This essay has been submitted by a student. This is not an example of the work written by our professional essay writers.
Objectives: The aim of this study was to identify T-cell subsets in healthy human skin using Real Time-PCR (RT-PCR). This method was tested to evaluate its usefulness in routinely studying the skin immune system (SIS). More specifically, to identify T cell subsets in normal skin relevant in the pathophysiology of atopic dermatitis (AD)
Methods: RNA from non-pathological skin was isolated from 15 healthy controls using Qiagen's RNeasy kit. RNA from PBMCs was isolated from 3 healthy controls. RT-PCR was performed on T-cell surface glycoprotein CD3 (CD3E) and transcription factors for T-helper 1 cells (T-BET), T-helper 2 cells (GATA-3), T-helper 17 cells (RORGT) and T-Regulatory cells (FOXP-3).
Results: All skin tissue samples demonstrated expression of CD3e and GATA-3. However, T-BET was not detected in any of the samples and FOXP-3 was only expressed in eight of the fifteen samples. All T-cell subsets were displayed in RT-PCR in PBMCs.
Conclusion: Because of the inability to demonstrate important T-cell subsets in skin, RT-PCR is currently not useful for evaluating T-cell subsets in healthy skin. Because of the high amount of inter donor variability no conclusive statement can be made on T-cell distribution in PBMC based on MRNA profiling. Further study using RT-PCR on lesional skin in AD patients might improve its usefulness for evaluating therapy.
Abbreviations: RT-PCR (Real-time PCR); SIS (Skin Immune System) ; AD (Atopic Dermatitis); SCIT (Subcutaneous Immunotherapy); Th1 (T-helper 1); Th2 (T-helper 2); Th9 (T-helper 9); Th17 (T-helper 17); Th22 (T-helper 22); Treg (T regulatory cell); SCORAD (Scoring Atopic Dermatits); Fluorescence-activated cell sorting (FACS); OD (optical density); RIN (RNA integrity number)
The skin has its own immune system consisting of numerous interacting immune cells.1 It is known as the skin immune system (SIS) The most important function of the skin immune system is defense against tumors and pathogens, while at the same time maintaining tolerance to auto-antigens.2 Sometimes the SIS dysfunctions leading to the generation of auto-reactive T-cells which can initiate inflammatory skin disease like psoriasis. It is also involved in another skin disease, more specifically atopic dermatitis. Atopic dermatitis is one of the most frequent occurring inflammatory skin diseases with its prevalence rising each year. 3 Its pathological process is influenced by different T-cell subsets like T-helper 1 (Th1), T-helper 2 (Th2), T-helper 9 (Th9), T-helper 17 (Th17). T-helper 22 (Th22) and T regulatory cells (Treg).4 Acute AD lesions are characterized by Th2 predominant inflammatory responses, while in chronic AD Th1 inflammation is more prominent5 The mechanism of the switch from Th2 in acute AD to Th1 in chronic AD is still not well understood. In AD skin Th17 cells have been detected in lesional skin as well as in peripheral blood, correlating with disease severity6 IL-9 mRNA produced by Th9 cells in AD skin has been detected upon allergen provocation.7 Il-9 MRNA expression levels correlate with the number of eosinophils . Recently Th22 cells have been distinguished from Th17 cells, as they produce Il-22 but no Il-17.8 In AD patients serum IL-22 levels showed a correlation with disease severity.9
Treg cells control immune homeostasis and regulate the immune response during inflammation. However, the roll of Treg cells in AD is still unclear. Treg cells can be divided in natural Foxp-3+ cells (nTregs) and Foxp-3 negative induced cells (iTregs), including Foxp3- Il-10+ Tr1 cells.10 Some studies show an increase of Foxp3+ cells in peripheral blood in AD patients,11 while others do not.12 In skin both the presence and absence of Foxp3+ cells has been reported, as well as the expression of Tr1 cells.12,13 This inconsistency suggests the role of Treg cells has to be further investigated.
Traditional treatment of AD consists of symptomatic anti-inflammatory drugs or anti allergic local or systemic therapy. Within treatment options for allergic disease, only subcutaneous immunotherapy (SCIT) gives long term improvement of allergic symptoms.14 However this therapy has only been proven for patients with asthma and allergic rhinitis and is still unproven in AD.15, Nonetheless, new papers suggest AD patients might benefit from SCIT as well. 16,17 However, these studies measure immunological outcome based on serum cytokine and immunoglobulin levels and do not investigate immunological changes within the SIS itself. Moreover SCIT outcome is measured using the Scoring Atopic Dermatitis (SCORAD) which is a subjective questionnaire and sensitive to intra- and inter observer variation.
The interest of this study is to evaluate the effects of SCIT on the SIS itself by investigating immunological changes in T-cell subsets in skin relevant in the pathophysiology of AD. This could lead to objective markers to evaluate SCIT therapy. However, the composition of the SIS in non-pathological skin needs to be investigated first.
Current methods for assessing T-cell subsets in skin are immunohistochemistry and Fluorescence-activated cell sorting (FACS) which is a type of flow cytometry.18,19 , .However, both techniques have disadvantages. Immunohistochemistry is difficult to interpret and has a low percentage of reproducibility. FACS analysis requires expensive equipment, specialized training and sometimes requires the stimulation of cells which could bias results. This article examines the use of Real-Time PCR (RT-PCR) for determining T-cell subsets in healthy skin as an alternative to these techniques. RT-PCR offers the advantage that it can be used directly on isolated RNA without stimulating any cells and at the same time provide a highly reproducible and less laborious technique.
However RT-PCR comes with difficulties. First it has proven to be challenging to get high quality RNA from skin due to high degradation of RNA during the extraction process.20 Second, because skin is 'though' tissue it is proven to be difficult to achieve full tissue homogenization using conventional pestle and mortar or homogenizer leading to variable RNA yields.21
Therefore this paper will first compare three techniques for RNA isolation and two manners for tissue homogenization in order to isolate the method with the highest RNA quality suitable for RT-PCR.
Subsequently, RT-PCR will be performed for relevant T-cell subsets on all RNA skin isolates. Additionally, RT-PCR will performed on RNA isolated from PBMCS for comparison. Finally it will be determined if RT-PCR is suitable for determining T-cell subsets in skin and consequently eligible for studying changes in SIS following immunotherapy in AD.
Materials and Methods
Three methods for RNA isolation were used. The trizol method, the Qiagen RNeasy mini kit and a combination approach using techniques from both methods. Detailed protocols for the three isolation methods can be found in the supplementary section.
Two manners of tissue disruption were utilized. One using a tissue crusher developed by Torik Ayoubi (former employee of the clinical genetics department of Maastricht University, the Netherlands) followed by beat beating and one in which the samples were only processed by bead beating without crushing the samples. The instrument used for beating was the Mini-BeadBeater Bead Homogenizer (Biospec, California, USA).
1mm glass beads were added to each sample before beating. Beating consisted of processing each sample in the bead-beater at a setting of 3 (3 meters per second) for 20 seconds and were then placed on ice slurry for 60 seconds. This was repeated 15 times. The tissue crush method involved fully immersing the tissue crusher in liquid nitrogen, after which the tissue was crushed one time and placed in sterile 2ml micro tubes.
Materials were treated with RNAseZap (Invitrogen, Bleiswijk Netherlands)before each new sample and cleaned with DEPC treated water afterwards.
Skin tissue samples from 15 plastic surgery patients were obtained In compliance with local review board institutional policies. Skin from abdominal, breast and unknown sites was used (Table 1)., Before RNA isolation all sample were snap-frozen and stored at -80â°.
Table 1. Origin of skin used for RNA isolation
Skin sample nr.
Skin sample nr.
PBMCS were isolated from 3 healthy controls by gradient centrifugation of heparinised blood using Ficoll-Hypaque (Sigma-Aldrich Chemie B.V., Zwijndrecht The Netherlands). PBMCS were snap-frozen and stored at -80â° until RNA isolation.
RNA quantity and quality
RNA quantity and quality of RNA were measured using a Nanodrop (Witec AG,Littau,Switzerland). The optical density (OD) 260/280 ratio, the OD 260/230 and the RNA concentration in ng/Âµl were measured. RNA quality was further determined using the Agilent 2100 bioanalyzer (Agilent, Palo Alto,USA). The RNA 6000 Nano Chip was used for analysis.
RNA samples were copied to cDNA using reversetranscriptase (MMLV-RT; Invitrogen, Bleiswijk
Netherlands) with Oligo (dT)15 primers according to the manufacturer's instructions. The negative control didn't contain MMLV-RT. Subsequently DEPC treated water was added to each cDNA sample to establish a concentration of 4 ng/Âµl.
Real-time PCR was performed using the Bio-Rad IQ5 (Bio-Rad, CA, USA) SYBR-Green from the Bioline SensiMix SYBR & Fluorescein Kit (GC biotech, Alphen aan den Rijn, The Netherlands) was used as fluorescent dye. The cDNA was subjected to primer pairs for GAPDH,CD3Î•, T-Bet, GATA-3, RORgT and FOXP3. Primer sequences were obtained from qPrimerDepot and ordered from Sigma-Aldrich (Sigma-Aldrich Chemie B.V., Zwijndrecht the Netherlands) (Table 3). The reaction mix was composed of 10Âµl of Sensimix, 0,6Âµl of 10ÂµM forward primer, 0,6Âµl of 10ÂµM reverse primer, 3.8Âµl of DEPC treated water and 5Âµl of cDNA. Cycling conditions were 95Â°C for 10 min, followed by 40 cycles of 95Â°C for 15s, 60Â°C for 20s, 72Â°C for 20s.
Table 3. Primer design
GCT CTC CAG AAC ATC ATC CCT GCC
CGT TGT CAT ACC AGG AAA TGA GCT T
GGG GCA AGA TGG TAA TGA AG
CCA GGA TAC TGA GGG CAT GT
AAA ATG AAC GGA CAG AAC CG
CAC GTC CAC AAA CAT CCT GT
CTG CTG AGA AGG ACA GGG AG
TCT GAC AGT TCG CAC AGG AC
GGG GCA AGA TGG TAA TGA AG
GGT GAT AAC CCC GTA GTG GA
GAA ACA GCA CAT TCC CAG AGT TC
ATG GCC CAG CGG ATG AG
For quantification, the target genes were normalized to the internal standard gene GAPDH. Furthermore the results these were similarly normalized for Cd3Î•. To assess relative gene expression, the Pfaffl method was used to calculate relative fold changes.23 De Pfaffl method uses the following formula:
Expression ratio (folds) = [(EGOI)( CT (GOI, calibrator) - CT (GOI, test))]/[ (Eref)( CT (ref, calibrator) - CT(ref, test))]
RNA quantity and quality
In this experiment three different methods of RNA isolation were compared. The Trizol method which is the most widespread method for the isolation of total RNAs.22 Second, the Qiagen RNeasy mini kit which is a classical spin column based kits used for total RNA isolation. Third, a combination approach using techniques from both methods. For evaluating the different methods of RNA isolation the skin from a single individual patient had to be used. The fresh tissue was divided in 18 comparable skin samples. Six of these were allocated to each RNA isolation method. Within each technique three skin samples were assigned to the tissue crush method followed by bead beating and three to the bead beating disruption method (Fig. 1 Appendix).
RNA quantity for all samples differs greatly among the different methods ranging from an average of 3.0Âµg tot 12.2Âµg (Table 4). The added crushing procedure does not seem to increase RNA yield in the Trizol and combination method. However, it seems to decrease the yield in the RNeasy procedure. The RNA quality as measured in 260/280 ratio ranges from 1.65 to 2.12 suggesting high quality RNA for 4 of the 6 methods. An OD 260/280 of 2.0 is optimal and accepted as pure RNA.24 The 260/230 ratio ranges from 0.64 to 2.06 suggesting contaminants in 5 of the 6 isolation methods.
Table 4. RNA quantitation for RNA isolation methods
RNA average quantity per biopsy (Âµg)
RNA average 260/280 ratio
RNA average 260/230 ratio
3.0 (Â±1.0 Âµg)
RNeasy + Crush
1.6 ( Â±0.4 Âµg)
Trizol + crush**
12.2 (Â±2.7 Âµg)
6.7 (Â±3.9 Âµg)
Combination + crush
6.6 (Â±1.6 Âµg)
All quantities are Â±SD.
The Bioanalyzer uses computer algorithms to assign a RNA integrity value (RIN) to each sample, which quantifies the amount of RNA degradation. The RIN is a numerical value between 1 (totally degraded RNA) and 10 (fully intact RNA). A RIN higher than 5 is recommended as good total RNA quality and a RIN higher than 8 as perfect total RNA quality for downstream RT-PCR application.25
RIN values range from 2.5 to an average of 5.6 showing great differences in RNA integrity amongst the methods (Table 5). The RNeasy kit shows the highest RIN value suggesting it is the most appropriate technique for the isolation of high quality RNA suitable for RT-PCR.
Table 5. Bioanalyzer RNA integrity (RIN) Values for RNA isolation methods
The two samples in each method with the highest 260/280 ratio were selected
RNeasy + Crush
Trizol + crush
Combination + crush
All quantities are Â±SD.
Real Time -PCR
To further strengthen the claim the RNeasy kit without crushing was the best option for RNA isolation it was decided to use RT-PCR to determine to expression of the housekeeping gene Glyceraldehyde-3-phosphate dehydrogenase (GAPDH) and CD3Î• as a marker for T-cell expression for all 11 RNA samples regardless of RIN-value (Table 6). Both Trizol methods show CT values greater than 35 for GAPDH and CD3E, meaning they are not suitable RNA isolation techniques as MRNA could not be detected. The cut-off is set at CT 35 because it is the last cycle completely without back-ground signal. The combination method without crushing was not able to detect CD3E, leaving the three other methods as possible options. Though CT values for GAPDH and CD3E for RNeasy, RNeasy+crush and combination + crush are comparable, the RNEASY has the lowest CT-value for GAPDH and the highest RIN-value compared to the other two methods. Thus, it was decided this was the optimal method for RNA isolation from skin.
Table 6. RT-PCR CT values for GAPDH and CD3E for different RNA isolation methods
RNeasy + Crush
Trizol + crush
Combination + crush
All quantities are Â±SD.
Table 7 shows average CT-values for GAPDH, CD3E, T-BET, GATA-3, RORGT and FOXP-3 for all skin samples. All samples show expression of GAPDH, CD3E and GATA-3. GAPDH has an average expression of CT 23.24 with a range of 21.90-25.04. This suggests it was able to extract good quality RNA from each skin sample with the RNeasy method. CD3E has an average expression of CT 32.56 with a range of 30.42-33.95 confirming T-cells are present in skin. GATA-3 expression shows an average CT expression 26,89with ranges from to 24,30 - 28.32 verifying the presence of TH-2 cells.
FOXP-3 was identified in eight samples and RORGT in two samples; however it is worth to mention these positive results are in fact very close to the detection limit and most negative results are only slightly above the detection limit. T-bet was not detected in any sample. Because not all transcription factors could be detected no relative distribution can be calculated.
Because it would be interesting to compare MRNA expression levels in skin with blood, RNA from PBMCSs was isolated form three donors using the same method as used for skin samples. RT-PCR was performed for the same T-cell subsets as skin (Table 8). The GAPDH average CT level of 24,52 and ranges from 23,10 - 25,72 are highly comparable to the levels found in skin once again proving the RNA isolation method is adequate for skin. All T-cell subsets are represented in PBMCS, which is expected as PBMCS are concentrated mononuclear cells and will thus express T-cell subsets more abundantly. Furthermore CT values for CD3E are much lower when compared to skin suggesting more T-cells in PBMCs than in skin.
Table 8. RT-PCR average CT-values for T-cell transcription markers in PBMCS
23,10 - 25,72
22,86 - 25,79
25,74 - 29,15
25,14 - 27,78
29,06 - 30,40
27,96 - 31,18
However, the only transcription markers which can be compared are CD3E and GATA-3 as the others were not or only barley represented in skin. Using the Pfaffl method relative fold changes were calculated (Table 9). GATA-3 expression levels were first corrected for GAPDH and that value was corrected for CD3E. This table reconfirms the lower amount of CD3E positive cells in skin compared to PBMCs. What is similarly remarking is the much higher expression of GATA-3 in skin.
Relative fold change for CD3e and GATA-3 in PBMC versus Skin
One of the goals of this study was to present a relative distribution of T-cell subsets in skin. Unfortunately due to the fact RT-PCR is not able to demonstrate all subsets, this can be only done for MRNA expression in PBMC's (Table 10). What is clearly demonstrated by the substantial ranges in all subsets, that there is a large degree of interdonor variability and no strong statement on its distribution can be made. However, it is striking RORGT seems to highly represented compared to the others.
Table 10. Relative distribution of T-cell subsets in PBMCs
13,14 - 30.19
6.26 - 10,10
50,08 - 76,45
4,15 - 18,83
The goal of this article was to determine whether RT-PCR on skin was suitable for identifying T-cell subsets in skin relevant to the pathology of AD. First the most optimal way of RNA extraction from skin had to be determined by comparing three isolation and two disruption methods. Second, RT-PCR was performed on 15 RNA skin isolates for relevant subsets demonstrating only CD3E and GATA-3 expression. Third, RT-PCR was performed on 3 RNA isolates from PBMCs displaying the expression of all subsets. Fourth, relative fold changes were calculated and showed lower CD3E expression, but higher GATA-3 expression in skin compared to PBMCs. Fifth, the relative distribution of T-cell subsets in PBMC was analyzed, but high inter donor variability make it difficult to make valid claims.
The first difficult part in this research set up was tissue homogenization. Crushing, using a home-made alternative to a pestle mortar followed by beat-beating was compared to beat-beating alone. Homogenizing using a mechanical homogenizer was not evaluated, because of the fact that tissue often becomes trapped within the probe and RNA yields are significantly lower compared to beat-beating.26 Collagenase has been described for though tissue homogenization.27 While this might be beneficial for isolating individual cells it requires fresh tissue to be used. Collagenase might improve RNA yield by better tissue homogenization, but the procedure also increases the risk of exposing samples to exogenous RNAses decreasing RNA quality. Nevertheless, this study already demonstrated RNA quantity and quality with the RNeasy isolation method were high enough for RT-PCR as GAPDH CT levels for skin and PBMC's were comparable, negating the need for increased homogenization.
It was hypothesized that crushing would lead to a higher homogenization rate thereby increasing the amount of RNA isolated. However, it was demonstrated crushing had not effect on RNA quantity in the Trizol and combination methods and even decreased the RNA yield in the RNeasy method. However, this was probably due small number of samples used for this experiment and slight variances in the size of the skin samples. During the isolation procedure two samples in the Trizol + crush and one in the Trizol were lost due to technique error. Unfortunately, these samples could not be replaced. Lack of these samples might bias results and might overpower the other methods. However, the OD 260/280 ratio for the RNeasy methods is higher than 2.0 for both implying pure RNA. Lower ratios indicate the presence of proteins, possibly RNAse, which might interfere with downstream reactions.24 On the other hand, the OD 260/230 for the RNeasy methods is significantly lower compared to the other methods and needs to be 2.0 for a pure sample as well. This indicates contamination with carbohydrates, peptides, phenols or aromatic compounds and in the case of RNeasy the buffer RLT used in the process. 24 In a test set-up, extra cleaning steps increased the OD 260/230 but significantly lowered RNA quality making it unusable for RT-PCR. Although samples with a low OD 260/230 ratio may interfere with downstream process , but only a few studies using RT-PCR have taken the OD 260/230 ratio into account when evaluating the RNA sample purity. Furthermore, there is no generally accepted cutoff value of the OD 260/230 ratio which marks the borderline for use in RT-PCR.28 Therefore the RIN value had to be calculated to determine the method with the highest RNA quality. The RNeasy method without crush displayed the highest RIN value, making it the method of choice for further RNA isolation. However, it is worthy to mention RNA can degrade op to 90% when frozen and thereby decrease the RIN value.29 All samples used in this study were snap frozen and thus RNA deteriorated. Even though current RIN values were sufficient to obtain good quality CDNA, fresh skin is preferred in RNA isolation In the future.
Difficulty arises in the interpretation of MRNA expression levels. MRNA expression for CD3E and GATA-3 was demonstrated in skin, while T-BET was not displayed. RORGT was expressed in two samples and FOXP-3 in 8 samples. However, It is difficult to compare and interpret these expression levels, as studies for t-cell subsets in skin using RT-PCR have not been done before. Therefore it was decided to compare these MRNA levels to those in PBMCs. Unfortunately, PBMC samples from only three patients were stored. As expected all T-cell subsets were exhibited in PBMCs. When taking in mind GAPDH CT values for PBMC's and skin are comparable, the gives the strong suggestion that the expression of most T-cell subsets in skin is just too low and therefore cannot be demonstrated by RT-PCR.
CD3E and GATA-3 were compared in terms of relative fold changes. As expected CD3E was more highly expressed in PBMC's. This demonstrates RT-PCR for all T-cell subsets is usable when CD3E levels are high enough. Though not significant, GATA-3 expression levels seem much higher in skin compared to blood. This might indicate Th2 activity in skin is higher than in blood. Non published data from this research group show a 3.7 time increase in Il-4 producing cells in skin when compared to PBMC's using FACS analysis among the same 15 patients in this study.30 This confirms Th2 cells play a more active role in skin compared to blood.
This study tried to describe the relative distribution of T-cell subsets in PBMC, but is unable to do so because of high inter donor variability. However, a comparison of T-cell subset distribution in PBMC and unpublished FACS data from this research group on PBMC in the same patients was performed (Table 1, Appendix). RORGT expression is particularly high in PBMCs, while FACS data for the same samples show a much lower expression of Th17 cells. This suggests a down regulating role for RORGT in terms of Th17 cytokine expression.
In conclusion, RNeasy is the most optimal method of acquiring high quality RNA from skin. However, RT-PCR is not suitable to determine most T-cell subsets in healthy skin and other techniques as FACS analysis and immunohistochemistry are preferred. To truly make a better comparison of MRNA expression levels non-pathological skin needs to be compared to pathological skin from AD patients. Furthermore, future research should include the transcription factors for Th22 and Th9 as they play a role in the pathogenesis of AD, but have not been investigated in this study paper.