The Huntington’s disease (HD) is a dominantly inherited neurodegenerative disease caused by a polyglutamine expansion in the N-terminus of the huntingtin protein. Greater than 36-38 CAG repeats in huntingtin will cause HD and longer CAG repeat lengths correlate with earlier onset of the disease
Currently there is no cure for HD. Treatments alleviate symptoms but do not prevent or delay disease progression
We have utilized a recently established HD-specific induced pluripotent stem cell (iPSC) line to generate a human HD cell model with a CAG expansion mutation in the endogenous huntingtin gene. The HD-specific iPSC (HD-iPSC) line was originally derived from a HD patient with a 72-repeat CAG tract by Park
We cultured the HD-iPSCs and normal iPSCs ( from Dr. George Daley) in the same conditions of conventional human ESCs and immunostained these cells periodically to ensure expression of hESC markers. Previous work demonstrated that the HD-iPSCs were pluripotent and capable of multilineage differentiation (8) . After extended culture of the HD-iPSCs on either MEF feeder cells or Matrigel (around 60 passages) these cells still stained positive for classic hESC markers including OCT4, NANOG, SOX2 and SSEA4 [Figure 1], indicating HD-iPSCs retained stem cell markers during passaging. We did note that the normal-iPSCs and H9 ESCs when compared to HD-iPSCs had distinct levels of ERK activation in response to bFGF and the level of ERK was significantly dampened in HD-iPSCs. This is consistent with previous reports in the literature of HD cell culture models of altered ERK activation
HD-iPS cells were cultured in the same conditions as human ES cells and immunostained for classic ES cell markers OCT4, NANOG, SOX2 and SSEA4.
HD-iPSCs, normal iPSCs and H9 cells were treated with 0, 4 or 20 ng/ml bFGF 24h after previous feeding. 30 min after bFGF treatment cells were harvested for protein lysates. Western blotting was performed to detect phospho-p44/42 (ERK1/2) (Thr202/Tyr204) as well as total p44/42 (ERK1/2) in these cells. The two bands represent p44 and p42 (either phosphor form or total protein) respectively.
We differentiated HD-iPSCs into neural lineages with a previously established method that utilizes the formation of embryoid body (EB) intermediates. Neural rosettes appeared after attachment of EBs onto poly-ornithine/laminin (pO/L) coated surfaces in neural differentiation medium supplemented with bFGF. We manually isolated the rosette cells, disrupted rosettes into smaller pieces mechanically and plated these rosette cells on pO/L coated surfaces in neural proliferation medium supplemented with bFGF. After the first passage with 0.05% trypsin these monolayer cells were cultured routinely as NSCs. The yield for this step was greater than 95%. Immunofluorescent staining showed that these cells expressed NSC markers Nestin, SOX1 and PAX6 but not the ESC marker OCT4 [Figure 3].
The HD-NSCs derived from HD-iPS cells immunostained positive for NSC markers Nestin, SOX1 and PAX6 while negatively for ES cell marker OCT4.
Next we further developed conditions to produce striatal neurons from the HD-NSCs. We modified a previously published protocol based on work of Aubry et al.
A (Stage 1) Immature striatal neurons express neuronal marker βIII-tubulin (Tuj1), GABAergic neuron marker GABA and striatal marker calbindin but not medium spiny neuron (MSN) marker DARPP-32. B (Stage 2) Mature striatal neurons express DARPP32 as well as well as βIII-tubulin, GABA and Calbindin.
The CAG expansion mutation of HD is sometimes unstable and may contract or expand spontaneously
Total DNA was extracted from normal iPS cells, HD-iPS cells, NSC and striatal neurons derived from HD-iPS cells, as well as HD-fibroblasts from which the HD-iPS cells were generated. A PCR reaction amplifying the CAG trinucleotide tract containing huntingtin sequence was performed with total DNA samples. cDNAs containing known number of CAG repeats were also included. Unlike Normal iPS cells, HD-iPS cells and cells derived from HD-iPS cells have the exact number of CAG repeat as the HD-Fibroblasts.
One of the most important applications of a HD cell model is to screen for therapeutic compounds with cellular assays. One such assay utilized in the field is the measurement of caspase activity. Caspase activity is elevated in HD cell culture models, mouse models and postmortem tissue when compared to controls
Both HD and normal NSCs were subject to a caspase-3/7 activity assay 24h after growth factor withdrawal. HD-NSCs showed elevated toxicity while normal NSCs did not. (** P<0.01 and *** P<0.001 in paired t-test) The upper panel the normal NSCs were derived from human H9 embryonic stem cells. The lower panel the NSCs were derived from iPS with normal CAG repeat length (HD-iPS line 1) .
We have demonstrated that HD patient-specific pluripotent stem cells can be cultured over multiple passages without losing CAG repeat length or pluripotent markers such as SSEA-4, NANOG, OCT4 and SOX2. These markers are absent in the fibroblasts in which the iPSCs were derived (data not shown). As might be expected, we do detect altered signalling when comparing the normal iPSCs to the HD iPSCs. In this case, we detect altered ERK phosphorylation in resting cells and in response to bFGF.
A characteristic property of pluripotent cells is their ability, when plated in suspension culture, to form embryoid bodies (EB). We found both the normal and HD iPSCs formed EBs. The hope is that HD iPSCs can be differentiated into the disease relevant cell types. We found that from EBs, HD iPSCs can be differentiated into neuronal precursor cells in high yield. These HD-NSCs expressed markers nestin, SOX1 and PAX6. The HD NSCs could be further differentiated into cells expressing GABAergic neuron marker GABA and a subset of these stained with neuron marker calbindin. Further differentiation yielded DARPP-32 positive neurons. Currently we are optimizing the yield of the DARPP-32 positive cells. However, even this mixed culture may yield clues to cell selective neuronal vulnerability in HD.
Finally, we found that normal NSCs when compared to HD NSCs have altered levels of caspase activity during serum withdrawal. This phenotype could be utilized for screening of therapeutic compounds or to optimize differentiation procedures. Although the possible differences in ERK activation and caspase induction are promising they will need to be confirmed in multiple HD iPS from distinct patient fibroblasts.
Many insights into molecular pathways in HD come from analysis of post-morterm HD tissue or HD mouse models. With the opportunity to utilized patient specific iPS cells, new tools will be generated to understand mechanisms of HD.
All cell culture reagents were from Invitrogen unless otherwise mentioned. HD-iPSCs were cultured like ES cells on either Matrigel (BD) or irradiation inactivated mouse embryonic fibroblasts (MEFs). When HD-iPSCs were grown on MEFs, the ES culture medium was knockout DMEM/F12 supplemented with 20% knockout serum replacement, 2.48mM L-glutamine, 1X nonessential amino acid, 15.4mM HEPES, 50μM β-mercaptoethanol, 100U/ml penicillin, 100μg/ml streptomycin (Cellgro) and 4ng/ml bFGF (Peprotech). When HD-iPSCs were grown on Matrigel, the ES medium conditioned by MEFs was used. The HD-iPSCs were regularly passaged with collagenase. HD-NSCs and WT-NSCs were cultured on plates coated with 20μg/ml poly-ornithine (Sigma) for 1h at 37 o C followed by 5μg/ml mouse laminin (Sigma) for 1h at 37 o C. The neural proliferation medium for NSCs was ENStem-A Neural Expansion Medium (Millipore, NeuroBasal medium with 1X B27, 10ng/ml LIF) supplemented with 2mM L-glutamine, 100U/ml penicillin, 100μg/ml streptomycin, and 25ng/ml bFGF. NSCs were regularly passaged with Accutase (Sigma).
HD-iPSCs, normal iPSCs or H9 cells cultured on Matrigel were fed with bFGF containing MEF conditioned medium daily. On the day of experiment, without feeding cells with fresh medium different doses of bFGF (4ng/ml or 20ng/ml) were added to these cells which had not been fed for 24h. 30 min after bFGF treatment cells were scraped off, pelleted and washed once with DPBS (Cellgro). Cell pellets were lyzed by sonication in mammalian protein extraction reagent (M-PER, from Thermo Scientific) containing protease inhibitors (one Complete mini tablet per 10ml, Roche) and 1% phosphatase inhibitor cocktail set II (Calbiochem). Protein concentration was measured with Pierce BCA protein assay kit (Thermo Scientific) to ensure equal sample loading. Protein samples were run on 4-12% bis-tris gel (Invitrogen), transferred to nitrocellulose membrane (Whatman), probed with anti-phospho-p44/42 (ERK1/2) (Thr202/Tyr204) antibody (Cell Signaling) and reprobed with anti-p44/42 (ERK1/2) antibody (Cell Signaling).
HD-NSCs were derived from HD-iPSCs with EB method. Briefly HD-iPSCs were passaged with collagenase and cell clumps were cultured in a low attachment petri-dish (Kord-Valmark) in ES medium without bFGF. Medium was replaced every 2 days and at each time of medium change 25% more ES medium was replaced by EB differentiation medium (DMEM supplemented with 20% fetal bovine serum, 1X nonessential amino acid, 50μM β-mercaptoethanol, 100U/ml penicillin and 100μg/ml streptomycin). After 8 days the medium was 100% EB differentiation medium. After 10 days the EBs in suspension were attached onto pO/L coated plates in neural differentiation medium (DMEM/F12 supplemented with 1X N2, 100U/ml penicillin and 100μg/ml streptomycin) with 25ng/ml bFGF. Medium was replaced every 2 days. After 10-12 days rosettes were manually picked, triturated by P1000 tip and plated on pO/L coated plates in neural proliferation medium. The first passage was performed with 0.05% Trypsin and the following passages were done with Accutase (Sigma). WT-NSCs were derived from H9 human ESCs by the same procedure. The striatal differentiation of HD-NSCs was induced by changing neural proliferation medium to neural differentiation medium supplemented with 250ng/ml SHH (R&D Systems), 100ng/ml DKK1 (R&D Systems), 20ng/ml BDNF (Peprotech) and 10μM Y27632 (Calbiochem). After 8-10 days in the condition above (Stage 1), cells were exposed to 0.5mM dibutryl-cyclic AMP (Sigma), 0.5μM valpromide (Alfa Aesar), 20ng/ml BDNF and 10μM Y27632 for an additional 1-3 days (Stage 2).
Total DNA was extracted from different cell samples with DNeasy kit (Qiagen) according to manufacturer’s instructions. The primers used to amplify the CAG trinucleotide containing fragment are: Forward 5’-CCT TCG AGT CCC TCA AGT CCT TC-3’, Reverse 5’-GGC GGG GGC GGC TGC GGC TGA G-3’.
The caspase activity assay was performed with Apo3 HTS kit (Cell Technology). Either HD-NSCs or WT-NSCs were grown in 24 well plates. For growth factor deprived samples cells were first washed once with neural proliferation medium without LIF and bFGF, then cultured in this growth factor free medium for 24h. After 24h of growth factor withdrawal, medium was removed and 150μl 1X lysis buffer was added into each well. The plate were placed on an orbital shaker at 700 rpm for 10 min. Then, 30μl of cell lysate was dispensed into one well of a 96 well plate in triplicate for each sample. 70μl of substrate mix (1X lysis buffer with 1X Apo3 HTS Caspase3/7 detection reagent and 20mM DTT) was added into each well and the plate was shaken at 700rpm for 30s. Subsequently the plate was loaded on Fusion-Alpha Universal Microplate Analyzer (Perkin Elmer) for the fluorescence based reading (Ex: 485nm, Em: 530nm). For each sample 10μl of lysate was dispensed into another 96 well plate in triplicate for protein concentration measurement with Pierce BCA protein assay kit. The caspase activity was normalized against protein concentration for each sample.
The authors have declared that no competing interests exist.
We thank Dr. Daley for providing us the normal and HD-iPSCs. Correspondence should be sent to [email protected], Buck Institute for Age Research, Novato, CA 94945