Clinical Immunology (2013) 147, 105-119
`
`available at www.sciencedirect.com
`
`Clinical Immunology
`
` www.elsevier.com/locate/yclim
`
`
`
`Modulation of immune function occurs within hours of
`therapy initiation for multiple sclerosis
`Chris L. Ayers?’ '»7, Jason P. Mendoza®’ '’?, Sushmita Sinha?’ ",
`Khrishen Cunnusamy*, Benjamin M. Greenberg”,
`Elliot M. Frohman?®, Nitin J. Karandikar °*
`
`* Department of Pathology, UT Southwestern Medical Center, 6000 Harry Hines Blvd., Dallas, TX 75390-9072, USA
`5 Department of Neurology, UT Southwestern Medical Center, 6000 Harry Hines Blvd., Dallas, TX 75390-9072, USA
`
`Received 15 January 2013; accepted with revision 15 February 2013.
`Available online 6 March 2013
`
`KEYWORDS
`Glatiramer acetate;
`Multiple sclerosis;
`immunomodulation
`
`
`
`Daily administration of FDA-approved glatiramer acetate (GA) has beneficial effects
`Abstract
`on clinical course of relapsing remitting multiple sclerosis (RRMS). Although mechanisms of
`GA-action have been widely investigated andpartially understood, immediate immune dynamics
`following GA-therapy are unknown.
`In the present study, we characterized the immediate
`effects of GA on phenotype, quantity and function of immune cells in MS patients. Prominent
`changesin immune cells were detected within 4-12 h post-first GA-injection. T-cell modulation
`included significantly decreased CD4/CD8 ratio, perturbed homeostasis of predominantly CD8 +
`T-cells, significant enhancement in CD8+ T-cell mediated suppression and inhibitory potentialof
`induced CD4-suppressors. Changes in APC were restricted to monocytes and included reduced
`stimulatory capacity in MLR and significantly increased IL-10 and TNF-« production. Our study
`provides the first evidence that GA treatment induces rapid immunologic changes within hours
`of first dose. Interestingly, these responses are not only restricted to innate immunecells but
`also include complex modulation of T-cell functionality.
`© 2013 Elsevier Inc. All rights reserved.
`
`
`* Corresponding author at: University of lowa Healthcare, Carver
`College of Medicine, lowa City, IA 52240, USA. Fax: +1 319 384 9613.
`E-mail address: nitin-karandikar@uiowa.edu (N.J. Karandikar).
`* Equal contribution (co-first authors).
`2 Currently at: Opexa Therapeutics, The Woodlands, TX 77381,
`USA.
`Currently at: Teva Pharmaceuticals, Kansas City, MO 64131, USA.
`
`1. Introduction
`
`Underlying immune-pathogenesis of multiple sclerosis (MS)
`involves multi-focal demyelination and axonalloss caused by
`infiltration of myelin-reactive T cells into the CNS. However,
`the disease is much more complex and tissue injury in active
`CNS lesions is also associated with activated microglia and
`macrophages. Glatiramer acetate (GA) is an FDA-approved
`immunomodulatory agent for the treatment of relapsing
`remitting-MS (RRMS). GA is a synthetic polypeptide composed
`of glutamic acid,
`lysine, alanine and tyrosine in a defined
`
`1521-6616/5 - see front matter © 2013 Elsevier Inc. All rights reserved.
`
`http: //dx.doi.org/10.1016/j.clim.2013.02.015
`
`Page 1 of 15
`
`Biogen Exhibit 2227
`Mylanv. Biogen
`IPR 2018-01403
`
`Page 1 of 15
`
`Biogen Exhibit 2227
`Mylan v. Biogen
`IPR 2018-01403
`
`

`

`106
`C.L. Ayers et al.
`
`molar ratio. Although the precise mechanism of GA action is
`still unclear, several immunologic effects of GA therapy
`have been described,
`including 1)
`immune deviation of
`CD4+ T cell responses from Th1 to Th2 [1-5], 2) induction
`of CD4+CD25+FoxP3+ regulatory T cells [6], 3) induction
`of oligoclonal, cytotoxic/suppressor CD8+ T cells [7-10],
`4) induction of immune modulatory Type II monocytes [11], and
`5) induction of IL-10 producing B cells [12,13]. However, the
`sequenceofthese interactions and the immediate immunologic
`effects of GA that may determine downstream effects on
`disease remain poorly understood.
`In fact, there has been no
`systemic dissection of the effects of early GA administration on
`immune cell phenotype and function. Several studies have
`documented altered immunecell effector functions upon short
`term exposure to GA. In accordance with this, GA is capable of
`inducing changes in antigen presenting cell
`(APC) function
`within hours of overnight in vitro cultures. Similarly, even in the
`absenceof prior GA therapy, GAis able to inducein vitro CD4+
`and CD8+ T cell responses from PBMC derived from healthy
`subjects and MS patients within a few days of culture [7,9].
`Therefore it
`is conceivable that
`following the first
`few
`injections, GA would show immediate in vivo immuneeffects
`that might dictate the eventual ability to develop a sustained
`immuneregulatory response.
`The present study is a novel comprehensive evaluation of
`immunealterations induced in T cell and APC populations
`during the first 72 h of GA therapy. Treatment naive RRMS
`patients initiating GA therapy were recruited for the study.
`Phenotypic and functional assays were performed on CD4+ T
`cell, CD8+ T cell, CD14+ monocyte, CD19+ B cell, BDCA1+
`myeloid dendritic cell
`(MDC) and BDCA4+ plasmacytoid
`dendritic cell (PDC) populations. The results were compared
`to the control subjects comprising of healthy donors (HD) as
`well as untreated-treatment naive RRMS patients, all of
`whom underwent a mock admission for specimen collection.
`We found that GA induces prominent phenotypic and func-
`tional changes in not only innate APC populations but also
`complex changes in T cells, particularly in the functional
`status of CD8+ T cells as early as 12 h after the first injection.
`These studies provide important insights into the timeline of
`immunealterations and emphasize the need for longitudinal
`studies to assess their significance in determining long-term
`immune and clinical consequences.
`
`2. Materials and methods
`
`2.1. Patients and control subjects
`
`After obtaining informed consent, 7 healthy donors, 8
`treatment-naive RRMS patients initiating glatiramer ace-
`tate (GA) therapy, and 4 “untreated” treatment naive RRMS
`patients were recruited for the study. At the time of
`monitoring, MS patients were free of steroid therapyfor at
`least 3 months, and had no record of acute relapse within
`3 months. None had a history of disease modifying therapy.
`All participants were admitted to the Clinical and Transla-
`tional Research Center (CTRC) for overnight blood draws
`(0 h baseline usually between 6 and 8 PM, followed by 4, 12
`and 24h post-first
`injection). The 24h collection was
`performedprior to the second daily GA injection. Partici-
`pants were then released and asked to return for a 72 h post-
`
`Page 2 of 15
`
`baseline blood draw (before their fourth daily shot of GA).
`Treatment decisions were determined by routine standard of
`care and patients were provided injection training during their
`first two GA injections. The healthy subjects and the untreated
`subjects served as important cohorts to control for potential
`diurnal variation of measured parameters. Thus, only the
`parameters that changed in the GA-treated cohort but not in
`the othertwo cohorts were considered an effect of GA therapy.
`All studies were approved by the UT Southwestern IRB according
`to Declaration of Helsinki principles.
`
`2.2. Cell preparation and bead sorting
`
`PBMC were isolated from whole blood using Ficoll Hypaque
`(GE Healthcare Biosciences, Pittsburg, PA) density gradient.
`In all cases, the 0 h, 4 h and 12 h specimens were processed
`simultaneously and the 24h and 72h specimens were
`processed independently. This design was based on initial
`stability studies for ex vivo subset quantification (not shown).
`From PBMC preparations, purified CD8+, CD14+ and CD19+
`cells wereisolated using respective Miltenyi microbeadpositive
`selection kits. The CD19 depleted fraction wasused forpositive
`selection of BDCA1+ (MDC), and BDCA4+ (PDC) populations
`using respective microbeads. “Untouched” CD4+ T cells were
`then isolated using negative selection kits. CD25+ T-cells were
`positively sorted from the purified CD4+ fraction using CD25
`microbeads. To prepare third party Te(CD4+CD25-) cells
`and APC, PBMC wereisolated from buffy coats of healthy
`donors using Ficoll Hypaque. APC fraction was prepared by
`depleting CD3+ T cells from PBMC using CD3+ microbeads.
`CD4+CD25-— (responder) cells were obtained by negative
`sorting for CD4+ T cells followed by depletion of CD25+
`cells. Both respondercells and APC werestored in freezing
`media in liquid nitrogen until further use in multiple assays.
`All magnetic microbeads were purchased from Miltenyi
`Biotech
`(Auburn,
`CA)
`and
`used
`according
`to
`the
`manufacturer's instructions, resulting in population purities
`> 90-95%.
`
`2.3. CFSE staining
`
`Third party CD4+CD25— respondercells used in suppression
`assays were labeled with carboxyfluorescein diacetate suc-
`cinimidyl ester (CFSE) (Invitrogen Molecular Probes, Eugene,
`OR), as described previously [14]. Briefly cells were suspended
`at 1 x 10® cells/mL and incubated for 7 min at 37 °C with
`0.25 iM CFSE, then washed twice with media containing 5%
`heat inactivated (HI) human serum.
`
`2.4. Flow cytometry based suppression assays
`
`(Terr) cells
`Allogeneic 0.1 x 10° CFSE-stained CD4+CD25-
`and APC were co-cultured with varying ratios of CD4+CD25+
`or CD8+ cells isolated from study participants in an anti-CD3
`coated 96 well round bottom plates (Costar, Corning, NY) for
`6 days. Cells were subsequently washed and stained for flow
`cytometry, as described later. Suppressive potential of
`regulatory T cells was assessed by normalizing the percent
`proliferative responses (%CFSE low CD4+) to the response
`without suppressors (defined as 100%), as described before
`[15]:
`
`Page 2 of 15
`
`

`

`GA-induced early immune changes
`107
`
`2.5. Induced suppression assay
`
`2.9. Statistical analysis
`
`Induced suppressors from project participants were gener-
`ated by culturing 2 x 10° CD4+CD25- T cells with plate
`bound anti-CD3 for 6 days, as described previously [16]. At
`the end of
`the culture period,
`the induced cells were
`harvested and washed twice and then co-cultured in varying
`ratios with CFSE-labeled allogeneic third party CD4+CD25-
`responder cells and APC in anti-CD3 coated round bottom
`plates for 6 days. Cells were subsequently washed and
`stained for flow cytometry, as described below. Inhibitory
`potential of induced suppressors was evaluated by normal-
`izing the percent proliferative responses (%CFSE low CD4+)
`to the response without suppressors (defined as 100%).
`
`2.6. Mixed lymphocyte reaction
`
`Antigen presenting cells (CD14+ & CD19+) and responder
`cells were isolated from PBMC by using microbeadsas described
`earlier. APC were collected from project participants, while
`the respondercells were obtained from third party donors. For
`the allogeneic response assay, APC and respondercells were
`incubated togetherat a 1:10 (APC:Ta4:) ratio in a 96-well plate
`(Costar, Corning, NY), as described previously [17]. The MLR
`was harvested after 6 days in culture and stained for flow
`cytometry. APC stimulatory capacity was evaluated by enu-
`merating CD25 expression and proliferation on responding T
`cells [17]. Additionally, respondercells isolated from project
`participants were co-cultured with APC from third party donors
`to assess the T cell activation potential of MS patients on GA
`therapy as compared to controls.
`
`2.7. Flow cytometric antibody staining
`
`For all the subjects, PBMC subset expression markers were
`evaluated on freshly isolated samples (1 x 10® cells/tube).
`Anti-human antibodies used for multi-color flow cytometric
`analysis included: CD4-PerCp-Cy5.5, CD8-AmCyan, CD45RO-
`Pacific Blue, CD27-FITC, CD19-PE-Cy7, CD14-PerCP, BDCA1-
`APC, and BDCA4-PE. All antibodies were obtained from BD
`Biosciences (San Jose, CA). Cells from suppression assays
`and MLR were washed with 0.1% (w/v) sodium azide/
`phosphate-buffered saline (Mediatech Cellgro) and stained
`with anti-human CD4-PE & CD25-APC, then resuspendedin
`1% paraformaldehyde
`(Electron Microscopy Sciences,
`Hatfield, PA). Flow cytometric data were acquired on a
`4-Laser LSRII using FACSDiva software (Becton Dickinson).
`Data were analyzed using FlowJo (Tree Star, Ashland, OR).
`
`2.8. Antigenic stimulation
`
`210° B cells, monocytes and MDC were separately
`placed in 96-well
`round bottom plates and stimulated
`with Staphylococcus aureus (pansorbin) (1:1000 dilution),
`LPS (5 g/mL) and Poly I:C (1 g/mL) respectively, for 48 h
`in activation media (RPMI + 5% HI human serum and 5 mM
`Pen/Strep). All culture supernatants were harvested and were
`stored at —80 °C for cytokine determination using multiplex
`MSD cytokine kits (Meso Scale Discovery, Gaithersburg, MD).
`Cytokines evaluated included TNF-«, IL-10, GMCSF,IL-8, IL-1,
`IL-6, IL-12, andIL-2.
`
`Page 3 of 15
`
`Data between the groups was analyzed with unpaired two-
`tailed Student's t-test and p < 0.05 was consideredsignificant.
`Paired two-tailed Student's t-test was used to compare data
`within a group and p < 0.05 wasconsideredsignificant.
`
`3. Results
`
`3.1. Subject characteristics and study design
`
`This study was designedas thefirst comprehensive evaluation
`of longitudinal changes in phenotype, quantity, and effector
`functions of immunecells in peripheral blood of MS patients
`within the first 72 h of GA therapy. The subjects comprised 12
`RRMSpatients and 7 healthy donors (HD). Both the GA treated
`(n = 8) and untreated (n = 4) MS patients werein the quiescent
`phase of relapsing remitting disease (not in active relapse)
`and were treatment naive (no history of disease modifying
`therapy). The healthy subjects and the untreated subjects
`served as important cohorts to control for potential diurnal
`variation of measured parameters. Thus, only the parameters
`that changed in the GA-treated cohort but notin the other two
`cohorts were considered an effect of GA therapy.
`The MS groupsconsisted of 2 males/8 females and 4 females
`respectively, with an age range and mean age of 22-59
`(42 years) and 38-56 (46.3 years). The HD cohort included 3
`males/4 females, 27-54 years (mean: 37.3 years). Peripheral
`blood was collected just before treatmentinitiation and this
`was considered time point O(h). Patients in the GA cohort
`initiated their GA therapy as part of an overnight admission in
`the UT Southwestern CTRC. Clinical oversight with injection
`training was provided as part of routine care procedures. Blood
`samples were then collected at 4h, 12h and 24h from
`the time of GA injection, as part of the admission. The
`24 h time point was collected before the administration of
`the second injection. Subjects then returned for the 72 h
`time point before their fourth dose of GA. Throughout the
`study, untreated and GA-treated MS patients were pooled
`together for baseline (0 h) comparisons with HD. At baseline,
`percentages and absolute numbers of CD4+, CD8+, CD14+,
`CD19+, myeloid dendritic cells (MDC) and plasmacytoid
`dendritic cells (PDC) were not different between HD and MS
`patients (Figs. 1A, B and C). We then sought to evaluate
`whether GA induced any phenotypic, quantitative, and/or
`functional changes in these cell populations within 72 h of
`therapyinitiation.
`
`3.2. GA induces a rapid increase in CD8+ T cell
`percentagesin the blood of MS patients
`
`The percentage of CD4+ T lymphocytes in peripheral blood
`of GA-treated MS patients remained stable during thefirst
`72 h of therapy, except for a slight but significant increase
`at 4h (GA-CD4+ T cells, Oh vs. 4h, p< 0.01, Fig. 2A). Of
`note, a similar increase in the percentage of CD4+ T cells
`was also seen in the peripheral blood of healthy donors
`(HD-CD4+ T cells, Oh vs. 4h, p<0.01, Fig. 2A), and
`therefore did not represent a GA-selective effect. Interest-
`ingly, GA-treated MS patients showed a significant increase
`
`Page 3 of 15
`
`

`

`
`
`108 C.L. Ayers et al.
`
`=
`3
`« 60
`o
`£
`mad
`3
`=
`220
`&
` »
`3
`
`+
`a
`oO
`
`+
`&
`Oo
`
`+
`2
`a
`3
`
`en
`°
`se’
`Q
`= 19
`a
`a
`51
`=
`
`+
`-
`Q
`Oo
`
`8
`
`8
`
`@ HD
`oO MS
`
`1.
`
`3
`s
`= 10
`~ 0.5
`g a. oA
`£
`2
`0.025
`<
`
`+
`+
`+
`+
`382228
`a
`a
`oO
`Oo
`o
`o
`
`p<0.01
`
`Similar percentages and absolute numbersof T, B and antigen presenting cells at baseline. PBMC from healthy donors and
`Figure 1
`treatment-naive MS patients were analyzed by flow cytometry for determining the frequency (A & B) and absolute numbers (C) of
`CD4+, CD8+, CD14+, CD19+, BDCA1+ (myeloid dendritic cells) and BDCA4+ (plasmacytoid dendritic cells) cells at the 0 h time point.
`A
`HD
`MS(Untreated)
`MS (GA)
`™)
`=
`2 8 6
`os
`Of
`4
`1S 40
`AE 20
`o2
`=
`
`p<o.ot
`
`o
`ims *
`tes nS cage sage
`4
`v
`+
`
`80
`60
`
`407
`201
`
`.
`+
`og
`s+ . oe UE
`,
`*
`
`@
`*
`
`Oh
`
`4h
`
`12h
`
`24h
`
`72h
`
`Oh
`
`4h
`
`12h
`
`24h
`
`72h
`
`Oh
`
`4h
`
`12h
`
`«24h
`
`«72h
`
`a
`a
`art
`
`¢
`*
`vfy «f+
`:
`‘.
`:
`
`12h
`
`24h
`
`72h
`
`25
`
`20
`15
`104
`5
`P
`
`a
`
`a
`4h
`
`«©
`
`e
`Oh
`
`oe
`oa
`<
`+ + ae
`yy
`-
`*
`-
`12h
`
`24h
`
`72h
`
`Oh
`
`4h
`
`12h
`
`24h
`
`72h
`
`20
`
`15
`
`10
`
`B
`25
`
`&2
`
`3” .
`o
`a
`oO 0 15
`re : a
`+ 210
`.
`BE
`e
`Oo 5
`PS
`‘
`
`Oh
`
`4h
`
`Cc
`© 20
`
`15
`
`10
`
`= ==
`
`o 9-
`
`aQ
`
`a
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`0
`oO 0
`Oh 24h=72h4h 12h 24h 72h Oh 4h 12h 24h 72h Oh 4h 12h
`
`
`
`
`
`
`
`
`
`E) Ms (GA)
`D) Ms (GA)
`210 ar
`p<0.05
`aa
`a
`—_——-
`2 $20
`o — =
`L315
`3 6
`= =o bk pe
`é = 10
`to 4
`o> eS a ——
`O- 5
`O02
`sf
`=
`0
`Bo
`°
`
`oh
`
`12h
`
`oh
`
`42h
`
`GA treatment reduces CD4/CD8 T cell ratio in the peripheral blood 12 h post injection. CD4+ (panel A), CD8+ (panel B)
`Figure 2
`percentages and CD4/CD8Tcell ratio (panel C) were determined in the peripheral blood of HD, untreated and GA-treated MS patients
`at indicated time points. Each point represents a single subject. Panels D and E show trends for CD8+ T cell percentages and CD4/CD8
`ratio, respectively, at 0 vs. 12 h in the GA-treated group. Significant p values are indicated.
`
`Page 4 of 15
`
`Page 4 of 15
`
`

`

`109
`GA-induced early immune changes
`
`in CD8+ T cell percentages at 12 h post-first injection when
`Suppl. Fig. 1. InCD4+T cells of MS patients treated with GA, the
`compared to Oh time point (p< 0.01, Figs. 2B and D).
`percentages of EM and Ter CD4+ T cells were significantly
`Concomitantly, the CD4/CD8 T cell ratio was significantly
`increased at 12 h (after first GA injection) when compared to
`reduced at 12h in the peripheral blood of GA-treated
`the 0 h time point (GA-EM CD4+ T cells, p < 0.05; GA-Ter CD4+
`patients (p < 0.05, Figs. 2C and E). No changes in CD8+T cell
`T cells, p < 0.05, Fig. 3C). While HD also showeda significant
`percentages were seen in the peripheral blood of HD or
`increase in EM CD4+ T cells at 12 h as compared to 0 h (HD-EM
`untreated MS patients. Within the patient cohort, increased
`CD4+T cells, p < 0.05, Fig. 3A), the increase in Ter CD4+T cells
`CD8+ T cell percentages and reduced CD4/CD8Tcell ratio at
`at 12h was selective to GA-treated MS patients. The only
`12 h were seen in the blood of 6/7 GA-treated MS patients
`diurnal fluctuation in CD4+T cell compartmentof untreated MS
`(Figs. 2D and E, respectively). At the 24h and 72h time
`patient was a significant decrease in EM population at 24h
`points, both representing “trough” points for drug (right
`when compared with 0 h time point (MS-untreated EM CD4+ T
`before daily shots), the cell percentages were somewhat
`cells, 0 h vs. 12 h, p < 0.05, Fig. 3B).
`elevated but not significantly different from 0 h.
`GA-treated MS patients, selectively, showed interesting
`changes in the peripheral blood CD8+ T cell homeostasis
`within 12 h of therapy initiation. Naive CD8+ T cells were
`3.3. GA predominantly alters CD8 + T cell homeostasis
`significantly decreased at 12 h when compared to 0hin the
`in the peripheral blood 12 h post injection
`blood of GA-treated MS patients (GA-Naive CD8+ T cells, 0h
`vs. 12 h, p < 0.05, Fig. 4C). Further, at 4 h, EM CD8+ T cells
`showed a decreasing trend (as compared to 0h) in GA
`treated and untreated MS patients and this was in contrast to
`increasing trend of EM CD8+ T cells from HD (Figs. 4A vs. B & C).
`While the decreasing trend in the EM CD8+ T cell population in
`untreated MS patients becamesignificant at 12 h (MS-untreated
`EM CD8+ T cells, 0h vs. 12 h, p < 0.05, Fig. 4B), notably, the
`
`To further characterize the early effects of GA on peripheral T
`cells, percentages of naive (CD27+CD45RO-), central memory
`(CM; CD27+CD45RO+), effector memory (EM; CD27—CD45RO +)
`and terminal (Ter; CD27—CD45RO-) CD4+ and CD8+ T-cells
`were examined in the peripheral blood of MS patients treated
`with GA. Gating strategy for these populations is shown in
`
`A) HD
`=Soo
`oo
`
`Naive
`
`<0.05
`
`== =
`Nnoo
` o
`
`ao%InCD4+Tcells
`
`B) Ms (Untreated)
`100
`8 90
`.
`2 a
`Ee
`sek Et VET
`Sol ET
`oO
`~
`.
`5
`£ 20
`= 6
`
`¥
`
`‘*
`
`‘
`
`
`Oh
`4h
`12h
`24h
`72h
`
`60
`
`40
`
`20
`
`0
`
`a
`
`a
`
`v
`
`Oh
`
`4h
`
`12h
`
`24h
`
`72h
`
` 12h
`
`
`
`
`5
`°
`pw St
`x wie
`+S
`v*
`0
`
`
`Oh 24h=72h4h
`12h
`24h
`72h
`Oh
`4h
`Oh 24h=72h4h 12h 24h) (72h Oh 4h 12h
`
`
`
`
`
`
`
`
`
`Terminal ¥
`
` Oh
`p<0.05 12h
`
`4h
`
`12h
`
`24h
`
`72h
`
`Oh
`
`4h
`
`12h
`
`24h)
`
`(72h
`
`p<0.05
`p<0.05
`
`C) Ms (GA)
`100
`=
`80
`aSo
`
`%InCD4+Tce 88 oS
`
`60
`
`°
`
`=
`
`'.
`
`a*
`
`Ss
`
`a
`
`a
`
`¥
`v"
`
`¥
`v
`
`
`
`0
`4h
`Oh
`4h
`Oh
`72h
`24h
`12h
`72h
`24h
`12h
`Oh
`4h
`
`
`12h 24h=72h Oh 4h
`
`24h==72h
`
`Figure 3
`No GA-induced changes in CD4+ T cell compartmentduringthefirst 72 h of therapy. Fluorescent conjugated anti-human CD27
`and CD45RO wereusedfor flow cytometry analysis to further differentiate CD4+ T cells into naive (CD27+CD45RO-), central memory
`(CM, CD27+CD45RO+), effector memory (EM, CD27—CD45RO+) and terminal (Ter, CD27—CD45RO-) populations (Suppl. Fig. 1). Panels A, B
`and C represent percentagesof naive, CM, EMand terminal CD4+ T cells in HD, untreated, and GA-treated MS patients, respectively. Except
`for the terminal CD4+ T cells, similar diurnal fluctuations in CD4+ T cell subpopulations were presentin all the study groups.
`
`Page 5 of 15
`
`Page 5 of 15
`
`

`

`
`
`C.L. Ayers et al.
`
`80
`
`60
`
`Terminal
`
`p<0.05
`
`Oh
`
`4h
`
`12h
`
`24h
`
`72h
`
`4h
`
`12h
`
`24h
`
`72h
`
`%InCD8+Tcell
`
`B) MS (Untreated)
`100
`8 30
`E
`‘+ 608
`
`re
`
`.
`
`2
`
`*
`aol
`°
`
`ae
`
`rt
`
`Q 40
`oO
`£ 20
`x
`
`Oh
`
`4h
`
`#12h
`
`24h
`
`72h
`
`Oh
`
`4h
`
`12h
`
`24h
`
`72h
`
`Oh
`
`4h
`
`12h
`
`24h
`
`72h
`
`Oh
`
`4h
`
`12h
`
`
` Oh
`
`72h
` Oh
`
`
`C) MS (GA)
`w 100
`=
`80
`
`%InCD8+Tce
`
`60
`feo
`
`p<0.05
`
`
`*,
`<"
`ee
`sy
`Rn ¥s
`
`.
`ry
`¥
`+
` p<0.001
`
`12h
`24h
`72h
`Oh
`4h
`
`50
`40
`
`30
`20
`10
`
`
`
`
`Tae TS
`
`Oh
`
`4h
`
`12h
`
`24h
`
`4h
`
`12h
`
`24h
`
`72h
`
`Figure 4
`Frequency of CD8+ T cell subpopulations is rapidly altered in the blood post GA therapy. Fluorescent conjugated
`anti-human CD27 and CD45RO wereused for flow cytometry analysis to further differentiate CD8+ T cells into naive (CD27+CD45RO-),
`central memory (CM, CD27+CD45RO+), effector memory (EM, CD27—CD45RO+) and terminal (Ter, CD27—CD45RO-) populations
`(Suppl. Fig. 1). Panels A, Band C represent percentagesof naive, CM, EMand terminal CD4+T cells in HD, untreated, and GA-treated
`MSpatients, respectively. GA induced significant decreasein naive (panel C) and increase in EM (panel C) CD8+T cells in the blood as
`early as 12 h after therapy initiation. No such changes in CD8+ T cell subpopulations were seen in HD (panel A) or untreated MS
`patients (panel B). Significant p values are indicated.
`
`initial decrease in EM CD8+ T cells in GA-treated MS patients
`was followed by a significant rebound at 12 h (GA-EM CD8+ T
`cells, 4h vs. 12 h, p < 0.01) post first GA injection (Fig. 4C).
`Fluctuations in the terminal CD8+ T cell compartment were
`seen in all three groups (Figs. 4A, B and C).
`
`3.4. Ex vivo suppression of allogeneic responder/
`effector T cells (Ter) by CD4+CD25+ T,.,s obtained
`from GA-treated MS patients remains unaltered
`during the first 72 h of therapy
`
`To analyze the effects of GA on suppressive function of
`CD4+CD25+ regulatory T cells (T,egs) during the first 72 h of
`therapy, Ter (CD4+CD25—) and irradiated APC, obtained
`from third-party healthy volunteers, were co-cultured with
`Tregs isolated from peripheral blood of either MS patients
`treated with GA or control cohorts. Representative dot plots
`are shown in Fig. 5A. Surprisingly, at Oh the mean sup-
`pression index of CD4+CD25+ T,,. cells in MS patients and HD
`was not significantly different
`(Suppl. Fig. 2). Moreover,
`there were no changes in percent suppression of anti-CD3
`stimulated allogeneic Ter cells by CD4+CD25+ T,.95 obtained
`
`either from GA-treated MS patients or controls during the
`first 72 h of therapy (Fig. 5B).
`
`3.5. CD8 suppressorability is significantly improved
`as early as 12 h after the first GA injection
`
`Expansion of GA-specific suppressor CD8+ T cell responses
`has been shown in MS patients on GA therapy [7,9],
`implicating a role of CD8+ T cells in GA-mediated immune
`modulation of MS. To test whether improvement in global
`suppressive function of CD8+ T cells is initiated as early
`immune change during GA therapy, suppression of anti-CD3-
`stimulated allogeneic Tor cells was carried out with CD8 + T
`cells sorted from GA-treated MS patients or control subjects.
`Representative dot plots of suppression of allogeneic Tar,
`cells by CD8 T cells obtained from a GA-treated patient are
`shown in Fig. 6A. Similar to CD4+CD25+ Tress, at Oh, CD8+
`suppressors from MS patients and HD displayed no significant
`difference in the mean suppression indices (Suppl. Fig. 2).
`Importantly, there was a significant enhancement of CD8+ T
`cell-mediated suppression at 12h after first GA injection
`(%suppression GA, 0h vs. 12 h, p < 0.05, Figs. 6B, C and D).
`
`Page 6 of 15
`
`Page 6 of 15
`
`

`

`GA-induced early immune changes
`
`111
`
`A
`
`1:0
`
`GA 0h
`
`GA 12h
`
`,
`
`mm
`
`
`
`
`Proliferation: 36%
`Proliferation: 32%
`Proliferation: 46% Proliferation: 38%
`
`
`
`Suppression: 31%
`Suppression: 17%
`Suppression: 22%
`
`ceepenercecepeepeeal
`
`
`
`
`
`
`
`CD4 :
`
`Proliferation: 42%
`
`Proliferation: 36%
`
`Proliferation: 31%|| Proliferation: 28%
`
`Suppression: 15%
`
`
`Suppression: 26%|: Suppression: 33%
`
`a
`
`= Soo%Suppression o *
`
`CFSE
`
`-_ eoo
`
`Oh
`
`4h
`
`12h
`
`24h
`
`72h
`
`MS (GA)
`
`
`
`a
`.
`v
`oA
`
`fet so
`
`
`Oh
`
`4h
`
`12h
`
`24h
`
`72h
`
`Oh
`
`4h
`
`12h
`
`24h
`
`+
`
`
`
`72h
`
`Ex vivo suppression by CD4+CD25+ T-egs remains unaffected during the first 72 h of GA therapy. Increasing numbers of
`Figure 5
`CD4+CD25+ T,.,5,
`isolated from the indicated cohorts were used in anti-CD3 stimulated CFSE-based suppression assays using
`third-party Ter cells and irradiated APC. Representative dot plots from a GA-treated patient are shown in panelA. Proliferation
`percentagesin the dotplots indicate gated percentages of CD4+/CFSE low (proliferating) cells. Percent proliferative responses were
`normalized to the response without suppressors (defined as 100%), indicated as 1:0 in the dot plots and %suppression wascalculated
`(also indicated). Panel B represents %suppression by CD4+CD25+ Treg cells isolated from HD, untreated and GA-treated MSpatients.
`
`Additionally, suppression remained elevated in the GA group
`at 24 and 72 h (as compared to 0 h), although the difference
`did not reach statistical significance at these “trough” time
`points early in therapy (mean %suppression GA, 0 hvs. 24h &.
`72 +h, 21.5 + 45.9 vs. 34.9 + 19.0 &. 33.4 + 29.8). At 12 h, 6/7
`GA treated patients had improvement in CD8+ suppressive
`potential as shownin Fig. 6D. The control cohorts showed no
`differences in CD8 suppressive ability over time (Fig. 6C).
`
`3.6. Significant improvement in the inhibitory
`potential of induced suppressors at 12 h post
`GA therapy
`
`Several studies from us and others have shown that human
`CD4+CD25— T cells, upon activation, can be converted into
`CD4+CD25+FOXP3+ cells and acquire suppressive potential
`[16,18,19]. To test whether GA therapy altered the ability
`of induced T,eg generation, we assessed the ability of anti-CD3-
`stimulated CD4+CD25 — T cells from GA-treated MS patients and
`controls to suppress allogeneic Tar, cells. CD4+CD25— T cells
`sorted from the peripheral blood werecultured with irradiated
`(third party) APC in the presence of plate bound anti-CD3.
`Cells were subsequently washed and used as suppressors ina
`
`secondary in vitro suppression assay. Representative dot
`plots of induced suppressors from a GA-treated patient
`co-cultured with allogeneic Ter; cells are shown in Figs. 7A
`and B. Interestingly, at the untreated state (0 h), induced
`suppressors generated from MS patient's CD4+CD25-— cells
`had significantly reduced suppressive potential towards
`allogeneic Tar; cells, compared to healthy donors (Fig. 7C).
`Importantly, the suppressive ability of CD4+CD25+ Tyegs
`generated from CD4+CD25— T cells from GA-treated MS
`patients wassignificantly improved at 12 h (%suppression
`GA, Oh vs. 12 h, p < 0.05, Fig. 7D). Within the cohort, 5/6
`patients showed improvement in induce T-eg suppression
`ability (Fig. 7E). Induced suppressors generated from the
`control cohort CD4+CD25— Tar cells displayed no significant
`differences over time in suppressing third party Ter, cells
`(Fig. 7D).
`
`3.7. Similar activation profiles of CD4+CD25-T cells
`from GA-treated MS patients and HD whencultured
`with third party allogeneic APC
`
`The effect of GAon the activation potential of CD4+CD25— Ter,
`cells in response to allogeneic APC wastested in an MLR. CD25
`
`Page 7 of 15
`
`Page 7 of 15
`
`

`

`112
`
`C.L. Ayers et al.
`
`B) GA 12h
` “100
`%Suppression24.88
`
`
`‘%Suppression
`
`= °Qo
`
`A) GA 0h
`
`%@o88
`
`MS (Untreated)
`
`MS (GA)
`p<0.05
`
`D
`
`°
`
`*
`
`*
`
`=
`fEPteR
`a
`e
`.
`a
`
`Oh
`
`4h
`
`12h
`
`24h
`
`72h
`
`.
`
`Oh
`
`4h
`
`42h
`
`24h
`
`72h
`
`Oh
`
`4h
`
`42h
`
`24h
`
`(72h
`
`oh
`
`eh
`
`Figure 6
`Significant improvement in CD8+ T cell suppressive potential within 12 h of GA therapyinitiation. Ex vivo purified CD8+
`cells were used in anti-CD3-stimulated CFSE-based suppression assays using third-party Ter cells and irradiated APC. Panels A and B
`show representative dot plots from a GA-treated patient at 0h and 12h, respectively. Percentages in dot plot indicate gated
`percentages of CD4+/CFSE low (proliferating) cells. Percent proliferative responses were normalized to the response without
`suppressors (defined as 100%), indicated as 1:0 in the dot plots. Panel C represents %suppression by CD8+ T cells isolated from HD,
`untreated, and GA-treated MS patients. GA-treated patients showedsignificant improvement in CD8+T cell mediated suppression at
`12 h (D). No such change was seen in HD and untreated MS patients. Per patient trends in GA treated group at 12 h are shownin D.
`
`upregulation response of CD4+ T cells, after co-culture with
`allogeneic APC, from a representative GA-treated MS patient
`is shown as a dotplot in Suppl. Fig. 3A. At baseline, CD4+ T
`cells from MS patients and HD showed similar activation by
`allogeneic APC (Suppl. Fig. 3B). Further, following GA therapy,
`there were no significant changes in the MLR responses of
`CD4+ T cells from any of the cohorts (Fig. 8).
`
`3.8. Reduced stimulatory potential of MS patient
`monocytes within 12 h of GA treatment
`
`Several reports have implicated immune modulatory effects
`of GA on APC populations such as reduced production of
`proinflammatory cytokines IL-12, TNF-a and elevated anti-
`inflammatory IL-10 from monocytes isolated from GA-treated
`MS patients [20,21]. GA therapy-induced increase in IL-10
`from B cells has also been reported. We sought to evaluate
`GA-induced phenotypic and functional changes in APC during
`the first 72 h of therapy. Immunophenotyping revealed no
`significant differences in total monocytes, B cell, MDC, and PDC
`percentages or absolute numbers in MS patients at baseline,
`when compared to HD (Figs. 1B & C). Moreover, unlike the
`dynamic changesin certain T cell subsets, the percentages of
`APC subsets in the peripheral blood remained relatively stable
`in all three groups when compared to baseline levels (Fig. 9),
`withno statistically significant increase or decrease at any time
`point.
`
`In addition to numbers, we also tested the capacity of
`monocyte and B cells to stimulate MLR responses to evaluate
`whetherGA induces functional changes in the APC activating
`potential within 72 h of treatment. Representative dot plots
`of CD25 upregulation by third party CD4+CD25— Tarr cells
`when cultured with GA patient's monocytes and B cells are
`shown in Figs. 10A and B, respectively. Monocytes from HD
`vs. untreated MS patients at baseline (0h) displayed no
`significant differences in their stimulatory capacity (Suppl.
`Fig. 3C). Importantly, while control groups did not show any
`variation in the monocytes stimulatory capacity over time,
`there was an appreciable trend of reduced activation potential
`of monocytes in GA-treated MS patients at 12 h (GA monocyte
`MLR, Oh vs. 12h, p = 0.06, Fig. 10C). Within the cohort, the
`activation potential of monocytes showed a downward trend in
`7/8 patients at 12 h post-first GA injection (Fig. 10E). Similar to
`monocytes, B cells from MS patients and HD at 0 h showed no
`significant difference in overall stimulatory capacity (Suppl.
`Fig. 3C). Furthermore, there were no significant changes in B
`cell stimulatory potential over time when comparedto baseline
`levels after GA therapy (Fig. 10D).
`
`3.9. Significant increase in IL-10 and TNF-a from
`monocytes early during GA therapy
`
`Cytokine levels were determined in culture supernatants of
`monocytes, B cells and MDC obtained from GA-treated MS
`
`Page 8 of 15
`
`Page 8 of 15
`
`

`

`GA-induced early immune changes
`
`113
`
`Pies
`
`| ees
`
`|
`
`1:0.125
`
`|
`
`§ 60
`
`3
`
`HD
`
`MS
`
`A) GA 0h
`
`1:0
`
`|
`!
`
`1:0.125
`
`35%
`|e
`bey
`
`|
`
`al
`
`{
`
`B) GA 12h
`1: