complex. Proc Natl Acad Sci U S A 1983, 80:5022e5026 14. Jarman AP, Nicholls RD, Weatherall DJ, Clegg JB, Higgs DR: Molecular characterisation of a hypervariable region downstream of the human a-globin gene cluster. EMBO J 1986, 5:1857e1863 15. Blattner A, Brunner-Agten S, Ludin K, Hergersberg M, Herklotz R, Huber AR, Röthlisberger B: Detection of germline rearrangements in patients with a- and b-thalassemia using high resolution array CGH. Blood Cells Mol Dis 2013, 51:39e47 16. Phylipsen M, Chaibunruang A, Vogelaar IP, Balak JR, Schaap RA, Ariyurek Y, Fucharoen S, den Dunnen JT, Giordano PC, Bakker E, Harteveld CL: Fine-tiling array CGH to improve diagnostics for a- and b-thalassemia rearrangements. Hum Mutat 2012, 33:272e280 17. Chan OT, Westover KD, Dietz L, Zehnder JL, Schrijver I: Comprehensive and efficient HBB mutation analysis for detection of b-hemoglobinopathies in a pan-ethnic population. Am J Clin Pathol 2010, 133:700e707 18. Shaji RV, Edison ES, Poonkuzhali B, Srivastava A, Chandy M: Rapid detection of b-globin gene mutations and polymorphisms by temporal temperature gradient gel electrophoresis. Clin Chem 2003, 49: 777e781 19. Vrettou C, Traeger-Synodinos J, Tzetis M, Malamis G, Kanavakis E: Rapid screening of multiple b-globin gene mutations by real-time PCR
8
20.
21.
22.
23.
24.
25.
26.
27.
28.
29. 30.
31.
32.
on the LightCycler: application to carrier screening and prenatal diagnosis of thalassemia syndromes. Clin Chem 2003, 49:769e776 Wang W, Kham SK, Yeo GH, Quah TC, Chong SS: Multiplex minisequencing screen for common Southeast Asian and Indian bthalassemia mutations. Clin Chem 2003, 49:209e218 Xiong F, Huang Q, Chen X, Zhou Y, Zhang X, Cai R, Chen Y, Xie J, Feng S, Wei X, Xiao Q, Zhang T, Luo S, Yang X, Hao Y, Qu Y, Li Q, Xu X: A melting curve analysisebased PCR assay for one-step genotyping of b-thalassemia mutations: a multicenter validation. J Mol Diagn 2011, 13:427e435 Craig JE, Barnetson RA, Prior J, Raven JL, Thein SL: Rapid detection of deletions causing db thalassemia and hereditary persistence of fetal hemoglobin by enzymatic amplification. Blood 1994, 83:1673e1682 De Andrade TG, Saad STO, Sonati MF, Costa FF: Simple fluorescent PCR method for detection of large deletions in the b-globin gene cluster. Am J Hematol 2003, 72:225e227 Colosimo A, Gatta V, Guida V, Leodori E, Foglietta E, Rinaldi S, Cappabianca MP, Amato A, Stuppia L, Dallapiccola B: Application of MLPA assay to characterize unsolved a-globin gene rearrangements. Blood Cells Mol Dis 2011, 46:139e144 Harteveld CL, Voskamp A, Phylipsen M, Akkermans N, den Dunnen JT, White SJ, Giordano PC: Nine unknown rearrangements in 16p13.3 and 11p15.4 causing a- and b-thalassaemia characterised by high resolution multiplex ligation-dependent probe amplification. J Med Genet 2005, 42:922e931 Kipp BR, Roellinger SE, Lundquist PA, Highsmith WE, Dawson DB: Development and clinical implementation of a combination deletion PCR and multiplex ligation-dependent probe amplification assay for detecting deletions involving the human a-globin gene cluster. J Mol Diagn 2011, 13:549e557 Phylipsen M, Vogelaar IP, Schaap RA, Arkesteijn SG, Boxma GL, van Helden WC, Wildschut IC, de Bruin-Roest AC, Giordano PC, Harteveld CL: A new a0-thalassemia deletion found in a Dutch family (eAW). Blood Cells Mol Dis 2010, 45:133e135 Ghedira ES, Lecerf L, Faubert E, Costes B, Moradkhani K, Bachir D, Galacteros F, Pissard S: Estimation of the difference in HbF expression due to loss of the 50 d-globin BCL11A binding region. Haematologica 2013, 98:305e308 Trent RJ, Svirklys L, Jones P: Thai (db)0-thalassemia and its interaction with g-thalassemia. Hemoglobin 1988, 12:101e114 Sankaran VG, Menne TF, Xu J, Akie TE, Lettre G, Van Handel B, Mikkola HK, Hirschhorn JN, Cantor AB, Orkin SH: Human fetal hemoglobin expression is regulated by the developmental stagespecific repressor BCL11A. Science 2008, 322:1839e1842 Galarneau G, Palmer CD, Sankaran VG, Orkin SH, Hirschhorn JN, Lettre G: Fine-mapping at three loci known to affect fetal hemoglobin levels explains additional genetic variation. Nat Genet 2010, 42: 1049e1051 Farrell JJ, Sherva RM, Chen ZY, Luo HY, Chu BF, Ha SY, Li CK, Lee AC, Li RC, Li CK, Yuen HL, So JC, Ma ES, Chan LC, Chan V, Sebastiani P, Farrer LA, Baldwin CT, Steinberg MH, Chui DH: A 3-bp deletion in the HBS1L-MYB intergenic region on chromosome 6q23 is associated with HbF expression. Blood 2011, 117: 4935e4945
jmd.amjpathol.org
-
The Journal of Molecular Diagnostics
FLA 5.4.0 DTD JMDI458_proof 20 November 2015 8:03 pm EO: JMD15_0091
931 932 933 934 935 936 937 938 939 940 941 942 943 944 945 946 947 948 949 950 951 952 953 954 955 956 957 958 959 960 961 962 963 964 965 966 967 968 969 970 971 972 973 974 975 976 977 978 979 980 981 982 983 984 985 986 987 988 989 990 991 992
993 994 995 996 997 998 999 1000 1001 1002 1003 1004 1005 1006 1007 1008
Supplemental Figure S1 Sanger sequencing chromatograms of the breakpoints of b-globin deletions characterized by array comparative genomic hybridization (CGH) that had not been finely mapped previously. PCR primers were designed based on the breakpoints as determined by array CGH. The resulting PCR products were sequenced by Sanger sequencing. Vertical black lines indicate where the breakpoints occur. Orange bars indicate sequences that are not present in the hg38 reference human germline sequence. Tracings are for subject 4 (A, Thai/Vietnamese (db)0 thalassemia), subject 5 (B, SEA-HPFH), subject 7 (C, Kurdish deletion; note that there are two breakpoints for this deletion shown), subject 8 (D, Irish/Romanian), and subject 9 (E, Belgian). Sequence data for subjects 6 and 10 can be found in the supplemental material for reference 5. Because the breakpoints for the 2.1-Mb deletion (subject 11) occur in low-density tiled regions of chromosome 11, it was not possible to design PCR primers and sequence the breakpoints. Supplemental Figure S2 Array comparative genomic hybridization (CGH) characterization of HBA single-gene deletions causing a-thalassemia. These include heterozygosity for the ea3.7 deletion (A), heterozygosity for the ea4.7 deletion (B), homozygosity for the ea3.7 deletion (C), and compound heterozygosity for the ea3.7 and ea4.7 deletions (D). Numbers at the bottom of the figure represent distance from the chromosome 16p telomere. In DNA samples heterozygous for either the ea3.7 or the ea4.2 deletion, a loss of signal of <50% can be seen in the region of the HBA genes (A and B, respectively). With DNA containing a homozygous ea3.7 deletion, the loss in signal is closer to the expected 50% in the stretch spanning nucleotides chr16:173,301177,104, including the 30 end of the HBA2 gene to the 50 end of the HBA1 gene (C). Again, although there is loss of genetic material from both alleles in this region, there is only effectively a 50% loss because there are two remaining a-globin genes, which hybridize to the probes for the deleted region. A similar result is obtained with DNA from a compound heterozygote for ea3.7 and ea4.2 (D).
FLA 5.4.0 DTD JMDI458_proof 20 November 2015 8:03 pm EO: JMD15_0091
1009 1010 1011 1012 1013 1014 1015 1016 1017 1018 1019 1020 1021 1022 1023 1024
The Journal of Molecular Diagnostics, Vol. -, No. -, - 2015
jmd.amjpathol.org
Characterization of Deletions of the HBA and HBB Loci by Array Comparative Genomic Hybridization Q11 Q1
Daniel E. Sabath,* M.A. Bender,yz Vijay G. Sankaran,x Esther Vamos,{ Alex Kentsis,k Hye-Son Yi,* and Harvey A. Greisman* From the Departments of Laboratory Medicine* and Pediatrics,y University of Washington, Seattle, Washington; the Clinical Research Division,z Fred Hutchinson Cancer Research Center, Seattle, Washington; the Division of Hematology/Oncology,x Boston Children’s Hospital and Department of Pediatric Oncology, Dana-Farber Cancer Institute, Harvard Medical School, Boston, Massachusetts; the Departments of Pediatrics and Genetics,{ Université Libre de Bruxelles, Brussels, Belgium; and the Department of Pediatrics,k Memorial Sloan Kettering Cancer Center, New York, New York Accepted for publication July 28, 2015.
Q4
Q5
Address correspondence to Daniel E. Sabath, M.D., Ph.D., and Harvey A. Greisman, M.D., Ph.D., Department of Laboratory Medicine, University of Washington, Box 359743, Harborview Medical Center, 3NJ345.1, 325 Ninth Ave., Seattle, WA 98104 E-mail: dsabath@ uw.edu or hagreisman@gmail. com.
Thalassemia is among the most common genetic diseases worldwide. a-Thalassemia is usually caused by deletion of one or more of the duplicated HBA genes on chromosome 16. In contrast, most b-thalassemia results from point mutations that decrease or eliminate expression of the HBB gene on chromosome 11. Deletions within the HBB locus result in thalassemia or hereditary persistence of fetal Hb. Although routine diagnostic testing cannot distinguish thalassemia deletions from point mutations, deletional hereditary persistence of fetal Hb is notable for having an elevated HbF level with a normal mean corpuscular volume. A small number of deletions accounts for most a-thalassemias; in contrast, there are no predominant HBB deletions causing b-thalassemia. To facilitate the identification and characterization of deletions of the HBA and HBB globin loci, we performed array-based comparative genomic hybridization using a custom oligonucleotide microarray. We accurately mapped the breakpoints of known and previously uncharacterized HBB deletions defining previously uncharacterized deletion breakpoints by PCR amplification and sequencing. The array also successfully identified the common HBA deletions –SEA and –FIL. In summary, comparative genomic hybridization can be used to characterize deletions of the HBA and HBB loci, allowing high-resolution characterization of novel deletions that are not readily detected by PCR-based methods. (J Mol Diagn 2015, -: 1e8; http://dx.doi.org/10.1016/j.jmoldx.2015.07.011)
Thalassemia is one of the most common genetic diseases and is responsible for significant morbidity worldwide. a-Thalassemia is most commonly caused by deletion of one or more of the duplicated HBA genes on chromosome 16. In contrast, most b-thalassemia results from point mutations that cripple or completely inactivate expression from the HBB gene on chromosome 11. These mutations commonly result in a decrease in HBB and a relative increase in HBD protein expression, resulting in an elevated HbA2 level. There is frequently a relative increase in HBG1/2 expression as well, with a corresponding increase in HbF. Deletions within the HBB locus occur and can eliminate HBB expression. If the HBD and HBG genes are deleted along with the HBB gene, the characteristic increases in HbA2 and HbF will not be seen, making it impossible to distinguish from a-thalassemia on the basis of HbA2 quantification alone. Deletions resulting in levels of HBD and HBG polypeptide production insufficient to balance that
from the HBA locus result in db- or gdb-thalassemia with a decreased mean corpuscular volume (MCV). In contrast, some deletions result in elevated levels of HBG sufficient to balance that from the HBA locus, resulting in hereditary persistence of fetal Hb (HPFH), the hallmark of which is a normal MCV. In both of these cases, an HBB locus deletion is assumed. HBB deletions causing thalassemia may be more common than generally thought because they may not be recognized during routine thalassemia diagnostic procedures, especially in the rare deletions that do not result in HbF elevation. Several well-characterized deletions account for most a-thalassemia; thus, gap-PCRebased methods are routinely used to identify HBA gene deletions.1 In contrast, HBB Supported by NIH grant DK44746 (M.A.B.) and funds from the Q2 Department of Laboratory Medicine, University of Washington. Q3 Disclosures: None declared.
Copyright ª 2015 American Society for Investigative Pathology and the Association for Molecular Pathology. Published by Elsevier Inc. All rights reserved. http://dx.doi.org/10.1016/j.jmoldx.2015.07.011
FLA 5.4.0 DTD JMDI458_proof 20 November 2015 8:03 pm EO: JMD15_0091
63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124
125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186
Sabath et al Table 1
PCR and Sequencing Primers Used in This Study
Subject no. Forward PCR primer 4 5 7*
8 9
0
Reverse PCR primer 0
Additional sequencing primers (if used)
0
0
5 -TAGATCCCTTTGCCATTATG-3 50 -AGCCTCATGGTAGCAGAATC-30 50 -CCTTTTTCTTGTGGTAAATGCTT-30
5 -TTGGGTTTCTGATAGGCACTG-3 50 -TGGTATCTGCAGCAGTTGCC-30 50 -TTTCCTTGTGTTTGAAAGTGCT-30
50 -ATGCCAGTGCTCTCCACAAT-30
50 -ATAATAAGCCTGCGCCCTTC-30
50 -ACCCTTTGAGTAATAGTTTCCTGA-30 (forward) 50 -GCAAATAAGCACACATATATTCCAA-30 (reverse) 50 -AGTTTCTTGTTTACTCTGGA-30 (reverse) 50 -GGCCATATAGGGTAACTTTCTGAC-30 (reverse) 50 -AAAAGTGTGCCATGGTTTTAATG-30 (reverse)
50 -CATCCACCACTTTCTGATAGG-30 50 -TAGCATGCATGAGCAAATTAAGA-30 50 -GCCACATGGTATGGGAGGTA-30 50 -TGTTACTTGTCTGGTGTGGCTAA-30 *Upper line shows sequences for telomeric breakpoint; lower line shows sequences for centromeric breakpoint.
deletions, when they occur, are diverse, so routine PCRbased methods are not efficient or routinely available. This is especially true in the United States, where no set of mutations predominates because of the ethnic diversity of this population. To provide a rapid inexpensive and efficient platform to identify common, previously described, and unique deletions of either the HBA or HBB loci, we developed a microarray-based comparative genomic hybridization (CGH) assay. This method can be used to genotype patients with deletional forms of a- or b-thalassemia, HPFH, or complex mixed HBA and HBB globin deletions, whose genotype is not apparent from protein-based studies, such as electrophoresis and high-performance liquid chromatography (HPLC). We found the use of this array to characterize both previously characterized and novel deletions of the HBB locus. This array has been used in patients with apparent thalassemia trait but no evidence of a-thalassemia and no increase in HbA2, providing diagnostic information that allows accurate genetic counseling.
Materials and Methods Patient Specimens and Diagnostic Studies
Q6
Q7
Patient specimens were sent to the University of Washington Red Cell Disorders laboratory for routine clinical testing. Clinical testing at the University of Washington, Helena, included qualitative Hb analysis by isoelectric focusing and HPLC (Bio-Rad Variant II). HbF and HbA2 were quantified by HPLC. Detection of HbH inclusions was performed by brilliant cresyl blue staining as previously described.2 Iron deficiency was assessed by determining the zinc protoporphyrin/heme ratio3; serum iron, ferritin, total iron binding capacity, and transferrin saturation were determined in a subset of cases. For samples collected at the University of Washington, DNA was isolated using a Corbett automated DNA extractor. PCR to detect HBA deletions was performed
2
as previously described.4 Additional specimens were obtained from Boston Children’s Hospital. All studies were approved by the University of Washington Human Subjects Division.
Microarray Design Custom arrays were designed essentially as described, including the software used for probe design and repetitive sequence masking.5 Genome coordinates used throughout this article are from the human hg38 assembly (December 2013). For chromosome 11, probes were tiled at a median density of 60 bp across an approximately 450-kbp region surrounding the HBB locus (coordinates chr11:5,022,271-5,472,133), and at a median density of 57 to 62 kbp on either side of this region for the remainder of chromosome 11 (coordinates chr11:218,3645,022,271 and chr11:5,472,133-134,998,453). For chromosome 16, probes were tiled at a median density of 64 bp across an approximately 400-kbp region surrounding the HBA loci (coordinates chr16:11,177-414,936) and a median density of 56,261 bp for the remainder of chromosome 16 (chr16:429,413-90,081,925). An additional 883 probes representing the other chromosomes (median spacing of 1.8 Mbp) were included for the purpose of signal normalization. The array design is accessible in the National Center for Biotechnology Information Gene Expression Omnibus (http://www. ncbi.nlm.nih.gov/geo; accession number GPL20136).
Array Hybridization, Scanning, and Data Analysis Subject genomic DNA was labeled with Cy5-dUTP, and normal human genomic DNA (Promega, Madison, WI) was labeled with Cy3-dUTP using the Agilent Genomic DNA Labeling kit PLUS (Agilent). Array hybridization, Q8 scanning, and data analysis were performed as described previously.5 The raw data are accessible at the National Center for Biotechnology Information Gene Expression
jmd.amjpathol.org
-
The Journal of Molecular Diagnostics
FLA 5.4.0 DTD JMDI458_proof 20 November 2015 8:03 pm EO: JMD15_0091
187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248
249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310
Array CGH for Globin Deletions Table 2 Q9
Clinical Data for Patients with HBB Deletions
Subject no.
Age
Sex
MCV
HbA2
HbF
Hb
1 2 3 4 5 6 7 8 9 10 11
2 months 20 months 26 years 25 years 25 years 36 years 30 years 29 years 12 years 4.5 years 2 months
F F F F F M F F M F F
93 76 82 69 79 83 65 72 60.5 69.1 66
0.9 3 16.5 ND 4.3 1.9 8.2 7 3 2.6 2.1
91.1 34.1 11.7 44.2 20.6 40.8 1.2 9.5 ND 16.8 26.3
11.3 11.6 13.1
16.6 10.2 11.6 12 7.4
Hct
35 35 50.6 34
36.8
Electrophoresis
Ethnicity
HbF, HbS HbS, HbF HbA, Hb Kenya HbE, HbF HbA, HbF, HbA2 HbA, HbF, HbA2
African African African Cambodian Vietnamese Sri Lankan Kurdish Irish/Romanian Belgian Sri Lankan Caucasian
HbA, HbA, HbA, HbA,
HbF, HbA2 HbA2 HbF, HbA2 HbF, HbA2
F, female; Hct, hematocrit; M, male; MCV, mean corpuscular volume; ND, not determined.
Omnibus (http://www.ncbi.nlm.nih.gov/geo; number GSE68587).
accession
Confirmation of Deletion Breakpoints PCR primers flanking breakpoints identified by CGH were designed using Primer 3 software version 1.1.0 (http://bioinfo. ut.ee/primer3, last accessed June 1, 2009). DNA sequencing was performed using standard BigDye Terminator reagents (Applied Biosystems, Foster City, CA) using the same PCR ½T1 primers (Table 1).
Results Deletions within the HBB Locus Subjects at the University of Washington were suspected of having a deletional form of b-thalassemia if they were microcytic without evidence of iron deficiency, had normal levels of HbA2, and had no evidence of a-thalassemia trait (eg, negative HbH inclusion body assay). Other subjects were identified because of unexplained elevations of HbF or unusually high HbA2. Hematologic findings are detailed in ½T2 Table 2. For array CGH, Cy5-labeled subject DNA was combined with Cy3-labeled reference DNA, and the labeled DNA was hybridized to a microarray containing oligonucleotide probes spanning the HBA and HBB loci. The density of probes varied (see Materials and Methods for genomic coordinates), with a high density of probes spanning structural genes of the loci and their upstream regulatory elements (approximately one probe per 60 to 64 bp) and a low density of probes in the flanking regions of the loci (approximately one probe per 56 to 57 kbp). Additional probes from the other chromosomes were included as normalization controls. Figure 1 shows the results of array CGH for four DNA ½F1 specimens from patients (subjects 1 to 4) with previously characterized HBB locus deletions (a generous gift of Dr. David H.K. Chui, Boston University, Boston, MA), including HPFH1 (subject 1) (Figure 1A), HPFH-2 (subject 2) (Figure 1B), and Hb Kenya (subject 3) (Figure 1C). The HPFH-1 deletion is
The Journal of Molecular Diagnostics
-
shown as a region of Cy5:Cy3 ratio of log2 1 extending from chromosome 11 coordinate chr11:5,153,221 within a cluster of olfactory receptor genes at the telomeric end of the locus to coordinate chr11:5,238,138 between the HBD and HBG1 genes (Figure 1A), resulting in deletion of the HBB and HBD genes. This deletion is most commonly found in populations of African ancestry and results in very high HbF levels (25% to 30%) in heterozygotes.6 The HPFH-2 deletion is very similar to the HFPH-1 deletion, with breakpoints just centromeric to the HPFH-1 breakpoints at both ends (telomeric breakpoint at chr11:5,158,438, centromeric breakpoint at chr11:5,242,749). The HPFH-2 deletion is also common in individuals of African ancestry, with HbF levels similar to those seen with HPFH-1.6 Hb Kenya (Figure 1C) is a deletion that excises sequences between the HBB and HBG2 genes (telomeric breakpoint chr11:5,226,631 to centromeric breakpoint at chr11:5,249,422), resulting in a HBG1-HBB fusion gene and protein.7 The abnormal fusion protein
Figure 1 Examples of array comparative genomic hybridization showing previously characterized HBB deletions. Deletions identified include hereditary persistence of fetal Hb (HPFH)-1 (subject 1, A), HPFH-2 (subject 2, B), and Hb Kenya (subject 3, C). Numbers at the bottom of the figure represent distance from the chromosome 11p telomere. The ratio of Cy5 to Cy3 is 1 in the absence of copy number change, reflected in a log2 of 0. Heterozygous deletions are identified by a 50% loss (log2 change of 1) of Cy5 signal in a group of contiguous probes. Homozygous deletions result in a log2 change of N.
jmd.amjpathol.org
FLA 5.4.0 DTD JMDI458_proof 20 November 2015 8:03 pm EO: JMD15_0091
3
Q10
311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372
Sabath et al 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395 396 397 398 399 400 401 402 403 404 405 406 Figure 2 Array comparative genomic hybridization used to characterize deletions in patient samples that had not been defined molecularly. These samples 407 include a deletional hereditary persistence of fetal Hb co-inherited with HbE trait in a Cambodian subject (subject 4, A), an HBB deletion in a Vietnamese 408 subject (subject 5, B), a deletion causing HPFH in a Sri Lankan individual that is identical to HPFH-3 (subject 6, C), a noncontiguous deletion removing the 409 HBBP1 and HBB genes while sparing the intervening HBD gene in a Kurdish individual (subject 7, D, expanded in inset), a novel deletion causing 410 db-thalassemia in an individual of Irish/Romanian ancestry (subject 8, E, expanded in inset), a deletion causing b-thalassemia that includes the b-globin locus control region and the HBE1, HBG1/2, and HBD genes that spares the linked HBB gene (subject 9, F), a deletion causing db-thalassemia in a Sri Lankan 411 individual (subject 10, G), and a very large (>2 Mbp) deletion causing b0-thalassemia with an inset showing an overview of the 11-Mb region containing the 412 deletion (subject 11, H). Numbers at the bottom of the figure represent distance from the chromosome 11p telomere. Note that the inset for H shows a 413 sequence span of approximately 10 Mb. 414 415 416 co-elutes with HbA2 on HPLC, explaining the elevated adult Cambodian patient with HbE, markedly elevated HbF, 417 level of HbA in subject 3; the intact HBG2 gene in cis is and no HbA, suggesting compound heterozygosity for HbE 418 2 419 expressed at an elevated level as well. and a HBB gene deletion (subject 4). The lack of anemia, 420 ½F2 Figure 2 shows the results obtained from subjects with very high HbF level, and MCV consistent with HbE trait 421 HPFH or deletional b-thalassemia whose HBB locus suggested a deletional HPFH rather than b0-thalassemia trait 422 deletions had not been characterized molecularly (subjects 4 on the non-E allele. This patient was found to have a 12.6-kbp 423 to 11). All these deletion breakpoints were sequenced by deletion from telomeric breakpoint chr11:5,220,623 to 424 designing PCR primers that flanked the breakpoints and centromeric breakpoint chr11:5,233,234, resulting in deletion 425 determining the sequence of the resulting PCR products of the HBB gene and a small segment of the 30 end of the HBD 426 using Sanger sequencing other than subject 11, whose gene. This deletion has been termed the Thai/Vietnamese 427 breakpoints occurred in sequences tiled at low density. (db)0 deletion.8 Figure 2B shows a 27.4-kbp deletion (telo428 Breakpoint sequence data for deletions not shown elsewhere meric breakpoint chr11:5,201,647 to centromeric breakpoint 429 can be found in Supplemental Figure S1 and Supplemental chr11:5,229,059) in a Vietnamese patient with an elevated 430 431 Table S1. A diagram showing the extent of the deletions HbF level, slightly elevated HbA2 level, and mild microcytosis 432 with respect to the globin genes on chromosome 11 is (subject 5) having the SEA HPFH deletion.9 The HBB gene is 433 ½F3 shown in Figure 3. Figure 2A is a DNA specimen from an deleted, but the HBD and HBG genes are intact, resulting in a 434
4
jmd.amjpathol.org
-
The Journal of Molecular Diagnostics
FLA 5.4.0 DTD JMDI458_proof 20 November 2015 8:03 pm EO: JMD15_0091
435 436 437 438 439 440 441 442 443 444 445 446 447 448 449 450 451 452 453 454 455 456 457 458 459 460 461 462 463 464 465 466 467 468 469 470 471 472 473 474 475 476 477 478 479 480 481 482 483 484 485 486 487 488 489 490 491 492 493 494 495 496
497 498 499 500 501 502 503 504 505 506 507 508 509 510 511 512 513 514 515 516 517 518 519 520 521 522 523 524 525 526 527 528 529 530 531 532 533 534 535 536 537 538 539 540 541 542 543 544 545 546 547 548 549 550 551 552 553 554 555 556 557 558
Array CGH for Globin Deletions
relative increase in both HBD and HBG expression. Figure 2C shows a 49.8-kbp HBB-HBD deletion (telomeric breakpoint chr11:5,194,453 to centromeric breakpoint chr11:5,244,225) consistent with HPFH-310,11 in a patient from Sri Lanka (subject 6). Figure 2D shows a complex deletion in a Kurdish family that was previously described but only partially characterized12 (subject 7). This is a discontinuous deletion that removes the HBB (deleted nucleotides chr11:5,223,9605,231,570) and HBBP1 (deleted nucleotides chr11:5,238,7595,246,317) genes but spares the HBD gene and results in microcytosis consistent with b-thalassemia trait but with more elevated HbA2 and normal HbF levels. Figure 2E shows the results from a patient of Irish/Romanian ancestry (subject 8) with a previously unreported 1518-bp deletion of the HBB promoter region (telomeric breakpoint chr11:5,225,894 to centromeric breakpoint chr11:5,227,412), resulting in a b-thalassemic phenotype with relatively high levels of both HBG and HBD. Figure 2F shows a 59.9-kbp deletion (telomeric breakpoint chr11:5,238,245 to centromeric breakpoint chr11:5,298,126) in a Belgian male with microcytosis and normal HbF level (subject 9). This deletion spares the HBB gene but removes the remaining globin genes and the locus control region. Figure 2G shows a 38.2-kbp deletion (telomeric breakpoint chr11:5,194,461 to centromeric breakpoint chr11:5,232,686) in a Sri Lankan patient, causing db-thalassemia with increased HbF (subject 10). Of note, this deletion shares a common breakpoint downstream of the locus with the HPFH-3 deletion but differs in the breakpoint within the locus, located immediately telomeric to the HBD gene and has been termed Sri Lankan (db)0 thalassemia.10 Finally, Figure 2H shows a deletion of >2 Mbp in a 2-month-old girl with microcytosis (subject 11). Because the resolution of the array is lower at large distances from the structural HBB genes, we were not able to precisely map the breakpoints of this deletion. However, the telomeric breakpoint is between chr11:4,109,643-4,188,151, and the centromeric breakpoint is between chr11:6,384,408-6,422,730. Of interest, this patient had early neurologic developmental abnormalities and white matter changes, raising the possibility that this large deletion is responsible for more than the hematologic changes observed.
The Journal of Molecular Diagnostics
-
559 560 561 562 563 564 Figure 3 Comparison of deletions for subjects with HBB locus deletions. Coordinates of chromo565 some 11 are shown at the bottom. Deletions not 566 previously described in the literature are indicated 567 by asterisks. 568 569 570 571 572 573 574 Deletions Causing a-Thalassemia 575 576 To determine the ability to characterize HBA deletions by 577 array CGH, DNA samples that had been characterized 578 previously by gap-PCR were studied using the globin array. 579 Figure 4 shows results from patients with cis (two-gene, ½F4 580 a-thalassemia-1) deletions. A sample heterozygous for the 581 582 Filipino deletion (aa/–FIL) yielded a 50% decrease in signal 583 spanning nucleotides chr16:150,821-182,671, including the 584 HBZ, the two HBA, and the HBQ1 genes (Figure 4A), 585 consistent with accurate mapping of the mutation. A 586 Southeast Asian deletion (aa/–SEA) sample similarly resul587 ted in the expected 50% decrease in signal from nucleotides 588 589 590 591 592 593 594 595 596 597 598 599 600 601 602 603 604 605 606 607 608 609 610 611 612 Figure 4 Array comparative genomic hybridization characterization of 613 HBA two-gene deletions in cis-causing a-thalassemia. These deletions 614 include heterozygosity for the common Filipino aa/–FIL deletion (A), 615 SEA heterozygosity for the common Southeast Asian deletion aa/– (B), and 616 SEA homozygosity for the – deletion (C). The arrows in B and C indicate the 617 SEA inter-z hypervariable region, which is shorter in individuals with the – 618 deletion than in individuals without the deletion. Numbers at the bottom 619 of the figure represent distance from the chromosome 16p telomere. 620
jmd.amjpathol.org
FLA 5.4.0 DTD JMDI458_proof 20 November 2015 8:03 pm EO: JMD15_0091
5
621 622 623 624 625 626 627 628 629 630 631 632 633 634 635 636 637 638 639 640 641 642 643 644 645 646 647 648 649 650 651 652 653 654 655 656 657 658 659 660 661 662 663 664 665 666 667 668 669 670 671 672 673 674 675 676 677 678 679 680 681 682
Sabath et al chr16:165,401-184,701, sparing the HBZ gene, and removing the HBM, the two HBA, and the HBQ genes (Figure 4B). Finally, a case of a homozygous –SEA deletion, associated with Hb Bart hydrops fetalis, reveals a complete loss of signal in the deleted region (Figure 4C). Of interest, the –SEA specimens (heterozygous and homozygous) had a shorter a-globin inter-z hypervariable region (HVR)13 when compared with the reference DNA, telomeric to the large deletion. This finding appears to be in linkage disequilibrium with the –SEA deletion because it is present in all –SEA specimens examined for this study. Similarly, the –FIL specimen appears to have a shorter a-globin 30 HVR14 than the reference DNA, seen as a decrease in signal just centromeric to the deleted region (Figure 4A) and the –SEA homozygous specimen apparently has a longer 30 HVR as evidenced by increased signal (Figure 4C). We also investigated the utility of this array to detect the common single HBA gene deletions. Interpretation of the plots is difficult due to the sequence homology between the two HBA genes because DNA from the remaining (nondeleted) HBA gene in cis hybridizes to the probes spanning the deleted HBA gene, as well as its own probes. Although it is not possible to determine the extent of these deletions from array CGH a priori, the patterns of signal loss for each deletion are reproducible. Examples are shown in Supplemental Figure S2.
CGH produces reproducible patterns for deletional athalassemia. Two other groups have reported using similar technology to identify HBA and HBB deletions,15,16 but sequencing of breakpoints was not reliable. In contrast, the higher resolution of our array has facilitated one-step PCR and sequencing when breakpoints occur in the high-density region of the array.
b-Globin Deletions
Molecular diagnosis of thalassemia is a useful adjunct to the more common techniques of Hb electrophoresis and HPLC when genetic counseling or prenatal diagnosis is a concern. It is especially important when standard methods cannot be used for a definitive diagnosis, for example, in the setting of b-thalassemia or HbE traits, where a-thalassemia can be masked, or in the setting of deletional b-thalassemia, where the characteristic increase in HbA2 is often not present. In this study, we found that array CGH can be accurately used to diagnose patients as having deletions of the HBA and HBB loci. We designed a custom oligonucleotide array representing 400- to 450-kbp genomic segments surrounding the HBA and HBB loci with a median probe spacing of 60 to 64 bp, enabling high-resolution mapping of deletions involving these loci. As a demonstration of the ability of this technique to map known deletions, we accurately map the 50 and 30 breakpoints of the previously characterized HPFH-1, HPFH-2, and Hb Kenya deletions (Figure 1). We then reveal that deletions that were mapped previously using lower-resolution DNA blots can be mapped at high resolution using the CGH array, allowing a definitive molecular characterization after the design of a single set of PCR primers and a single Sanger sequencing run. This technique allows the rapid identification of novel mutations and facilitates their rapid sequencing. Finally, we reveal that
Diagnosis of deletional b-thalassemia can be challenging. Most b-thalassemias are due to point mutations of the HBB structural or regulatory sequences and are routinely characterized by DNA sequencing.17 In populations where specific mutations are very common, methods are available to detect these mutations using allele-specific reagents.18e21 Similarly, gap-PCR methods, in which primers are used that flank specific deletion breakpoints, may be used to confirm specific mutations when suspected from less specific studies or family history.22 To detect unknown deletions, DNA blot analysis has been the standard approach but is not commonly used in clinical laboratories because it is labor intensive, low resolution, and slow. Quantitative PCR using primer sets spaced throughout the HBB locus followed by capillary electrophoresis has been reported for characterizing HBB deletions.23 More recently, multiplex ligation-dependent probe amplification has been used to characterize deletions of both the HBA and HBB loci.24e27 These latter techniques, although useful to reveal the presence of deletions, do not have the resolution available with the CGH method presented in this study. Our CGH approach was able to map previously uncharacterized HBB locus deletions with high enough resolution to design PCR primers that could amplify the DNA sequences flanking the deletion breakpoints and obtain definitive sequence analysis in a single run. This finding resulted in the identification and characterization of novel HPFH and (g) (d)b0-thalassemia deletions as well as definitive sequencing of prior mutations that were only imprecisely mapped. Subject 4, an adult Cambodian woman, was a compound heterozygote for HbE trait and bthalassemia with 44.2% HbF. CGH array analysis revealed chromosome 11 breakpoints that correspond to previously described Vietnamese/Laotian/Thai HPFH deletions.8,28,29 In a previously reported case,28 the HbF level was 21.3%. In our case, the HbF level is roughly twice that in previous reports. This can be accounted for by the fact that expression of bE is approximately half that of bA. Both subjects 6 and 10, of Sri Lankan ethnicity, had elevated HbF levels but with quite different phenotypes. Subject 6, a 36-year-old man, has a very high HbF level of 40.8% with low-normal MCV. CGH revealed the 49,772-bp HPFH-3 deletion.10 Subject 10, a 4.5-year-old girl, was microcytic with a moderately elevated HbF of 16.8% and a novel 38,225-bp deletion we define as Sri Lankan (db)0-thalassemia.10 The telomeric breakpoints were almost
6
jmd.amjpathol.org
Discussion
-
The Journal of Molecular Diagnostics
FLA 5.4.0 DTD JMDI458_proof 20 November 2015 8:03 pm EO: JMD15_0091
683 684 685 686 687 688 689 690 691 692 693 694 695 696 697 698 699 700 701 702 703 704 705 706 707 708 709 710 711 712 713 714 715 716 717 718 719 720 721 722 723 724 725 726 727 728 729 730 731 732 733 734 735 736 737 738 739 740 741 742 743 744
745 746 747 748 749 750 751 752 753 754 755 756 757 758 759 760 761 762 763 764 765 766 767 768 769 770 771 772 773 774 775 776 777 778 779 780 781 782 783 784 785 786 787 788 789 790 791 792 793 794 795 796 797 798 799 800 801 802 803 804 805 806
Array CGH for Globin Deletions identical for these two patients (chr11:5,194,453 for subject 6 versus chr11:5,194,461 for subject 10), whereas the centromeric breakpoints were 11,539 bp apart (chr11:5,244,225 for subject 6 versus chr11:5,232,686 for subject 10). The region between chr11:5,244,225 and chr11:5,232,686, which is immediately centromeric to the HBD gene, contains sequences that repress HBG expression based on the large difference in HbF levels between these two subjects, and chromatin immunoprecipitation has reveasled a binding site for the key HbF silencing factor BCL11A in this interval,30 along with binding sites for corepressors and repressive histone and chromatin marks.10 This site has also been associated with regulation of HBG expression in genome-wide association studies.31,32 Subject 7, a 30-year-old Kurdish woman, has b-thalassemia with typical microcytosis and a significantly elevated HbA2 level but no significant increase in HbF. CGH analysis revealed two noncontiguous deletions removing the HBB and HBBP1 genes but sparing the HBD gene. The retained sequences include this BCL11A binding site.10 Other mechanisms must exist for elevated HBG expression, however, because the deletions in subjects 4 and 5 do not remove the BCL11A binding site centromeric to the HBD gene, yet these subjects have elevated HbF levels of 44.2% and 20.6%, respectively. Perhaps the increased HbF in these subjects is due to a telomeric enhancer brought into proximity of the HBG genes. One other interesting feature of subject 10’s Sri Lankan (db)0 thalassemia deletion is that the HBD gene is intact, yet there is no increase in HbA2. The breakpoint occurs approximately 200 bp 30 (telomeric) to the HBD structural sequences and 30 untranslated region. This result suggests the presence of an HBD-positive regulatory element located somewhere telomeric to this breakpoint.
of >0.5 (Supplemental Figure S2). Standard gap-PCR is adequate for detecting the common HBA deletions leading to a-thalassemia,1 but the array CGH method still has potential utility to identify unusual deletions not tested for by gap-PCR and novel deletions of the HBA locus. Although we did not have samples to test, array CGH should also be able to identify and characterize triplicated HBA genes.
Conclusions We found that CGH using a high-density custom oligonucleotide array can be used to rapidly and efficiently define both previously described and unique deletions of the HBA and HBB loci. This method will be a useful adjunct to current diagnostic methods for patients suspected of having thalassemia where standard diagnostic techniques are inconclusive. The typical patient would be microcytic with normal iron studies and have a normal level of HbA2. If gap-PCR to detect the common HBA deletions is negative, an HBB or uncommon HBA deletion is likely. The lack of HbA2 elevation makes an HBB point mutation unlikely, so Sanger sequencing of HBB would not likely be diagnostic, and thus proceeding directly to array CGH would be the most efficient diagnostic pathway. The other clinical scenario where HBB deletion analysis would be useful is HPFH, where there is a large variety of deletions that are not amenable to gap-PCR, particularly in an ethnically diverse population. Next-generation sequencing should eventually replace CGH and most other genetic technologies for clinical testing, although currently CGH represents a less labor- and data-intensive way to characterize deletions, such as those in the globin loci, and thus could remain more accessible to laboratories with limited resources.
a-Globin Deletions
Acknowledgment
We also report the utility of array CGH for detecting HBA gene deletions. For subjects heterozygous for large deletions of the HBA locus, such as the –SEA and –FIL deletions, the expected 50% decrease in signal intensity is observed throughout the deleted region. CGH is also able to characterize some of the polymorphic repeat regions in the HBA locus, including the inter-z HVR, a polymorphic region on chromosome 16 that lies between the HBZ and HBZP1 genes13 and the 30 HVR centromeric to the HBQ gene.14 It appears that the inter-z HVR is smaller and the 30 HVR is larger for the –SEA allele than for the corresponding regions of the reference DNA used for the CGH study (Figure 4, A and B). The –FIL allele appears to have a shorter 30 HVR (Figure 4A). Identification of alleles lacking one of two HBA genes, such as the -a3.7 and -a4.2 deletions, is more difficult because the two HBA genes are highly homologous, and when only one is deleted, labeled DNA fragments from the nondeleted HBA genes (and their homologous pseudogenes) cross-hybridize to probes covering the deleted regions, resulting in a signal ratio of subject to reference DNA
HBB locus deletions were a generous gift of Dr. David H.K. Chui (Boston University, Boston, MA).
The Journal of Molecular Diagnostics
-
Supplemental Data Supplemental material for this article can be found at http://dx.doi.org/10.1016/j.jmoldx.2015.07.011.
References 1. Chong SS, Boehm CD, Higgs DR, Cutting GR: Single-tube multiplexPCR screen for common deletional determinants of a-thalassemia. Blood 2000, 95:360e362 2. Sabath DE, Cross ST, Mamiya LY: An improved method for detecting red cells with hemoglobin H inclusions that does not require glass capillary tubes. Clin Lab Haematol 2003, 25:87e91 3. Labbe RF, Rettmer RL: Zinc protoporphyrin: a product of irondeficient erythropoiesis. Semin Hematol 1989, 26:40e46 4. Chong SS, Boehm CD, Cutting GR, Higgs DR: Simplified multiplexPCR diagnosis of common Southeast Asian deletional determinants of a-thalassemia. Clin Chem 2000, 46:1692e1695
jmd.amjpathol.org
FLA 5.4.0 DTD JMDI458_proof 20 November 2015 8:03 pm EO: JMD15_0091
7
807 808 809 810 811 812 813 814 815 816 817 818 819 820 821 822 823 824 825 826 827 828 829 830 831 832 833 834 835 836 837 838 839 840 841 842 843 844 845 846 847 848 849 850 851 852 853 854 855 856 857 858 859 860 861 862 863 864 865 866 867 868
869 870 871 872 873 874 875 876 877 878 879 880 881 882 883 884 885 886 887 888 889 890 891 892 893 894 895 896 897 898 899 900 901 902 903 904 905 906 907 908 909 910 911 912 913 914 915 916 917 918 919 920 921 922 923 924 925 926 927 928 929 930
Sabath et al 5. Greisman HA, Hoffman NG, Yi HS: Rapid high-resolution mapping of balanced chromosomal rearrangements on tiling CGH arrays. J Mol Diagn 2011, 13:621e633 6. Tuan D, Feingold E, Newman M, Weissman SM, Forget BG: Different 30 end points of deletions causing db-thalassemia and hereditary persistence of fetal hemoglobin: implications for the control of gglobin gene expression in man. Proc Natl Acad Sci U S A 1983, 80: 6937e6941 7. Huisman TH, Wrightstone RN, Wilson JB, Schroeder WA, Kendall AG: Hemoglobin Kenya, the product of fusion of g and b polypeptide chains. Arch Biochem Biophys 1972, 153:850e853 8. Chalaow N, Thein SL, Viprakasit V: The 12.6 kb-deletion in the bglobin gene cluster is the known Thai/Vietnamese (db)0-thalassemia commonly found in Southeast Asia. Haematologica 2013, 98: e117ee118 9. Xu XM, Li ZQ, Liu ZY, Zhong XL, Zhao YZ, Mo QH: Molecular characterization and PCR detection of a deletional HPFH: application to rapid prenatal diagnosis for compound heterozygotes of this defect with b-thalassemia in a Chinese family. Am J Hematol 2000, 65: 183e188 10. Sankaran VG, Xu J, Byron R, Greisman HA, Fisher C, Weatherall DJ, Sabath DE, Groudine M, Orkin SH, Premawardhena A, Bender MA: A functional element necessary for fetal hemoglobin silencing. N Engl J Med 2011, 365:807e814 11. Henthorn PS, Mager DL, Huisman TH, Smithies O: A gene deletion ending within a complex array of repeated sequences 30 to the human b-globin gene cluster. Proc Natl Acad Sci U S A 1986, 83:5194e5198 12. Oner C, Oner R, Gurgey A, Altay C: A new Turkish type of b-thalassaemia major with homozygosity for two non-consecutive 7.6 kb deletions of the jb and b genes and an intact d gene. Br J Haematol 1995, 89:306e312 13. Goodbourn SE, Higgs DR, Clegg JB, Weatherall DJ: Molecular basis of length polymorphism in the human z-globin gene complex. Proc Natl Acad Sci U S A 1983, 80:5022e5026 14. Jarman AP, Nicholls RD, Weatherall DJ, Clegg JB, Higgs DR: Molecular characterisation of a hypervariable region downstream of the human a-globin gene cluster. EMBO J 1986, 5:1857e1863 15. Blattner A, Brunner-Agten S, Ludin K, Hergersberg M, Herklotz R, Huber AR, Röthlisberger B: Detection of germline rearrangements in patients with a- and b-thalassemia using high resolution array CGH. Blood Cells Mol Dis 2013, 51:39e47 16. Phylipsen M, Chaibunruang A, Vogelaar IP, Balak JR, Schaap RA, Ariyurek Y, Fucharoen S, den Dunnen JT, Giordano PC, Bakker E, Harteveld CL: Fine-tiling array CGH to improve diagnostics for a- and b-thalassemia rearrangements. Hum Mutat 2012, 33:272e280 17. Chan OT, Westover KD, Dietz L, Zehnder JL, Schrijver I: Comprehensive and efficient HBB mutation analysis for detection of b-hemoglobinopathies in a pan-ethnic population. Am J Clin Pathol 2010, 133:700e707 18. Shaji RV, Edison ES, Poonkuzhali B, Srivastava A, Chandy M: Rapid detection of b-globin gene mutations and polymorphisms by temporal temperature gradient gel electrophoresis. Clin Chem 2003, 49: 777e781 19. Vrettou C, Traeger-Synodinos J, Tzetis M, Malamis G, Kanavakis E: Rapid screening of multiple b-globin gene mutations by real-time PCR
8
20.
21.
22.
23.
24.
25.
26.
27.
28.
29. 30.
31.
32.
on the LightCycler: application to carrier screening and prenatal diagnosis of thalassemia syndromes. Clin Chem 2003, 49:769e776 Wang W, Kham SK, Yeo GH, Quah TC, Chong SS: Multiplex minisequencing screen for common Southeast Asian and Indian bthalassemia mutations. Clin Chem 2003, 49:209e218 Xiong F, Huang Q, Chen X, Zhou Y, Zhang X, Cai R, Chen Y, Xie J, Feng S, Wei X, Xiao Q, Zhang T, Luo S, Yang X, Hao Y, Qu Y, Li Q, Xu X: A melting curve analysisebased PCR assay for one-step genotyping of b-thalassemia mutations: a multicenter validation. J Mol Diagn 2011, 13:427e435 Craig JE, Barnetson RA, Prior J, Raven JL, Thein SL: Rapid detection of deletions causing db thalassemia and hereditary persistence of fetal hemoglobin by enzymatic amplification. Blood 1994, 83:1673e1682 De Andrade TG, Saad STO, Sonati MF, Costa FF: Simple fluorescent PCR method for detection of large deletions in the b-globin gene cluster. Am J Hematol 2003, 72:225e227 Colosimo A, Gatta V, Guida V, Leodori E, Foglietta E, Rinaldi S, Cappabianca MP, Amato A, Stuppia L, Dallapiccola B: Application of MLPA assay to characterize unsolved a-globin gene rearrangements. Blood Cells Mol Dis 2011, 46:139e144 Harteveld CL, Voskamp A, Phylipsen M, Akkermans N, den Dunnen JT, White SJ, Giordano PC: Nine unknown rearrangements in 16p13.3 and 11p15.4 causing a- and b-thalassaemia characterised by high resolution multiplex ligation-dependent probe amplification. J Med Genet 2005, 42:922e931 Kipp BR, Roellinger SE, Lundquist PA, Highsmith WE, Dawson DB: Development and clinical implementation of a combination deletion PCR and multiplex ligation-dependent probe amplification assay for detecting deletions involving the human a-globin gene cluster. J Mol Diagn 2011, 13:549e557 Phylipsen M, Vogelaar IP, Schaap RA, Arkesteijn SG, Boxma GL, van Helden WC, Wildschut IC, de Bruin-Roest AC, Giordano PC, Harteveld CL: A new a0-thalassemia deletion found in a Dutch family (eAW). Blood Cells Mol Dis 2010, 45:133e135 Ghedira ES, Lecerf L, Faubert E, Costes B, Moradkhani K, Bachir D, Galacteros F, Pissard S: Estimation of the difference in HbF expression due to loss of the 50 d-globin BCL11A binding region. Haematologica 2013, 98:305e308 Trent RJ, Svirklys L, Jones P: Thai (db)0-thalassemia and its interaction with g-thalassemia. Hemoglobin 1988, 12:101e114 Sankaran VG, Menne TF, Xu J, Akie TE, Lettre G, Van Handel B, Mikkola HK, Hirschhorn JN, Cantor AB, Orkin SH: Human fetal hemoglobin expression is regulated by the developmental stagespecific repressor BCL11A. Science 2008, 322:1839e1842 Galarneau G, Palmer CD, Sankaran VG, Orkin SH, Hirschhorn JN, Lettre G: Fine-mapping at three loci known to affect fetal hemoglobin levels explains additional genetic variation. Nat Genet 2010, 42: 1049e1051 Farrell JJ, Sherva RM, Chen ZY, Luo HY, Chu BF, Ha SY, Li CK, Lee AC, Li RC, Li CK, Yuen HL, So JC, Ma ES, Chan LC, Chan V, Sebastiani P, Farrer LA, Baldwin CT, Steinberg MH, Chui DH: A 3-bp deletion in the HBS1L-MYB intergenic region on chromosome 6q23 is associated with HbF expression. Blood 2011, 117: 4935e4945
jmd.amjpathol.org
-
The Journal of Molecular Diagnostics
FLA 5.4.0 DTD JMDI458_proof 20 November 2015 8:03 pm EO: JMD15_0091
931 932 933 934 935 936 937 938 939 940 941 942 943 944 945 946 947 948 949 950 951 952 953 954 955 956 957 958 959 960 961 962 963 964 965 966 967 968 969 970 971 972 973 974 975 976 977 978 979 980 981 982 983 984 985 986 987 988 989 990 991 992
993 994 995 996 997 998 999 1000 1001 1002 1003 1004 1005 1006 1007 1008
Supplemental Figure S1 Sanger sequencing chromatograms of the breakpoints of b-globin deletions characterized by array comparative genomic hybridization (CGH) that had not been finely mapped previously. PCR primers were designed based on the breakpoints as determined by array CGH. The resulting PCR products were sequenced by Sanger sequencing. Vertical black lines indicate where the breakpoints occur. Orange bars indicate sequences that are not present in the hg38 reference human germline sequence. Tracings are for subject 4 (A, Thai/Vietnamese (db)0 thalassemia), subject 5 (B, SEA-HPFH), subject 7 (C, Kurdish deletion; note that there are two breakpoints for this deletion shown), subject 8 (D, Irish/Romanian), and subject 9 (E, Belgian). Sequence data for subjects 6 and 10 can be found in the supplemental material for reference 5. Because the breakpoints for the 2.1-Mb deletion (subject 11) occur in low-density tiled regions of chromosome 11, it was not possible to design PCR primers and sequence the breakpoints. Supplemental Figure S2 Array comparative genomic hybridization (CGH) characterization of HBA single-gene deletions causing a-thalassemia. These include heterozygosity for the ea3.7 deletion (A), heterozygosity for the ea4.7 deletion (B), homozygosity for the ea3.7 deletion (C), and compound heterozygosity for the ea3.7 and ea4.7 deletions (D). Numbers at the bottom of the figure represent distance from the chromosome 16p telomere. In DNA samples heterozygous for either the ea3.7 or the ea4.2 deletion, a loss of signal of <50% can be seen in the region of the HBA genes (A and B, respectively). With DNA containing a homozygous ea3.7 deletion, the loss in signal is closer to the expected 50% in the stretch spanning nucleotides chr16:173,301177,104, including the 30 end of the HBA2 gene to the 50 end of the HBA1 gene (C). Again, although there is loss of genetic material from both alleles in this region, there is only effectively a 50% loss because there are two remaining a-globin genes, which hybridize to the probes for the deleted region. A similar result is obtained with DNA from a compound heterozygote for ea3.7 and ea4.2 (D).
FLA 5.4.0 DTD JMDI458_proof 20 November 2015 8:03 pm EO: JMD15_0091
1009 1010 1011 1012 1013 1014 1015 1016 1017 1018 1019 1020 1021 1022 1023 1024