• We are sorry, but NCBI web applications do not support your browser and may not function properly. More information
Logo of physiolgenomicsPublished ArticleArchivesSubscriptionsSubmissionsContact UsPhysiological GenomicsAmerican Physiological Society
Physiol Genomics. Mar 2010; 41(1): 91–101.
Published online Dec 15, 2009. doi:  10.1152/physiolgenomics.00086.2009
PMCID: PMC2841496

Long human CHGA flanking chromosome 14 sequence required for optimal BAC transgenic “rescue” of disease phenotypes in the mouse Chga knockout

Abstract

Chromogranin A (CHGA) plays a catalytic role in formation of catecholamine storage vesicles and also serves as precursor to the peptide fragment catestatin, a catecholamine secretory inhibitor whose expression is diminished in the hypertensive individuals. We previously reported the hypertensive, hyperadrenergic phenotype of Chga−/− knockout (KO) mice and rescue by the human ortholog. In the present study, we compare two humanized CHGA mouse models. Into the Chga null background, by bacterial artificial chromosome transgenesis human CHGA transgene has been introduced. Both lines have the complete ~12 kbp CHGA gene integrated stably in the genome but have substantial differences in CHGA expression, as well as consequent sympathochromaffin biochemistry and physiology. A mouse model with longer-insert HumCHGA31 displays integration encompassing not only CHGA but also long human flanking sequences. This is in contrast to mouse model HumCHGA19 with limited flanking human sequence co-integrated. As a consequence, HumCHGA19 mice have normal though diminished pattern of spatial expression of CHGA, and 14-fold lower circulating CHGA, with failure to rescue KO phenotypes to normalcy. In the longer-insert HumCHGA31 mice, catecholamine secretion, exaggerated responses to environmental stress, and hypertension were all alleviated. Promoter regions of the transgenes in both HumCHGA19 and HumCHGA31 display minimal CpG methylation, weighing against differential “position effects” of integration, and thus suggesting that lack of cis elements required for optimal CHGA expression occurs in HumCHGA19 mice. Such “humanized” CHGA mouse models may be useful in probing the physiological consequences of variation in CHGA expression found in humans, with consequences for susceptibility to hypertension and cardiovascular disease.

Keywords: blood pressure, catecholamine, stress, hypertension, chromogranin A, bacterial artificial chromosome

chromogranin A (CHGA) is present with catecholamines in their storage vesicles/chromaffin granules and co-released by exocytosis in response to sympathoadrenal stimulation (37). It is a proprotein that is processed into several biologically active peptides, including the dysglycemic hormone pancreastatin, vasodepressor vasostatin, and catecholamine release-inhibitory peptide catestatin (12, 17, 20, 23, 35). The catestatin fragment of CHGA seems to be processed inefficiently in hypertensives, leading to lower levels in plasma (25). As a nicotinic cholinergic antagonist, catestatin inhibits secretion from dense core granules, thereby attenuating levels of circulating catecholamines (20, 21). It also augments the release of the vasodilator histamine from mast cells, thereby enhancing its hypotensive action (14, 16).

Chga−/− mice develop hypertension, along with disordered catecholamine storage and release, as evidenced by a decrease in number and size of chromaffin granules coupled with elevated catecholamine levels in the plasma (18). A central role for catestatin deficiency in the elevated blood pressure (BP) is supported by the “rescue” of BP elevation by exogenous catestatin (18). Here we report the effects of introducing human CHGA transgenes into knockout (KO) mice. The two “humanized” CHGA models we describe in this study differ in length of inserted chromosome 14 sequence containing CHGA. The Tg(CHGA:RPC11-862G15)31Smv (HumCHGA31) mice have a single copy/haploid genome of an entire ~211 kbp bacterial artificial chromosome (BAC, RPC11-862G15) insert stably integrated into the genome, while mouse model Tg(CHGA:RPC11-862G15)19Smv (HumCHGA19) integrated only ~72 kbp of the BAC RPC11-862G15.

The ability of these mice to rescue KO phenotypes varies dramatically. Long flanking stretches 5′ and 3′ of CHGA appear to be essential for the optimal CHGA expression and consequent rescue of the KO phenotypes. Such mouse models, with sufficient vs. insufficient levels of CHGA, may effectively mimic variability in CHGA expression in the human population (5, 6, 27, 33) and offer tools to study the mechanism of action of CHGA/catestatin and its role in the pathophysiology of developing hypertension.

METHODS

Generating BAC transgenic humanized CHGA mice.

Mice were utilized in the study according to a protocol approved by the Animal Subjects Committee of the University of California at San Diego, and research was conducted in accordance with institutional guidelines. RP11-862G15 BAC human genomic clone, obtained from Children's Hospital Oakland Research Institute, has a ~211 kbp insert that spans the ~12 kbp CHGA locus (human chromosome 14: 92,459,198-92,471,393) and was used to generate transgenic and humanized mice as described earlier (18). The founder mice Tg19 (with shorter integration) and Tg31 (with longer integration) were used to establish transgenic lines homozygous for the transgene: Tg19 (TgCHGA19+/+; Chga+/+) or Tg31 (TgCHGA31+/+; Chga+/+) and subsequently backcrossed with Chga−/− (C57BL6 × 129/SvJ background) mice to generate humanized transgenic mice TgCHGA+/+; Chga−/− with the human CHGA transgene and lacking mouse Chga alleles. Once longer HumCHGATg31 and shorter HumCHGATg19 lines were derived, PCR analysis using five primer sets to amplify the 12 kbp CHGA gene as well as seven primer sets for the ~44 kbp upstream (92,415,173–92,459,199) region and 10 sets for the downstream ~156 kbp region (92,471,394-92,626,041 bp) of the BAC insert, with respect to the ~12 kbp human CHGA boundaries established the limits of insertion of the RP11-862G15.

Mouse breeding.

The ability of the transgenes in Tg19 and Tg31 to rescue the reported partial prenatal lethality of mice with genotype Chga−/− was determined by two types of mating designs for each of the transgenes. Using the PCR method described earlier (18) pups derived from the matings were genotyped for presence or absence of the human transgene and the mouse Chga allele(s). The PCR assay also provides information on the copy number of the mouse alleles [0 copy in KO, 1 copy in heterozygote (Het), 2 copies in wild type (WT)]; however, the assay was not informative regarding copy number of human CHGA alleles. In the first type of mating, CHGATg+/−;Chga+/− was crossed with KO mice (CHGA−/−;Chga−/−). CHGA refers to transgene Tg19 or Tg31. The expected ratios and genotypes of the offspring are 1:1:1:1 for CHGATg+/−;Chga+/− : CHGATg+/−;Chga−/− : CHGATg−/−;Chga+/− (Het) : CHGATg−/−;Chga−/− (KO). Of 38 offspring, the observed ratio was 13:10:9:6.

The second type of mating was between mice heterozygous for both human and mouse alleles. CHGATg+/−;Chga+/− crossed with CHGATg+/−;Chga+/−. There are nine expected genotypes' however, the PCR distinguishes offspring into six groups (CHGA+/+;Chga+/+ and CHGA+/−;Chga+/+) : (CHGA+/+;Chga−/− and CHGA+/−;Chga−/−) : (CHGA+/+;Chga+/− and CHGA+/−;Chga+/−): WT (CHGA−/−;Chga+/+) : Het (CHGA−/−;Chga+/−) : KO (CHGA−/−; Chga−/−) with expected ratios of 3:3:6:1:2:1. The second cross CHGATg+/−;Chga+/− × CHGATg+/−;Chga+/−, in the instance of mice with the transgene Tg19, produced offspring in the genotypic ratio (described above) 9:6:26:2:19:3 (total 65 pups). For transgene Tg31, of a total of 74 offspring the observed ratio was of 14:12:34:2:9:3. Any deviation of the observed ratios from the expected genotype ratio was estimated by χ2 analysis.

DNA analyses: slot and Southern blot.

Genomic DNA was prepared from tails of founder or offspring mice using a Puregene DNA purification kit (Gentra Systems). The Minifold II apparatus (Schleicher & Schuell, Keene, NH) was used for slot blot analysis. Genomic DNA digests with XmnI (8-base recognition restriction enzyme not sensitive to methylation) were used for Southern analysis. Probe specific for human CHGA was used for both blots.

Real-time PCR, immunoblot, ELISA, and confocal microscopy.

The animals were humanely killed, and the tissues isolated from the dissected animal; particular brain regions were obtained by excision with a razor blade. The tissues were homogenized in TRIZOL reagent (Invitrogen, Carlsbad, CA) and processed according to manufacturer's instruction.

For RNA analyses, total RNA was then DNase digested and further purified using RNeasy mini-kit (Qiagen, Valencia, CA). The first-strand cDNA synthesis was carried out using 500 ng of total RNA as template, with the iScript cDNA synthesis kit (Bio-Rad, Hercules, CA). The primers employed were specific for human CHGA(NM_001275)562F:CTGAAAGAGGCGGTGGAAGA, CHGA(NM_001275)653R:TCTGTGGCTTCACCACTTTTCTC; or mouse Chga-exon-1–22F:CACCATCACCGC TGTTACCA, Chga-exon-1–100R:TGGGTCTGTAGTCTGGCTACCA. The values obtained were normalized against mouse 18S ribosomal RNA.

Analysis for expression of genes within BAC insert RP11-862G15, co-inserted and flanking CHGA in HumCHGATg31 or HumCHGATg19, was accomplished by RT-PCR. cDNAs from adrenal gland, brain (total), testis, lung, and spleen were prepared from mRNA, and PCR amplified. The primers were: for BC127782.1: F 5′-TTGTGTTTTTAGGAGAGACGGGG-3′ and R 5′-TTCTGCTTCTTACCTGCTGTGAGTC-3′; for BX647940.1: F 5′-AGGAAGGTTCTGGTGCTCACATC-3′ and R 5′-TGTCAGGAGGGGGCTATTTTG-3′; for AK093301.1: F 5′-GAATCTCAGGCACTGGAAGGTTAC-3′, and R 5′-TTAGGGCAGGGAGGGATGTAGTAG-3′; and for BC042084.1 F 5′-AGAGCGTTGCTGTCCTCATCTC-3′, and R 5′-CCGTTGTTACCCCCATTTCAC-3′ were employed. WT tissues were used as control.

For immunoblots, adrenal glands were homogenized, the protein concentration was determined using Bio-Rad assay reagent (Hercules), and 5 μg of total protein was loaded per well. The primary polyclonal antibody was generated against human catestatin domain of CHGA using the vendor Strategic Biosolutions. Sandwich ELISAs between monoclonal anti-chromogranin A (MAb B4E11) antibody and rabbit polyclonal anti-human CHGA/catestatin antiserum have previously been described in detail (7). Plasma samples of each animal were assayed in duplicate.

To analyze the microanatomy of adrenal glands, they were fixed overnight in fresh 3.5% paraformaldehyde (Electron Microscopy Sciences, Hatfield, PA) and then transferred to 30% sucrose in 1× PBS for 48 h. The gland was fixed in OCT compound (Tissue-Tek) and stored at −70°C until sectioning. Sections 10–15 μm thick were cut using Leica CM3050S. Once the medullary region of the gland was well defined by light microscopy, the subsequent sections were stained for indirect immunofluorescence. The primary antibody used is rabbit anti-recombinant human CHGA (gift from Dr. L. Taupenot). The secondary antibody, Alexa Fluor 488-conjugated goat anti-rabbit IgG, and TO-PRO iodine (2 μM in DMSO) used to stain nuclei were obtained from Molecular Probes (Eugene, OR). Samples were analyzed using a Zeiss Axiovert 100M laser-scanning microscope.

Measurements of BP and plasma catecholamines in mice.

BP of mice was measured using BP-2000 Blood Pressure Analysis System (Visitech Systems, Cary, NC). The instrument employs a noninvasive tail-cuff method. Mice were placed in individual rodent restraint holders on a preheated specimen platform at 38°C. BP values were recorded 10 times in rapid succession, and the results were averaged. This process was repeated three or four times daily between 10:00 and 12:00 AM, for 5 consecutive days. Over the first week the readings were not collected, so as to allow the mice to adapt. Over the following week, the data were analyzed. During measurements, data were continuously stored in a notebook computer running the BP Analysis software package (Visitech) via a PCMCIA data acquisition card. Averaged readings of systolic BP (SBP) having standard deviation of <10 mmHg were accepted for further analysis.

Catecholamines were measured in plasma drawn from mice euthanized by deep anesthesia in isoflurane (Baxter) chambers followed by cervical dislocation. High performance liquid chromatography coupled to an electrochemical detector (Waters 600E Multisolvent Delivery system and Waters 2465 Electrochemical Detector) was used. Separation was performed on an Atlantis dC18 column (2.1 × 150 mm, 3 μm) from Waters. The mobile phase used was a mixture composed of phosphate-citrate buffer (2 mM NaH2PO4, 268 μM Na2EDTA, 50 mM sodium citrate, 10 mM diethylamine hydrochloride, 0.072% 1-octanesulfonic acid, pH 3.1 adjusted using phosphoric acid and 2.2% N,N-dimethylacetamide) and acetonitrile at 95:5 (vol/vol). A flow rate of 0.25 ml/min was used with isocratic mobile phase. The electrode potential was set at +0.6 V. The data were analyzed using Empower software from Waters, and CA levels normalized according to the recovery of an internal 3,4-dihydroxybenzylamine (DHBA) standard. To 0.3 ml of plasma sample, 1 ng of DHBA and 0.125 mM sodium meta-bisulfite were added. Following the addition of 12 mg of alumina (aluminum oxide, activity grade: Super I, type WA-4, Sigma-Aldrich), the pH of the plasma was raised to pH 8.6 by adding Tris · HCl buffer, pH 8.9. After a 30 min incubation, the sample was centrifuged (5,000 rpm, 5 min), the supernatant was discarded, and the beads were washed with water. The catecholamines were eluted with 80 μl of 0.1 N HCl supplemented with 100 mM sodium meta-bisulfite, and injected for analysis.

Immobilization stress.

This experimental protocol had prior approval by the Animal Care and Use Committee of our university, following humane National Institutes of Health (NIH) guidelines. Gentle and safe immobilization stress of mice was carried out using the Universal Restrainer apparatus (Braintree Scientific, MA). Prior to start of the immobilization stress, the BP of the mice was measured as described above. For a period of 21 days, the 4-mo-old mice were immobilized daily for 2 h between 8:00 and 10:00 AM. The BP of the stressed mice was monitored twice weekly. The stressed mice were given a 2 h recovery period prior to BP measurement.

Human CHGA promoter CpG methylation assay.

CpG methylation was assayed using Pyrosequencing after bisulfite treatment and PCR amplification at EpigenDX (Worcester, MA). Genomic DNA was prepared from adrenal gland (does express CHGA) and liver (does not express CHGA) tissues of HumCHGATg31 and HumCHGATg19 mice. It was subjected to bisulfite conversion, and an amplicon of 156 bp, upstream to ATG and including the TATA box (−312 to −157), was PCR amplified from the region of human CHGA promoter in the transgene. Pyrosequencing analyzed five of the CpG sites in the amplicon for DNA methylation. In-house EpigenDx control human genomic DNA methylated at low, medium, and high degrees was also PCR amplified and Pyrosequenced. Low and high in vitro methylated DNA controls were mixed at different ratios, subjected to bisulfite modification, PCR amplified, and analyzed by Pyrosequencing. R2 was determined to be 0.984 for the PCR bias testing. As control, methylation status of mouse LINE-I (heavily methylated) sequences was estimated at five positions in amplicons generated from LINE-1 rich region of the mouse genome from liver tissue.

Statistics.

Statistical analyses were done using unpaired Student's t-test or one-way ANOVA, as appropriate. Results are presented as mean value ± one SE.

RESULTS

Humanized CHGA mouse models generated by BAC transgenes.

The human BAC clone RP11862G15 ~211 kbp insert (Fig. 1A) was used to microinject the F2 embryos of the hybrid strain CB6 (F1 hybrid between BalbC and C57BL6 mice). Of 58 microinjected, implanted embryos that were born, six had integration of the CHGA BAC insert. However, only founders Tg19 and Tg31 gave rise to stable transgene inheritance over several generations and were characterized further.

Fig. 1.
Structures of the integrated human chromogranin A (CHGA) transgenes. A: schematic depicting the ~220 kbp BAC (bacterial artificial chromosome) construct used to create “humanized” CHGA mice. The insert comprises the ~211 ...

Founders Tg19 and Tg31 differed in the lengths of human flanking sequence that was co-integrated along with the entire CHGA. Figure 1B shows a schematic representation of these differences. The integrity of the RP11862G15 BAC inserts (posttransgenesis) was determined by PCR amplification of the BAC insert region from mouse genomic DNA preparations. By a series of PCR amplicons, the limits of insertion of the BAC in the transgenic lines were determined. Tg19 contains ~72 kbp of stably integrated RP11-862G15 insert, including in addition to the ~12 kbp CHGA gene, ~41 kbp upstream (5′) and ~19 kb downstream (3′) of the gene. Tg31, on the other has a near-complete 210,868 bp of RP11-862G15 insert.

To determine the copy number of the transgene CHGA integrated in the genome of the founders, slot blot analysis was employed (Fig. 2A). A densitometric scan revealed that only one copy of the transgene is integrated in the genomes of the founder mice (Fig. 2B). Qualitative Southern blot probing also confirmed the presence of a single copy of transgene (Fig. 2C). The F1 pups obtained by a cross between the founder and inbred strain CB6F1 mice are hemizygous for the CHGA transgene and have two of the mouse Chga alleles. In these mice we probed expression patterns of CHGA by RT-PCR (Fig. 2D). In neuroendocrine tissues the human allele is expressed. No expression of the allele is detected in liver and spleen, indicative of the expected pattern of fidelity of expression specific to neuroendocrine tissues in both Tg31 and Tg19 mice. The expression of CHGA in the transgenic mice clearly indicates elevated expression in Tg31 compared with Tg19. All the data values lie left of the line of identity with the highest levels of expression in Tg31 adrenal gland.

Fig. 2.
Single copy human CHGA transgene integration into the mouse genome. A: slot blot analysis of DNA. 1 μg or 100 ng duplicate samples of genomic DNA from human (diploid) or from F1 pups of founders Tg19 and Tg31 were hybridized with a 500 bp human ...

Expression of chromogranin A in humanized CHGA mice.

The fully humanized CHGA mice lack both copies of the mouse Chga allele but have two copies of the human CHGA transgene. Once both HumCHGA31 and HumCHGA19 lines were generated, we compared the lines for expression of chromogranin A mRNA and protein (Figs. 3 and and4).4). Human CHGA is expressed in several neuroendocrine tissues (data not shown), including the adrenal gland and brain stem, and is detected in circulating plasma in both lines. Comparison of the HumCHGA31 and HumCHGA19 adrenal extracts to WT and KO revealed more CHGA expression in HumCHGA31 (Fig. 3A). A stronger signal (>3-fold) is seen in case of HumCHGA31 than WT adrenal extracts, since the immunoblotting uses polyclonal anti-human catestatin antibody (Fig. 3B). The HumCHGA19 and KO adrenal extracts did not give a signal. Figure 3C shows mRNA expression of CHGA in the HumCHGA31 and HumCHGA19 strains. HumCHGA19 has significantly lower CHGA expression, ~23-fold lower in adrenal and ~7-fold lower in the brain stem, compared with HumCHGA31. The ELISA employed is based on capturing antibody B4E11 and detecting polyclonal antibody raised against full length CHGA (Fig. 3D). Only the human epitope is detected: it does not cross-react with either WT or KO mouse plasma (7). The ELISA results show that HumCHGA31 has ~14-fold higher circulating levels of CHGA in the plasma compared with HumCHGA19. Immunohistochemistry of adrenal glands of HumCHGA31 and HumCHGA19 (Fig. 4) also supported the evidence by immunoblot and RT-PCR analysis, in that expression is significantly higher in HumCHGA31. Also notable was the fact that CHGA staining was restricted to the medulla of the adrenal gland in the humanized mice. It appears therefore that the ~198 kbp flanking sequence in HumCHGA31 mice (Fig. 1B) results in higher expression of CHGA than the ~61 kbp insulation in HumCHGA19 mice. Immunohistochemistry of adrenal glands of HumCHGA31 and HumCHGA19 (Fig. 4) demonstrated that CHGA staining was absent in KO (−/−) mice and appropriately restricted to the medulla of the adrenal gland in both WT (+/+) and humanized mice. Thus, cell type (i.e., chromaffin)-specific expression seemed to occur for both mouse and human CHGA.

Fig. 3.
Chromogranin A expression in “humanized” CHGA mice: protein and mRNA. A: human CHGA protein expression. Immunoblot analysis (representative blot). Adrenal protein extracts of HumCHGA31, knockout (KO), wild-type (WT), and HumCHGA19 were ...
Fig. 4.
Microanatomy of chromogranin A expression: confocal micrographs of mouse adrenal glands. As depicted in F, each of the pictures (A–E) has 4 subpanels: Top left, red nuclear staining (TO-PRO 3 iodine, 640λ); top right, green CHGA staining ...

Ability of the humanized CHGA insert to rescue the Chga null phenotype.

Both the transgenes rescued prenatal lethality observed in KO mice. Thus in case of Tg19 transgenic mice, the combined expected ratio of humanized (CHGA+/−:Chga−/− and CHGA+/−:Chga−/−) mice to rest of the offspring is 17:68 compared with the observed ratio of 8:77 (χ2 = 3.0 and P = 0.083). Similar analysis of Tg31 offspring displays a ratio of 22:90 and the expected ratio of 24:88. The χ2 (0.027) analysis indicates that the observed ratio is not significantly deviant (P = 0.87) from the expected ratios.

The dramatic phenotypic differences of the Chga KO mice compared with WT mice include elevated BP as well as plasma catecholamine levels (18). In this study we analyzed these two disease traits in both of the humanized CHGA lines. Shown in Fig. 5A is evidence that SBP is rescued to normalcy in HumCHGA31 mice, with final values comparable to WT. HumCHGA19, with lower levels of CHGA, was unable to completely rescue the SBP to normalcy. Indeed, while ANOVA across five strains [WT (+/+), KO (−/−), Het (+/−), HumCHGA31, and HumCHGA19] yielded significant effects of strain on both SBP (Fig. 5A, P < 0.0001) and DBP (Fig. 5B, P < 0.0001); post hoc Bonferroni-corrected comparisons were not significant for WT vs. HumCHGA31, KO vs. HumCHGA19, HumCHGA19 vs. Het, or Het vs. KO.

Fig. 5.
Preferential “rescue” of elevated blood pressure to normalcy by the longer human CHGA transgene Tg31. Higher systolic blood pressure (SBP) and diastolic blood pressure (DBP) are observed in KO, HumCHGA19, and heterozygote (mouse Chga +/−, ...

Circulating catecholamines parallel the BP changes in these mouse strains. One-way ANOVA analyses across the four strains for NE (P = <0.0001) and E (P = 0.0015) are significant (Fig. 6). The plasma levels of norepinephrine (NE) and epinephrine (E) are significantly higher in KO compared with WT mice (Fig. 6, A and B). There is no significant difference in either the NE or E levels between HumCHGA31 and WT mice (Fig. 6, C and D). On the other hand, both of these groups have significantly lower levels of NE and E compared with HumCHGA19. These findings are congruent with the observation that HumCHGA19 display elevated BP in contrast with WT and HumCHGA31 mice.

Fig. 6.
Rescue of catecholamine levels to normalcy by the longer human CHGA transgene Tg31. Plasma catecholamines were measured in 10–12 wk old mice, n = 8/group. A and B: comparison of norepinephrine and epinephrine levels in WT and KO mice. C and D ...

Thus we conclude that HumCHGA31 reverts to the phenotype of the WT mice and is thereby effectively rescuing Chga−/− disease phenotypes. By contrast, replacement in the KO by two copies of the Hum Tg19 transgene does not completely rescue either BP or catecholamine phenotypes. The difference in the two humanized lines resides in the length of the flanking sequence that gets inserted along with the entire CHGA locus. This difference results in lower levels of CHGA in HumCHGA19 compared with HumCHGA31; the insufficient levels of CHGA, in turn, result in failure to rescue the KO phenotype to normalcy in HumCHGA19.

Differential response to stress by HumCHGA31 (Tg31, longer) vs. HumCHGA19 (Tg19, shorter) mice.

Both lines of human CHGA replacement mice (HumCHGA19 and HumCHGA31) were subjected to chronic immobilization stress, with BP responses measured, as well as in a set of age- and sex-matched mice not subjected to immobilization stress. Older mice (4 mo old) were used in this experiment, compared with 2-mo-old mice in Figs. 5 and and6.6. At the end of the stress period, SBPs of WT and Chga−/− mice were comparably elevated (Fig. 7A); thus it appeared that both the −/− and +/+ mice had reached an upper limit or “ceiling” for ability to raise BP. Once again, transgene Tg31 mice had lower basal SBPs than Tg19 mice, though at the end of stress SBP values were similar in HumCHGA31 and HumCHGA19 mice (Fig. 7B), further suggesting an SBP ceiling to which all mice ascend during stress.

Fig. 7.
BP response to immobilization stress. WT vs. knockout SBP normal and poststress. Humanized CHGA transgenic mice: SBP changes in response to immobilization stress. Chronic stress causes the SBP to be elevated significantly in WT, KO, and HumCHGA31 mice. ...

Location and expression of CHGA flanking genes.

The transgene Tg31 represents the entire insert of the BAC clone RP11862G15 (Fig. 8A). In addition to the complete CHGA gene, the BAC insert spans AK093301, BC042084, BX647940, and BC127782. Their transcripts (usually as expressed sequence tags) have been detected, but their functions are unknown. In the shorter Tg19 only AK093301 and BC042084 are additionally present. AK024887 is present in Tg31 but likely gives rise to a noncoding RNA, while at ITPK1, gene the 5′ end is incomplete (Fig. 8B). The expression profile of these flanking genes was as expected (Fig. 8C). Except for AK093301, the other three transcripts are detected in HumCHGA31 by RT-PCR in all tissues tested. An exception is the absence of expression of BC042084 in lung and spleen.

Fig. 8.
HumCHGA19 and HumCHGA31 transgenes: expression and location. A: human CHGA. Local genomic region in the BAC insert. The image was created at http://genome.ucsc.edu using the assembly February 2009 human reference sequence (GRCh37). Displayed are local ...

To determine if degrees of expression of the two human CHGA transgenes could be attributed to differences in the extent of heterochromatin organization, the methylation status of five CpG sites in the proximal CHGA promoter was determined in adrenal DNA from the two BAC transgenic lines (Fig. 8D). The very low degree of human CHGA promoter CpG methylation in adrenal gland DNA for either HumCHGA19 (at 9.4 ± 4.2%) or HumCHGA31 (8.2 ± 1.6%) suggests that the integration site of either human transgenes into the mouse genome did not result in heterochromatinization with consequent silencing. In contrast (and as expected), endogenous mouse LINE-1 methylation was highly methylated in adrenal genomic DNA from both the HumCHGA19 line (at 89.7 ± 1.8%) and the HumCHGA31 line (89.9 ± 1.0%), confirming that this assay can discover hypermethylation in the same animals.

DISCUSSION

BAC transgenic mice are currently indispensable to address the in vivo role of the human CHGA gene. Aligning the 10,059 bp of the murine gene (ENSMUSG00000021194) located on chromosome 10 with the 12,194 bp of human CHGA (ENSG00000100604) on chromosome 14 using the MacVector 7.2 software reveals that only ~4,713 residues are conserved. Given sequence divergences, it is unlikely that these orthologous genes are absolutely identical in expression. Unlike conventional transgenics, BAC constructs integrated into the mouse genome are quite large. Therefore the entire panoply of cis-regulatory elements, including long-range elements required for faithful expression of the transgene, are co-integrated. The extensive flanking sequences may also “buffer” the transgene from “position effects.” It has been observed in Drosophila, yeast, and plants that when genes are located in close proximity to the heterochromatin region, their expression is unstable (30, 31). In mouse embryonic stem cells, transgenes located near the telomeres are silenced (26). Like conventional transgenics, the possibility also exists that essential mouse sequences at the site of the transgene integration are disrupted. HumCHGATg31 displayed ~14-fold higher circulating levels of CHGA compared with HumCHGATg19 (Fig. 3D), in conjunction with more extensive co-integration of flanking sequence (Fig. 1B). Enhanced CHGA expression likely results in superior rescue of the Chga (−/−) disease phenotypes of elevated BP and catecholamines to normalcy by the longer HumCHGATg31 transgene. These differences in expression of HumCHGATg31 and HumCHGATg19 likely result from a “positive” consequence of retained essential cis-acting sequences for human CHGA expression by longer HumCHGATg31. The difference in expression is unlikely to be a “negative” consequence of position effects (36) arising from inadequate “insulation” by incorporation of the shorter HumCHGATg19 into surrounding mouse heterochromatin, since the proximal CHGA promoters in both the transgenic lines display low methylation levels (Fig. 8D), consistent with active promoters. While methylation is known to be a lasting influence on the heterochromatin organization of a gene influencing in turn its expression, it appears here that the integration locus of the human CHGA transgene in both lines (HumCHGATg31 and HumCHGATg19) does not result in promoter methylation differences likely to alter gene expression (10, 13). Lack of adequate insulation of the transgene does not appear to be the cause for silencing of either HumCHGATg19 or HumCHGATg31.

While the region differentially integrated in HumCHGATg19 vs. HumCHGATg31 does contain several regions of conserved sequence across species (Fig. 8A), our experiments cannot resolve the relative contributions of positive cis-acting enhancer in HumCHGATg31.

Longer (Tg31) vs. shorter (Tg19) flanking sequences.

Both HumCHGA31 and HumCHGA19 represent lines that span the entire 12.1 kbp CHGA gene. While HumCHGA31 mice have ~44 kbp upstream and ~155 kbp downstream human chromosome 14 sequences, HumCHGA19 mice have ~41 kbp upstream but only ~19 kbp downstream sequences. Additional expressed genes (AK093301, BC042084, BX647940, and BC127782) in the longer transgene HumCHGA31 may influence the final phenotypic differences between the two lines. Thus far, the human CHGA transgenic lines do not appear to have detrimental developmental effects. The two humanized mouse models for CHGA (HumCHGA31 and HumCHGA19) were analyzed for rescue of Chga KO phenotypes. Phenotypes of Chga-KO (−/−) mice already suggest that Chga regulates SBP, catecholamine secretion, and perhaps catecholamine storage vesicle biogenesis (18). In both HumCHGA replacement lines, the spatial expression pattern is comparable to WT mouse Chga as previously reported (18, 19) and restricted to neuroendocrine tissues (Figs. 2D and and4,4, B, D, and E). However, only the longer HumCHGA31 transgene completely rescued the KO phenotypes to normalcy: mice with HumCHGATg31 regulate BP, catecholamines and response to stress, in a manner comparable to WT (+/+) mice. Mice heterozygous (+/−) for Chga deletion have been characterized earlier (18), and their BP values resemble those of the HumCHGA19 and KO mice; the three groups display similar BP elevations (Fig. 5, A and B) with catecholamine excess (Fig. 6, C and D).

Thus the most pertinent gene expression difference between the two humanized CHGA lines seems to be in amount (rather than the distribution) of expression. Why should this quantitative difference come to be? The shorter HumCHGA19 and longer HumCHGA31 transgenes are likely integrated at different sites in the mouse genome, but each seems to exist as a single copy per haploid genome (Fig. 2, A–C). While the region differentially integrated in Tg31 vs. Tg19 does contain several regions of conserved sequence across species (Fig. 8), our experiments cannot resolve the relative contributions of positive cis-acting enhancer in HumCHGA31.

CHGA level significantly influences blood pressure in humans (24, 32, 33). Hypertensive patients differ from the normotensives in their levels of circulating CHGA, with elevated levels seen in the former. These humanized mouse models with varying levels of CHGA therefore depict the general population and are useful tools to address effect of CHGA circulating level in regulating BP in humans.

Transgene rescue of mouse gene function by BAC-encoded transgenes has been reported for several loci. Some examples of murine BAC transgene rescue include that of neural tube defect by BAC transgene Ltap (15) mouse circadian Clock gene (2), pcd mouse defect by Nna1 bearing BAC, and tattered mouse phenotype by BAC transgene Ebp (22).

Human ortholog transgenes have also been demonstrated to rescue deficient function in mice. This approach differentiates differences in regulation of mouse and man genes as has been shown with MC1R BAC transgenesis in Mc1r-deficient mice (11). The mouse α-thalassemia phenotype was rescued by BAC containing human α-thalassemic gene (1), Krabbe disease rescue by BAC transgenesis employed BAC with GALC (9), and Abca1−/− deficient mice were compensated by ABCA1 BAC transgene (8). In each of these cases the inset integrated into the transgenic mouse contained the gene and its native flanking chromosomal sequences. The length of the flanking sequence varied. The transgene BRCA1 rescued embryonic lethality of Brca1 1700T mutant mice (4). Similarly, the role of BRCA2 in gametogenesis was delineated using a BAC transgene to rescue the Brca2 +/− phenotype (29), ABCA1 gene was sufficient for rescue of Abca−/− phenotype (8), and finally, an extremely large 413-kbp Gata2 BAC transgene fully rescued Gata2 null mutant embryonic lethality (3). In each of these rescue scenarios the mouse of human gene required a unique length of flanking native sequence to rescue the disease or mutant phenotype by BAC transgenesis.

Caveats and alternative approaches.

We were able to characterize the insertion boundaries of two large CHGA BAC inserts, ranging from ~72 kbp (shorter Tg19) to ~211 kbp (longer Tg31); these two flanking inserts, coupled with careful biochemical and physiological phenotyping, allowed us to draw inferences about required boundaries for optimal expression and phenotype rescue. One limitation of this study is the random (initially uncontrollable) site and extent of BAC transgene integration, and the subsequent development of only two transgenic lines in great detail. The precise site of integration of a BAC transgene cannot be well controlled, nor can the extent of human flanking sequences. Therefore, this study features one transgenic CHGA line with higher and the other with lower levels of CHGA expression. Random integration of the TgCHGA19 vs. TgCHGA31 into different sites in the mouse genome might result in interruption of an essential mouse gene in one case but not the other. Although the TgCHGA19 and TgCHGA31 strains displayed no readily detectable differences in morphology or breeding, we cannot exclude the possibility of such disruption at one or the other integration site. A refinement of the BAC transgenic procedure known as the “co-placement” strategy (34) allows placement of two different haplotypic variants of a gene within the same human BAC into a single chromosomal site within the mouse genome. However, even this approach cannot ensure that the entire BAC is integrated into the mouse genome, and the integration site cannot be controlled.

Conclusions and Perspectives

Human BAC transgenes, to complement loss of mouse gene function, are a feasible approach to complement loss of mouse gene function, and to humanize the target locus. Such “humanization” should permit more specific studies of the function of the human CHGA gene in vivo, under circumstances where studies in intact humans are inappropriate or impossible. Generation of humanized mouse models may also serve the purpose of differentiating between regulation of human and mouse genes. Finally, the results of replacement by two humanized lines reinforce the impression that effective expression of CHGA is important for optimal catecholamine release, BP and the response to environmental stress. Finally, since genetic variation at the human CHGA locus results in variability of CHGA expression as well as susceptibility to hypertension (5, 6, 27), these models may represent a useful approach to understanding the coupling between the CHGA gene and disease risk.

GRANTS

This work was supported by NIH Grants DK-069613 (to S. M. Vaingankar), R01 DA-011311 (to S. K. Mahata), DK-60702 (to D. T. O'Connor), and P01 HL-58120 (to S. K. Mahata and D. T. O'Connor) and by the Department of Veterans Affairs (S. K. Mahata and D. T. O'Connor).

DISCLOSURES

The authors declare no conflicts of interest.

REFERENCES

1. Al-Hasani K, Vadolas J, Knaupp AS, Wardan H, Voullaire L, Williamson R, Ioannou PA. A 191-kb genomic fragment containing the human alpha-globin locus can rescue alpha-thalassemic mice. Mamm Genome 16: 847–853, 2005. [PubMed]
2. Antoch MP, Song EJ, Chang AM, Vitaterna MH, Zhao Y, Wilsbacher LD, Sangoram AM, King DP, Pinto LH, Takahashi JS. Functional identification of the mouse circadian Clock gene by transgenic BAC rescue. Cell 89: 655–667, 1997. [PMC free article] [PubMed]
3. Brandt W, Khandekar M, Suzuki N, Yamamoto M, Lim KC, Engel JD. Defining the functional boundaries of the Gata2 locus by rescue with a linked bacterial artificial chromosome transgene. J Biol Chem 283: 8976–8983, 2008. [PMC free article] [PubMed]
4. Chandler J, Hohenstein P, Swing DA, Tessarollo L, Sharan SK. Human BRCA1 gene rescues the embryonic lethality of Brca1 mutant mice. Genesis 29: 72–77, 2001. [PubMed]
5. Chen Y, Rao F, Rodriguez-Flores JL, Mahapatra NR, Mahata M, Wen G, Salem RM, Shih PA, Das M, Schork NJ, Ziegler MG, Hamilton BA, Mahata SK, O′Connor DT. Common genetic variants in the chromogranin A promoter alter autonomic activity and blood pressure. Kidney Int 74: 115–125, 2008. [PMC free article] [PubMed]
6. Chen Y, Rao F, Rodriguez-Flores JL, Mahata M, Fung MM, Stridsberg M, Vaingankar SM, Wen G, Salem RM, Das M, Cockburn MG, Schork NJ, Ziegler MG, Hamilton BA, Mahata SK, Taupenot L, O'Connor DT. Naturally occurring human genetic variation in the 3′-untranslated region of the secretory protein chromogranin A is associated with autonomic blood pressure regulation and hypertension in a sex-dependent fashion. J Am Coll Cardiol 52: 1468–1481, 2008. [PMC free article] [PubMed]
7. Corti A, Gasparri A, Chen FX, Pelagi M, Brandazza A, Sidoli A, Siccardi AG. Characterisation of circulating chromogranin A in human cancer patients. Br J Cancer 73: 924–932, 1996. [PMC free article] [PubMed]
8. Coutinho JM, Singaraja RR, Kang M, Arenillas DJ, Bertram LN, Bissada N, Staels B, Fruchart JC, Fievet C, Joseph-George AM, Wasserman WW, Hayden MR. Complete functional rescue of the ABCA1−/− mouse by human BAC transgenesis. J Lipid Res 46: 1113–1123, 2005. [PubMed]
9. De Gasperi R, Friedrich VL, Perez GM, Senturk E, Wen PH, Kelley K, Elder GA, Gama Sosa MA. Transgenic rescue of Krabbe disease in the twitcher mouse. Gene Ther 11: 1188–1194, 2004. [PubMed]
10. Hashimoto K, Oreffo RO, Gibson MB, Goldring MB, Roach HI. DNA demethylation at specific CpG sites in the IL1B promoter in response to inflammatory cytokines in human articular chondrocytes. Arthritis Rheum 60: 3303–3313, 2009. [PMC free article] [PubMed]
11. Healy E, Jordan SA, Budd PS, Suffolk R, Rees JL, Jackson IJ. Functional variation of MC1R alleles from red-haired individuals. Hum Mol Genet 10: 2397–2402, 2001. [PubMed]
12. Helle KB, Corti A, Metz-Boutigue MH, Tota B. The endocrine role for chromogranin A: a prohormone for peptides with regulatory properties. Cell Mol Life Sci 64: 2863–2886, 2007. [PubMed]
13. Horike SI, Ferreira JC, Meguro-Horike M, Choufani S, Smith AC, Shuman C, Meschino W, Chitayat D, Zackai E, Scherer SW, Weksberg R. Screening of DNA methylation at the H19 promoter or the distal region of its ICR1 ensures efficient detection of chromosome 11p15 epimutations in Russell-Silver syndrome. Am J Med Genet A 149A: 2415–2423, 2009. [PubMed]
14. Kennedy BP, Mahata SK, O'Connor DT, Ziegler MG. Mechanism of cardiovascular actions of the chromogranin A fragment catestatin in vivo. Peptides 19: 1241–1248, 1998. [PubMed]
15. Kibar Z, Gauthier S, Lee SH, Vidal S, Gros P. Rescue of the neural tube defect of loop-tail mice by a BAC clone containing the Ltap gene. Genomics 82: 397–400, 2003. [PubMed]
16. Kruger PG, Mahata SK, Helle KB. Catestatin (CgA344–364) stimulates rat mast cell release of histamine in a manner comparable to mastoparan and other cationic charged neuropeptides. Regul Pept 114: 29–35, 2003. [PubMed]
17. Laslop A, Mahata SK. Neuropeptides and chromogranins: session overview. Ann NY Acad Sci 971: 294–299, 2002. [PubMed]
18. Mahapatra NR, O'Connor DT, Vaingankar SM, Hikim AP, Mahata M, Ray S, Staite E, Wu H, Gu Y, Dalton N, Kennedy BP, Ziegler MG, Ross J, Mahata SK. Hypertension from targeted ablation of chromogranin A can be rescued by the human ortholog. J Clin Invest 115: 1942–1952, 2005. [PMC free article] [PubMed]
19. Mahata SK, Mahapatra NR, Mahata M, Wang TC, Kennedy BP, Ziegler MG, O'Connor DT. Catecholamine secretory vesicle stimulus-transcription coupling in vivo. Demonstration by a novel transgenic promoter/photoprotein reporter and inhibition of secretion and transcription by the chromogranin A fragment catestatin. J Biol Chem 278: 32058–32067, 2003. [PubMed]
20. Mahata SK, Mahata M, Wakade AR, O'Connor DT. Primary structure and function of the catecholamine release inhibitory peptide catestatin [chromogranin A(344–364)]: identification of amino acid residues crucial for activity. Mol Endocrinol 14: 1525–1535, 2000. [PubMed]
21. Mahata SK, O'Connor DT, Mahata M, Yoo SH, Taupenot L, Wu H, Gill BM, Parmer RJ. Novel autocrine feedback control of catecholamine release. A discrete chromogranin a fragment is a noncompetitive nicotinic cholinergic antagonist. J Clin Invest 100: 1623–1633, 1997. [PMC free article] [PubMed]
22. Means GD, Boyd Y, Willis CR, Derry JM. Transgenic rescue of the tattered phenotype by using a BAC encoding Ebp. Mamm Genome 12: 323–325, 2001. [PubMed]
23. Montero-Hadjadje M, Vaingankar S, Elias S, Tostivint H, Mahata SK, Anouar Y. Chromogranins A and B and secretogranin II: evolutionary and functional aspects. Acta Physiol (Oxf) 192: 309–324, 2008. [PubMed]
24. O'Connor DT. Chromogranin A: implications for hypertension. J Hypertens Suppl 2: S147–S150, 1984. [PubMed]
25. O'Connor DT, Kailasam MT, Kennedy BP, Ziegler MG, Yanaihara N, Parmer RJ. Early decline in the catecholamine release-inhibitory peptide catestatin in humans at genetic risk of hypertension. J Hypertens 20: 1335–1345, 2002. [PubMed]
26. Pedram M, Sprung CN, Gao Q, Lo AW, Reynolds GE, Murnane JP. Telomere position effect and silencing of transgenes near telomeres in the mouse. Mol Cell Biol 26: 1865–1878, 2006. [PMC free article] [PubMed]
27. Rao F, Wen G, Gayen JR, Das M, Vaingankar SM, Rana BK, Mahata M, Kennedy BP, Salem RM, Stridsberg M, Abel K, Smith DW, Eskin E, Schork NJ, Hamilton BA, Ziegler MG, Mahata SK, O'Connor DT. Catecholamine release-inhibitory peptide catestatin [chromogranin A(352–372)]: naturally occurring amino acid variant Gly364Ser causes profound changes in human autonomic activity and alters risk for hypertension. Circulation 115: 2271–2281, 2007. [PubMed]
28. Ratti S, Curnis F, Longhi R, Colombo B, Gasparri A, Magni F, Manera E, Metz-Boutigue MH, Corti A. Structure-activity relationships of chromogranin A in cell adhesion. Identification of an adhesion site for fibroblasts and smooth muscle cells. J Biol Chem 275: 29257–29263, 2000. [PubMed]
29. Sharan SK, Pyle A, Coppola V, Babus J, Swaminathan S, Benedict J, Swing D, Martin BK, Tessarollo L, Evans JP, Flaws JA, Handel MA. BRCA2 deficiency in mice leads to meiotic impairment and infertility. Development 131: 131–142, 2004. [PubMed]
30. Siegal ML, Hartl DL. An experimental test for lineage-specific position effects on alcohol dehydrogenase (Adh) genes in Drosophila. Proc Natl Acad Sci USA 95: 15513–15518, 1998. [PMC free article] [PubMed]
31. Singh J, Freeling M, Lisch D. A position effect on the heritability of epigenetic silencing. PLoS Genet 4: e1000216, 2008. [PMC free article] [PubMed]
32. Takiyyuddin MA, Cervenka JH, Hsiao RJ, Barbosa JA, Parmer RJ, O'Connor DT. Chromogranin A. Storage and release in hypertension. Hypertension 15: 237–246, 1990. [PubMed]
33. Takiyyuddin MA, Parmer RJ, Kailasam MT, Cervenka JH, Kennedy B, Ziegler MG, Lin MC, Li J, Grim CE, Wright FA, O'Connor DT. Chromogranin A in human hypertension. Influence of heredity. Hypertension 26: 213–220, 1995. [PubMed]
34. Tanimoto K, Sugiura A, Kanafusa S, Saito T, Masui N, Yanai K, Fukamizu A. A single nucleotide mutation in the mouse renin promoter disrupts blood pressure regulation. J Clin Invest 118: 1006–1016, 2008. [PMC free article] [PubMed]
35. Taupenot L, Harper KL, O'Connor DT. The chromogranin-secretogranin family. N Engl J Med 348: 1134–1149, 2003. [PubMed]
36. Weiler KS, Wakimoto BT. Heterochromatin and gene expression in Drosophila. Annu Rev Genet 29: 577–605, 1995. [PubMed]
37. Winkler H, Fischer-Colbrie R. The chromogranins A and B: the first 25 years and future perspectives. Neuroscience 49: 497–528, 1992. [PubMed]

Articles from Physiological Genomics are provided here courtesy of American Physiological Society
PubReader format: click here to try

Formats:

Related citations in PubMed

See reviews...See all...

Cited by other articles in PMC

See all...

Links

Recent Activity

Your browsing activity is empty.

Activity recording is turned off.

Turn recording back on

See more...